Preface

Topics in feline surgery

Guest Editor

This issue of the Veterinary Clinics of North America: Small Animal Prac-

tice is devoted to feline disorders amendable to surgical treatments. With 73
million feline pets currently exceeding their 68 million canine counterparts
(US News & World Report July 2001), and the increasing number of exclu-
sively feline practices and practitioners, a need exists for a relevant review of
surgically related conditions aﬀecting cats. To reach this goal, information
has been collected ranging from ‘‘hot’’ topics such as pain management,
dental diseases, and sarcomas to recurrent clinical themes involving
hyperthyroidism, megacolon, and urethrostomy. Additionally, the standard
aﬄictions of the eyes, nasopharynx, abdominal viscera, bones, and joints are
covered in this issue.

Presentation of these topics has been geared for generalists and specialists

serving feline patients. Both hard and soft tissue procedures are detailed to
provide the reader with a broad knowledge base. Furthermore, contributors
were chosen from academic and private practices throughout the United
States in an eﬀort to diversify the scope of the text.

Most, if not all, feline ‘‘owners’’ and veterinarians will unequivocally pro-

claim that ‘‘cats are not small dogs’’ and should be treated as unique
patients. In terms of recognizing the needs of feline patients, much credit
should be given, from a historical perspective, to the classic texts of Drs.
J. Holzworth (Diseases of the Cat) and Robert Sherding (The Cat—Diseases

and Clinical Management) ﬁrst published in the late 1980s.

As the guest editor of this issue, I wish to thank the authors who

graciously took time from their professional positions to provide concise

Vet Clin Small Anim 32 (2002) xi–xii

Dr. Joseph Harari

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 2 6 - 8

information based on their extensive clinical experiences. Mr. John Vassallo,
at W. B. Saunders, deserves credit for the development and production of
this issue, our second enjoyable joint venture. Finally, I dedicate this publi-
cation to my current and former feline companions whose presence has
thoroughly enriched my life.

Dr. Joseph Harari

Veterinary Referral Services

21 East Mission Avenue

Spokane, WA 99202, USA

E-mail address: jharari103@aol.com

xii

J. Harari / Vet Clin Small Anim 32 (2002) xi–xii

Feline perioperative pain management

Leigh A. Lamont, DVM, MS

Department of Companion Animals, Atlantic Veterinary College, University of Prince
Edward Island, 550 University Avenue, Charlottetown, Prince Edward Island, Canada

Advances in veterinary surgery have allowed practitioners to successfully

perform increasingly complicated and invasive surgical procedures in an
eﬀort to prolong the life span and improve the quality of life of their pa-
tients. This issue is a testament to this considerable progress. Unfortunately,
our eﬀorts to treat pain eﬀectively have not necessarily kept pace [1–3].
There are many reasons for this apparent lag in implementing progressive
perioperative pain management programs. Simply accepting the philosophic
postulate that animal pain should and must be treated is not enough. The
clinician is then confronted with the problem of recognizing when an animal
is in pain and, ultimately, choosing from a long and rapidly growing list of
potential analgesic agents and techniques. Issues associated with analgesic
therapy include appropriate dosing, frequency, route of administration, po-
tential adverse side eﬀects, and lack of eﬃcacy.

Although these obstacles may seem daunting, the beneﬁts to implement-

ing a proactive pain management program in a veterinary practice are not
limited solely to the patients themselves but can greatly improve owner sat-
isfaction and boost morale among veterinary caregiving staﬀ members. As
with any pathologic process, a basic understanding of the pathophysiology
of pain is necessary before appropriate treatment can be undertaken. With
this knowledge, the fundamental principles of pain management can be
appreciated, and the pharmacology of analgesic agents and anatomic basis
of local and regional analgesic techniques can be used to generate rational
and eﬀective pain management strategies.

Pathophysiology of pain

Clearly, not all types of pain are pathologic. In fact, the day-to-day pain

that we all experience after exposure to various noxious stimuli is actually

Vet Clin Small Anim 32 (2002) 747–763

E-mail address: llamont@upei.ca (L.A. Lamont).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 2 8 - 1

protective, functioning as part of the body’s normal defense mechanisms
and warning of contact with potentially damaging insults [4]. This is referred
to as physiologic pain, and it is quite distinct from the pain arising from
overt damage to tissues and nerves. Surgical trespass or major trauma, con-
versely, produces ongoing changes within the nervous system, resulting in
pathologic or clinical pain. Pathologic pain serves no useful adaptive role;
in addition to the unpleasant experience mediated by the nervous system,
it also produces a variety of detrimental systemic eﬀects mediated by sympa-
thoadrenal activation, which may contribute to patient morbidity [5].

Nociception is the term used to describe the transduction, transmission,

and modulation of neural signals generated in response to an external nox-
ious stimulus [5]. When carried to completion, it is the physiologic process
that results in the conscious perception of pain. Because the anatomic struc-
tures and neurophysiologic mechanisms mediating pain are remarkably
similar in human beings and animals, it is appropriate to assume that a stim-
ulus that is painful to people, is capable of damaging tissues, and induces
behavioral avoidance responses in an animal and is indeed painful to that
animal [5,6].

The pain pathway begins in the periphery with the transduction of phys-

ical noxious input into electric activity at specialized nerve endings known as
nociceptors [7,8]. The impulse is then transmitted through the peripheral
nervous system via aﬀerent sensory ﬁbers to the dorsal horn [9]. Myelinated
A-delta ﬁbers conduct fast ‘‘ﬁrst’’ pain, and nonmyelinated C ﬁbers transmit
slower ‘‘second’’ pain [5]. Cell bodies of the aﬀerent axons are located in the
dorsal horn of the spinal cord, where much of the initial integration and
modulation of nociceptive input occurs [10–12]. There are three populations
of dorsal horn neurons involved in nociceptive processing: interneurons,
which may be either excitatory or inhibitory; propriospinal neurons, which
are involved in segmental reﬂex activity; and projection neurons, which
extend supraspinally to the brainstem and cortex [12]. Projections neurons
are further classiﬁed into nociceptive-speciﬁc and wide dynamic range sub-
types, and they ascend the white matter of the spinal cord in one of several
tracts [13]. Of these, the spinothalamic tract is the most important with
regard to pain processing [5]. These projections synapse with third-order
neurons located in portions of the medulla, pons, midbrain, thalamus, and
hypothalamus, where further modulation occurs [14]. The integrated signal
is conveyed by third-order neurons to the cerebral cortex, where pain is ulti-
mately perceived [15]. Noxious aﬀerent stimuli are also subject to an array of
diverse descending inhibitory inﬂuences, with modulation occurring within
the cortex, thalamus, brainstem, and spinal cord dorsal horn [16].

In pathologic pain states, this classic physiologic model of nociceptive

processing does not hold true. With marked tissue damage caused by major
trauma or surgical insult, inﬂammation at the site of injury initiates a cas-
cade of cellular and subcellular events leading to a reduction in the threshold
of peripheral nociceptors [17]. This peripheral sensitization manifests clini-

748

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

cally as an increased pain response to a given noxious stimulus (ie, hyper-
algesia) [5,18]. In addition to peripheral hypersensitivity, pathologic pain
may also result from an upregulation of dorsal horn neuron activity, causing
an intensiﬁed pain response in the tissue surrounding the primary site of
injury (ie, secondary hyperalgesia) as well as pain produced by stimuli that
are not normally noxious (ie, allodynia) [5,18]. This misinterpretation of
normal physiologic input by the central nervous system is referred to as cen-
tral sensitization.

Principles of pain management

An appreciation of the profound and complex changes that occur within

the nervous system in response to intense noxious input leads to the inevi-
table conclusion that our eﬀorts should focus on preventing sensitization
from occurring in the ﬁrst place. To this end, there are two basic strategies
that can be employed.

Preemptive analgesia

Preemptive analgesia refers to the practice of initiating analgesic treat-

ment before the anticipated noxious insult (eg, surgery). The rationale for
this approach is that pathologic changes in nervous system processing can
be minimized and the development of peripheral or central sensitization
can be averted [19,20]. It is important to remember that even though a
patient may be under general anesthesia and is thus unable to perceive
pain, nociceptive input to the nervous system continues throughout the
surgical procedure and may result in sensitization and enhanced pain sen-
sation that manifests in the postoperative period. Once this has occurred,
traditional analgesic interventions become signiﬁcantly less eﬀective in
treating pain, and higher doses of analgesic agents are required to achieve
the desired eﬀect. Simply put, pain is easier to prevent than it is to treat
retroactively.

Multimodal analgesia

Multimodal or balanced analgesia involves the simultaneous administra-

tion of two or more analgesic drug classes or treatment modalities to achieve
optimal pain control. This approach attempts to capitalize on the additive
or synergistic eﬀects that are obtained when analgesic agents targeting diﬀer-
ent points along pain pathways are combined [21–23]. This practice often
allows lower doses of each drug to be administered, resulting in fewer
adverse side eﬀects. In many ways, multimodal analgesia is analogous to
using several antineoplastic agents in an attempt to inhibit tumor metabo-
lism and replication through diﬀerent mechanisms.

749

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

Pain recognition in cats

Recognizing pain in any domestic species can be diﬃcult, but cats seem to

represent a particularly unique challenge [24]. Physiologic signs of acute
pain reﬂect sympathoadrenal activation and may include increased blood
pressure and heart rate as well as peripheral vasoconstriction, which mani-
fests as blanched mucus membranes [5,6,24]. The respiratory rate also typ-
ically increases, and muscle splinting may occur if pain is localized to the
thorax. A stress leukogram is often evident, and catabolic processes pre-
dominate. Feline behavorial changes may be subtle. Many cats attempt to
hide when in pain and may squint their eyes or refuse to move or change
body position. Overt vocalization is not common, although cats may growl
or purr when in pain. Like dogs, cats typically demonstrate guarding behav-
iors of the injured area when palpated and may lick or chew at the site of
pain. In addition, pain often reduces appetite, alters voiding behaviors, and
reduces grooming activity. Obviously, none of these signs is speciﬁc for pain.

Although there are a number of pain grading scales available that have

been adapted from human medicine, it must be emphasized that in veteri-
nary patients, by necessity, such evaluations are made subjectively by vet-
erinary caregivers [25]. As yet, no single, reliable, objective measure of
pain exists, and few correlations between subjective impressions and actual
laboratory indices of pain-induced stress have been documented [26]. Thus,
when there is potential for an animal to experience postoperative pain, anal-
gesics should be administered regardless of whether the animal exhibits
behaviors typically attributed to pain [6]. In most cases, the beneﬁts of man-
aging pain outweigh the risks associated with analgesic drug administration.
When in doubt, it often advisable to administer a ‘‘test dose’’ of an analgesic
agent and simply monitor the patient’s response to treatment [6,22].

Pharmacology of analgesic agents

Opioids

The opioids have been and continue to be the mainstays of perioperative

pain management in companion animals [21,22,27,28]. These drugs act at
pre- and postsynaptic receptors located throughout the peripheral and cen-
tral nervous systems. Although all opioids have a similar fundamental
mechanism of action, the activity of a given drug at a speciﬁc receptor sub-
type does vary. Currently, three major classes of opioid receptors known to
mediate analgesia have been cloned: OP1 (d), OP2 (j), and OP3 (l) [6]. In
addition, there are several subtypes of each of these receptors, and their rel-
ative expression varies among diﬀerent tissues. Opioid receptors may be
activated by either endogenous ligands or exogenous opioid agonists, result-
ing in neuronal hyperpolarization, which ultimately dampens aﬀerent noci-
ceptive transmission in both the peripheral and central nervous systems.

750

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

OP3 (l) agonists

A wide array of natural and synthetic opioid agonists with diﬀering

receptor proﬁles and clinical eﬀects are currently available to the veterinary
practitioner. Morphine, oxymorphone, hydromorphone, and fentanyl are
OP3 (l) receptor agonists that are equally eﬃcacious in treating moderate
to severe acute pain but diﬀer with regard to potency, duration of action,
cardiovascular eﬀects, and cost. Despite the historical concerns about dys-
phoria and excitement after the administration of pure opioid agonists to
cats, all of these drugs can be safely used in this species and are excellent
choices for management of feline perioperative pain [28]. When adminis-
tered preemptively in a pain-free animal, the potential for excitement can
be markedly reduced by using a conservative dose and coadministering a
sedative or tranquilizer agent, such as acepromazine or medetomidine. The
incidence of excitement after opioid administration in the postoperative
period is low if appropriate doses are given.

Vomiting is another adverse side eﬀect seen in cats and is most commonly

observed when morphine is administered to healthy animals as part of a pre-
anesthetic cocktail [29]. Vomiting is rarely seen in the postoperative period.
Other less common side eﬀects include histamine release associated with
intravenous administration of morphine or meperidine and respiratory
depression. In the past, these adverse eﬀects may have been overstated and
have prevented many practitioners from using the pure agonist opioids to
their fullest potential in companion animals [29]. In fact, histamine release
is extremely rare, and morphine can be safely administered intravenously
as long as it is diluted in saline and given slowly over several minutes. Further-
more, profound respiratory depression is also uncommon, and in many cir-
cumstances, untreated pain may impair ventilation more signiﬁcantly than
any respiratory depression induced by opioid administration.

In general, any feline patient that is going to undergo a surgical proce-

dure capable of producing moderate to severe pain should receive an OP3
(l) opioid agonist in the preanesthetic period. Morphine, oxymorphone,
or hydromorphone is a reasonable choice [30]. Fentanyl may also be used;
however, its duration of action when administered systemically is brief,
meaning that supplemental doses may need to be given during surgery. Pre-
emptive administration of opioid analgesics signiﬁcantly reduces the
requirements for basal anesthetics, and intraoperative supplemental doses
may be used to achieve a deeper plane of anesthesia without resorting to
high inhalant anesthetic vaporizer settings [31]. After surgery, morphine,
oxymorphone, or hydromorphone should be administered in small boluses
to achieve the desired analgesic eﬀect and then continued on a ﬁxed dosing
interval throughout the ﬁrst 12 to 24 hours after surgery. An alternative to
periodic dosing is to set up a continuous rate infusion of morphine (0.05–0.1
mg/kg/h of body weight per hour) or fentanyl (0.001–0.005 mg/kg/h). A
loading dose of either drug is administered intravenously, and a continuous
rate infusion is then started and adjusted according to the patient’s response

751

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

to treatment. This is a particularly eﬀective and economic way to manage
pain. Another convenient option is the fentanyl transdermal patch. This sys-
tem delivers a continuous dose of fentanyl for a prolonged period of time
(usually 3–5 days in cats) [32,33]. Remember, there is a lag time of 12 to
24 hours before signiﬁcant analgesic eﬀects are observed; thus, supplemental
opioids should be administered during this period if the cat seems to be in
pain [33]. In addition, drug absorption from fentanyl patches may vary con-
siderably from animal to animal; thus, the adequacy of analgesia needs to be
assessed frequently, and additional analgesics should be administered if
deemed necessary.

OP2 (j) agonists/OP3 (l) antagonists

The opioid agonist/antagonists, including butorphanol and buprenor-

phine, are popular among veterinary practitioners for treating feline peri-
operative pain. This is because these drugs are associated with a lower
incidence of dysphoria and excitement and are even less likely to cause sig-
niﬁcant respiratory depression compared with the pure opioid agonists [34].
Unfortunately, they are not superior in their analgesic eﬃcacy and should be
reserved for the treatment of mild to moderate types of pain. Butorphanol is
known to be an OP2 (j) receptor agonist and an OP3 (l) receptor antago-
nist [35]. Although the traditional view that a given drug always behaves as
either an agonist or an antagonist at a particular receptor class is probably a
gross oversimpliﬁcation, the current recommendation from many veterinary
anesthesiologists is still to avoid to coadministration of butorphanol with a
pure opioid agonist (including the fentanyl patch), because this at least par-
tially reverses the OP3 (l)–mediated analgesia.

Buprenorphine is classiﬁed as a partial OP3 (l) receptor agonist with a

slow onset of action and a prolonged duration of action. The dose-response
curve of buprenorphine seems to be bell shaped such that higher doses may
actually antagonize OP3 (l) receptors. The drug’s unique receptor binding
aﬃnity means that pure opioid agonists used to supplement analgesia or
pure opioid antagonists used to reverse analgesia if necessary are relatively
ineﬀective. Consequently, this narrow therapeutic window means that
buprenorphine should be reserved for the management of mild to moderate
pain only. Table 1 provides recommendations regarding opioid administra-
tion in cats.

OP3 (l) antagonists

An important advantage of the opioids is that their eﬀects can be partially

or completely reversed if necessary [34]. Complete reversal is rarely indi-
cated; however, partial reversal with low doses of naloxone (0.001–0.002
mg/kg) titrated intravenously may be administered to reverse excessive
opioid-induced sedation, respiratory depression, or dysphoria while main-
taining residual analgesia. Because of its weak OP3 (l) receptor antagonism,
butorphanol (0.1–0.2 mg/kg) can also be used intravenously to reverse

752

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

sedative eﬀects in part after large doses of pure opioid agonists. Nalbuphine
is another opioid agonist-antagonist that may be used to partially reverse
opioid agonist activity.

Nonsteroidal anti-inﬂammatory agents

Although traditionally reserved for the management of chronic osteo-

arthritic pain in companion animals, the nonsteroidal anti-inﬂammatory drugs
(NSAIDs) are increasingly being advocated for use during the perioperative
period for the management of acute surgical pain [21,22,27]. Several of the
recently developed NSAIDs actually compare favorably (or even seem to be
superior) to opioids for the treatment of moderate to severe pain [36].

Most NSAIDs function, to varying degrees, through competitive inhibi-

tion of the enzyme cyclooxygenase (COX) early on in the arachidonic acid
cascade, leading to diminished synthesis of prostaglandins in a variety of tis-
sues [37]. This peripheral anti-inﬂammatory eﬀect is thought to account for
most of the analgesia attributed to NSAIDs. Approximately a decade ago, it
was recognized that there are, in fact, two distinct isoforms of COX, resulting

Table 1
Recommendations for the perioperative use of opioids in cats

Opioid

Dose/Route

Indications

Duration*

Morphine

0.05–0.1 mg/kg IM, SC

Moderate to severe

pain

3–4 hours

Morphine continuous

rate infusion

0.1 mg/kg loading dose

IV, then 0.05–0.1 mg/
kg/h IV

Moderate to severe

pain

Infusion

Oxymorphone

0.05–0.1 mg/kg IV, IM,

Moderate to severe

pain

2–4 hours

Hydromorphone

0.05–0.1 mg/kg IV, IM,

Moderate to severe

pain

2–4 hours

Fentanyl

0.001–0.005 mg/kg IV,

IM, SC

Mild to severe pain,

inadequate duration
of action from a
single bolus injection

20–30 minutes

Fentanyl continuous

rate infusion

0.002–0.003 mg/kg

loading dose IV, then
0.001–0.005 mg/kg/h
IV

Moderate to severe

pain

Infusion

Fentanyl patch

0.005 mg/kg/h

transdermal

Mild to severe pain

3–5 days

Butorphanol

0.2–0.4 mg/kg IV, IM,

Mild to moderate pain

1–3 hours

Buprenorphine

0.005–0.015 mg/kg IV,

IM, SC

Mild to moderate pain

3–8 hours

* Duration of action varies with the dose and route of administration. In general, IV

administration results in a more rapid onset and shorter duration of action than listed above.
Use lower end of dose range for initial IV administration.

¼ intramuscular; SC ¼ subcutaneous; IV ¼ intravenous.

753

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

in the designation of COX-1 and COX-2. COX-1 is the constitutive form of
the enzyme and is part of a cell’s normal enzymatic complement, mediating
the production of prostaglandins, which subserve vital physiologic functions.
Conversely, COX-2 is apparently induced by inﬂammatory mediators [38].
This distinction has led to the ‘‘good COX/bad COX’’ concept, and those
NSAIDs that seemed to have COX-2 inhibitory selectivity were presented
as the solution to all problems associated with NSAID toxicity. Unfortu-
nately, this black and white classiﬁcation is probably a gross oversimpliﬁca-
tion, and it seems that COX-2 speciﬁc inhibition alone cannot exclusively
account for either a given NSAID’s analgesic eﬃcacy or its potential for
adverse side eﬀects.

In addition to the mediation of peripheral anti-inﬂammatory processes,

there is also evidence demonstrating that NSAIDs may contribute to anal-
gesia by inhibiting prostaglandin synthesis in the central nervous system
[38]. This suggests that NSAIDs may potentially reduce the perception of
pain at supraspinal structures and diminish the development of central sen-
sitization in response to intense noxious input.

Numerous NSAIDs are available to today’s veterinary practitioner, and

several may have applications for acute postsurgical pain in feline patients
[31]. Both ketoprofen and tolfenamic acid are approved for use in cats in
Europe and Canada, and meloxicam is currently undergoing trials in cats
and seems to be an eﬀective analgesic in this species [39,40]. At this time,
none of these NSAIDs is approved for use in cats in the United States.
Although their potency and toxicity proﬁles diﬀer, the administration of any
NSAID should be reserved for well-hydrated normotensive cats with normal
renal and hepatic function, with no hemostatic abnormalities, with no pre-
disposition for development of gastric ulceration, and not receiving concur-
rent treatment with corticosteroids or other NSAIDs [39].

Although use of NSAIDs in the perioperative period is increasing, con-

troversy exists regarding when to initiate therapy. From a preemptive anal-
gesia standpoint, it would seem beneﬁcial to start treatment before surgery
to minimize the development of peripheral inﬂammatory changes. There is
currently no evidence to suggest that this produces any real beneﬁt after sur-
gery, however, and it may increase the risk of compromised renal perfusion
if the patient undergoes a period of hypotension or hypovolemia while
under general anesthesia [29]. Therefore, at this time, many investigators
and clinicians recommend waiting until recovery from general anesthesia
is complete before initiating NSAID therapy in feline surgical patients. It
is safe to combine opioids and NSAIDs in the postoperative period, because
this provides improved pain control while allowing lower doses of each drug
to be administered. In general, when prescribing meloxicam, ketoprofen, or
tolfenamic acid in cats, it is advisable to restrict the duration of treatment to
3 to 5 days only. Carprofen has also been used to manage surgical pain in
feline patients, but it is recommended as a ‘‘one time only’’ dose. Table 2
provides recommendations regarding NSAID administration in cats.

754

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

Local anesthetics

The incorporation of local anesthetic agents into a variety of local and

regional analgesic techniques is beginning to gain widespread acceptance
in companion animal practice as a feasible and extremely eﬀective way to
manage acute surgical pain. Local anesthetic molecules bind to ion-selective
channels located in nerve membranes and interfere with conformational
changes necessary to activate the channel. Once suspended in this inacti-
vated state, increases in channel sodium permeability are inhibited, and this
slows the rate of depolarization such that membrane threshold potential is
not achieved and an action potential is not propagated [41,42].

The most commonly administered local anesthetics in companion ani-

mals are lidocaine and bupivacaine. Cats are considered to be more suscep-
tible than dogs to toxicity secondary to local anesthetic administration, but
problems are uncommon if appropriate doses are given and accidental intra-
venous injection is guarded against. Toxic symptoms usually involve the car-
diovascular and central nervous systems and may include muscle twitching,
convulsions, and cardiac arrhythmias [42].

Perineural deposition is the most common route of administration, and

numerous techniques can be employed depending on the surgical procedure
to be performed. In general, it is recommended to perform local and region-
al analgesic techniques before surgery, because this markedly reduces anes-
thetic and postoperative analgesic requirements while minimizing the
development of central sensitization. In addition, lidocaine has proven to
be safe when administered intravenously at low doses to supplement analge-
sia and maintain gastrointestinal function after surgery. In contrast, because
of its increased potential for cardiovascular toxicity, bupivacaine should
never be administered intravenously at any dose. Table 3 provides recom-
mendations regarding local anesthetic administration in cats.

Table 2
Recommendations for the perioperative use of Nonsteroidal Anti-inﬂammatory drugs in cats

Nonsteroidal Anti-
inﬂammatory Drug

Dose/Route

Indications

Duration

Ketoprofen

1.0–2.0 mg/kg once after

surgery, then 0.5–1.0 mg/
kg for up to 3–5 days IV,
IM, SC, PO

Mild to moderate

pain

24 hours

Meloxicam

0.1–0.2 mg/kg once after

surgery, then 0.05–0.1
mg/kg for up to 3–5 days
IV, IM, SC, PO

Mild to moderate

pain

24 hours

Tolfenamic acid

2–4 mg/kg, for up to 3 days

only SC, PO

Mild to moderate

pain

24 hours

Carprofen

1.0 mg/kg once only PO

Mild to moderate

pain

24 hours

¼ intravenous; IM ¼ intramuscular; SC ¼ subcutaneous; PO ¼ oral.

755

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

Analgesic adjuvant agents

By deﬁnition, an analgesic adjuvant is a drug that has primary indications

other than pain but may enhance or potentiate analgesia in some painful
conditions [43]. In the context of feline perioperative pain management,
there are several agents that fall under this classiﬁcation and may prove use-
ful in this setting.

-adrenoceptor agonists

The a

-adrenoceptor agonists, including medetomidine and xylazine,

bind to speciﬁc receptors located pre- and postsynaptically in the dorsal
horn of the spinal cord. This results in decreased release of a variety of neu-
rotransmitters and neuropeptides (eg, glutamate and substance P) as well as
hyperpolarization of ion channels. The net eﬀect of these events is dimin-
ished ascending transmission of nociceptive input, which contributes to
analgesia [43].

In the past, the a

-adrenoceptor agonists have been used alone in large

doses to produce a state approaching general anesthesia. We now recognize
that this is not the optimal way to capitalize on the unique properties of
these drugs; instead, medetomidine and xylazine should be regarded as
adjuncts to a balanced anesthetic and analgesic protocol. Even when doses
much lower than those listed on the label are used, profound hemodynamic
eﬀects are still evident after administration of medetomidine or xylazine. For
this reason, this class of drugs should not be used in patients that are
adversely aﬀected by decreases in cardiac output; increases in cardiac after-
load; systemic hypertension; increases in vagal tone, which may cause severe
bradycardia; and increases in intra-abdominal, intraocular, or intracranial
pressure, which may occur secondary to vomiting.

Despite these side eﬀects, the a

-adrenoceptor agonists are useful when

administered at low doses (0.01–0.02 mg/kg) in combination with an opioid
in the preanesthetic period. This signiﬁcantly enhances the preemptive anal-
gesic eﬀect, minimizes the incidence of opioid-induced excitement, and dra-
matically decreases requirements for induction and maintenance anesthetic
agents. In cats, although heart rates typically fall by as much as 50% to
60%, concurrent anticholinergic administration is not always necessary and
may actually do little to improve hemodynamics. Each case should be judged
individually, and monitoring of cardiopulmonary parameters throughout the

Table 3
Recommendations for the perioperative use of local anesthetics in cats

Local anesthetic

Dose/Route

Indications

Duration

Lidocaine 1%, 2%

up to 4.0 mg/kg perineural

Mild to severe pain

1–2 hours

Lidocaine constant rate

infusion

0.025 mg/kg/min

intravenous

Mild to severe pain

Infusion

Bupivacaine 0.25%, 0.5%

up to 1.0 mg/kg perineural

Mild to severe pain

2–6 hours

756

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

procedure is recommended. In addition to presurgical use, low doses (0.001–
0.003 mg/kg) may be administered after surgery if the patient is experiencing a
period of dysphoria or excitement secondary to opioid administration or if
pain is refractory to opioid treatment alone.

Ketamine

Ketamine has traditionally been viewed as a poor analgesic agent.

Although this is certainly the case when it is used to treat acute visceral pain,
recent work has renewed interest in this old drug and revealed its potential
as an analgesic adjuvant. Ketamine is classiﬁed as a dissociative agent, and
its label use is for chemical restraint and short-term anesthesia. More
recently, it has also been recognized as an N-methyl-

-aspartate receptor

antagonist in the central nervous system, and by blocking the eﬀects of the
excitatory amino acid glutamate, ketamine inhibits the development of cen-
tral sensitization and modulates nociceptive transmission. In this context,
ketamine can be used in subanesthetic doses in combination with opioids
and may be a useful analgesic adjunct in feline patients undergoing thora-
cotomies, limb amputations, or neurosurgical procedures [43].

Acepromazine

Acepromazine has no inherent analgesic properties of its own and should

not be used as a substitute for adequate perioperative pain control. When
combined with an opioid, however, acepromazine prolongs and potentiates
opioid-induced analgesia in cats [35]. It is most commonly administered in
the preanesthetic period to provide tranquilization and minimize the poten-
tial for excitement associated with concurrent opioid administration. After
surgery, low doses may also be useful in cats if the patient is dysphoric or
agitated [43]. Before initiating treatment with acepromazine, ensure that the
issue of analgesia has been addressed with administration of suitable anal-
gesic agents and that symptoms and behavior patterns are not simply a man-
ifestation of untreated pain. Table 4 provides recommendations regarding
analgesic adjuvant administration in cats.

Analgesic techniques

Local and regional analgesic techniques are an integral part of a multi-

modal pain management strategy. These techniques can be used safely and
to great eﬀect in a variety of perioperative settings in the cat.

Epidural anesthesia/analgesia

In the cat, epidural techniques may be useful to manage pain associated

with hindlimb orthopedic procedures, tail injuries, caudal abdominal proce-
dures, intrathoracic procedures, and even forelimb procedures [44–46]. In
general, the anatomic landmarks for epidural injection are easily palpated

757

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

in cats, because cats often have less dorsal subcutaneous tissue and fat than
dogs. With the cat placed in either sternal or lateral recumbency, the lumbo-
sacral space can be palpated along the dorsal midline at a level immediately
caudal to the wings of the ilia [6]. The skin overlying the space is clipped and
surgically prepared. Sterile technique is absolutely mandatory. A 22-gauge
1.0- or 1.5-in spinal needle with a stylet is introduced at an angle perpendic-
ular to the contour of the animal’s dorsum and is slowly advanced until a
‘‘popping’’ sensation or loss of resistance is appreciated, indicating that the
ligamentum ﬂavum has been punctured. If bone is struck, the needle is with-
drawn to the subcutaneous tissue and then redirected. Correct needle place-
ment can be conﬁrmed by removing the stylet and performing a test
injection with 0.5 mL of air. If the tip of the needle is in the epidural space,
no resistance is encountered on injection. It is important to observe carefully
the hub of the needle for ﬂow of cerebrospinal ﬂuid or blood. If blood is
encountered, the needle should be removed and ﬂushed, and the stylet
should be replaced before reinserting the needle. If cerebrospinal ﬂuid is
present, this means that the subarachnoid space has been entered (remem-
ber, the spinal cord in cat extends farther caudally than it does in dogs).
If this is the case, a subarachnoid injection of analgesic can be administered
if desired, but the dose volume should be reduced by at least 50%. Once the
needle tip is conﬁrmed to be in the correct location, a slow injection of anal-
gesic agent over 60 to 90 seconds is made. Contraindications for epidural
anesthesia/analgesia include inﬂammation, coagulopathy, or any other
pathologic ﬁndings in the area of the lumbosacral junction.

The most commonly used drugs for epidural administration are local

anesthetics, opioids, or a combination of the two. Local anesthetics have the

Table 4
Recommendations for the perioperative use of analgesic adjuvants in cats

Analgesic Adjuvant

Dose/Route

Indications

Duration*

Medetomidine

0.01–0.02 mg/kg IV,

IM, SC

Mild pain or to supplement

opioid analgesia in

moderate to severe pain

0.5–1.5 hours

Xylazine

0.1–0.5 mg/kg IV,

IM, SC

Mild pain or to supplement

opioid analgesia in
moderate to severe pain

0.5–1.0 hours

Ketamine

0.1–1.0 mg/kg IV,

IM, SC

Mild to moderate pain,

often in combination
with opioids

0.5–4.0 hours

Acepromazine

0.02–0.2 mg/kg IV,

IM, SC

Potentiates/prolongs

analgesia achieved with
other drugs

6–12 hours

* Duration of action varies with the dose and route of administration. In general, IV

administration results in a more rapid onset and shorter duration of action than listed above.
Use lower end of dose range for initial IV administration.

¼ intravenous; IM ¼ intramuscular; SC ¼ subcutaneous.

758

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

unique advantage of blocking all sensory input to the dorsal horn; thus, they
are especially useful when administered in the presurgical period. Epidural
opioids typically produce long-lasting analgesia while minimizing potential
systemic side eﬀects. These two classes of drugs are often coadministered
to provide optimal perioperative pain management. Table 5 provides recom-
mendations regarding epidural drug and dose selection in cats.

Brachial plexus block

Presurgical blockade of the brachial plexus in cats is a useful technique to

manage pain associated with orthopedic manipulations or major soft inju-
ries of the forelimb distal to the midhumerus. The anatomic landmarks are
the point of the shoulder, the ﬁrst rib, and the transverse processes of the
cervical vertebrae. An area cranial and dorsal to the point of the shoulder
is clipped and surgically prepared. With the head and neck placed in a neu-
tral position, the cervical transverse processes form a line that, if continued,
usually traverses the proximal brachial plexus at approximately the level of
the ﬁrst rib. The ﬁrst rib is palpated and followed as far dorsally as possible.
A 1.5-in sterile needle is inserted (a spinal needle or catheter stylet works
well) and advanced toward the ﬁrst rib to a point just caudal to it. The nee-
dle should pass beneath the scapula but remain outside the thoracic cavity.
Once in place, a sterile syringe is attached, and aspiration is attempted to
check for accidental puncture of a blood vessel. Approximately one quarter
of the analgesic solution is then injected, and the needle is withdrawn 0.5 cm.
The procedure of aspirating and injecting is repeated until the needle is just
ready to exit the skin. If the surgical procedure is lengthy, the block may be
repeated before awakening the animal from general anesthesia.

Table 5
Drugs and dosages suitable for epidural administration in cats

Epidural Agent*

Dose

Duration

Lidocaine 2% with

or without epinephrine

4.0 mg/kg (or approximately 1.0 mL

per 4.5 kg of body weight)

1–2 hours

Bupivacaine 0.5% with

or without epinephrine

1.0 mg/kg (or approximately 1.0 mL

per 4.5 kg of body weight)

4–6 hours

Morphine

0.1 mg/kg (combined with 1.0 mL of

saline or local anesthetic per 4.5 kg
of body weight)

8–24 hours

Oxymorphone

0.05 mg/kg (combined with 1.0 mL of

saline or local anesthetic per 4.5 kg
of body weight)

6–10 hours

Fentanyl

0.005 mg/kg combined with 1.0 mL of

saline or local anesthetic per 4.5 kg
of body weight)

2–4 hours

* Local anesthetics and opioids are often combined as indicated above to enhance epidural

analgesia.

759

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

Either lidocaine or bupivacaine is suitable for brachial plexus blockade.

Lidocaine has a shorter onset of action and a shorter duration, whereas
bupivacaine takes longer to reach full eﬀect but lasts longer. The dose can
sometimes be cut in half, and smaller volumes of both drugs can be com-
bined and administered together to obtain a quicker onset of action and
to maintain an adequate duration of eﬀect. Table 3 provides drug and dos-
age recommendations in cats.

Radial, ulnar, and median nerve blocks

For procedures involving the carpus or areas distal to it, blockade of

branches of the radial, ulnar, and median nerves can be easily accomplished.
This technique is particularly useful for declawing, takes approximately 1
minute to perform, and requires only a 25-gauge needle and sterile syringe.
The median nerve and palmar branch of the ulnar nerve can be blocked
medial to the accessory carpal pad with 0.1 mL of local anesthetic solution;
the dorsal branch of the ulnar nerve can be similarly blocked lateral and
proximal to the accessory carpal pad; and, ﬁnally, the superﬁcial branches
of the radial nerve can be blocked at the dorsomedial aspect of the proximal
carpus.

Lidocaine or bupivacaine can be used for this technique, but remember

that bupivacaine typically takes at least 20 minutes to achieve full eﬀect;
thus, if the procedure is to be performed shortly after the blocks have been
administered, lidocaine may be a better choice from a preemptive analgesia
standpoint. Note the maximum total local anesthetic doses recommended in
Table 3, and ensure that these are not exceeded.

Dental nerve blocks

Dental nerve blocks are quick and relatively simple to perform, and they

are extremely eﬀective in preempting and managing pain associated with
tooth extractions or surgical procedures involving the mandible or maxilla.
These blocks are usually done after the animal has been anesthetized but
before the surgical intervention. The infraorbital nerve and its branches sup-
ply sensory innervation to the upper dental arcade, whereas the mandibular
nerve supplies the lower dental arcade. To block the infraorbital nerve, the
infraorbital foramen is palpated through the buccal mucosa just ventral to
the eye at the junction of the zygomatic arch and maxilla. If the rostral
branches of the nerve are to be blocked, 0.1 to 0.3 mL of analgesic solution
is inﬁltrated at the point that the nerves exit the foramen using a 25-gauge
needle. If the procedure to be performed involves the caudal maxilla, the
needle is advanced into the infraorbital canal in a caudodorsal direction
with the syringe parallel to the zygomatic arch. Always attempt to aspirate
before administering any injection to be sure that the needle has not inadver-
tently entered a vessel.

760

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

To block the mandibular nerve, palpate the mandibular foramen intra-

orally. It is located on the ventromedial aspect of the ramus of the mandible
dorsal and rostral to the angular process. The injection can then be made
through a small area of skin that has been clipped and prepared at the angle
of the mandible. The needle is directed toward the foramen by intraoral pal-
pation. Alternatively, if the procedure to be performed only involves the ros-
tral tip of the mandible, the nerve can be blocked distally as it exits the
mental foramen ventral to the ﬁrst premolar. As always, be sure to aspirate
before injecting to ensure that a vessel has not been penetrated. The total
dose of local anesthetic should not exceed that listed in Table 3.

Intercostal nerve blocks

Selective intercostal nerve blockade can be used to control pain associ-

ated with thoracic trauma such as rib fractures as well as for perioperative
pain management for patients undergoing intercostal thoracotomy. As such,
these blocks can be performed before, during, or after surgery. A minimum
of two intercostal spaces both cranial and caudal to the incision site should
be blocked because of overlapping nerve supply. A 22-gauge needle is intro-
duced percutaneously using aseptic technique at the caudal border of the rib
near the level of the intervertebral foramen. A more direct approach is pos-
sible if the blocks are performed during a thoracotomy. Depending on the
size of the cat, 0.1 to 0.3 mL of local anesthetic solution is deposited per site.
Pneumothorax, or rarely pulmonary laceration, is a possible complication if
the thoracic cavity is punctured; thus, animals should be closely monitored
for at least 20 to 30 minutes after performing the blocks. Table 3 provides
maximum local anesthetic dosage recommendations in cats.

Interpleural anesthesia/analgesia

In cats, this technique is most often used to manage postthoracotomy

pain in patients in which indwelling chest tubes have been placed. In this
case, the chest tube can be used to deliver local anesthetic solution into the
interpleural space after the chest has been evacuated of air and ﬂuid but
before recovery from anesthesia. Sterile saline can be used to dilute the local
anesthetic to a total volume of 3 to 5 mL to facilitate dispersion of the sol-
ution and improve the quality of the block. After injection, the chest tube
can be cleared by injection of 1 to 2 mL of air or saline before being closed
oﬀ and secured to the chest wall. If possible, the patient should be placed
with the surgical site located dependently to allow ﬂow of the local anes-
thetic solution to the area of damaged tissue. The chest tube should not
be aspirated for approximately 60 minutes unless the animal begins to exper-
ience signs of respiratory distress. For details regarding the technique for
chest tube placement, the reader is referred elsewhere. The local anesthetic
dosages should not exceed those listed in Table 3.

761

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

References

[1] Dohoo SE, Dohoo IR. Postoperative use of analgesics in dogs and cats by Canadian

veterinarians. Can Vet J 1996;37:546–51.

[2] Hansen B, Hardie E. Prescription and use of analgesics in dogs and cats in a veterinary

teaching hospital: 258 cases (1983–1989). JAVMA 1993;202:1485–94.

[3] Lascelles BDX, Capner CA, Waterman-Pearson A. Current British veterinary attitudes to

perioperative analgesia for cats and small mammals. Vet Rec 1999;145:601–4.

[4] Woolf CJ. Somatic pain—pathogenesis and prevention. Br J Anaesth 1995;75:169–76.
[5] Lamont LA, Tranquilli WJ, Grimm KA. Physiology of pain. Vet Clin North Am Small

Anim Pract 2000;30:703–28.

[6] Tranquilli WJ, Grimm KA, Lamont LA. Pain management for the small animal prac-

titioner. Jackson, WY: Teton NewMedia; 2000.

[7] Raﬀe M. Recent advances in our understanding of pain: How should they aﬀect man-

agement? Semin Vet Med Surg 1997;12:75–9.

[8] Raja SN, Meyer RA, Campbell JN. Transduction properties of the sensory aﬀerent ﬁbers.

In: Yaksh TL, Lynch C III, Zapol WM, et al, editors. Anesthesia: biologic foundations.
Philadelphia: Lippincott-Raven; 1997. p. 515–30.

[9] Raja SN, Meyer RA, Ringkamp M, et al. Peripheral neural mechanisms of nociception.

In: Wall PD, Melzack R, editors. Textbook of pain. 4th edition. Edinburgh: Churchill-
Livingstone; 2000. p. 11–58.

[10] Doubell TP, Mannion RJ, Woolf CJ. The dorsal horn: state-dependent processing,

plasticity and the generation of pain. In: Wall PD, Melzack R, editors. Textbook of pain.
4th edition. Edinburgh: Churchill-Livingstone; 2000. p. 165–81.

[11] Millan MJ. The induction of pain: an integrative review. Prog Neurobiol 1999;57:1–164.
[12] Sorkin LS, Carlton SM. Spinal anatomy and pharmacology of aﬀerent processing. In:

Yaksh TL, Lynch C III, Zapol WM, et al, editors. Anesthesia: biologic foundations. Phila-
delphia: Lippincott-Raven; 1997. p. 577–609.

[13] Heinricher MM. Organization of supraspinally mediated responses to nociceptive input.

In: Yaksh TL, Lynch C III, Zapol WM, et al, editors. Anesthesia: biologic foundations.
Philadelphia: Lippincott-Raven; 1997. p. 643–62.

[14] Craig AD, Dostrovsky JO. Medulla to thalamus. In: Wall PD, Melzack R, editors.

Textbook of Pain. 4th edition. Edinburgh: Churchill-Livingstone; 2000. p. 183–214.

[15] Craig AD, Dostrovsky JO. Processing of nociceptive information at supraspinal levels. In:

Yaksh TL, Lynch C III, Zapol WM, et al, editors. Anesthesia: biologic foundations.
Philadelphia: Lippincott-Raven; 1997. p. 625–42.

[16] Fields HL, Bausbaum AI. Central nervous system mechanisms of pain modulation. In:

Wall PD, Melzack R, editors. Textbook of Pain. 4th edition. Edinburgh: Churchill-
Livingstone; 2000. p. 309–29.

[17] Dray A. Inﬂammatory mediators of pain. Br J Anaesth 1995;75:125–31.
[18] Siddall PJ, Cousins MJ. Neurobiology of pain. Int Anesthesiol Clin 1997;35:1–26.
[19] Abram SE, Yaksh TL. Morphine, but not inhalation anesthesia, blocks post-injury

facilitation. The role of preemptive suppression of aﬀerent transmission. Anesthesiology
1993;78:713–21.

[20] Woolf CJ, Chong MS. Preemptive analgesia—treating postoperative pain by preventing

the establishment of central sensitization. Anesth Analg 1993;77:362–79.

[21] Hellyer PW. Management of acute surgical pain. Semin Vet Med Surg 1997;12:106–14.
[22] Hellyer PW, Gaynor JS. Acute post-surgical pain in dogs and cats. Compend Contin Educ

Pract Vet 1998;20:140–53.

[23] Kehlet H, Dahl JB. The value of ‘‘multimodal’’ or ‘‘balanced analgesia’’ in post-operative

pain treatment. Anesth Analg 1993;77:1048–56.

[24] Mathews KA. Pain assessment and general approach to management. Vet Clin North Am

Small Anim Pract 2000;30:729–56.

762

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

[25] Cambridge AJ, Tobias KM, Newberry RC, et al. Subjective and objective measurements of

post-operative pain in cats. JAVMA 2000;217:685–90.

[26] Smith JD, Allen SW, Quandt JE, et al. Indicators of post-operative pain in cats and

correlation with clinical criteria. Am J Vet Res 1996;209:1674–8.

[27] Hansen B. Acute pain management. Vet Clin North Am Small Anim Pract 2000;30:899–916.
[28] Lascelles BDX, Waterman A. Analgesia in cats. In Practice 1997;19:203–13.
[29] Papich MG. Pharmacologic considerations for opiate and nonsteroidal anti- inﬂammatory

drugs. Vet Clin North Am Small Anim Pract 2000;30:815–38.

[30] Pascoe PJ. Perioperative pain management. Vet Clin North Am Small Anim Pract

2000;30:917–32.

[31] Taylor PM. Newer analgesics: nonsteroidal anti-inﬂammatory drugs, opioids and

combinations. Vet Clin North Am Small Anim Pract 1999;29:719–35.

[32] Franks JN, Boothe HW, Taylor L, et al. Evaluation of fentanyl patches for analgesia in

cats undergoing onychectomy. JAVMA 2000;217:1013–20.

[33] Lee DD, Papich MG, Hardie EM. Comparison of pharmacokinetics of fentanyl after

intravenous and transdermal administration in cats. Am J Vet Res 2000;61:672–7.

[34] Pascoe PJ. Opioid analgesics. Vet Clin North Am Small Anim Pract 2000;30:757–72.
[35] Briggs SL, Sneed K, Sawyer DC. Antinociceptive eﬀects of oxymorphone-butorphanol-

acepromazine combination in cats. Vet Surg 1998;27:466–72.

[36] Slingsby LS, Waterman-Pearson AE. Comparison of pethidine, buprenorphine and

ketoprofen for post-operative analgesia after ovariohysterectomy in the cat. Vet Rec 1998;
143:185–9.

[37] Livingston A. Mechanism of action of nonsteroidal anti-inﬂammatory drugs. Vet Clin

North Am Small Anim Pract 2000;30:773–82.

[38] Vane JR, Botting RM. New insights into the mode of action of nonsteroidal anti-

inﬂammatory drugs. Inﬂamm Res 1995;44:1–10.

[39] Mathews KA. Nonsteroidal anti-inﬂammatory analgesics: indications and contraindica-

tions for pain management in dogs and cats. Vet Clin North Am Small Anim Pract 2000;
30:783–804.

[40] Slingsby LS, Waterman-Pearson AE. Post-operative analgesia in the cat after ovariohys-

terectomy by use of carprofen, ketoprofen, meloxicam or tolfenamic acid. J Small Anim
Pract 2000;41:447–50.

[41] Butterworth JF, Strichartz GR. Molecular mechanisms of local anesthesia: a review.

Anesthesiology 1990;72:711–34.

[42] Lemke KA, Dawson SD. Local and regional anesthesia. Vet Clin North Am Small Anim

Pract 2000;30:839–58.

[43] Lamont LA, Tranquilli WJ, Mathews KA. Adjunctive analgesic therapy. Vet Clin North

Am Small Anim Pract 2000;30:805–14.

[44] Golder FJ, Pascoe PJ, Bailey CS, et al. The eﬀect of epidural morphine on the minimum

alveolar concentration of isoﬂurane in cats [abstract]. J Vet Anaesth 1998;25:52.

[45] Pascoe PJ. Local and regional anesthesia and analgesia. Semin Vet Med Surg 1997;12:

94–105.

[46] Torske KE, Dyson DH. Epidural anesthesia and analgesia. Vet Clin North Am Small

Anim Pract 2000;30:859–74.

763

L.A. Lamont / Vet Clin Small Anim 32 (2002) 747–763

Ocular surgeries in cats

Ralph E. Hamor, DVM, MS

Department of Veterinary Clinical Medicine, College of Veterinary Medicine,

University of Illinois at Urbana-Champaign, 1008 West Hazelwood Drive,

Urbana, IL 61802, USA

The approach to the surgical treatment of feline adnexal, ocular, and orbi-

tal diseases varies among clinicians depending on their training as well as
their personal experience. Quite often, the surgical approaches and recom-
mendations for the feline eye are similar to those for the canine eye. Never-
theless, there are some diseases as well as anatomic or pathophysiologic
diﬀerences between the feline and canine eye that are important and must
be emphasized. The approaches and techniques presented here are those that
are performed most commonly in small animal practice and have worked
most successfully for me. Every attempt is made to educate the reader on
those surgical procedures that can be performed routinely in the average
practice, those that may require more specialized instrumentation, and those
that may necessitate referral to a specialist. As with any surgery, appropriate
instrumentation is critical. Without the necessary instruments, even a simple
ophthalmic surgical procedure is cumbersome. Excellent references regard-
ing microsurgical instrumentation and principles can be found in a previous
issue of Veterinary Clinics of North America: Small Animal Practice [1,2].

Surgery of the adnexa

Eyelids

Trauma

Trauma to the eyelids is frequently encountered in clinical practice, espe-

cially in cats. Because the eyelid has a rich vascular and nervous supply,
most wounds resolve quickly and heal well even if repaired many days after
the original injury. After minimal debridement, the wound can be closed
directly in two layers. The conjunctival layer is closed with 5-0 or 6-0

Vet Clin Small Anim 32 (2002) 765–790

E-mail address: hamor@uiuc.edu (R.E. Hamor).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 3 1 - 1

absorbable suture, being careful to bury the knot and place the suture bites
within the eyelid tissue to keep the suture material from rubbing on the cor-
nea. The skin is closed with 4-0 to 5-0 nonabsorbable suture. Some clinicians
prefer a braided suture, such as silk, because it is soft, but if it is left in the
eyelids too long, it can cause a suture reaction. I prefer monoﬁlament, such
as nylon, because there is much less suture reaction, but care must be taken
to ensure that the suture does not rub on the cornea. The lid margin should
then be closed using a cruciate pattern to allow for appropriate eyelid mar-
gin apposition without a knot rubbing on the cornea (Fig. 1). The remainder
of the skin wound is closed with simple interrupted sutures placed approx-
imately 2 mm apart.

Entropion

Developmental or primary entropion is an uncommon ﬁnding in cats,

with the possible exception of the mild medial lower lid entropion found
in many brachycephalic cats, which is similar to that found in brachyce-

Fig. 1. Full-thickness lid resection or wedge resection. This technique also applies to any
procedure where the lid margin is cut and needs to be sutured or repaired. The ﬁgure-of-eight or
cruciate suture at the eyelid margin is crucial because it allows for precise apposition of the lid
margin and decreases the chance of a suture knot rubbing on the cornea. (Courtesy of Gheorghe
Constantinescu, DVM, PhD, Columbia, MO.)

766

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

phalic dogs. This does not often cause any clinical problem other than mild
excessive tearing. This can occur if the lower punctal opening is functionally
closed as a result of the mild entropion or if eyelid hairs are touching the
cornea. If the cat has signiﬁcant tearing or secondary corneal disease, a
medial canthal V-plasty procedure is usually suﬃcient to correct the prob-
lem (Fig. 2). A triangle-shaped area of skin is excised with the base of the
triangle centered over the lower puncta parallel to the lid margin and 3
mm from the lid margin. The length of the incision varies depending on the
amount of the entropion. The apex of the triangle is usually about 5 to 6 mm
from the lid margin, again, depending on the severity of the entropion. Be
careful to remove skin only so that the nasolacrimal apparatus is not com-
promised. The skin incision is closed with 4-0 to 5-0 nonabsorbable suture,
starting with an initial suture placed from the center of the base of the
triangle to the apex of the triangle.

Fig. 2. Medial canthal V-plasty. A triangular-shaped area of skin is excised from the lower
medial lid margin. The base of the triangle is centered over the lower puncta parallel to the lid
margin and 3 mm from the lid margin. The length of the incision varies depending on the
amount of the entropion. (Courtesy of Gheorghe Constantinescu, DVM, PhD, Columbia, MO.)

767

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Secondary entropion is far more common in cats. It most commonly

occurs secondary to ocular pain (spastic entropion) or eyelid scarring (cica-
tricial entropion). In most cases of spastic entropion, I recommend eyelid
tacking until the original cause of the ocular pain has been treated and has
resolved. If tacking is not suﬃcient, a Hotz-Celsus procedure may be
required to keep the eyelid in an appropriate position.

Eyelid tacking. A 4-0 to 5-0 nonabsorbable suture material with a three-
eighths to one-half circle-cutting needle is used. Multiple mattress sutures
are placed in the aﬀected portion of the eyelid (Fig. 3). The initial suture bite
is placed perpendicular to and approximately 3 mm from the eyelid margin.
The suture should penetrate the skin and orbicularis muscle but should not
penetrate the full thickness of the eyelid. Each suture bite should incorpo-

Fig. 3. Eyelid tacking. Temporary eversion of the lower eyelid (A,B) and both the upper and
lower eyelids (C,D). Multiple mattress sutures are placed in the aﬀected portion of the eyelid.
The initial suture bite is placed perpendicular to and approximately 3 mm from the eyelid
margin. The entire suture is then tightened with a surgeon’s knot to create two ridges of skin
with a central furrow (dashed line). The suture is tightened until the eyelid margin is slightly
everted. A small amount of surgical glue placed between the ridges of skin created by the
mattress sutures takes tension oﬀ the sutures and prolongs the amount of time the sutures
remain eﬀective. (Courtesy of Gheorghe Constantinescu, DVM, PhD, Columbia, MO.)

768

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

rate approximately 5 mm of skin tissue. The second bite is placed over the
orbital rim, penetrating the skin and subcutaneous tissue. The entire suture
is then tightened with a surgeon’s knot to create two ridges of skin with a
central furrow. The suture is tightened until the eyelid margin is slightly
everted. Additional sutures are similarly placed about 4 to 5 mm apart until
the aﬀected area of entropion is everted. A small amount of surgical glue
placed between the ridges of skin created by the mattress sutures takes ten-
sion oﬀ the sutures and prolongs the amount of time the sutures remain
eﬀective. Postoperative care includes topical ointment applied three to four
times daily to protect the cornea and a restraint collar to keep the patient
from rubbing at the sutures. Antibiotic ointment only should be used if the
cornea is ulcerated, and antibiotic/steroid ointment can be used if the cornea
is intact.

Hotz-Celsus procedure. If tacking is unsuccessful or the entropion is recur-
rent, surgical correction is necessary, and a simple lid eversion is almost
always suﬃcient, especially in cats (Fig. 4). An eyelid plate (or back of a
blade handle) is placed into the conjunctival fornix and behind the aﬀected
eyelid to stabilize the skin. A number 15 Bard Parker (BP) blade is used to
create the eyelid incision. The initial incision is always placed parallel to and
3 mm from the aﬀected eyelid margin. The design and shape of the second
incision are decided based on the location of the entropion and the amount
of eyelid tissue that needs to be removed. The medial and lateral aspects of
the second incision are tapered to blend into the initial incision. The depth
of the incision should be to the level of the orbicularis oculi muscle, but it is
usually not necessary to remove that muscle. The conjunctival tissue should
never be removed with this procedure. The skin defect is removed with small
scissors. The defect is closed in one layer with 4-0 to 5-0 nonabsorbable
suture, with sutures placed 2 to 3 mm apart. It is often helpful to split the
incision into two or three sections and then ﬁll in the remaining spaces.
When the sutures are trimmed, the end pointing toward the cornea should
be cut close to the knot to keep the suture from inadvertently rubbing on
the cornea. The opposite end is left about 5 to 8 mm long to facilitate suture
removal. Sutures should be removed in 10 to 14 days. Again, a restraint col-
lar is beneﬁcial, and topical ointment may help but is not required in most
cases unless the patient has concurrent corneal disease. In diﬃcult patients,
absorbable sutures may be used but they do cause more swelling, tissue reac-
tion, and scarring.

Reconstructive procedures of the eyelids

Wedge resection

A wedge resection is a relatively simple reconstructive procedure and can

be used to repair a full-thickness eyelid defect created by an eyelid mass
removal. A simple triangle- or ‘‘house’’-shaped incision can be used (see

769

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Fig. 1). The surgeon must be careful, however, not to remove too much eye-
lid margin. In the average cat, only one quarter of the eyelid margin can be
safely removed with this procedure. If a triangular shape is used, the base of
the triangle is placed at the eyelid margin, and the sides of the triangle need
to be at least twice the length of the base to allow for adequate closure of the
wound. A partial-thickness skin incision is created with a number 15 BP
blade and handle. Scissors are used to ﬁnish the full-thickness eyelid inci-
sion. The eyelid defect is closed in two layers in a fashion similar to that used
for an eyelid laceration.

Sliding H-plasty

For eyelid masses greater that one quarter of the upper or lower eyelid

margin or for traumatic lesions where large amounts of tissue have been

Fig. 4. Hotz-Celsus procedure. An eyelid plate (or back of a blade handle) is placed into the
conjunctival fornix and behind the aﬀected eyelid to stabilize the skin. The initial incision is
always placed parallel to and 3 mm from the aﬀected eyelid margin. The design and shape of
second incision are decided based on the location of the entropion and the amount of eyelid tissue
that needs to be removed. The medial and lateral aspects of the second incision are tapered to
blend into the initial incision. The defect is closed in one layer with 4-0 to 5-0 nonabsorbable
suture with sutures placed 2 to 3 mm apart. It is often helpful to split the incision into two or three
sections and then ﬁll in the remaining spaces. (Courtesy of Gheorghe Constantinescu, DVM,
PhD, Columbia, MO.)

770

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

lost, a sliding H-plasty works well (Fig. 5). This procedure can be used for
both split- and full-thickness eyelid defects. In any grafting procedure, tissue
must be meticulously harvested to ensure adequate blood supply, and care
must be taken to harvest enough replacement tissue so that the graft does
not have any tension placed on it. In large full-thickness defects, replace-
ment conjunctival tissue can be harvested from the opposite eyelid to line
the defect before placement of the skin graft. Replacement of conjunctival

Fig. 5. Sliding H-plasty. This procedure can be used for split- and full-thickness eyelid defects.
In large full-thickness defects, replacement conjunctival tissue can be harvested from the
opposite eyelid to line the defect before placement of the skin graft. Conjunctival tissue is
sutured with 6-0 absorbable suture, being careful to ensure that no suture rubs on the cornea.
Care must also be taken to ensure that sutures placed at the eyelid margin (5-0, nonabsorbable)
do not rub on the cornea. A split-thickness temporary tarsorrhaphy that incorporates the H-
plasty and normal eyelid margin minimizes graft tension and protects the cornea until the graft
sutures can be removed. (Courtesy of Gheorghe Constantinescu, DVM, PhD, Columbia, MO.)

771

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

tissue provides better eyelid motility and function as well as improved cos-
mesis. Conjunctival tissue is sutured with 6-0 absorbable suture, being careful
to ensure that no suture rubs on the cornea. Care must also be taken to
ensure that sutures placed at the eyelid margin (5-0, nonabsorbable) do not
rub on the cornea. A split-thickness temporary tarsorrhaphy that incorpo-
rates the H-plasty and normal eyelid margin minimizes graft tension and
protects the cornea until the graft sutures can be removed.

Eyelid agenesis rotating/pedicle graft

This grafting technique works well for repair of eyelid agenesis in cats but

can also be used for other large narrow eyelid defects of 5 mm or less (Fig. 6)
[3]. Repair of wider defects may result in ectropion of the opposite eyelid.
I ﬁnd that it is easier to harvest and place the conjunctival graft before sutur-
ing the eyelid graft in place. The conjunctival graft is harvested from the
anterior surface of the third eyelid. It is important to rotate or ﬂip the con-
junctival graft so that the epithelial surface of the conjunctival graft and the
cornea are in direct contact. The use of a rotational conjunctival graft
instead of a free conjunctival graft may provide better blood supply, but the
base of the conjunctival graft may need to be severed once healing is com-
plete. This is easily performed under topical anesthesia.

Third eyelid surgery

Cartilage eversion/third eyelid gland prolapse

This is a reasonably rare occurrence in cats but has been reported in Bur-

mese cats [4]. Usually, only the cartilage is everted, but the gland may also
be prolapsed. I have also seen a case of bilateral prolapse of the gland of the
third eyelid in a Scottish Fold cat. If only the cartilage is everted, only the
bent portion of the cartilage needs to be removed (Fig. 7). If the gland is
prolapsed, it is repaired similar to that of gland prolapse in a dog (Fig. 8)
[5]. I generally recommend that an imbricating suture pattern (6-0, absorb-
able) be used and that both the initial and ﬁnal knot be anchored on the
anterior surface of the third eyelid to eliminate the possibility of suture or
suture knots rubbing on the cornea. It is important to incise only the con-
junctiva on either side of the prolapsed gland without any subconjunctival
dissection and that the ends of the conjunctival incisions do not meet. This
allows the gland to remain in place and facilitates drainage of conjunctival
secretions.

Surgery of the cornea

Trauma

After any trauma, damage to the corneal surface is usually apparent, but

careful evaluation of the intraocular structures is also warranted. Any dam-

772

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

age to the anterior uveal tract should be noted and treated. With any pen-
etrating trauma, particular attention should be paid to the anterior lens cap-
sule. A tear in the anterior lens capsule can lead to severe anterior uveitis
and blindness even in the face of aggressive topical and systemic anti-inﬂam-
matory therapy [6]. If the anterior lens capsule is compromised, the most
eﬀective therapy is surgical removal of the lens (phacoemulsiﬁcation and
aspiration) along with repair of the corneal defect. A lesion of this magni-
tude may require referral to an ophthalmologist.

Fig. 6. Eyelid agenesis rotating/pedicle graft. This procedureR is used to repair full-thickness
upper eyelid defects (A–F). I ﬁnd that it is easier to harvest and place the conjunctival graft
before suturing the eyelid graft in place. The conjunctival graft is harvested from the anterior
surface of the third eyelid. The use of a rotational conjunctival graft instead of a free conjunctival
graft may provide a better blood supply, but the base of the conjunctival graft may need to be
severed once healing is complete. This is easily performed under topical anesthesia. (Courtesy of
Gheorghe Constantinescu, DVM, PhD, Columbia, MO.)

773

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Any corneal ulcer should be treated with a topical broad-spectrum anti-

biotic and topical atropine ointment if the eye is painful. In cats, topical
atropine ointment is usually recommended, because the drops are more
likely to drain from the nasolacrimal system into the mouth and cause
hypersalivation as a result of the bitter taste. Whenever the anterior cham-
ber is penetrated, however, topical drops should be used, because topical
ointments may seep into the anterior chamber and cause irritation. Also, the
addition of systemic broad-spectrum antibiotics is recommended for 7 to 10
days, and systemic anti-inﬂammatory agents may also be indicated if there is
any evidence of anterior uveitis. Although systemic anti-inﬂammatory
agents may cause some physiologic delay in wound healing, the beneﬁt of
decreasing the anterior uveitis far outweighs the potential risks.

Fig. 7. Cartilage eversion. After making cAonjunctival incisions, the aﬀected cartilage is bluntly
dissected, undermined, and removed with scissors. The conjunctival incision does not need to be
sutured. (Courtesy of Gheorghe Constantinescu, DVM, PhD, Columbia, MO.)

774

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Foreign bodies may be imbedded in the cornea or anterior chamber.

Before attempting removal of any corneal foreign body, the depth of the for-
eign body’s penetration should be carefully evaluated. After removal, the
cornea should be treated like any other ulcer. Superﬁcial foreign bodies may
be removed in the examination room after topical anesthesia. If you are
unsure of the depth, or the penetration is deep, general anesthesia is neces-
sary. Some superﬁcial foreign bodies ‘‘angle’’ across the corneal surface and
also require general anesthesia. These angled foreign bodies are best
removed by making a sharp incision through the cornea over the foreign
body and then gently lifting the foreign body out of the cornea. The bed
or trough in the cornea is copiously lavaged with a dilute povidone
iodine–saline solution (ratio of 1:20–1:50). If the foreign body is just ‘‘pulled
out,’’ remnants may remain buried in the cornea. If the foreign body pene-
trates the cornea, it is carefully removed, and the wound may require direct
suturing as with a full-thickness laceration.

Partial- or full-thickness lacerations are not unusual and often result

from another cat’s claw. Partial-thickness lacerations may only require med-
ical therapy as for any corneal ulceration: a topical broad-spectrum anti-
biotic with or without topical atropine. If the partial-thickness laceration
is deep (two thirds to three quarters of the corneal thickness), direct suturing
may be required. Simple interrupted sutures placed 1 to 2 mm apart with 7-0
to 8-0 absorbable suture are usually suﬃcient.

Fig. 8. Morgan technique for third eyelid prolapse. The gland is secured into a conjunctival
pocket formed on the posterior surface of the third eyelid. It is important to incise only the
conjunctiva on either side of the prolapsed gland without any subconjunctival dissection and
that the ends of the conjunctival incisions do not meet. This allows the gland to remain in place
and facilitates drainage of conjunctival secretions. I generally recommend that an imbricating
suture pattern (6-0, absorbable) be used and that both the initial and ﬁnal knot be anchored on
the anterior surface of the third eyelid to eliminate the possibility of suture or suture knots
rubbing on the cornea. (Courtesy of Gheorghe Constantinescu, DVM, PhD, Columbia, MO.)

775

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Full-thickness lacerations are sutured similarly. Care must be taken to

place sutures accurately. Sutures should be placed no more than 2 mm apart
and may incorporate two thirds to three quarters of the corneal depth with-
out penetrating the entire thickness of the cornea. If possible, the sutures
should enter and exit the cornea 1 to 2 mm from the wound edge. If the cor-
nea is edematous, this distance may need to be increased. If the corneal
defect is large or irregularly shaped, the wound may need to be divided into
smaller sections to achieve appropriate closure with minimal scarring.
Prolapse of the iris, with or without a ﬁbrin clot, often accompanies a full-
thickness laceration and may provide a temporary seal for the corneal
defect. If it is present, care must be taken not to disturb the temporary seal
until surgery. At surgery, the prolapse must be replaced or excised. Whenever
possible, replacement is preferable, but necrotic tissue should be removed. If
amputation is necessary, this can be accomplished with small sharp scissors
or wet-ﬁeld electrocautery. Because the iris is extremely vascular, wet-ﬁeld
electrocautery is preferable. A 1:10,000 epinephrine solution may control
mild to moderate intraocular hemorrhage. The prolapsed iridal tissue usually
needs to be gently teased from the corneal wound with an iris spatula or other
small blunt instrument. Viscoelastic material is beneﬁcial in most large or
complicated corneal lacerations to reform the anterior chamber, protect and
separate ocular tissues, decrease the formation of synechia, and decrease
intraocular hemorrhage. With appropriate instrumentation, simple corneal
lacerations can often be managed in private practice. Large and complicated
corneal defects or those with known or suspected intraocular damage often
necessitate referral, however.

Corneal ulceration

Linear grid keratotomy

Refractory or indolent corneal ulcers are much less common in cats than

in dogs, but they do occur with some regularity. These ulcers have a typical
appearance consisting of a superﬁcial ulcer (only through the corneal epithe-
lium) with loose redundant epithelium at the ulcer edges. These ulcers typ-
ically are not as painful as other superﬁcial ulcers and have a delayed
vascular response. In cats, these may be seen most commonly with super-
ﬁcial ulceration secondary to feline herpes virus, but other causes, such as
lagophthalmos, chronic irritation from entropion or cilia, or other infection,
can occur. If a primary cause cannot be identiﬁed and treated, a linear grid
keratotomy is recommended (Fig. 9). After administration of topical anes-
thesia with 0.5% proparacaine, the loose epithelium is debrided with a dry
cotton-tipped applicator. Quite often, this procedure enlarges the ulcer sig-
niﬁcantly, but the ulcer should be debrided until the edges are ﬁrm. Once the
applicator gets moist, it should be replaced with a dry one. Once the ulcer
edges are debrided, a linear grid keratotomy is performed with the tip of
a small 22- to27-gauge needle. I ﬁnd that bending the tip of the bevel of the

776

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

needle increases my conﬁdence in an awake patient, because the needle can-
not penetrate the cornea. A grid pattern is made across the entire surface of
the ulcer with vertical and horizontal scratches separated by approximately
1 mm. The scratches should leave a visible line, but it is not necessary to
leave a groove in the ulcer surface. These scratches in the basement mem-
brane and superﬁcial stroma of the cornea provide scaﬀolding for the newly
formed epithelium to adhere to the superﬁcial corneal stroma. It may take
multiple grid procedures for an indolent ulcer to heal, but it is best to allow
at least 10 to 14 days between each procedure. Topical antibiotics should be
prescribed to prevent infection, and a restraint collar may be needed to pre-
vent rubbing. If the superﬁcial corneal stroma is particularly edematous,
topical hyperosmotic agents may be beneﬁcial but are usually not necessary.
Although multiple other surgical procedures and medical therapies
have been recommended, none of them facilitates healing of an indolent
ulcer as well as debridement and linear keratotomy in routine cases. It is
imperative to point out that recommendation of this procedure is limited
to superﬁcial indolent ulcers. If the ulcer is invading even the anterior one

Fig. 9. Linear grid keratotomy for an indolent ulcer. Loose redundant epithelium at the ulcer
margin is debrided (A) and enlarges the ulcer (B). A linear grid keratotomy is performed with a
22- to 27-gauge needle to create a grid of vertical and horizontal cuts (approximately 1 mm
apart) in the superﬁcial stroma of the cornea (C). The tip of the needle can be bent to ensure
that the needle does not penetrate the cornea too deeply, especially in an awake patient.

777

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

quarter of the corneal stroma, this procedure is not beneﬁcial and can be
hazardous if the ulcer has invaded deep into the cornea.

Conjunctival grafts

Conjunctival grafts are easily harvested and indicated in any corneal

ulcer that has not responded to appropriate medical therapy, in any melting
corneal ulcer, or in any corneal ulcer that is deeper than one half of the cor-
neal depth. Conjunctival grafts provide mechanical support, a direct blood
supply, and antimicrobial/anticollagenase properties [7]. Many conjunctival
grafting procedures have been described, but a simple rotational pedicle
graft is easy to harvest and allows both the surgeon and the patient to see
around the graft. Before harvesting the graft, the ulcer bed must be prepared
by gently debriding any necrotic cornea and removing any corneal epithe-
lium that may have begun to migrate across the ulcer bed. This is easily per-
formed with a number 15 BP blade. Bulbar conjunctiva is most appropriate
and is most easily harvested from the dorsotemporal aspect of the globe. A
lateral canthotomy provides exposure and facilitates harvesting. Small blunt
scissors are used to create a small hole in the conjunctiva and to undermine
the conjunctiva, taking care to bluntly dissect the conjunctiva from the epis-
cleral tissues (Fig. 10). A good conjunctival graft is thin enough to read
newspaper through. Once the area to be harvested has been undermined, the
perilimbal edge and then the outside edge are cut with blunt scissors. The
length and width of the graft is determined by the size of the ulcer bed, being
careful to ensure that the base of the graft remains at least as wide as the tip.
The length of the graft should be positioned to lie perpendicular to the eyelid
margin to ensure that natural blinking does not serve to ‘‘roll’’ the graft oﬀ
the cornea. It is also best if the length of the graft is suﬃcient to allow the tip
of the graft to cover the ulcer bed without requiring a suture. If the graft
needs to be ‘‘pulled into place’’ with corneal sutures, the graft has too much
tension and may contract before incorporating with the cornea. The graft is
rotated into the ulcer bed (cut surface of graft facing the ulcer bed) and
sutured in place using simple or continuous sutures of 7-0 to 8-0 absorbable
suture. The donor site can be sutured, and two tension-relieving sutures are
placed at the limbus. The lateral canthotomy is closed as with any eyelid
repair. Postoperative care includes routine medical ulcer management and
a restraint collar to prevent rubbing. The graft is left in place for at least
4 to 6 weeks depending on the severity of the ulcer and the response of the
cornea to the graft. After administering topical anesthesia, the conjunctival
graft can be trimmed with blunt scissors (Fig. 11). Only the portion of the
graft that is not attached to the cornea should to be trimmed. Usually, one
cut is made at the dorsal aspect of the previous ulcer bed, and another is
made at the limbus. Any redundant conjunctival tissue can be trimmed from
the previous ulcer bed. The goal of trimming is to sever the blood supply of
the graft, but do not attempt to remove the attached graft from the former
ulcer bed. Once the blood supply is gone, the remaining graft atrophies and

778

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

leaves a scar. The severity of the scar depends on the severity of the initial
ulcer. Topical antibiotic therapy can be continued for an additional 5 to 7
days, but topical corticosteroids are not usually necessary and do carry some
risk.

Superﬁcial keratectomy

A keratectomy involves the removal of a portion of the corneal epithe-

lium and a portion of the underlying corneal stroma. In cats, the most com-
mon indications for the use of this procedure include removal of a corneal
sequestrum or a corneal dermoid [8]. Without appropriate microsurg-
ical instrumentation, this procedure is diﬃcult and may necessitate referral.

Fig. 10. Rotational conjunctival pedicle graft. (A) The bulbar conjunctiva is elevated and
incised with small blunt scissors. (B) The entire area of conjunctiva to be harvested is bluntly
separated from the episclera and is then incised as indicated in cuts 1 through 3. The base
remains attached at the limbus. (C) The graft is sutured directly to the cornea and limbus, and
the donor site is closed.

779

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Initially, the abnormal corneal tissue to be removed is outlined with a number
64 Beaver blade or a number 15 BP blade to a depth necessary to allow
removal of all the abnormal corneal tissue (Fig. 12). Once the area to be
removed has been outlined, the edge of the corneal defect is grasped with
small forceps (Colibri forceps work well), and the defect is undermined with a
Martinez corneal dissector, a number 64 Beaver blade, or a number 15
BP blade. The corneal dissector is recommended because it was developed
for this procedure and allows the surgeon to easily remain on the same
lamellar plane. If a blade is used, the surgeon must pay close attention to
the corneal depth to avoid deeper penetration of the cornea. Once the
corneal defect has been completely undermined, the remainder of the defect
is excised with small scissors or a blade. Corneal defects of up to one third to
one half of the corneal depth can be removed and allowed to heal as an
ulcer. Deeper defects usually require the addition of a conjunctival or lamellar
graft for best results. Postoperative care includes routine medical ulcer
management and a restraint collar to prevent rubbing.

Corneal-scleral-conjunctival transposition

Corneal-scleral-conjunctival transpositions are a type of lamellar kerato-

plasty indicated for deep corneal ulcers or corneal perforations or to repair
the cornea after a deep keratectomy. This procedure repairs the defect and
transposes healthy peripheral cornea to the central portion of the cornea to
provide a clear central cornea. For best results, the peripheral cornea needs
to be healthy, and the defect to be repaired should be no larger than 25% to
30% of the corneal diameter. In most cases, a lateral canthotomy provides

Fig. 11. Trimming a conjunctival graft. After topical anesthesia, the portion of the conjunctival
graft not adhered to the cornea is severed and removed with small forceps and scissors. This
requires that two cuts to be made: one at the dorsal aspect of the previous ulcer bed and one at
the limbus.

780

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

increased exposure. Initially, the corneal defect is removed, or the recipient
bed is debrided to remove as much abnormal tissue as possible (Fig. 13). The
donor corneal tissue is harvested similar to a keratectomy with a blade and
corneal dissector peripherally to the limbus at a depth of one half of the cor-
neal thickness. If possible, it is easiest to harvest the donor corneal and con-
junctival tissue from the 12 o’clock position. The corneal incisions are
extended to include the bulbar conjunctiva, which is elevated similar to har-
vesting a conjunctival graft. The only remaining attachment of the graft is
the scleral attachment at the limbus. This attachment is dissected free with
small scissors, taking care to keep the donor cornea attached to the bulbar
conjunctiva. The leading edge of the graft is advanced into the central cor-
nea and may need to be trimmed to ﬁt easily into the recipient bed. The graft
is sutured into place with a combination of simple interrupted and continu-
ous sutures of 7-0 to 8-0 absorbable suture with the sutures placed at one
half to two thirds of the corneal depth. This procedure can be modiﬁed to
harvest donor corneal tissue only and to place it in the recipient bed as a free
lamellar graft. In either case, postoperative care includes routine medical
ulcer management and a restraint collar to prevent rubbing.

Fig. 12. Superﬁcial keratectomy. (A) The area to be excised is outlined with a blade to an
appropriate depth to remove the lesion. (B) A Martinez corneal dissector is used to undermine
the lesion. (C) A blade or scissors are used to excise the aﬀected cornea.

781

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Other lamellar or penetrating keratoplasty surgeries have been described

[9]. These procedures usually require rather specialized instrumentation and
expertise that necessitates referral to a specialist.

Surgery for blind painful globes

Although glaucoma is much less common in cats than in dogs, cats do

regularly develop secondary glaucoma that necessitates surgical interven-
tion. Cats also may develop severe blinding uveitis or suﬀer a traumatic ocu-
lar injury that results in blindness. So long as the globe is comfortable, it

Fig. 13. Corneo-scleral-conjunctival transposition. (A) After the recipient bed is debrided, two
half-thickness corneal incisions are made extending from the lesion toward the limbus. (B) A
Martinez corneal dissector is used to undermine the corneal graft to the limbus. The incisions
are extended into the conjunctiva, the conjunctiva is undermined, and the scleral attachment is
cut with scissors, taking care to keep the donor cornea attached to the bulbar conjunctiva.
(C) The graft is advanced into the recipient bed and sutured in place with simple interrupted
sutures or a combination of simple interrupted and continuous sutures.

782

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

need not be removed. If globe becomes blind and painful, some surgical pro-
cedure is warranted to keep that patient comfortable.

Pharmacologic ablation of the ciliary body

An intravitreal injection of gentamicin and dexamethasone has been used

as a means to decrease intraocular pressure through toxic necrosis of the
uveal tract [10]. This procedure is quick and usually eﬀective but can require
additional injections or result in phthisis bulbi. It is best performed under a
short general anesthetic episode but can be performed under sedation and
topical anesthesia in geriatric or compromised patients. This procedure
should not be used for patients with suspected intraocular neoplasia or
infection. Initially, a 20-gauge needle is inserted 6 to 8 mm behind the limbus
and directed toward the optic nerve to avoid the lens. Using gentle pressure
and a gentle back and forth motion, the needle is inserted into the vitreous.
Once into the vitreous, approximately 0.5 to 1.0 mL of vitreous is removed
depending on the volume of gentamicin (25 mg) and dexamethasone (1 mg)
to be used. It is best to remove at least as much vitreous volume as is
replaced by the injection. It is important, especially in small patients, not
to inject more gentamicin than the patient’s systemic toxic dose of 4.4 mg/
kg of body weight per day [10]. The globe is then treated medically with a
topical antibiotic or antibiotic-steroid for 2 to 3 weeks. Some ophthalmolo-
gists do not recommend this procedure in cats, because the development of
feline posttraumatic ocular sarcoma has been linked to a history of penetrat-
ing ocular trauma [11].

Transconjunctival enucleation

A lateral canthotomy is used to facilitate exposure. A 360

° perilimbal

incision is made through the bulbar conjunctiva adjacent to the limbus with
small scissors (Fig. 14). Small toothed forceps are used to grasp the limbus,
and curved Metzenbaum or enucleation scissors are used to separate epis-
cleral tissue from the sclera and to sever the extraocular muscles from their
scleral attachments. Because of hemorrhage at surgery, it is often diﬃcult
(and unnecessary) to visualize each extraocular muscle. By gently manipu-
lating the globe within the orbit, the surgeon can ‘‘feel’’ what attachments
need to be severed with the blunt scissors. It is imperative, however, that the
surgeon does not place too much traction on the globe until the optic nerve
has been transected. Excessive traction on the globe can result in damage to
the optic chiasm and blindness in the opposite globe [12,13]. The optic nerve
is severed blindly deep within the orbit with curved scissors. Do not attempt
to clamp or ligate the optic nerve before severing it, because this may also
result in damage to the optic chiasm. The orbital lacrimal gland (located in
the dorsolateral orbit adjacent to the orbital ligament) is typically removed
along with the globe. The third eyelid is grasped with forceps and completely
removed with scissors. Once removed, the third eyelid should be examined

783

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

to ensure that the gland of the third eyelid has been removed. The eyelid
margins (4–5 mm) are then removed, taking care to remove all the eyelid
margin and the meibomian glands. The most diﬃcult portion of the eyelid
margin to remove is at the medial canthus, because the medial canthal
tendon is strong and intimately associated with the medial orbit. Any
remaining conjunctival tissue is dissected free from the subcutaneous tissue.
Failure to remove all glandular and conjunctival tissue can result in the

784

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

development of an orbital cyst. If desired, a mesh (Fig. 15) or silicone orbital
implant (see Fig. 14I) can be used to improve cosmesis by occupying poten-
tial ‘‘dead space’’ in the orbit [14]. In cats, a mesh orbital implant is usually
more than suﬃcient. The orbital implant is anchored in place with 3-0 to 4-0
suture across the orbital septum in a continuous pattern. Subcutaneous tis-
sues are closed over the orbital septum or orbital implant with 4-0 absorb-
able suture material in a continuous pattern. If an orbital implant is used, it
is imperative that subcutaneous tissue completely covers the orbital implant
so as to decrease the chance of implant extrusion. Skin is closed with 4-0 to
5-0 nonabsorbable suture material. If an orbital implant is planned, preop-
erative and perioperative systemic antibiotics are recommended [14].
A restraint collar is often indicated to prevent rubbing.

Transpalpebral enucleation

This procedure is indicated if the conjunctival sac or globe is contami-

nated, because the orbit is never directly exposed to these tissues. Initially,
the eyelid margins are sutured or clamped together. A number 10 BP blade
is used to incise the skin parallel to and 4 to 5 mm from the eyelid margin. A
combination of careful sharp and blunt dissection is used to dissect through
the subcutaneous tissue to the level of the conjunctiva. The medial and lat-
eral canthal tendons are then bluntly or sharply dissected to completely free
the eyelid margin. Blunt dissection is used between the conjunctival sac and
the orbital margin to the level of the sclera and the extraocular muscles.

Fig. 14. Transconjunctival enucleation. (A) A lateral canthotomy is made to increase exposure.
The eyelids are retracted and a 360

° perilimbal incision is made in the conjunctiva. Curved

Metzenbaum scissors are used to undermine the conjunctiva posterior to the level of the rectus
muscles (B) and to transect the rectus muscles and fascial attachments (C). This is usually done
‘‘blindly’’ by gently pressing the blunt curved scissors against the sclera and transecting the
attachments. (D) Curved Metzenbaum or enucleation scissors are used to transect the optic
nerve. This is done blindly by following the curve of the sclera with curved scissors. The optic
nerve is transected deep within the orbit without any attempt to ligate or clamp the optic nerve.
(E) The third eyelid is grasped and excised close to its orbital attachment. The gland of the third
eyelid is palpated before and after excision to ensure that the gland has been completely
removed. (F) Any remaining conjunctiva is removed with scissors. (G) The lateral upper eyelid
margin is grasped with forceps, and 4 to 5 mm of the eyelid margin is trimmed from the lateral
to medial aspect, leaving it attached at the medial canthus. The same procedure is repeated for
the lower eyelid. (H) Both eyelid margin pedicles are grasped with forceps, and sharp dissection
is used to transect the medial canthus tendon to complete the removal of the entire eyelid
margin. (I) If desired, the anterior one quarter to one third of the anterior surface of a silicone
sphere is excised with a scalpel to form an implant with a ﬂat and beveled surface. (J) The
implant is inserted into the orbit with the ﬂat surface facing forward, and absorbable suture is
anchored to the orbital rim fascia to create a meshwork that bridges the orbital opening over
the implant. (K) Subcutaneous tissues are closed with absorbable suture in a continuous
pattern. It is important to completely cover the implant with subcutaneous tissue. (L) The skin
is sutured with nonabsorbable suture in a simple interrupted pattern.

785

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Again, gentle manipulation and traction on the eyelid margin allow the sur-
geon to feel what attachments need to be severed with blunt curved scissors,
and the optic nerve is severed blindly deep within the orbit so as to protect
the optic chiasm. With this procedure, the eyelids, conjunctiva, third eyelid,
orbital lacrimal gland, and globe are removed together. Closure of the sur-
gical site and postoperative care are similar to those used for transconjunc-
tival enucleation.

Intraocular prosthesis

A more cosmetic option for a blind painful globe than enucleation is the

placement of an intraocular prosthesis. This procedure should not be per-
formed in any patient with suspected or conﬁrmed intraocular infection
or neoplasia. This procedure is facilitated by a lateral canthotomy. An inci-
sion is then made through the dorsolateral conjunctiva and sclera directly
over the ciliary body (Fig. 16). The uveal tract is gently teased from its
scleral attachments at the ciliary body with blunt dissection. A cyclodialysis
spatula works well for this. The entire anterior uveal tract, lens, and retina
are then removed through the scleral incision. A silicone sphere (matching
the diameter of the normal-sized globe) is placed into the scleral incision
with a sphere inducer. The sclera and conjunctiva are closed separately with
5-0 to 6-0 absorbable suture in a continuous pattern. If the globe is buph-
thalmic, a silicone sphere that approximates the normal globe is implanted,
and the buphthalmic globe shrinks around the silicone sphere. Medical ther-
apy consists of preoperative and postoperative systemic antibiotics, topical
antibiotics, and a restraint collar to prevent rubbing. A split-thickness tem-
porary tarsorrhaphy may be considered for 1 to 2 weeks to protect the
cornea.

Fig. 15. Mesh orbital implant. In the absence of a silicone orbital implant, a mesh implant
created with nonabsorbable suture can be used to decrease orbital concavity after healing. The
nonabsorbable suture is anchored to the orbital rim fascia to create a ‘‘tennis racket’’ over the
orbital rim before routine closure.

786

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Surgery for cataracts

Cataracts are not common in cats compared with dogs and are usually

secondary to some other problem, such as chronic anterior uveitis. If cata-
racts are diagnosed and the globe is determined to be a good candidate,
cataract removal is warranted. As with dogs, phacoemulsiﬁcation and aspi-
ration is the best technique and is performed similar to the technique used in
dogs [15].

Surgery of the orbit

The most common reason for orbital surgery in a cat is secondary to orbi-

tal neoplasia. Orbital neoplasia is not uncommon in cats; however, unlike the
case in dogs, it is usually secondary rather than primary [16]. Because of this,
it is even more important for appropriate diagnostic steps be taken before
surgery. Any patient with suspected orbital neoplasia (nonpainful exophthal-
mos) should have thoracic radiography and abdominal ultrasonography per-
formed to evaluate for primary neoplasia or metastasis. A complete blood
cell count, serum biochemical proﬁle, and urinalysis are also strongly recom-
mended. Although orbital ultrasonography has improved in recent years, an
orbital magnetic resonance imaging or computed tomography scan usually

Fig. 16. Intraocular prosthesis. (A) After a lateral canthotomy, the globe is positioned to expose
the dorsolateral sclera. (B) A 140

° to160° conjunctival incision is made parallel to the limbus over

the ciliary body. (C) The sclera is then incised with a blade to the level of the ciliary body. (D) The
entire lens, uveal tract, and retina are removed. (E) The silicone prosthesis is inserted. (F) The
sclera and conjunctiva are closed with absorbable suture in separate layers. (Modiﬁed from
Severin GA. Glaucoma. In: Severin’s Veterinary Ophthalmology Notes, 3rd edition. Fort Collins
(CO): Design Pointe Communications, Inc.; 1995;16:453–76.)

787

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

provides more useful information to determine the tumor location and to
give a more accurate prognosis to the owner. Magnetic resonance imaging
does a better job of delineating soft tissue densities, whereas a computed
tomography scan better delineates bony abnormalities; however, each
modality has its own limitations [17].

Exenteration

An exenteration is the total removal of all the orbital contents and is most

commonly indicated in cases of orbital neoplasia or invasive cutaneous neo-
plasia adjacent to the orbit. An exenteration can be performed through a
transconjunctival or transpalpebral approach depending on surgeon prefer-
ence. In most cases, a transpalpebral approach is recommended. The proce-
dure is begun as for a transpalpebral enucleation until the orbital rim is
encountered. Instead of dissecting to the sclera and removing the globe, the
entire orbital contents are removed by bluntly dissecting all tissues from the
orbital margin. This is quite simple in the dorsal, temporal, and medial
aspects of the orbit, because those areas have a bony margin. More care
should be taken in the ventral and ventromedial aspects of the orbit, because
those margins are deﬁned by muscle. Once the orbital margins are deﬁned
and the orbital contents are separated from them, the orbital contents
remain attached only at the orbital apex in the area of the optic foramen.
This attachment is excised with curved scissors. As with an enucleation, care
must be taken not to place excessive traction on the optic nerve until it has
been severed. If all orbital infection or neoplasia has been removed, an orbi-
tal implant may be considered to improve cosmesis. Closure of the surgery
site and postoperative care are similar to those used for enucleation.

For localized orbital neoplasia where the globe may be salvaged, multiple

orbitotomy procedures have been described [13,18–20]. These procedures
usually require some specialized equipment and expertise and may be best
suited to a referral practice.

Laser surgery

As laser surgery is becoming more available and cost-eﬀective, more prac-

titioners are purchasing or strongly considering the purchase of a laser.
Although lasers have a solid place in ophthalmic surgery, care must be taken
not to use a laser just because we can. In my experience, carbon dioxide
lasers do not consistently oﬀer any true advantages over a scalpel blade and
appropriate cautery. Ophthalmic lasers do oﬀer consistent advantages in the
treatment of intraocular disease. Both neodymium:yttrium, aluminum,
garnet and diode lasers have well-documented success in the treatment of
glaucoma, intraocular neoplasia, and retinal separation. In cats, the most
common use of a laser is in the treatment of suspected or conﬁrmed intra-
ocular neoplasia, speciﬁcally feline anterior uveal melanoma [21,22].

788

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Traumatic ocular proptosis

Whereas traumatic proptosis is much more common in dogs than in cats,

it does occur with some frequency in cats. Because cats have a deep orbit,
more substantial trauma is necessary to cause a proptosis than in the aver-
age dog. This decreases the prognosis for retention of vision as well as for
retention of a cosmetically acceptable globe. In cats, the most common asso-
ciated complications were facial bone fractures, hyphema, corneal perfora-
tion, and ocular desiccation [23]. In the same study, avulsion of three or
more extraocular muscles was also found to be a negative prognostic indi-
cator. Although the prognosis for cats is much less favorable than for dogs,
I usually recommend replacement of the globe with a temporary tarsorrha-
phy if the orbit and globe are relatively intact (Fig. 17). Topical antibiotic or

Fig. 17. Replacement of a proptosed globe. (A) Lateral view of a proptosis demonstrating the
rostral displacement of the globe with entrapment of the eyelid margins behind the equator of
the globe. (B) A lateral canthotomy helps to release the entrapped eyelid margin to allow proper
placement of eyelid sutures. (C) Nonabsorbable sutures (4-0–5-0) are preplaced in the eyelid
margin in a split-thickness simple interrupted or split-thickness horizontal mattress pattern. The
split-thickness pattern allows for direct apposition of the eyelid margin without danger of
sutures rubbing on the cornea. (D) A scalpel handle can be used to protect the cornea from the
preplaced sutures as the sutures are tightened. (E) The tarsorrhaphy sutures are tightened,
leaving a space at the medial canthus for placement of topical medication. Stents can be used to
protect the eyelid margin but are not usually necessary. (F) The lateral canthotomy is closed in
two layers. (F) A side view demonstrates how the split-thickness pattern closes the eyelid margin
and protects the cornea from the sutures.

789

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

antibiotic/steroid medications can be used depending on the health of the
cornea. Systemic antibiotics and anti-inﬂammatory agents are usually war-
ranted to limit infection and decrease orbital inﬂammation.

References

[1] Grevan VL. Ophthalmic instrumentation. Vet Clin North Am Small Anim Pract 1997;27:

963–86.

[2] Nasisse MP. Principles of microsurgery. Vet Clin North Am Small Anim Pract 1997;27:

987–1010.

[3] Dziezyc J, Millichamp NJ. Surgical correction of eyelid agenesis in a cat. JAVMA 1989;25:

513–6.

[4] Albert RA, Garrett PD, Whitley RD. Surgical correction of everted third eyelid in two cats.

JAVMA 1982;180:763–6.

[5] Morgan RV, Duddy JM, McClurg K. Prolapse of the gland of the third eyelid in dogs:

a retrospective study of 89 cases (1980–1990). J Am Anim Hosp Assoc 1993;29:56–60.

[6] Davidson MG, Nasisse MP, Jamieson VE, et al. Traumatic anterior lens capsular

disruption. J Am Anim Hosp Assoc 1991;27:410–4.

[7] Hakanson NE, Meredith RE. Conjunctival pedicle grafting in the treatment of corneal

ulcers in the dog and cat. J Am Anim Hosp Assoc 1987;23:641–8.

[8] Morgan RV. Feline corneal sequestration: a retrospective study of 42 cases (1989–1991).

J Am Anim Hosp Assoc 1994;30:24–8.

[9] Wilkie DA, Whittaker C. Surgery of the cornea. Vet Clin North Am Small Anim Pract 1997;

27:1067–107.

[10] Bingaman DP, Lindley DM, Glickman NW, et al. Intraocular gentamicin and glaucoma:

a retrospective study of 60 dog and cat eyes (1985–1993). Vet Comp Ophthalmol 1994;4:113–9.

[11] Dubielzig RR, Hawkins KL, Toy KA, et al. Morphologic features of feline ocular sar-

comas in 10 cats: light microscopy, ultrastructure, and immunohistochemistry. Vet Comp
Ophthalmol 1994;4:7–12.

[12] Barnett KC, Crispin SM. Fundus. In: Barnett KC, Crispin SM, editors. Feline oph-

thalmology. 1st edition. London: WB Saunders; 1998. p. 146–68.

[13] Ramsey DT, Fox DB. Surgery of the orbit. Vet Clin North Am Small Anim Pract 1997;27:

1215–64.

[14] Hamor RE, Roberts SM, Severin GA. Use of orbital implants after enucleation in dogs,

horses, and cats: 161 cases (1980–1990). JAVMA 1993;203:701–6.

[15] Glover TD, Constantinescu GM. Surgery for cataracts. Vet Clin North Am Small Anim

Pract 1997;27:1143–73.

[16] Gilger BC, McLaughlin SA, Whitley RD, et al. Orbital neoplasms in cats: 21cases (1974–

1990). JAVMA 1992;201:1083–6.

[17] Ramsey DT, Gerding PA, Jr, Losonsky JM, et al. Comparative value of diagnostic imaging

techniques in a cat with exophthalmos. Vet Comp Ophthalmol 1994;4:198–202.

[18] Gilger BC, Whitley RD, McLaughlin SA. Modiﬁed lateral orbitotomy for removal of

orbital neoplasms in two dogs. Vet Surg 1994;23:53–8.

[19] Slatter DH, Abdelbaki Y. Lateral orbitotomy by zygomatic arch resection in the dog.

JAVMA 1979;175:1179–82.

[20] Slatter DH, Wolf ED. Orbit. In: Slatter D, editor. Textbook of small animal surgery. 2nd

edition. Philadelphia: WB Saunders; 1993. p. 1245–63.

[21] Cook CS, Wilkie DA. Treatment of presumed iris melanoma in dogs by diode laser

photocoagulation: 23 cases. Vet Ophthalmol 1999;2:217–25.

[22] Hamor RE, Hinkle KM. Ocular neoplasia. In: Bonagura JD,

editor. Kirk’s current

veterinary therapy XIII. 13th edition. Philadelphia: WB Saunders; 2000. p. 1094–7.

[23] Gilger BC, Hamilton HL, Wilkie DA, et al. Traumatic ocular proptoses in dogs and cats:

84 cases (1980–1993). JAVMA 1995;206:1186–90.

790

R.E. Hamor / Vet Clin Small Anim 32 (2002) 765–790

Feline odontoclastic resorptive lesions

An unsolved enigma in veterinary dentistry

Alexander M. Reiter, Dipl Tzt

Krista A. Mendoza, DVM

Department of Clinical Studies, Veterinary Hospital of the University of Pennsylvania,

3900 Delancey Street, Philadelphia, PA 19104–6010, USA

Veterinary Specialty Hospital of the Carolinas, 305-C Ashville Avenue,

Cary, NC 27511, USA

Feline odontoclastic resorptive lesions (FORLs) are a common and frus-

trating ﬁnding in small animal practice and represent the most common den-
tal disease in cats. Sophisticated dental treatments promise neither cure nor
permanent improvement of aﬀected teeth. Research on the etiology of
FORLs has not been rewarding in recent years, and the causative factors
contributing to the development of FORLs are still unknown. Despite the
clear evidence of a resorptive process, this condition in cats was often termed
caries or erosion. Furthermore, the anatomic connotation of the term neck
lesion and the use of various other terms often led to confusion and misdiag-
nosis of FORLs. These lesions initially start in the periodontal ligament and
cementum below the gingival margin and not at the cervical portion of the
tooth crown. The authors of this article favor the term feline odontoclastic
resorptive lesions because it most accurately reﬂects the disease process.

Tooth resorption is considered to be either internal or external [1–5].

FORLs are characterized by progressive defects of the calciﬁed substance
of permanent teeth [6], resulting from the destructive activity of odonto-
clasts on the cemental, or external, tooth surface (Figs. 1–3) [7,8]. Internal
resorption results from the clastic activity of cells lining the pulp chamber
or root canal wall and is seen on radiographs as an irregular oval-shaped
enlargement of the pulp cavity. It is important to distinguish between the
forms of resorption, because their indicated treatments are diﬀerent; internal
resorption requires endodontic therapy [4]. If internal resorption occurs in

Vet Clin Small Anim 32 (2002) 791–837

* Corresponding author.

E-mail address: Reiter@vet.upenn.edu (A.M. Reiter).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 2 7 - X

the crown, pinkish discoloration (‘‘pink tooth of Mummery’’) and enamel
perforation may take place [9,10]. A pinkish discoloration is also possible
in the external resorptions that undermine crown dentin and enamel, or sim-
ply when pulpitis is present in younger animals with thin crown walls.

Fig. 1. Multiple feline odontoclastic resorptive lesions in a domestic cat. Stage 2 lesion of the
left maxillary third premolar with gingival hyperplasia (*), stage 3 lesion of the left mandibular
third premolar with inﬂamed granulation tissue ﬁlling the defect (small arrows), stage 4 lesion of
the left maxillary fourth premolar with extensive structural damage (stippled line outlines
original shape of crown), and stage 5a lesion of the left mandibular canine with crown fracture
(large arrow). The left maxillary canine appears to extrude abnormally (dotted line depicting
cementoenamel junction).

Fig. 2. Clinical picture of stage 3 feline odontoclastic resorptive lesion of the right mandibular
fourth premolar. Note the highly vascular granulation tissue ﬁlling the defect (*). There is a
small bulge in the area of a missing right mandibular third premolar (arrows).

792

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Etiopathogenesis

Extension of periodontal disease

Chronic inﬂammation of the periodontium was suggested to be a cause of

FORLs [11–17]. Plaque accumulation can cause inﬂammation of periodon-
tal tissues, leading to local immune responses and release of inﬂammatory
elements (eg, cytokines) and bacterial products (eg, lipopolysaccharides)
that stimulate diﬀerentiation and migration of clastic cells [18–20]. Cyto-
kines, such as interleukin (IL)-1 and IL-6, are secreted locally by epithelial
and endothelial cells as well as by inﬂammatory cells [21]. One author
described FORLs as ‘‘necrobacillosa dentis,’’ an infectious disease transmit-
ted from cat to cat through saliva and nonsterile dental instruments, making
lipopolysaccharides of gram-negative anaerobic Fusobacterium and Bacte-
roides species responsible for the development of FORLs [22]. Antibodies
reactive to Actinobacillus actinomycetemcomitans (A.a.), the cause of juve-
nile periodontitis in human beings, were found in higher concentration in
cats with oral inﬂammation than in cats with little or no oral inﬂammation
[23]. Pasteurella species (commonly found in the mouths of cats) and A.a.
antigens cross-react, and A.a. antibodies in serum of cats may be Pasteurella
species antibodies [18]. One study failed to demonstrate the presence of A.a.
in cats with or without FORLs [24].

Although most FORLs are associated with inﬂammatory cells, the early

lesion, covered by an apparently normal gingiva, does not seem to be. This
led some researchers to believe that the appearance of inﬂammatory cells is a
secondary rather than primary event during development of FORLs [25]. In

Fig. 3. Dental radiograph of area shown in Figure 2. There is a large scalloping defect visible in
the mesial crown portion of the right mandibular fourth premolar (large arrow). Several other
lesions are present (small arrows). Remnants of the right mandibular third premolar appear as
ghost roots (*).

793

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

a necropsy study, including the clinical examination, histology and radio-
graphy, no signiﬁcant correlation between periodontitis and FORLs in feline
mandibles could be demonstrated [26]. A longitudinal analysis of radio-
graphs revealed that FORLs appeared on some teeth, although plaque and
gingival indexes remained low, whereas teeth with greater plaque and gingi-
val indexes were free of FORLs; there was no signiﬁcant correlation between
FORLs and gingivitis [27]. It is currently not known whether FORLs devel-
op without inﬂammation and the present inﬂammation clinically results
from the rough plaque-retentive surface of the resorptive defect or whether
the inﬂammation is the initial etiopathogenetic factor.

Anatomic peculiarities of feline teeth

The predilection of FORLs for certain sites of the teeth may be a reﬂec-

tion of the tooth anatomy in domestic cats [28]. Unusual structures in feline
permanent teeth are most commonly observed in the coronal dentin, root
furcation area, and planes facing the interradicular septa [18]. The occur-
rence of vasodentin and osteodentin may be evidence of deﬁcient dentin
mineralization [29]. Vasodentin has been observed in ﬁsh, rabbit, armadillo,
prairie dog, and other normal mammalian teeth as well as in human teeth
after trauma and in teeth with osteogenesis imperfecta [30]. In cats, it was
found in 3 of 10 control teeth and in 6 of 49 teeth with FORLs [31]. Another
study found vasodentin approximately equally in teeth with or without
FORLs, although the locations of vasodentin and FORLs diﬀered
[19,21,29]. In vasodentin, the dentinal tubules run randomly with penetra-
tion of canals. Rounded cells or ﬂattened endothelial-like cells were
observed inside the canals [29]. It is not known whether peripheral blood cir-
culates in these canals or if these canals are connected to a vascular system
(originating from the periodontal ligament) and could participate in calcium
regulation [18]. Osteodentin (intermediate cementum) was most often
observed in root dentin close to the root canal and at furcation areas [19].
Although FORLs were not more commonly found in teeth with osteodentin
as compared to teeth with no detectable osteodentin, FORLs tend to occur
in areas of the tooth in which osteodentin is most typically found [19,21].

Microhardness of enamel and dentin has been shown to be lower in cats

than in dogs and human beings [32]. It also decreases signiﬁcantly from the
upper part of the crown to the cervical area and from the outer layer to the
inner layer [33]. Enamel thins noticeably closer to the cementoenamel junc-
tion. Chemical analysis of teeth in dogs and cats showed that magnesium
concentrations were signiﬁcantly higher in canine dentin, whereas calcium
concentrations and the calcium-to-phosphorus quotient were signiﬁcantly
higher in feline dentin [34]. Although lateral and accessory canals are un-
common in pet carnivores, lateral canals have been observed in permanent
premolar and molar teeth in cats but not in their canine teeth [35,36]. Fur-
cation canals connecting the pulp chamber and periodontal ligament were

794

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

found in deciduous premolar teeth in kittens; after experimental pulp injury,
resorption of dental tissues and alveolar bone in the furcation area took
place [37]. In a more recent study, patent furcation canals were found in
27% of maxillary and mandibular permanent carnassial teeth in adult cats;
nonapical ramiﬁcations from the main canal (lateral and secondary canals)
were found in 9% of the examined roots [38]. Preliminary results of a study
on the histologic appearance of normal feline root cementum indicated that
the middle and apical portions of root cementum appeared similar to
cementum in human beings; however, three types of cervical root cementum
were found: cementum as in human beings, cementum with cementoid of
varying thickness, and a thick ledge of cellular cementum overlaid by
cementoid [39]. The signiﬁcance of these ﬁndings is as yet unknown. Cemen-
tum may also be thin in furcation areas of feline teeth [36]. Cementum abuts
the enamel in 30% of human teeth, overlaps it in 60%, and is separated from
it by a gap in approximately 10%, although a wide range of variation is
found in diﬀerent teeth and diﬀerent studies [40]. Furcation canals, apical
and nonapical ramiﬁcations, and the cementoenamel junction could be areas
that favor the development of FORLs if the protective cemental layer fails
to cover the dentin. Mineralized dentin would thus be exposed and could
attract odontoclasts.

Mechanical trauma

Acute and chronic mechanical trauma can induce root resorption, partic-

ularly apical root resorption [9,19]. The more apical location of FORLs
often found in canine teeth and the fact that these front teeth are more sub-
ject to mechanical trauma would ﬁt with this concept. Teeth of domestic cats
are usually not submitted to high mechanical trauma, however. In one
study, inﬂammatory root resorption apical to the junctional epithelium was
noted in feline teeth after removal of buccal bone and placement of a foreign
body at the periphery of the periodontal ligament [41]. The so-called
occlusal stress (tooth ﬂexure) theory was created in an attempt to explain
noncarious cervical lesions in human teeth [42,43]. Repeated compressive
and tensile forces caused by tooth ﬂexure during mastication and malocclu-
sion may disrupt the chemical bonds between enamel rods, resulting in
abfraction of enamel and exposure of dentin. Occlusal stress was considered
a contributing factor in the development of FORLs [44,45], although
FORLs may develop on any teeth and any tooth surfaces and not just on
those exposed to occlusal or shearing forces.

Risk factors, diet texture and components

Cats that gulped their food compared with those that chewed their food

well had an increased prevalence of FORLs [47]. A history of abscesses,
feeding noncommercial cat food and treats, and consumption of cheese,

795

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

butter, and table foods may represent an increased risk for FORLs. A
decreased risk may be explained by hunting prey or mixing noncommercial
and commercial cat foods or treats [48]. Cats with FORLs may also be
older, female, taking medications, drinking city (versus well) water, and play-
ing less often with toys [49]. Cats without FORLs were more likely to have
owners who cleaned their teeth daily or twice a week and fed diets with higher
magnesium, calcium, phosphorus, and potassium contents [49]. A history of
dental disease, city residence, and being an exclusively indoor cat was
associated with an increased risk for FORLs, whereas consumption of com-
mercial treats seemed to be protective for FORLs [50]. Cats that vomit reg-
ularly may have a greater number of FORLs [47]. The low pH of gastric
ﬂuids deposited around the gingival margin after vomiting is thought to
damage enamel and cementum and may predispose to development of
FORLs [28,51,52]. Coating the surface of dry kibbles with an acidic sub-
stance to preserve the food and enhance its palatability does not predispose
the teeth to development of FORLs [53,54]. No correlation was found
between speciﬁc oral bacteria and the pH of the tooth surface in cats [55].

Although the prevalence of tooth resorption in the stray, feral, and exotic

cat population is low, several studies demonstrated resorptive defects of per-
manent teeth in stray and feral small cats [10,11,47,56–60] and in large cats
(Table 1) [11,61–71]. The texture of a ‘‘natural diet’’ may not totally protect
permanent teeth from resorption, leading some investigators to believe that
a nonnatural diet has no eﬀect on the pathogenesis of FORLs [72]. The diet
of zoo carnivores in North America largely consists of food composed of
ground meat. Animals on such a soft diet suﬀer from periodontal disease
similar to that seen in the domestic pet population [64,73]. Resorption of
permanent teeth in carnivores may also be common in North American
zoos. In Europe, where the diet of large carnivores is composed of large
pieces of ﬁbrous meat (‘‘self-cleansing diet’’), periodontal disease and
resorption of teeth are rare and may be less than 2% [65]. Pet foods are
nutritionally balanced; however, they lack the ﬁbrous consistency of a nat-
urally self-cleansing diet. Dry food can maintain healthy gums and teeth in
cats [74]. Food form (size and shape) does aﬀect dental substrate accumula-
tion in cats, and a highly signiﬁcant reduction in plaque and calculus accu-
mulation can be achieved through a maintenance diet with enhanced
textural and other physical characteristics [75]. Although feeding a dry diet
compared with a soft diet may be associated with less accumulation of
plaque and calculus in cats [76,77], a dry food diet is not associated with
a lower prevalence of FORLs [78].

FORLs were found to be associated with excessive feeding of raw liver

[25,79,80]. Both retinol and retinoic acid may stimulate clastic cell activity
directly. In clinical and experimental studies, researchers reported ‘‘caries,’’
tooth loss, and retention and displacement of incisors in cats with chronic
hypervitaminosis A [81–85]. There were also cats fed predominantly with
raw bovine liver that did not develop FORLs, however [86]. An increased

796

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

ble

evalence

felin

ontoclast

reso

rptive

lesio

stray/

feral

small

and

exotic

cats

Year

reported

ountry

Mat

erial

Perc

entage

RLs

mber

cats

exa

mined

Dia

gnosis

Signiﬁ

cant

ﬁndin

eth

oft

aﬀecte

Numbe

teeth

aﬀected

hlup

and

Stich

[10]

1982

SWI

Skulls

(str

and

feral)

n/r

hlup

and

Stich

[10]

1982

SWI

Skulls

(str

ay)

244

n/r

erstraete

[60]

1996

SAF

Skulls

(feral)

301

Visua

exa

minatio

Perio

dontitis

62%

all

cats,

root

remna

nts

not

recorde

FORL

anines

and

carnas

sials

1.4

vin

[58]

1996

Liv

cap

tive

exo

tic

cats

1–2

Retrosp

ectiv

udy

(dental

and

medical

rec

ords)

n/r

vin

[58]

1996

Eutha

nized

feral

cats

Probin

rad

iograp

Perio

dontitis

55%

all

cats,

100%

all

cats

with

FORL

Inc

isors

and

molars

1.4

larke

and

Cam

eron

[47]

1997

Liv

feral

cats

Oral

examin

ation

All

cats

exa

mined

were

less

than

months

age

n/r

Gio

[63]

2001

Liv

cap

tive

exotic

cats

36,

Anesth

esia

puma

and

jaguars

had

FORL

n/r

FORL

Feline

odontoc

lastic

resorp

tive

lesion;

SWI

Switz

erland;

SAF

South

Africa

;

AUS

Aus

tralia;

USA

Unit

State

erica;

Brazil;

not

rep

orted.

797

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

prevalence of FORLs was signiﬁcantly associated with a calcium-deﬁcient
diet, decreased radiopacity of lamina dura and alveolar bone, and horizon-
tal alveolar bone loss [86]. A calcium-deﬁcient diet can lead to a nutritional
secondary hyperparathyroidism with increased synthesis and release of para-
thyroid hormone [87] that stimulates osteoclasts to mobilize calcium from
bone. There is no evidence that permanent teeth undergo resorption in cats
and dogs with nutritional secondary hyperparathyroidism [87–98]. In a
long-term study, the pathogenesis of periodontal disease was investigated
in laboratory dogs with induced nutritional secondary hyperparathyroidism
[99]. After 12 months, the test dogs showed signiﬁcantly decreased radiopac-
ity of the jaw bones and accelerated alveolar bone resorption, and ‘‘cemento-
lysis was accentuated’’ [99]. No changes were observed in the dentin,
however [99]. A study on the interrelation between calciotropic hormones
and related factors in cats with FORLs is underway [100].

Endocrine and metabolic imbalances

The roots of teeth show a remarkable resistance to detectable resorption,

even with systemic diseases that can cause signiﬁcant bone resorption. Renal
dystrophy results in an increased oxalate concentration in the blood and
precipitation in the hard tissues, which can cause root resorption in human
beings [101]. Genetic linkage is likely, because external root resorption of no
apparent cause has been found in members of the same family [101]. Apical
lesions related to hypoparathyroidism, pseudohypoparathyroidism, and
hypothyroidism may not represent true tooth resorption but simply short
blunted roots caused by early apical closure, because this may also occur
with radiation exposure [9]. Root resorption has not been reported in dogs
or cats with primary hyperparathyroidism [102–105] or renal secondary
hyperparathyroidism [106–110]. Because resorption of permanent teeth is
often observed in females, abnormalities of sexual hormones may play a role
in root resorption [109]. Estradiol deﬁciency caused by menopause or
removal of the ovaries results in accelerated bone loss and may also be
responsible for postmenopausal osteoporosis [111]. Estrogens are mainly
produced in the ovaries, in the placenta during pregnancy, and in small
amounts in the adrenal cortex and the testicles. Regular neutering of domes-
tic cats was not found to be associated with the development of FORLs
[112–114]. Stress-induced hypocalcemia was suggested to be a cause of root
resorption in captured or domesticated mammals [69].

Local and systemic viral infections

Calicivirus has been implicated in the development of FORLs and the so-

called supereruption of canine teeth of domestic cats [13,115]. Feline calici-
virus (FCV) was isolated from oral swabs in 19% of randomly selected cats
and in 61% of cats with a speciﬁc history of oral inﬂammation [116].
Approximately 20% of the random cat population throughout the world

798

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

may chronically shed FCV from their oral cavities [117]. In one study, 8 of
11 cats (73%) with FORLs had severe chronic stomatitis; 5 of these cats were
found to be positive for FCV by means of polymerase chain reaction anal-
ysis from gingival samples [118]. Other studies found chronic stomatitis in
only 3% [112] and 17% [31] of cats with FORLs. Conditions that lead to dys-
function of the immune system or immune suppression play a role in the
cause of periodontal disease. All immunosuppressive viruses in cats may
produce infections with oral manifestations and lead to lifelong persistence
[72]. Many cats infected with feline immunodeﬁciency virus (FIV) or feline
leukemia virus (FeLV) show chronic oral inﬂammation. The incidence of
FORLs in cats infected with FIV was found to be higher (6 of 10 [60%])
compared with age-matched control cats free of FIV (3 of 9 [33%]) [119].
Interpretation of these ﬁndings is diﬃcult because of the small number of
cats studied and the investigative methods used (dental radiographs to verify
the presence of FORLs were taken in only 1 cat). Another study found
FORLs in 13 of 30 cats (43%) with chronic oral inﬂammation [120]. Sys-
temic immunosuppressive diseases may aggravate the condition; however,
it is unlikely that they alone initiate FORLs [19]. In the authors’ own expe-
rience, few cats with FORLs are positive for FIV or FeLV; however, cats
with chronic oral inﬂammation may be more likely to have FORLs com-
pared with cats that have healthy gums [121].

A new approach

Unlike bone that undergoes resorption and apposition as part of a con-

tinual remodeling process, the roots of permanent teeth are normally not
resorbed. Only the resorption of deciduous teeth can be considered physio-
logic. Tooth resorption of no apparent cause is termed idiopathic resorption,
which more accurately reﬂects our limited understanding of the causative
factors than the absence of a causative factor in tooth resorption. Two con-
ditions must be present in local root resorption [122]. First, the protective
covering of the root must be missing or altered. Second, a stimulus for the
clastic cells must be present. As soon as the stimulus is removed, resorption
stops and healing follows. If additional stimulation is not present, the
resorption is transient. If the stimulus is sustained, the resorption is progres-
sive [123]. The roots of permanent teeth are resistant to resorption both on
external surfaces and internally on pulpal aspects. The exact mechanism by
which the resorption process is inhibited is as yet unclear. Cementoblasts
and precementum as well as odontoblasts and predentin are uncalciﬁed
organic components and may have resorption-inhibiting characteristics
[101,124–128]. Odontoclasts may be attracted only to or can attach only
to mineralized tissue. If mineralized tissue is not present, the odontoclast
is not attracted to the root surface. It is postulated that either removal or
mineralization of the organic matrix of bone or the root covering makes
it possible for clastic cells to recognize the mineral component [101,129].

799

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Retrospective examinations of zoologic and museum collections of feline

skulls revealed a low incidence of FORLs before the 1970s (Table 2)
[56,130–133]. The increase of FORLs recognized in the 1970s may be asso-
ciated with aspects of domestication, such as altered feeding practices and
the introduction of neutering and vaccination programs [5,28,79,86,134].
It may be that aspects of domestication and environmental agents have
disturbing inﬂuences on the development of feline teeth. The resorption-
inhibiting characteristics of the outer tooth surface may be altered, inviting
inﬂammation caused by periodontal disease and other possible sources of
local inﬂammatory activity to attack not only alveolar bone but also teeth
by stimulated odontoclasts.

Prevalence and predisposition

A signiﬁcant increase of FORLs has been recognized since the 1960s,

with the frequency varying from 2% to 75% depending on the population

Table 2
Prevalence of feline odontoclastic resorptive lesions in retrospective examinations of feline
skulls

Death of cats

Percentage
of FORLs

Number
of cats
examined

Source of skulls

Colyer [130]

Before 1936

140

Royal College of Surgeons

Odontologic Collection
and British Museum of
Natural History
(London, United
Kingdom)

Dobbertin [56]

Before 1940

Germany and Switzerland*

Dobbertin [56]

1940–1949

Germany and Switzerland*

Dobbertin [56]

1950–1959

Germany and Switzerland*

Harvey and

Alston [132]

Before 1959

Royal College of Surgeons

Odontologic Collection
and British Museum of
Natural History
(London, United
Kingdom)

Dobbertin [56]

1960–1969

Germany and Switzerland*

Dobbertin [56]

1970–1979

339

Germany and Switzerland*

Lu¨ps [59]

1971,

1973–1974

Not

reported

257

Stray cats killed in

Switzerland

Dobbertin [56]

1980–1989

Germany and Switzerland*

Dobbertin [56]

Not known

(before 1993)

Germany and Switzerland*

Okuda et al [114]

Before 1994

138

Japan

* Zoologisches Institut Hamburg (random selection), Institut fur Haustierkunde Kiel

(random selection), and Naturhistorisches Museum Bern (stray cats killed in 1971, 1973–1974,
and 1976).

800

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

of cats studied and the investigative methods applied (Table 3). The popu-
lations are described as ‘‘random,’’ meaning there was no speciﬁc reason for
the veterinary examination; ‘‘mixed,’’ meaning that cats were presented to a
veterinarian for treatment; and ‘‘dental,’’ meaning that cats were examined
during a procedure scheduled for examination or treatment of oral disease
[135]. Prevalence in these studies is reported based on clinical observation
only, with or without a dental explorer, or by radiographic examination
under general anesthesia in addition to the clinical examination. Studies
without dental radiographs greatly underestimate the actual number of
FORLs present [136] as may studies of subjects with missing teeth that may
have been lost as a result of FORLs [113,135]. The only statistically
supported observations made in more than one report are that FORLs
are rarely seen in cats younger than 2 years of age [79,80,114] and that there
is an increasing prevalence of FORLs with increasing age [26,31,
47,49,80,113,133,135,137–141]. Onset of the disease is usually at 4 to 6 years
of age [22,26,49,80,113,133,137]. The number of teeth that have FORLs also
increases with age. FORLs may occur more often in purebred cats
[47,80,133,140,142], although the current data are insuﬃcient to verify a
breed predisposition [135]. Persian Longhair cats seem to develop FORLs
at a younger age compared with other breeds [31,133,142]. Neutering, gen-
der, and age at neutering may not aﬀect the prevalence of FORLs [112–114].
Only one study found an increased prevalence of FORLs in male cats [133],
whereas two studies revealed female cats to be more commonly aﬀected
[49,114]. FORLs are reported to occur more often on the buccal and labial
tooth surfaces of premolar and molar teeth and less commonly on canines
and incisors [47,49,52,80,133,137,138]. The maxillary and mandibular third
premolars, the maxillary fourth premolar, and the mandibular ﬁrst molar
are most commonly aﬀected [17,78,80,113,121,133,140,143].

Classiﬁcation

FORLs are clinically and radiographically classiﬁed into ﬁve stages [52],

although earlier reports combined stages 4 and 5 for therapeutic reasons
[144]. Stage 1 lesions extend into the cementum only (Fig. 4). They do not
enter the dentin and are not sensitive. They may be diﬃcult to detect because
of their microscopic size. The authors of the present article think that most
lesions diagnosed as stage 1 lesions may already be stage 2. Stage 2 lesions
progress through cementum into crown or root dentin and become painful
when dentinal tubules are exposed (Fig. 5). One author further divides stage
2 lesions into stage 2a and 2b lesions. Stage 2a lesions extend into the dentin
but not near the pulp, whereas stage 2b lesions extend into the dentin and
near the pulp [145]. Hyperplastic gingiva and inﬂammatory granulation tis-
sue may overlie these defects. Stage 3 lesions advance through the dentin
into the pulp chamber or root canal and are painful (Fig. 6). Bleeding from

801

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Table

Studie

prevale

nce

felin

ontoclast

reso

rptive

lesio

dome

stic

cats

Yea

repor

ted

Country

Percen

tage

FORL

cats

examined

Numbe

cats

examined

Source

cats

Diagnosti

tools

ommen

eth

often

aﬀe

cted

ean

numbe

teeth

aﬀe

cted

per

aﬀe

cted

cat

Hopewell-Sm

ith

[57]

1930

USA

n/r

200

n/r

ost

cats

were

youn

with

good

dentitio

few

cats

had

FORL

Tekeli

[154]

1973

SWI

420

Random

Probin

(limited

radiog

raphy)

oth

cavit

ations

described

‘‘cari

es’’

Schlup

[80]

1982

SWI

200

Random

Probin

›

age

›

purebre

cats

and

hen

fed

raw

liver

emola

and

molars

2.9

Reichart

[157]

1984

GER

(premolars

and

lars)

Random

selected

head

(necro

psy

study)

Probin

radiog

raphs

Perio

dontiti

77%

all

premolar

and

molars

Frost

and

Will

iams

[79]

1986

USA

n/r

Cats

with

chronic

ora

inﬂam

mation

(AVDS

survey

)

n/r

802

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Harvey

[167]

1988

USA

(inc

isors),

(man

d/max

P3)

(man

n/r

Mixed

n/r

Zetne

[120]

1989

AUT

Cats

with

chron

stom

atitis

Exami

nation

radi

ographs

n/r

Zetne

[86]

1990

AUT

Dental

(chr

onic

ora

disease

)

Anesth

esia,

prob

ing,

radi

ographs

›

signi

ﬁcantly

associa

ted

with

calcium-

deﬁc

ient

diet

decre

ased

radiop

acity

lam

ina

dura

and

alve

olar

bone,

and

horizon

tal

alveola

bone

loss

Coles

[137]

1990

AUS

Mixed

Anesth

esia,

prob

ing

›

age

emola

and

molars

3.2

Cros

sley

[112]

1991

152

Dental

n/r

FORL

50%

cats

with

periodo

ntitis,

periodo

ntitis

32%

cats

with

FORL

eﬀec

neute

ring

3.5

(conti

nued

page

)

803

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Table

(conti

nued

)

Year

reported

ountry

Perc

entage

RLs

cats

exa

mined

Numbe

cats

examined

Sour

cats

Dia

gnostic

too

Comments

Teeth

most

often

aﬀected

ean

numbe

teeth

aﬀe

cted

per

aﬀe

cted

cat

Remeeus

[142]

1991

306

ntal

n/r

›

pre

Persian

(aﬀe

cted

youn

ger

age)

and

Siam

ese

n/r

Wessum

[133]

1992

432

ntal

Anesth

esia

pro

bing

›

age

;

›

pre

Persia

(aﬀe

cted

youn

ger

age),

Siame

se,

and

Aby

ssinian;

›

mal

Max

and

mand

2.8

Wessum

[133]

1992

ental

Anesth

esia

pro

bing,

rad

iograp

›

age

RLs

suprab

ony,

8.5%

cre

stal

bone

area

Premolars

4.1

Harvey

[138]

1992

796

ixed

Prob

ing

›

age

Max

P3,

max

P4,

mand

and

mand

2.3

Wiggs

[unpu

blished]

1993

605

ixed

n/r

FORL

38%

cats

ithout

ronic

oral

inﬂam

matio

n/r

3.5

Wiggs

[unpu

blished]

1993

214

Cat

without

chronic

oral

inﬂamma

tion

n/r

›

age

eﬀ

ect

utering

n/r

2.1

804

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

bbertin

[56]

1993

GER

Prese

nted

for

utering

Exami

nation

prob

ing

Periodon

titis

35.1%

teeth

with

RLs

(inc

luding

103

skulls

ith

FORL

s),

roo

remna

nts

not

recorde

FORL

and

M1,

max

P4,

and

max

2.2

(whe

includin

103

sku

lls

with

FORL

brescu

[22]

1994

GER

299

n/r

FORL

37.7%

cats

with

periodo

ntal

disease

n/r

Oku

[114]

1994

JAP

Dental

n/r

eﬀec

neute

ring;

FORL

61%

females

but

only

31%

mal

(me

age

7.4

years

for

fem

ales

and

5.7

years

for

mal

es)

and

2.5

ngler

[26]

1995

USA

Rando

mly

sele

cted

man

dibles

(nec

ropsy

udy)

Probin

rad

iograp

›

age

(bone

and

too

resorp

tion)

n/r

(continue

page

)

805

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Table

(con

tinued

)

Year

repor

ted

Country

Percent

age

FORL

cats

examined

Numbe

cats

examined

Sour

cats

Diagnost

tools

Comments

eth

often

aﬀe

cted

Mean

numb

teeth

aﬀecte

per

aﬀecte

cat

Ande

rson

[121]

1996

USA

Most

cats

had

FORL

Cats

with

ronic

omatitis

Anesth

esia,

prob

ing,

rad

iograp

Cats

with

RLs

had

tter

resolu

tion

oral

inﬂam

mation

after

tooth

extractio

comp

ared

cats

witho

FORL

and

premolar

n/r

Levin

[58]

1996

MEX

8.3

n/r

Rando

Anesth

esia,

prob

ing,

limited

rad

iograp

All

cats

non—

comm

ercial

diet

n/r

Lukman

[31]

1996

SLO

253

Dental

Anesth

esia,

prob

ing

›

age

stomatit

16.7%

cats

with

FORL

80.6%

cats

with

stomat

itis

and

M1,

mand

P3,

and

max

5.7

Roes

[140]

1996

GER

152

Dental

Anesth

esia,

prob

ing,

rad

iograp

›

age

Asian

Shor

thair

with

highe

ave

rage

numbe

FORL

per

mouth

(4.5)

P3,

mand

M1,

mand

P3,

and

max

3.4

806

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Clark

and

ameron

[47]

1997

AUS

168

Dental

Probin

rad

iograp

›

age

signiﬁc

antly

RLs

purebre

cats

(Aby

ssinian),

FORL

cats

that

gulpe

food

and

vomited

regular

emola

and

molars

n/r

Hennet

[180]

1997

FRA

Cats

with

ronic

omatitis

Anesth

esia,

prob

ing,

rad

iograp

10%

cats

had

chatt

ering

reﬂex

witho

any

clinically

radio-

graphi

cally

evident

FORL

n/r

Lund

[49]

1998

USA

145

Anesth

esia

for

any

reason

Exami

nation

Risk

facto

for

FORL

age

fem

ale,

medic

ations,

city

water,

toys,

brush

ing,

ﬂ

die

tary

magne

sium,

calcium,

phos

phor

us,

and

potassiu

and

premolar

3.9

Gioso

[237]

1999

BRA

60,

67,

126,

39,

n/r

Anesth

esia,

prob

ing,

rad

iograp

n/r

Lom

mer

and

erstraete

[227]

2000

USA

265

Dental

Radiog

raphs

Periapic

lucencie

13.7%

cats

with

RLs

(no

sign

iﬁcant

associa

tion

)

3.5

(continue

page

)

807

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

ble

(con

tinue

)

Year

reported

Country

Perc

entage

RLs

cats

exa

mined

Numbe

cats

examin

Sour

cats

Dia

gnostic

too

ommen

Teeth

most

often

aﬀected

Mean

numb

teeth

aﬀecte

cat

Har

vey

[166]

2000

USA

162

Dental

Rad

iograp

entral

point

canin

teeth

apical,

coronal

alveola

crest

mand

premolar

and

molars

Mand

M1,

max

and

mand

canin

es,

mand

4.4*

Verh

aert

[141]

2000

BEL

753

Mixed

Anest

hesia

52%

cats,

pro

bing,

limited

rad

iograp

›

age,

soft

food

double

FORL

times

periodo

ntal

disease

and

times

stomatit

cats

with

FORL

Premolars

and

molars

n/r

Lom

mer

and

Verstra

ete

[139]

2001

USA

147

Dental

Rad

iograp

›

age

n/r

Cogn

[118]

2001

FRA

Colony

cats

Anest

hesia,

pro

bing

vere

chronic

gingivo

stomatit

73%

cats

with

RLs

n/r

808

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Ingham

[113]

2001

228

Ran

dom

Ane

sthesia,

prob

ing,

radiog

raphs

›

age

FORL

asso

ciated

with

miss

ing

teeth

eﬀec

neute

ring

Man

n/r

The

only

study

that

counts

lesions

per

tooth

aﬀecte

cat

inste

aﬀected

teeth

per

aﬀecte

cat.

n/r

not

reported

;

mand

ibular;

max

maxillary

;

third

premo

lar;

fourt

pre

molar;

ﬁrst

molar;

feline

odont

oclastic

resorpt

ive

lesio

›

age

increa

sed

prevale

nce

FORL

with

increa

sed

age;

›

increa

sed

prevale

nce

FORL

random

spe

ciﬁc

reason

for

the

veterinary

exa

minatio

mix

cats

presented

for

treatmen

dental

cats

exa

mined

durin

proce

dure

sched

uled

for

examin

ation

treatm

ent

ora

disease;

USA

ited

States

erica;

SWI

Swit

zerland;

Germany;

AUT

Aust

ria;

Australia;

Unit

Kin

gdom;

NET

The

Netherlan

ds;

JAP

Japan

;

MEX

Mexico

;

Slove

nia;

FRA

Fra

nce;

Brazil;

BEL

Belgium;

AVDS

erican

eterinary

Dental

Societ

Adapte

from

van

Wessu

Harvey

Hennet

Feline

ntal

resorp

tive

lesions.

Preva

lence

patte

rns.

Vet

Clin

North

Small

Anim

Pract

1992;22:1

405–

16;

with

permission.

809

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

pulp tissue and gingival or pulpal granulation tissue is evident on probing.
Spontaneous fracture of the tooth crown often occurs. Stage 4 lesions have
extensive structural damage, and dentoalveolar ankylosis of roots with
alveolar bone may occur (Fig. 7). These teeth are fragile and prone to frac-
ture. Stage 5 lesions classically have no crown, with only root remnants left,
having the appearance of a ‘‘bulge’’ in the open and inﬂamed gingiva or in
the intact and not inﬂamed gingiva. In some cases, the crown may be intact,
whereas the root structure is widely lost because of extensive root replace-
ment resorption. For this reason, the authors of the present article suggest
that stage 5 lesions be subdivided into a stage 5a lesion for a tooth with
no crown but retained or ‘‘ghost root(s)’’ (Figs. 8–10) and a stage 5b lesion
for a tooth with a more or less intact crown but no or almost no roots(s)
(Figs. 11–13).

A less commonly used classiﬁcation system is the type A through F clas-

siﬁcation [146]. Type A and B lesions are conﬁned to the crown. Type C and
D lesions are conﬁned to the cementoenamel junction. Type E and F lesions
are conﬁned to the root. Type A, C, and E lesions show no endodontic
involvement, whereas type B, D, and F lesions expose pulp tissue. Another
classiﬁcation system divides FORLs into two diﬀerent types based on an

Fig. 4. Illustration of stage 1 feline odontoclastic resorptive lesion aﬀecting the cementum only.
(Courtesy of Rebecca Rae Bradford, Champaign, IL.)

810

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

elevation of aspartate aminotransferase and elastase concentrations in the
gingival crevicular ﬂuid [147]. Primary or type I FORLs include those seen
in cats with elevations in these enzyme concentrations before the develop-
ment of resorptive defects (cats with stomatitis) [147]. With the secondary
or type II FORLs, the elevation of enzyme concentrations would not be
present until the resorptive defect is clinically detectable [147]. Initiating
causes and predisposing factors for type II FORLs are still unknown, and
this category includes most FORLs found in domestic cats today.

Histopathologic appearance

FORLs were ﬁrst recognized and histologically diﬀerentiated from caries

in the 1920s [57,148], but they were not reported again in the veterinary liter-
ature until after the mid-1950s, when they were thought to be carious lesions
[149–155]. Two histopathologic studies in the 1970s again revealed that
‘‘neck lesions,’’ now referred to as FORLs, were not caries but a type of tooth
resorption [25,156]. Subsequent studies in the 1980s [10,80,157–159] and
1990s [11,19,21,29,31,61,160–165] provided new information to the veteri-
nary dental community, although they mainly conﬁrmed the microscopic

Fig. 5. Illustration of stage 2 feline odontoclastic resorptive lesion progressing into the dentin.
One lesion developed close to the cementoenamel junction. A second lesion developed in the
furcation area. (Courtesy of Rebecca Rae Bradford, Champaign, IL.)

811

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

picture of tooth resorption in human beings. Signiﬁcant discoveries did not
occur, and the true etiopathogenesis of FORLs remains speculative. There
is a need for further microscopic research to diﬀerentiate the histopathologic
ﬁndings of FORLs from normal histoanatomy in feline teeth [134].

Resorptive phase

The morphology of the resorptive defects on the tooth surface is similar

to that of shedding deciduous teeth, except that inﬂammatory cells are
numerous in permanent teeth aﬀected by FORLs [134]. The resorptive
defect in FORLs is described as a half-moon depression of varying depth
[10,79]. FORLs develop anywhere on the periodontal ligament and not just
close to the cementoenamel junction [26,166]. Most of the outer enamel sur-
face lies above the gingival margin and may not undergo direct odontoclas-
tic resorption. The resorptive process usually starts on root surfaces facing
the periodontal ligament, progressing from cementum coronally into crown
dentin as well as apically into root dentin [144,167]. As the lesion progresses
into crown dentin, the enamel becomes undermined (overhang of enamel)
[168]. The enamel is then resorbed (resorption of enamel) or loses its contact
with originally intact dentin and breaks oﬀ (abfraction of enamel).

Fig. 6. Illustration of stage 3 feline odontoclastic resorptive lesion entering the pulp cavity.
(Courtesy of Rebecca Rae Bradford, Champaign, IL.)

812

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Tartrate-resistant acid phosphatase is used to detect clastic cells, whereas

alkaline phosphatase is used as a marker for odontoblasts, cementoblasts,
and osteoblasts [19–21]. The recently discovered member of the tumor
necrosis factor family, receptor activator of nuclear factor-kappa B ligand
(RANKL), is released by osteoblasts and plays a profound role for osteo-
clastogenesis [169] and osteoclast activation [170]. The mRNA expression
of RANKL, IL-1, and IL-6 was studied in teeth with FORLs and in normal
teeth by reverse transcription polymerase chain reaction [171]. RANKL
mRNA expression was consistently higher in teeth with FORLs and associ-
ated gingiva. IL-6 mRNA expression seemed to increase in diseased teeth.
Both IL-1 and IL-6 mRNA was expressed at higher levels in diseased gin-
giva, indicating that the gingiva may harbor mechanisms that control clastic
cell activity in the tooth microenvironment [171]. Odontoclasts derive from
blood vessels in the periodontal ligament or alveolar bone. Odontoclasts
and precursors are also found in or close to blood vessels in the inﬂamed
gingiva of permanent feline teeth [19]. Inﬂammatory cells migrate to the
defect, and highly vascular granulation tissue matures [163,164]. It was
reported that 52% and 72% of the resorptive defects were ﬁlled with granu-
lation tissue [22,157]. The edges of the defect are lined regularly with odon-
toclasts [163,164] surrounded by inﬁltrated white blood cells, macrophages,

Fig. 7. Illustration of stage 4 feline odontoclastic resorptive lesion with extensive structural
damage and dentoalveolar ankylosis. (Courtesy of Rebecca Rae Bradford, Champaign, IL.)

813

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

and ﬁbroblasts and penetrated capillaries [19]. Lymphocytes (most often
B-cell type) were a minor population of inﬂammatory cells; macrophages,
plasma cells, and polymorphonuclear granulocytes were the predominant
inﬂammatory cells [57,157]. In addition to these cells, some rounded cells,
likely resting osteoblast-like cells, were present alongside odontoclasts. Odon-
toclasts resorb cementum and dentin, creating lacunae, lagoons, and resorp-
tion canals in the dental hard tissues [11,61,72]. The surface of the substrate
being resorbed seems to be ‘‘etched’’ on scanning electron microscopy [172].

Fig. 8. Illustration of stage 5a feline odontoclastic resorptive lesion with the crown absent and
retained root(s). (Courtesy of Rebecca Rae Bradford, Champaign, IL.)

Fig. 9. Clinical picture of a stage 5a feline odontoclastic resorptive lesion. A small gingival
bulge is visible in the area of a missing left mandibular third premolar (small arrows).

814

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Fig. 10. Dental radiograph of area shown in Figure 9. Although the mesial root remains
somewhat visible (arrows), the distal root of the left mandibular third premolar has undergone
root replacement resorption (*).

Fig. 11. Illustration of a stage 5b feline odontoclastic resorptive lesion with extensive root
replacement resorption and almost intact crown. (Courtesy of Rebecca Rae Bradford,
Champaign, IL.)

815

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Resorption occurs not only of teeth but also of alveolar bone, with the

destruction of the periodontal ligament [19]. The extent of periodontal bone
loss in cats is correlated statistically with the extent of resorptive tooth
destruction [86,139,173]. Others have reported that clastic cell activity at the
edges of alveolar bone is only occasionally observed [157] or far less often
present than with tooth resorption [163]. Studies on chronic adult periodon-
titis have shown that root resorption almost always accompanies periodon-
titis, suggesting that root resorption is associated with inﬂammation.
Although most of the resorptive defects in people did not extend beyond the
cementum, in severe stages of periodontitis (more than 33% alveolar bone
loss), the resorption had spread as far as the dentin [174].

Reparative phase

Most clinically evident FORLs appear histologically to be in resorptive

and reparative phases simultaneously. Seventy percent of resorptive lacunae
in the dentin showed active resorption, whereas 30% showed characteristics
of repair [157]. During the reparative phase, rounded cementoblast- or
osteoblast-like cells produce new hard tissue resembling osteoid, bone,
cementum, bone-cementum, and osteodentin to replace the lost dentin
[10,11,19,29,61,72,157,163,165]. In the human dental literature, repaired

Fig. 12. The crowns of both mandibular canines appear intact. The true extent of disease would
not be apparent without dental radiographs.

816

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

small resorptive lacunae localized in the cementum only have been termed
surface resorptions [122]. Resorption also occurs on the surface of newly
formed reparative tissue [19]. Odontoclasts prefer to attach to the intact den-
tin rather than the newly formed reparative tissue, however. This predispo-
sition to regular dentin was attributed to the diﬀerence in osteocalcin
quantity between dentin and reparative bone-cementum–like tissue. Osteo-
calcin may have chemotactic activity for monocytes, macrophages, and pre-
cursors of clastic cells [19]. The authors of the present article suggest that
odontoclasts may prefer intact dentin because it is more mineralized than
the newly formed reparative tissue. A thickening of the apical cemental
layers resembling hypercementosis was observed in all investigated teeth
in one study [157]. The signiﬁcance of this ﬁnding is unknown.

Pulp response

Pulp involvement is not seen until late in advanced stages of FORLs

[10,72]. It has been reported that odontoblasts in the pulp are arranged
regularly without any pathologic changes but that the layer of predentin
is thin or not present [29] or that only mild pulpitis is present even when

Fig. 13. Dental radiograph of area shown in Figure 12. The roots of both mandibular canines
have undergone extensive root replacement resorption (dotted line depicting original root
contours).

817

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

odontoclasts reach close to the predentin area [19,162]. Once odontoclastic
resorption attacks any area of predentin, however, odontoblasts degenerate
and pulpitis becomes gradually progressive [19]. Odontoblasts may react
with the production of reparative dentin (tertiary dentin), which is less min-
eralized than primary and secondary dentin and has fewer dentinal tubules,
which run in diﬀerent directions. Reparative dentin was only observed in
one histologic study [31,162]. It was found in 53% of aﬀected teeth, whereas
reparative bone-cementum was present in only 12% of teeth with FORLs.
The younger the animal, the more frequent was the presence of reparative
dentin. Sixty-nine percent of stage 2 lesions but only 31% of stage 3 lesions
showed evidence of reparative dentin. Reparative dentin was never found in
stage 1 lesions [31,162].

Adhesion molecules

Speciﬁc antibodies were used to examine adhesion molecules associated

with mineralized tissues, such as bone sialoprotein II and osteopontin (bone
sialoprotein I) as well as a clastic cell surface receptor (a

¼ vitronectin

receptor) linked with these molecules for their localization in FORLs
[165]. It was concluded that aggressive odontoclastic activity is accompa-
nied, at least in part, by increased concentrations of osteopontin and a

at resorption sites. So-called reversal lines (thin cemental layers where
bone-cementum–like tissue was newly formed on resorbed dentin surfaces)
showed bone sialoprotein II and osteopontin antibody reactions. Lack of
collagen expression by odontoblasts indicated that osteoblast-like cells
rather than odontoblasts play an active role in attempts at repair [165].

Diagnosis

Clinical signs

Although cats with FORLs can present with halitosis, dysphagia, ptya-

lism, anorexia, dehydration, weight loss, lethargy, and discomfort [5,
144,175], most aﬀected cats do not show distinct clinical signs. Head shak-
ing, sneezing, and excessive tongue movements have also been observed [46].
Cats may sometimes show spontaneous repetitive jaw motions while eating,
drinking, or grooming. Signs related to oral pain may include dropping food
while eating, refusing to eat hard food, ‘‘hissing’’ and running from the food
bowl when attempting to eat, or other behavioral changes, including aggres-
sion [51,79]. The pain associated with FORLs likely results from exposure of
sensitive nerve endings in the dentinal tubules [5]. As the resorption nears
the pulp, the pain may become more pronounced because of pulpal irrita-
tion [147]. The authors of the present article believe that FORLs are asymp-
tomatic as long as the resorptive process remains below the gingival
attachment (not exposed to oral microorganisms) and does not aﬀect pulp

818

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

tissue. Root resorption in human beings is usually asymptomatic until pul-
pitis develops secondary to a deep lesion or when the resorptive process
emerges at the gingival margin [176].

Oral examination

An examination of the oral cavity in cats can be challenging, because they

often resist orofacial manipulation. In addition, FORLs are often covered
with plaque, calculus, hyperplastic gingiva, and gingival or pulpal granula-
tion tissue [5,79], leading to the conclusion that an accurate oral exami-
nation can be performed only under general anesthesia [177]. Various
degrees of oral inﬂammation (gingivitis, periodontitis, or stomatitis) may be
present with FORLs [6]. Because FORLs are often associated with perio-
dontal inﬂammation, they have been included in the overall category of
periodontal disease in cats and may present a unique pathologic feature
of feline periodontitis [19].

Clinically, FORLs are most likely detected at the gingival margin near

the cementoenamel junction, [25,157] and in multirooted teeth, particularly
in the furcation area (Fig. 14) [19]. After cleaning the teeth, a dental explorer
is gently placed into the gingival sulcus, and each tooth is probed for any
surface irregularities [147]. FORLs are identiﬁed as rough deﬁcits of the
tooth structure, and the dental explorer does not ‘‘stick’’ (as it does in tooth
substance softened by caries) [10,25,178]. Inﬂamed granulation tissue bleeds
readily when examined with a dental explorer (Fig. 15) [5,177]. Furcation
exposure resulting from alveolar bone loss and the cementoenamel junction
apical to the cervical bulge are sometimes confused with FORLs [146]. In

Fig. 14. Multiple feline odontoclastic resorptive lesions of premolars and molars (small arrows),
missing teeth (large arrows), and supereruption phenomenon on maxillary and mandibular
canines (dotted line depicting cementoenamel junction).

819

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

advanced stages of a FORL, the lesion can penetrate into the pulp, fol-
lowed by fracture and loss of the crown with radiographically visible root
remnants [173]. Occasionally, a small gingival stoma (ﬁstula) can be
observed over the site of a retained root remnant [51]. When the dental
explorer is moved across or close to a lesion even under general anesthesia,
the lower jaw may ‘‘chatter’’ or twitch [6,10,79,179]. A longitudinal study
found no correlation between repetitive jaw motions and FORLs or gingival
inﬂammation, however [27]. Ten percent of cats with chronic gingivitis-
stomatitis showed repetitive jaw motions without any clinically or radiograph-
ically evident FORLs [180]. Innumerable studies have been performed on
pain mechanisms in experimental animals, particularly in cats. In one study
on anesthetized cats, reﬂex responses of the digastric and tongue muscles to
stimulation of the intact tooth crown, exposed dentin, and pulp were
recorded by electromyography. The results indicated that activation of both
A- and C-type pulp nerve ﬁbers induces muscle activation, resulting in
evoked licking movements of the tongue and jaw-opening reﬂexes [181].

Some cats seem to exhibit gingival recession around their canines, or their

canines seem to extrude abnormally [177,179]. This ‘‘supereruption’’ may be
the result of odontoclastic root resorption and inﬂammation [177] and

Fig. 15. (A) Feline odontoclastic resorptive lesion of the right mandibular ﬁrst molar (small
arrows). (B, C) A dental explorer is gently inserted into the lesion. Bleeding occurs on probing
the inﬂamed granulation tissue.

820

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

osteoblastic activity in the periapical alveolar bone area [179], leading to
excessive exposure of the tooth root surface. Painful thickening of alveolar
bone and local osteomyelitis are often associated with this phenomenon.

Radiographic ﬁndings

FORLs can develop anywhere along the root cementum and not just at

the cementoenamel junction close to the gingival margin [26]. In addition,
FORLs that appear small clinically may be extensive within the tooth struc-
ture or have advanced root resorption. Radiographs are necessary to deter-
mine whether the resorptive defect has penetrated into the pulp or if
periapical lucencies are present, which are important considerations in
restorative treatment [2]. Radiographs also provide helpful information
when extraction is the chosen treatment with respect to root remnants, den-
toalveolar ankylosis, or root replacement resorption. On clinical examina-
tion, 27% of FORLs were conﬁned to or above the gingival margin with
no pulp exposure compared with 12.5% based on dental radiographs
[133]. In a necropsy study, cats revealed 1.4 times more crown lesions and
2.4 times more root and alveolar bone lesions detected by dental radiogra-
phy than by clinical examination [26]. Full-mouth dental radiographs are
helpful in the diagnosis of FORLs. In one study, FORLs were not detected
on clinical examination but were subsequently diagnosed radiographically in
8.7% of cats [136].

Radiographic techniques are not easily performed in cats. The use of an

intraoral projection technique for the maxillary premolars either causes
overlap of dental structures with the zygomatic arch or intentional tooth
elongation. The extraoral near-lateral view of the maxillary premolars was
found to be easy to master, and superimposition of the zygomatic arch infre-
quently occurred [136]. FORLs along the buccal or lingual aspect of the
tooth may not be distinctly visible on a ﬁlm taken using a parallel technique,
but lesions on mesial or distal tooth surfaces or in the furcation area are
obvious [173]. Artifacts from cervical burnout should not be confused with
or interpreted as FORLs at the cementoenamel junction. Cervical burnout,
or ‘‘adumbration,’’ results from the diﬀerent densities at the neck of the
tooth, where the cervical region of the crown and cervical region of the root
(covered by alveolar bone) meet [182]. Early lesions are sometimes diﬃcult
to see radiographically but may appear as minute radiolucent defects [28].
Clinically evident FORLs are generally visible as notched radiolucencies
with sharp or scalloped margins, often found at the level of the cementoena-
mel junction [2]. ‘‘Dentoalveolar ankylosis’’ describes the condition of anky-
lotic fusion between bone and the root cementum surface or between bone
and shallow resorption lacunae, making extraction by conventional means
challenging. ‘‘Replacement resorption’’ is the process whereby the root den-
tin is replaced by bone. Dentoalveolar ankylosis is thought to be a delayed
form of replacement resorption in which the cementum layer signiﬁcantly

821

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

slows the resorption and replacement of the dentin with bone. In both cases,
the lamina dura disappears [182]. Invasive FORLs are more diﬀuse and give
the tooth a moth-eaten or striated appearance as is visible particularly in
canine teeth [173]. When replacement resorption occurs, the tooth structure
is gradually replaced by bone, and the contour of the root becomes irregular
or disappears (ghost roots). Retained root fragments may become seques-
trated or cause osteomyelitis and mandibular swelling [183,184].

Comparison with other defects of the dental hard substances

FORLs are often confused with hard tissue defects that are not resorptive

in nature [44]. ‘‘Attrition’’ is deﬁned as normal or excessive loss of tooth
substance caused by tooth-by-tooth frictional contact, usually as a result
of mastication or malocclusion [185]. ‘‘Abrasion’’ is deﬁned as tooth wear
caused by frictional contact of a tooth with a nondental material [185], such
as that found in teeth in carnivores (common in dogs with untreated pruritic
dermatoses; excessive chewing on hard toys, bones, and rocks; cage-biter
teeth of dogs and zoo carnivores). ‘‘Erosions’’ or ‘‘erosive lesions’’ are
deﬁned as loss of tooth substance caused by a chemical process without the
activity of bacteria [185]. The terms corrosion and stress corrosion have also
been proposed, because the term erosion in (natural) science terminology
only describes the result of mechanical forces, whereas corrosion is the result
of chemical or electrochemical inﬂuences [186]. Tooth lesions caused by
acidic damage are far more frequently seen in human beings than in pet ani-
mals, although there are some reports of erosive lesions of permanent teeth
in farm animals fed excessively acidic silage and supplements [187]. Caries is
a microbial disease characterized by demineralization of the inorganic
portion and destruction of the organic substance of the tooth secondary
to sugar fermentation by plaque bacteria [188], especially Streptococcus
mutans in human beings. True tooth caries has never been proven in cats
and is an uncommon problem in dogs, probably because of their lower car-
bohydrate diet, higher salivary pH, paucity of true occlusal anatomy, and
decreased food stagnation between teeth compared with human beings [64].

Tooth resorptions in nonfelidae

There are several reports describing canine odontoclastic resorptive

lesions in domestic dogs [1,189–202]. In one study, root resorption was
reported in 17.9% of randomly selected dogs [203]. In the authors’ own expe-
rience, the prevalence of canine odontoclastic resorptive lesions is low. Re-
sorptions of permanent teeth have been reported in the dolphin, Montana
black bear, and Rocky Mountain elk. Most of these lesions in captured
or domesticated mammals occurred in root cementum but were not similar
clinically to FORLs [69]. Tooth resorptions have also been described in

822

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

chinchillas [204] and in numerous studies on teeth in laboratory animals,
including rodents, lagomorphs, and monkeys. Resorption in the human den-
tal literature is classiﬁed on the basis of the site of origin and is referred to as
internal or external and as apical, lateral, or cervical root resorption [122].
The major local causative factors of root resorption in human beings are
deﬁned as excessive pressure from orthodontic tooth movement, impacted
teeth, tumors, or cysts and inﬂammation usually in response to endodontic
disease, tooth luxation and avulsion, and injury to the periodontal ligament
and cementum [4].

Treatment

The ultimate goal of any treatment option is to provide the cat with

a healthy pain-free mouth [177]. Determining the underlying etiology and
predisposing factors for FORLs is essential for making future recom-
mendations on preventative measures and successful treatment options.
Historically, a common approach for the treatment of FORLs included ﬂuo-
ride treatment of stage 1 lesions, restoration of stage 2 lesions, and extrac-
tion of teeth graded stage 3 or higher [15]. Fluoride treatment remains
controversial. Although it has commonly been recommended for prevention
of FORLs or treatment of stage 1 lesions, ﬂuoride has never been proven to
prevent or slow tooth resorption in cats. Fluoride has anticariogenic proper-
ties (increases the microhardness of enamel and dentin, desensitizes the
tooth, and inhibits plaque formation). It has also been found to possess
inhibiting eﬀects on the activity of isolated osteoclasts in vitro [205]. In
another study, the administration of ﬂuoride suppressed root resorption
induced by mechanical injuries of the periodontal soft tissues in rats [206].
If ﬂuoride treatment is to be used for stage 1 lesions, a routine prophylaxis
is performed, and the lesion is root-planed, polished, rinsed, and dried to
prepare the tooth for the application of a ﬂuoride varnish, ﬂuoride-releasing
dentinal bonding agent, or ﬂuoride-releasing pit-and-ﬁssure sealant [147].

Restoration

Stage 1 lesions are too small to restore. The restorations that were com-

monly endorsed in the past for stage 2 lesions and after endodontic treat-
ment for stage 3 lesions [160,168,207–217] are now rarely placed because
of poor long-term success rates. Possible causes of restorative failure are the
delicate nature of feline teeth, poor cavity preparation, improper classiﬁca-
tion of the lesion, use of inappropriate restorative materials, and lack of
home care and professional follow-up [145]. An even more important factor
in restorative failure may be that treatment of lesions at the cervical third of
the root and gingival third of the crown is technically diﬃcult. One study
revealed that only 4% of FORLs were localized entirely in a suprabony
location and that 8.5% of FORLs were conﬁned to the crestal bone area,

823

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

concluding that just 12.5% of FORLs are accessible for restorative treat-
ment without the need for extensive periodontal surgery [133].

In one American study of 58 cats, 33% of the glass ionomer ﬁllings in

stage 2 lesions were still present with no further evidence of resorption after
a mean period of 15 months [179]. In another American study, the glass ion-
omer restoration was in place, and no additional resorption was evident in
18% of treated teeth at 48 months (Bob Wiggs, DVM, Dallas, TX, unpub-
lished data, 1993). An Austrian study using composite restorations in 19
cats showed that only 10% of restored teeth at 24 months still had an intact
restoration and showed neither macroscopic nor radiographic progression
of the resorptive process [218]. A German thesis study in which 69 teeth were
treated with glass ionomer ﬁllings found only 20.3% of restored teeth with
no further macroscopic or radiographic enlargement of the original defect
after 9 to 15 months [140]. Another German study investigated long-term
results after ﬁlling 59 stage 2 and stage 3 lesions in 21 cats with a compomer
(combination of composite and glass ionomer); 32% of restored teeth at 18
months still had an intact restoration and showed no macroscopic progres-
sion of the resorptive process [17]. The results of this most recent study must
be interpreted with caution, because dental radiographs were taken in only 1
of 21 cats. Based on the results of these long-term studies (Table 4), restora-
tion as a treatment for FORLs cannot currently be justiﬁed.

Extraction

The progressive nature of FORLs combined with an unknown etiology

makes complete extraction of aﬀected teeth the most acceptable treatment

Table 4
Long-term results of restoration of teeth with feline odontoclastic resorptive lesions in domestic
cats

Percentage of success after

6–48 months

Year
reported

Restorative
material

Radiographs 6

9 12 15 18 24 30 36 48

Lyon [179]

1992

Glass

ionomer

Yes

Wiggs

[unpublished]

1993

Glass

ionomer

Yes

44 33 27 19 18

Zetner and

Steurer [218]

1995

Composite

Yes

Roes [140]

1996

Glass

ionomer

Yes

Schweighart-

Banzhaf and
Benz [17]

1997

Compomer

Percentage of success after a period of 9 to 15 months.

Radiographs of teeth with restorations taken in 1 of 21 cats.

824

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

option at this point in time [5,219,220]. Multirooted teeth should be sec-
tioned before extraction, and each section should be treated as a single-
rooted tooth extraction [177,221]. Extracting teeth with advanced FORLs
can be diﬃcult, because the teeth are brittle and break easily. It is also more
diﬃcult to extract roots once the crown has been fractured oﬀ [18]. Dentoal-
veolar ankylosis and root replacement resorption make diﬀerentiation
between root remnants and alveolar bone impossible, greatly complicating
complete extraction of all roots. This is especially true for stage 5a lesions,
which may be evident only on dental radiographs. At some point, the tooth
root becomes incorporated into the normal remodeling process of the alveo-
lus and is gradually replaced by bone [123]. Root remnants under intact gin-
giva and without periapical pathologic ﬁndings on dental radiographs may
be left where they are [147,222]. Often, they appear as a small gingival bulge
in the area of a missing tooth. Others recommend extraction of any root
remnants, because they may present a permanent inﬂammatory irritation
[86,223]. Persistent inﬂammatory responses associated with retained roots
can result in chronic oral inﬂammation, alveolar osteitis, bony sequestra,
and varying forms of osteomyelitis [184], which also supports complete
extraction of these teeth. If closed extraction is not possible, or if root rem-
nants need to be removed, a mucoperiosteal ﬂap is made to allow access to
the area. On completed extraction, the alveolus is ﬂushed with chlorhexidine
or sterile saline solution and sutured closed with absorbable suture material.
The authors of the present article decline routine administration of antibiot-
ics in the otherwise healthy dental patient.

‘‘Root pulverization (atomization)’’ is sometimes performed to remove

brittle or ankylosed root remnants [51,115,147,177,224]. A water-cooled
high-speed dental handpiece equipped with a small round burr is used to
touch retained root(s) several times. A chirping noise indicates that the root
remnants are crushed into many particles and washed out from the alveolar
socket [224]. Serious complications can occur with this technique, including
incomplete removal of the tooth root, alveolar bone damage, injury to infe-
rior alveolar and infraorbital neurovascular bundles, subcutaneous and sub-
lingual emphysema, air emboli, salivary extravasation syndrome by injuring
sublingual tissues, and transportation of root remnants into the mandibular
canal or nasal cavity [174,179,238].

Crown amputation with intentional root retention

The observation that root fragments from teeth undergoing resorption

often continue to resorb and resolve without evidence of discomfort, gingi-
val inﬂammation, or ﬁstulation [225] led to the idea of amputating the
crown in cats with FORLs and leaving the roots to be resorbed. This tech-
nique should only be used in cases that meet the following criteria: (1) an
absence of periodontal disease (no periodontal pockets on probing or ab-
normal mobility), (2) no radiographic evidence of endodontic disease or

825

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

periapical pathologic ﬁndings, (3) no clinical evidence of stomatitis, and (4)
patients that test negative for FIV and FeLV [226]. Otherwise, the entire
root structure must be extracted. After making a simple envelope ﬂap, a
round burr in a water-cooled high-speed handpiece is used to remove the
crown of the tooth to or slightly below the level of the alveolar crest. The
gingiva is sutured closed across the top of the remaining root(s). The study
initially proposing crown amputation with intentional root retention as
a treatment option for stage 2 through 5 lesions involved radiographic fol-
low-up of 51 roots for 5 to 36 months after the procedure [226]. Results
showed continued tooth resorption without surrounding alveolar bone reac-
tion in all but two cases. One cat had normal periodontal ligament remain-
ing 1 year after treatment, and the other subsequently developed stomatitis
and had all retained roots extracted [226]. A recent study that reviewed full-
mouth dental radiographs of 265 cats found that pulp involvement caused
by FORLs does not seem to be associated with radiographic evidence of
periapical lucencies [227]. Another study did not ﬁnd any association
between mandibular thickening and FORLs [60]. A 1-year prospective study
from a large institution found only 24 cats with mandibular swellings; of
those with benign lesions, the most frequent cause of osteomyelitis was
reported to be ‘‘retained tooth roots and external root resorptions’’ [184].
The results of all these studies suggest that except for the rare cases where
a FORL is associated with infection, crown amputation with intentional
root retention may be a suitable alternative to extraction in selected cats
with FORLs [227].

Laser

The poor success rates of restoring FORLs prompted a study in which

stage 1 and 2a lesions were treated with a neomydium:YAG laser, enamelo-
plasty, and gingivoplasty [145]. After 5 years, 5.6% of treated teeth devel-
oped FORLs around the edges of the lased lesions, 15.5% developed
FORLs at other areas on the tooth in question, and 78.9% had no further
lesion development [145]. Dental radiography and histopathologic examina-
tion are needed to substantiate a true absence of FORL progression and to
conﬁrm these promising results. Other recent studies using the same laser on
feline teeth warn of potential nerve injury and irreversible tissue damage in
the pulp [228,229].

Possible topical and systemic medical therapies

One author recommended supportive homeopathic treatment of stage 3

and 4 FORLs with a combination of galium-heel and china-homaccord (and
traumeel ad us. vet.) and treatment of stage 1 and 2 FORLs with calcium
ﬂuoratum-injeel/calcium carbonicum-injeel and engytol ad us. vet [230]. An
antibiotic/corticosteroid paste was found to be eﬀective for use in the
treatment of progressive root resorption in traumatically injured teeth in

826

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

monkeys [231]. Systemic tetracycline administration also signiﬁcantly
reduced root resorption of dried replanted teeth in dogs [232]. Improved
healing was seen in dogs whose teeth were topically treated with alendronate
before replantation [233]. The topical administration of risedronate was also
found to be useful in preventing root resorption during orthodontic tooth
movement in rats [234]. Another bisphosphonate used for treatment of post-
menopausal osteoporosis and gallium nitrate, which is used for treatment of
hypercalcemia of malignancy, have signiﬁcantly reduced root resorption in
vitro [235]. In vivo, however, bisphosphonate injections actually induced
root resorption in mice and rats [236]. Calcium hydroxide and calcitonin
were shown to be eﬀective medications for the intra-canal treatment of
inﬂammatory root resorption in dogs and monkeys [239,240].

Conclusion

Approximately one third of all domestic cats may develop FORLs during

their life span, and the risk of developing FORLs increases with age. Extrac-
tion is the current treatment of choice. The etiology of FORLs remains
unknown. Finding the causative factors that alter the resorption-inhibiting
characteristics of the outer tooth surface may be the clue to the enigma of
FORLs.

Acknowledgment

The authors thank Colin E. Harvey, BVSc, FRCVS, and Sandra Manfra

Marretta, DVM, for their time, eﬀort, and contribution in proofreading this
article.

References

[1] Arnbjerg J. Idiopathic dental root replacement resorption in old dogs. J Vet Dent 1996;

13:97–9.

[2] Bellows J. Radiographic signs and diagnosis of dental disease. Semin Vet Med Surg

1993;8:138–45.

[3] van Foreest A, van Zuilen D. Veterinaire tandheelkunde (10). Tandhalslaesies bij katten.

Etiologie–Klinische diagnostiek—Therapie. Tijdschr Diergeneeskd 1995;120:356–60.

[4] Ne RF, Witherspoon DE, Gutmann JL. Tooth resorption. Quintessence Int 1999;30:9–25.
[5] Reiter AM. Feline odontoclastic resorptive lesions (FORL): a review of literature. In:

Proceedings of the 11th Annual Veterinary Dental Forum. Denver, CO, 1997, p. 56–61.

[6] Harvey CE. Feline dental resorptive lesions. Semin Vet Med Surg 1993;8:187–96.
[7] Harvey CE. Inﬂammatory oral diseases of the cat. In: Harvey CE, Orr HS, editors.

Manual of small animal dentistry. Gloucestershire: British Small Animal Veterinary
Association; 1990. p. 49–54.

[8] van Zuilen D. Literatuuronderzoek: tandhalslaesies bij katten. Fakulteit Diergeneeskunde,

Vakgroep Geneeskunde van Gezelschapsdieren. Utrecht: Rijksuniversiteit Utrecht; 1993.

[9] Lyon KF. Tooth substance resorption: diagnosis and management. In: Proceedings of the

Annual Meeting of the American Veterinary Dental Society. Boston (MA); 1994. p. 20–3.

827

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

[10] Schlup D, Stich H. Epidemiologische und morphologische Untersuchungen am

Katzengebib. II. Mitteilung: Morphologische Untersuchungen der ‘‘neck lesions’’.
Kleintierpraxis 1982;27:179–88.

[11] Berger M, Schawalder P, Stich H, et al. ‘‘Neck Lesions’’ bei Grosskatzen: Untersuchun-

gen beim Leoparden (Panthera pardus). Kleintierpraxis 1995;40:537–49.

[12] Fahrenkrug P. Krankheiten der Za¨hne und des Kiefers. In: Kraft W, Du¨rr UM, editors.

Katzenkrankheiten. Klinik und Therapie. 4th edition. Hannover: Verlag Schaper; 1996.
p. 439–66.

[13] Floyd M. A new, easy method for feline ‘‘necklesion’’ repair and tooth restoration. In:

Proceedings of the 64th Annual Meeting of the American Animal Hospital Association.
San Diego (CA); 1997. p. 135–8.

[14] Legendre L. Cervical line lesions: an update. Can Vet J 1996;37:183–4.
[15] Mulligan TW. Feline cervical line lesions. Vet Med Rep 1990;2:343–9.
[16] Reichart P. Krankheiten der Za¨hne und des Kiefers. In: Kraft W, Du¨rr UM, editors.

Katzenkrankheiten. Klinik und therapie. 3rd edition. Hannover: Verlag Schaper; 1991.
p. 276–87.

[17] Schweighart-Banzhaf D, Benz C. ‘‘Resorptive lesions’’ bei der Katze. Eine klinische

Untersuchung zur Versorgung mit ‘‘Dyract.’’ Kleintierpraxis 1997;42:97–108.

[18] Okuda A, Asari M, Harvey CE. Challenges in treatment of external odontoclastic

resorptive lesions in cats. Compend Contin Educ Pract Vet 1995;17:1461–9.

[19] Okuda A, Harvey CE. Etiopathogenesis of feline dental resorptive lesions. Vet Clin North

Am Small Anim Pract 1992;22:1385–404.

[20] Reiter AM. Biology of alveolar bone and tooth resorbing cells. In: Proceedings of the 12th

Annual Veterinary Dental Forum. New Orleans, LA, 1998, p. 225–7.

[21] Okuda A, Harvey CE. Immunohistochemical distributions of interleukins as possible

stimulators of odontoclastic resorption activity in feline dental resorptive lesions.
In: Proceedings of the Sixth Annual Veterinary Dental Forum. Las Vegas (NV); 1992.
p. 41–3.

[22] Dobrescu MV. Nekrobazillose der Za¨hne bei der Katze. Tiera¨rztliche Umschav 1994;49:

498–501.

[23] Sims TJ, Moncla BJ, Page RC. Serum antibody response to antigens of oral gram-

negative bacteria in cats with plasma cell gingivitis-stomatitis. J Dent Res 1990;69:877–82.

[24] Harvey CE, Hammond BF, Whitaker EJ. Is there an association between dental

resorptive lesions and Actinobacillus actinomycetemcomitans in cats? In: Proceedings of
the Eighth Annual Veterinary Dental Forum. Philadelphia, PA, 1994, p. 48.

[25] Schneck GW, Osborn JW. Neck lesions in the teeth of cats [abstract]. Vet Rec

1976;99:100.

[26] Gengler W, Dubielzig R, Ramer J. Physical examination and radiographic analysis to

detect dental and mandibular bone resorption in cats: a study of 81 cases from necropsy.
J Vet Dent 1995;12:97–100.

[27] Harvey CE, Anderson JG, Miller B. Longitudinal study of periodontal health in cats. Part

2—radiographic examination and correlation with clinical data. In: Proceedings of the
Fifth European Congress of Veterinary Dentistry. Bordeaux, France, 1996, p. 1–2.

[28] Williams CA, Aller MS. Gingivitis/stomatitis in cats. Vet Clin North Am Small Anim

Pract 1992;22:1361–83.

[29] Okuda A, Harvey CE. Histopathological ﬁndings of features of odontoclastic resorptive

lesions in cat teeth with periodontitis. In: Proceedings of the Fifth Annual Veterinary
Dental Forum. New Orleans, LA, 1991. p. 141–4.

[30] Moss-Salentijn L, Moss ML. Studies on dentin 2—transient vasodentin in the incisor

teeth of a rodent (Perognathus longimembris). Acta Anat 1975;91:386–404.

[31] Lukman K, Pavlica Z, Juntes P. Prevalence patterns and histological survey of feline

dental resorptive lesions. In: Proceedings of the Eighth Annual Scientiﬁc Meeting of the
British Veterinary Dental Association. Birmingham, UK, 1996.

828

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

[32] Crossley DA. Tooth enamel thickness in the mature dentition of domestic dogs and

cats—preliminary study. J Vet Dent 1995;12:111–3.

[33] Hayashi K, Kiba H. Microhardness of enamel and dentine of cat premolar teeth. Jpn

J Vet Sci 1989;51:1033–5.

[34] Gauthier O, Rennou M, Pilet P, et al. Comparative study of the mineral composition of

dentin and enamel in canine, feline and human teeth. In: Proceedings of the 10th
European Congress of Veterinary Dentistry. Berlin, Germany, 2001, p. 32–3.

[35] Harvey CE, Emily PP. Endodontics. Small animal dentistry. St. Louis: Mosby; 1993.

p. 156–212.

[36] Orsini P, Hennet P. Anatomy of the mouth and teeth of the cat. Vet Clin North Am Small

Anim Pract 1992;22:1265–77.

[37] Winter GB, Kramer IRH. Changes in periodontal membrane and bone following

experimental pulpal injury in deciduous molar teeth in kittens. Arch Oral Biol 1965;
10:279–89.

[38] Negro VB, Hernandez SZ, Maresca BM. Endodontic system of the carnassial teeth in

cats. In: Proceedings of the Ninth European Congress of Veterinary Dentistry. Copen-
hagen, Denmark, 2000, p. 25–6.

[39] Gorrel C, Larsson A, Ingham K, et al. The histological appearance of normal feline root

cementum. In: Proceedings of the 15th Annual Veterinary Dental Forum. San Antonio,
TX, 2001, p. 365–367.

[40] Schroeder HE, Scherle WF. Cemento-enamel junction—revisited. J Periodontal Res 1988;

23:53–9.

[41] Tal H, Stahl SS. Periodontal attachment responses to surgical injury in the cat. J Clin

Periodontol 1986;13:45–51.

[42] Lee WC, Eakle WS. Possible role of tensile stress in the etiology of cervical erosive lesions

of teeth. J Prosthet Dent 1984;52:374–80.

[43] Lee WC, Eakle WS. Stress-induced cervical lesions: review of advances in the past 10

years. J Prosthet Dent 1996;75:487–94.

[44] Burke FJT, Johnston N, Wiggs RB, et al. An alternative hypothesis from veterinary science

for the pathogenesis of noncarious cervical lesions. Quintessence Int 2000;31:475–82.

[45] Johnston N. Acquired feline oral cavity disease. Part 2: feline odontoclastic resorptive

lesions. In Practice 2000;22:188–97.

[46] Chapman JK. Sneezing in a cat secondary to caries lesion in the left canine tooth. J Vet

Dent 1987;4:3–4.

[47] Clarke DE, Cameron A. Feline dental resorptive lesions in domestic and feral cats and the

possible link with diet. In: Proceedings of the Fifth World Veterinary Dental Congress.
Birmingham, UK, 1997, p. 33–4.

[48] Donoghue S, Scarlett JM, Williams CA, et al. Diet as a risk factor for feline external

odontoclastic resorption. J Nutr 1994;124 (Suppl):2693S–4S.

[49] Lund EM, Bohacek LK, Dahlke JL, et al. Prevalence and risk factors for odontoclastic

resorptive lesions in cats. JAVMA 1998;212:392–5.

[50] Scarlett JM, Saidla J, Hess J. Risk factors of odontoclastic resorptive lesions in cats. J Am

Anim Hosp Assoc 1999;35:188–92.

[51] Holmstrom SE. External osteoclastic resorptive lesions. Feline Pract 1992;20:7–11.
[52] Mulligan TW. Restorations for feline cervical line lesions: glass ionomers. In: Proceedings

of the Second Annual Veterinary Dental Forum. New Orleans, LA, 1988, p. 6–12.

[53] Anderson JG, Harvey CE, Flax B. Clinical and radiographic evaluation of external

odontoclastic resorptive lesions in cats. In: Proceedings of the 11th Annual American
College of Veterinary Internal Medicine Forum. Washington, DC, 1993, p. 947.

[54] Zetner K, Steurer I. The inﬂuence of dry food on the development of feline neck lesions.

J Vet Dent 1992;9:4–6.

[55] Zetner K, Steurer I. Pathogenesis and treatment of neck lesions. In: Proceedings of the

First European Congress of Veterinary Dentistry. Rome, Italy, 1992.

829

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

[56] Dobbertin F. Zur Pathologie der Zahn- und Zahnbetterkrankungen bei Felis silvestris

forma catus [dissertation]. Hamburg, Germany: Universita¨t Hamburg; 1993.

[57] Hopewell-Smith A. The process of osteolysis and odontolysis, or so-called ‘‘absorption’’

of calciﬁed tissues: a new and original investigation. Dental Cosmos 1930;72:1036–48.

[58] Levin J, Emily E, Frost P. The prevalence of odontoclastic resorptive lesions in feral,

domestic, and exotic felids. In: Proceedings of the Eighth Annual Scientiﬁc Meeting of the
British Veterinary Dental Association. Birmingham, UK, 1996.

[59] Lu¨ps P, Gebib- und Zahnvariationen an einer Serie von 257 Hauskatzen (Felis silvestris f.

catus L., 1758). Zoologische Abhandlungen Staatliches Museum fu¨r Tierkunde in
Dresden 1976;34:155–65.

[60] Verstraete FJM, van Aarde RJ, Nieuwoudt BA, et al. The dental pathology of feral cats

on Marion Island, part II: periodontitis, external odontoclastic resorption lesions and
mandibular thickening. J Comp Pathol 1996;115:283–97.

[61] Berger M, Schawalder P, Stich H, et al. Feline dental resorptive lesions in captive and wild

leopards and lions. J Vet Dent 1996;13:13–21.

[62] Cook RA, Stoller NH. Periodontal status in snow leopards. JAVMA 1986;189:1082–3.
[63] Gioso MA. Prevalence of oral lesions in Puma concolor and Panthera onca in the state of

Sao Paulo. In: Proceedings of the 15th Annual Veterinary Dental Forum. San Antonio,
TX, 2001, p. 200.

[64] Kertesz P. Dental diseases and their treatment in captive wild animals. In: A colour atlas of

veterinary dentistry and oral surgery. Aylesbury, UK: Wolfe Publishing; 1993. p. 215–81.

[65] Kertesz P. Large carnivore dental surgery. In: Proceedings of the Fifth World Veterinary

Dental Congress. Birmingham, UK, 1997, p. 192–4.

[66] Levin J. Tooth resorption in a Siberian tiger. In: Proceedings of the 10th Annual

Veterinary Dental Forum. Houston, TX, 1996, p. 212–4.

[67] Lobprise HB. Exotic animal dentistry. In: Proceedings of the Seventh Annual Veterinary

Dental Forum. Auburn, AL, 1993, p. 39–41.

[68] Mendoza KA, Manfra Marretta S, Klippert LS. Odontoclastic resorptive lesion of a

mandibular right ﬁrst molar in a cougar. J Vet Dent 2000;17:173–6.

[69] Myrik AC. Is tissue resorption and replacement in permanent teeth of mammals caused by

stress-induced hypocalcemia? In: Davidovitch Z, editor. The biological mechanisms
of tooth eruption and root resorption. Birmingham (AL): EBSCO Media; 1988. p. 379–89.

[70] Scheels JL. Periodontal disease associated with ‘‘feline cervical erosion’’ resulting in

oronasal ﬁstula and osteomyelitis in a Siberian tiger. J Vet Dent 1989;6:24.

[71] Wiggs RB, Lobprise HB. Exotic animal oral disease and dentistry. In: Veterinary

dentistry principles and practice. Philadelphia: Lippincott-Raven; 1997. p. 538–58.

[72] Berger M, Schawalder P, Stich H, et al. Diﬀerentialdiagnose von resorptiven

Zahnerkrankungen (FORL) und Karies. Schweiz Arch Tierheilkd 1996;138:546–51.

[73] Kertesz P. Periodontal disease and treatment. In: A colour atlas of veterinary dentistry

and oral surgery. Aylesbury, UK: Wolfe Publishing; 1993. p. 73–99.

[74] Studer E, Stapley RB. The role of dry foods in maintaining healthy teeth and gums in the

cat. Vet Med Small Anim Clin 1973;68:1124–6.

[75] Logan EI, Boyce EN, Berg M, et al. Eﬀects of dietary form on plaque and calculus

accumulation and gingival health in cats—methodology and results. In: Proceedings of
the Fifth World Veterinary Dental Congress. Birmingham, UK, 1997. p. 28–32.

[76] Boyce EN. Feline experimental models for control of periodontal disease. Vet Clin North

Am Small Anim Pract 1992;22:1309–21.

[77] Harvey CE. Feline oral pathology, diagnosis and management. In: Crossley DA, Penman

S, editors. Manual of small animal dentistry. 2nd edition. Gloucestershire: British Small
Animal Veterinary Association; 1995. p. 129–38.

[78] Harvey CE. Are dental resorptive lesions in cats caused by dental plaque-induced gingival

inﬂammation? In: Proceedings of the 10th Annual Veterinary Dental Forum. Houston,
TX, 1996, p. 121–3.

830

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

[79] Frost P, Williams CA. Feline dental disease. Vet Clin North Am Small Anim Pract

1986;16:851–73.

[80] Schlup D. Epidemiologische und morphologische Untersuchungen am Katzengebib.

I. Mitteilung: Epidemiologische Untersuchungen. Kleintierpraxis 1982;27:86–94.

[81] Gialamas J. Zur deformierenden cervikalen Spondylose in Verbindung mit Vitamin-A-

Hypervitaminose bei Katzen. Zentralbl Veterinarmed A 1977;24:160–76.

[82] Seawright AA, English PB. Hypervitaminosis A and deforming cervical spondylosis of the

cat. J Comp Pathol 1967;77:29–39.

[83] Seawright AA, Hrdlicka J, Pathogenetic factors in tooth loss in young cats on a high daily

oral intake of vitamin A. Aust Vet J 1974;50:133–41.

[84] Seawright AA, Hrdlicka J, Severe retardation of growth with retention and displacement

of incisors in young cats fed a diet of raw sheep liver high in vitamin A. Aust Vet J
1974;50:306–15.

[85] Seawright AA, Steele DP, Clark L. Hypervitaminosis A of cats in Brisbane. Aust Vet J

1968;44:203–6.

[86] Zetner K. Neck lesions bei der Katze. Diagnostisch-a¨tiologische Untersuchungen u¨ber

Zusammenha¨nge zwischen Ro¨ntgenbefund und Fu¨tterung. Waltham Report 1990;30:
15–23.

[87] Crager CS, Nachreiner RF. Increased parathyroid hormone concentration in a Siamese

kitten with nutritional secondary hyperparathyroidism. J Am Anim Hosp Assoc
1993;29:331–6.

[88] Bray NC. Nutritional secondary hyperparathyroidism in a kitten. Feline Pract

1984;14:31–5.

[89] Ghergariu S, Greblea A, Giuglea M, et al. Periodontopathie carencielle chez les chiens.

Zentralbl Veterinarmed A 1975;22:696–703.

[90] Kene ROC. Nutritional secondary hyperparathyroidism in the dog and cat. Trop Vet

1992;10:99–107.

[91] Krook L, Barrett RB, Usui K, et al. Nutritional secondary hyperparathyroidism in the

cat. Cornell Vet 1963;53:224–40.

[92] Pedersen NC. Nutritional secondary hyperparathyroidism in a cattery associated with the

feeding of a fad diet containing horsemeat. Feline Pract 1983;13:19–26.

[93] Saville PD, Krook L, Gustafsson P, et al. Nutritional secondary hyperparathyroidism in a

dog. Cornell Vet 1969;59:155–67.

[94] Scott PP. Nutritional secondary hyperparathyroidism in the cat. Vet Med Small Anim

Clin 1967;62:42–4.

[95] Slusher R, Bistner SJ, Kircher C. Nutritional secondary hyperparathyroidism in a tiger.

JAVMA 1965;147:1109–15.

[96] Svanberg G, Lindhe J, Hugoson A, et al. Eﬀect of nutritional hyperparathyroidism on

experimental periodontitis in the dog. Scand J Dent Res 1973;81:155–62.

[97] Tomsa K, Glaus T, Hauser B, et al. Nutritional secondary hyperparathyroidism in six

cats. J Small Anim Pract 1999;40:533–9.

[98] Voorhout G. Rontgenologische bevindingen bij alimentaire secundaire hyperparathyr-

eoidie bij de opgroeiende hond en kat. Tijdschr Diergeneeskd 1981;106:317–24.

[99] Henrikson P.-A. Periodontal disease and calcium deﬁciency. An experimental study in the

dog. Acta Odontol Scand 1968;26 (Suppl 50):1–132.

[100] Reiter AM. Evaluation of serum levels of calciotropic hormones in cats with feline

odontoclastic resorptive lesions (FORL)—preliminary results of an ongoing study.
In: Proceedings of the 12th Annual Veterinary Dental Forum. New Orleans, LA, 1998,
p. 3–4.

[101] Gold SI, Hasselgren G. Peripheral inﬂammatory root resorption. A review of the

literature with case reports. J Clin Periodontol 1992;19:523–34.

[102] Berger B, Feldman EC. Primary hyperparathyroidism in dogs: 21 cases (1976–1986).

JAVMA 1987;191:350–6.

831

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

[103] Gaddis RS. Oral eﬀects in a case of secondary renal hyperparathyroidism. In: Proceedings

of the 12th Annual Veterinary Dental Forum. New Orleans, LA, 1998, p. 152.

[104] Hertog den E, Goossens MMC, Linde-Sipman JS, et al. Primary hyperparathyroidism

in two cats. Vet Q 1997;19:81–4.

[105] Kallet AJ, Ritcher KP, Feldman EC, et al. Primary hyperparathyroidism in cats: seven

cases (1984–1989). JAVMA 1991;199:1767–71.

[106] Capron KM. Secondary renal hyperparathyroidism. In: Proceedings of the Seventh

Annual Veterinary Dental Forum. Auburn, AL, 1993, p. 181–2.

[107] Carmichael DT, Williams CA, Aller MS. Renal dysplasia with secondary hyper-

parathyroidism and loose teeth in a young dog. J Vet Dent 1995;12:143–6.

[108] Mattson A, Fettman MJ, Grauer GF. Renal secondary hyperparathyroidism in a cat.

J Am Anim Hosp Assoc 1993;29:345–50.

[109] Sarkiala EM, Dambach D, Harvey CE. Jaw lesions resulting from renal hyper-

parathyroidism in a young dog—a case report. J Vet Dent 1994;11:121–4.

[110] George DI, Miller RL. Idiopathic resorption of teeth. A report of three cases. Am

J Orthod Dentofacial Orthop 1986;89:13–20.

[111] Horowitz MC. Cytokines and estrogen in bone: anti-osteoporotic eﬀects. Science 1993;

260:626–7.

[112] Crossley DA. Survey of feline dental problems encountered in a small animal practice

in NW England. Br Vet Dent Assoc J 1991;2:3–6.

[113] Ingham KE, Gorrel C, Blackburn J, et al. Prevalence of odontoclastic resorptive lesions

in a population of clinically healthy cats. J Small Anim Pract 2001;42:439–43.

[114] Okuda A, Inoue E, Fukase T, et al. Prevalence of feline resorptive lesions in Japan. In:

Proceedings of the Eighth Annual Veterinary Dental Forum. Philadelphia, PA, 1994,
p. 44.

[115] Cukjati JF. Cervical line lesions (neck lesions) in the cat. In: Proceedings of the Second

Annual Veterinary Dentistry Forum. New Orleans, LA, 1988. 26–8.

[116] Reubel GH, Hoﬀmann DE, Pedersen NC. Acute and chronic faucitis of domestic cats: a

feline calicivirus-induced disease. Vet Clin North Am Small Anim Pract 1992;22:1347–60.

[117] Pedersen NC. Inﬂammatory oral cavity diseases of the cat. Vet Clin North Am Small

Anim Pract 1992;22:1323–45.

[118] Cognet R, Mesnard E, Stambouli F, et al. Chronic gingivo-stomatitis and viral infections

in a population of 54 cats. In: Proceedings of the Ninth European Congress on Veterinary
Dentistry. Copenhagen, Denmark, 2000, p. 15–16.

[119] Hofmann-Lehmann R, Berger M, Sigrist B, et al. Feline immunodeﬁciency virus (FIV)

infection leads to increased incidence of feline odontoclastic resorptive lesions (FORL).
Vet Immunol Immunopathol 1998;65:299–309.

[120] Zetner K, Kampfer P, Lutz H, et al. Vergleichende immunologische and virologische

Untersuchungen von Katzen mit chronischen oralen Erkrankungen. Wiener Tiera¨rztliche
Monatsschrift 1989;76:303–8.

[121] Anderson JG. Periodontal and radiographic ﬁndings in cats with chronic lymphocytic-

plasmacytic gingivitis-stomatitis complex: a review of 22 cases. In: Proceedings of the 10th
Annual Veterinary Dental Forum. Houston, TX, 1996, p. 106–8.

[122] Trope M, Chivian N. Root resorption. In: Cohen S, Burns RC, editors. Pathways of the

pulp. 6th edition. St Louis: Mosby-Year Book; 1994. p. 486–512.

[123] Tronstad L. Root resorption—etiology, terminology and clinical manifestations. Endod

Dent Traumatol 1988;4:241–52.

[124] Andreasen JO. Review of root resorption system and models. Etiology of root resorption

and the homeostatic mechanisms of the periodontal ligament. In: Davidovitch Z, editor.
The biological mechanisms of tooth eruption and root resorption. Birmingham: EBSCOP
Media; 1988. p. 9–21.

[125] Lindskog S, Blomlo¨f L, Hammarstro¨m L. Comparative eﬀects of parathyroid hormone

on osteoblasts and cementoblasts. J Clin Periodontol 1987;14:386–9.

832

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

[126] Lindskog S, Hammarstro¨m L. Evidence in favor of an anti-invasive factor in cementum or

periodontal membrane of human teeth. Scand J Dent Res 1980;88:161–3.

[127] Polson AM, Caton J. Factors inﬂuencing periodontal repair and regeneration.

J Periodontol 1982;53:617–25.

[128] Wedenberg C. Evidence for a dentin-derived inhibitor of macrophage spreading. Scand

J Dent Res 1987;95:381–8.

[129] Chambers TJ. Phagocytic recognition of bone by macrophages. J Pathol 1981;135:1–7.
[130] Colyer F. Variations and diseases of the teeth of animals. London: John Bale, Sons &

Danielson; 1936.

[131] Gorlin RJ, Barron CN, Chaudhry AP, et al. The oral and pharyngeal pathology of

domestic animals. A study of 487 cases. Am J Vet Res 1959;20:1032–61.

[132] Harvey CE, Alston W. Dental diseases in cat skulls acquired before 1960. In: Proceedings

of the Fourth Annual Veterinary Dental Forum. Las Vegas, NE, 1990, p. 41–3.

[133] van Wessum R, Harvey CE, Hennet P. Feline dental resorptive lesions. Prevalence

patterns. Vet Clin North Am Small Anim Pract 1992;22:1405–16.

[134] Reiter AM. Etiopathology of feline odontoclastic resorptive lesions (FORL). In:

Proceedings of the 12th Annual Veterinary Dental Forum. New Orleans, LA, 1998,
p. 228–34.

[135] Harvey CE. Feline odontoclastic resorptive lesions—update on prevalence and develop-

ment. In: Proceedings of the Sixth World Veterinary Dental Congress. Hobart, Australia:
Australian Veterinary Dental Society; 1999.

[136] Verstraete FJM, Kass PH, Terpak CH. Diagnostic value of full-mouth radiography in

cats. Am J Vet Res 1998;59:692–5.

[137] Coles S. The prevalence of buccal cervical root resorptions in Australian cats. J Vet Dent

1990;7:14–6.

[138] Harvey CE. Epidemiology of periodontal conditions in dogs and cats. In: Proceedings of

the Sixth Annual Veterinary Dental Forum. Las Vegas, NE, 1992, p. 45–6.

[139] Lommer MJ, Verstraete FJM. Radiographic patterns of periodontitis in cats: 147 cases

(1998–1999). JAVMA 2001;218:230–4.

[140] Roes F. Pathogenese, Diagnostik und Therapie bei ‘‘neck lesions’’ der Katze unter

Verwendung von Glas–Ionomer–Zementen [dissertation]. Berlin, Germany: Freie
Universita¨t; 1996.

[141] Verhaert L. Survey of oral disease in Flemish cats. In: Proceedings of the 14th Annual

Veterinary Dental Forum. Albuquerque, NM, 2000, p. 148–150.

[142] Remeeus PGK. Tandhalslaesies bij de kat. Dier en Arts 1991;7:223–6.
[143] Hennet PR, Harvey CE, Flax BM. Feline odontoclastic resorptions: a retrospective study.

In: Proceedings of the Second European Congress of Veterinary Dentistry. Berlin,
Germany: European Veterinary Dental Society; 1993.

[144] Lyon KF. Feline dental disease—classiﬁcation and treatment. Vet Focus 1990;2:27–9.
[145] Anthony J. The use of Nd:YAG laser for treatment of feline odontoclastic resorptive

lesions. J Am Anim Hosp Assoc 2001;37:17–20.

[146] Harvey CE, Emily PP. Restorative dentistry. In: Small animal dentistry. St Louis: Mosby;

1993. p. 213–65.

[147] Wiggs RB, Lobprise HB. Domestic feline oral and dental disease. In: Veterinary dentistry

principles and practice. Philadelphia: Lippincott-Raven; 1997. p. 487–517.

[148] Reiter AM. Feline ‘‘odontolysis’’ in the 1920s: the forgotten histopathological study of

feline odontoclastic resorptive lesions (FORL). J Vet Dent 1998;15:35–41.

[149] Builder PL. Opening paper. Vet Rec 1955;67:386–92.
[150] Catcott EJ. Diseases of the digestive system. In: Feline medicine and surgery. Santa

Barbara, CA: American Veterinary Publications; 1964. p. 177–88.

[151] Joshua JO. The clinical aspects of some diseases of cats. London: W. Heinemann; 1965. p. 51.
[152] Prescott CW. Some oral lesions in the cat. Aust Vet J 1971;47:41–5.
[153] Schneck GW. Caries in the dog. JAVMA 1967;150:1142–3.

833

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

[154] Tekeli O

¨ . Statistische Erhebungen u¨ber die Zahnkrankheiten bei der Katze. Veteriner

Fakultesi Dergisi 1973;20:157–67.

[155] Wilkinson GT. Diseases of the digestive system. In: Diseases of the cat. Oxford: Pergamon

Press; 1966. p. 35–92.

[156] Kerebel B, Daculsi G. Histologie et histopathologie dentaires du chat. Science et

Recherche Odontostomatologiques 1971;7:29–32.

[157] Reichart PA, Du¨rr U.-M., Triadan H, et al. Periodontal disease in the domestic cat.

A histopathologic study. J Periodontal Res 1984;19:67–75.

[158] Schlup D. Epidemiologische und morphologische Untersuchungen am Katzengebib

[dissertation]. Bern, Switzerland: Universita¨t Bern; 1981.

[159] Vickendey G. Klinische und histologische Untersuchungen am gingivoparodontalen

Komplex der Katze [dissertation]. Hannover, Germany: Universita¨t Hannover; 1981.

[160] Colmery BH. Restorations for neck lesions: amalgam/composites—dental ‘‘caries’’. In:

Proceedings of the Second Annual Veterinary Dental Forum. New Orleans, LA, 1988, p. 14–15.

[161] Gauthier O, Boudigues S, Pilet P, et al. Scanning electron microscopy description of

cellular implication and mineral changes in feline odontoclastic resorption lesions. In:
Proceedings of the Eighth European Congress of Veterinary Dentistry. Lyon, France,
1999, p. 17–20.

[162] Lukman K, Pavlica Z, Juntes P. A histologic evaluation of reparative dentin in feline teeth

with odontoclastic resorptive lesions—a preliminary report. In: Proceedings of the Third
European Congress of Veterinary Dentistry. Amsterdam, The Netherlands: European
Veterinary Dental Society; 1994.

[163] Ohba S, Kiba H, Kuwabara M, et al. A histopathological study of neck lesions in feline

teeth. J Am Anim Hosp Assoc 1993;29:216–20.

[164] Ohba S, Kiba H, Kuwabara M, et al. Contact microradiography analysis of feline tooth

resorptive lesions. J Vet Med Sci 1993;55:329–32.

[165] Shigeyama Y, Grove TK, Strayhorn C, et al. Expression of adhesion molecules during

tooth resorption in feline teeth: a model system for aggressive osteoclastic activity. J Dent
Res 1996;75:1650–7.

[166] Harvey CE, Orsini P, McLahan C, et al. Mapping the central point of feline odontoclastic

resorptive lesions. In: Proceedings of the 14th Annual Veterinary Dental Forum.
Albuquerque, NM, 2000, p. 142–4.

[167] Harvey CE. Pathogenesis—neck lesion. In: Proceedings of the Second Annual Veterinary

Dental Forum. New Orleans, LA, 1988, p. 2–4.

[168] Gammon RL. Pathogenesis, detection and treatment of neck lesions. In: Proceedings of

the Second Annual Veterinary Dental Forum. New Orleans, LA, 1988, p. 23.

[169] Itonaga I, Sabokbar A, Murray DW, et al. Eﬀect of osteoprotegerin and osteoprotegerin

ligand on osteoclast formation by arthroplasty membrane derived macrophages. Ann
Rheum Dis 2000;59:26–31.

[170] Burgess TL, Qian Y.-X, Kaufman S, et al. The ligand for osteogrotegerin (OPGL) directly

activates mature osteoclasts. J Cell Biol 1999;145:527–38.

[171] DeLaurier A, DeFlandre C, Allen S, et al. Osteoclastic resorptive lesions of cat teeth are

associated with changes in the expression of RANKL, IL-1 and IL-6 mRNAs [abstract].
J Bone Miner Res 2000;15 (Suppl 1):389.

[172] Cleland WP, Longhofer SL, Braswell K. Electron photomicrographic studies of tooth

surfaces aﬀected by feline odontoclastic resorptive lesions. In: Proceedings of the 14th
Annual Veterinary Dental Forum. Albuquerque, NM, 2000, p. 136.

[173] Harvey CE, Flax BM. Feline oral-dental radiographic examination and interpretation.

Vet Clin North Am Small Anim Pract 1992;22:1279–95.

[174] Crespo Abelleira AC, Rodriguez Cobos MA, Fuentes Boquete IM, et al. Morphological

study of root surfaces in teeth with adult periodontitis. J Periodontol 1999;70:1283–91.

[175] Dupont G. Feline resorptive lesions. In: Proceedings of the Sixth World Veterinary

Dental Congress. Hobart, Australia: Australian Veterinary Dental Society; 1999.

834

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

[176] Heithersay GS. Clinical, radiologic, and histopathologic features of invasive cervical

resorption. Quintessence Int 1999;30:27–37.

[177] Debowes LJ. Odontoclastic resorptive lesions in cats. Waltham Focus 1994;4:2–8.
[178] Emily P, Penman S. Restorative dentistry. In: Handbook of small animal dentistry. 2nd

edition. Oxford: Pergamon Press; 1994. p. 55–69.

[179] Lyon KF. Subgingival odontoclastic resorptive lesions. Classiﬁcation, treatment and

results in 58 cats. Vet Clin North Am Small Anim Pract 1992;22:1417–32.

[180] Hennet PR. Chronic gingivo-stomatitis in cats: long-term follow-up of 30 cases treated

by dental extractions. J Vet Dent 1997;14:15–21.

[181] Narhi M, Hirvonen T, Juvasjarvi E, et al. Reﬂex responses in the digastric and tongue

muscles to stimulation of intradental nerves in the cat. Proceedings of the Finnish Dental
Society 1989;85:383–7.

[182] Mulligan TW, Aller MS, Williams CA. Acquired defects: caries and regressive changes.

In: Aller MS, editor. Atlas of Canine and Feline Dental Radiography. Trenton (NJ):
Veterinary Learning Systems; 1998. p. 153–69.

[183] Holmstrom SE. Surgical treatment for osteomyelitis with dental resorption. Veterinary

Medicine 1985;80:48–9.

[184] Kapatkin AS, Manfra Maretta S, Patnaik AK, et al. Mandibular swellings in cats:

prospective study of 24 cats. J Am Anim Hosp Assoc 1991;27:575–80.

[185] Kertesz P. Glossary. In: A colour atlas of veterinary dentistry and oral surgery.

Aylesbury, UK: Wolfe Publishing; 1993. p. 285–303.

[186] Grippo JO, Simring M. Dental ÔerosionÕ revisited. J Am Dent Assoc 1995;126:619–30.
[187] Rogers GM, Poore MH, Ferko BL, et al. In vitro eﬀects of an acidic by-product feed on

bovine teeth. Am J Vet Res 1997;58:498–503.

[188] West-Hyde L, Floyd M. Dentistry. In: Ettinger SJ, Feldmann EC, editors. Textbook of

veterinary internal medicine. 4th edition. Philadelphia: WB Saunders; 1995. p. 1097–124.

[189] Bergstro¨m A. Root resorption in a dog [abstract]. Br Vet Dent Assoc J 1992;5:5.
[190] Eikenberg S, Loheide H, Arens FC. Treatment of asymptomatic internal resorption of a

maxillary premolar tooth in a military working dog. J Vet Dent 1998;15:175–8.

[191] Gorrel C, Tutt C, Derbyshire S. Idiopathic external root resorption in a dog. In:

Proceedings of the 14th Annual Veterinary Dental Forum. Albuquerque, NM, 2000, p. 183.

[192] Harvey CE. Dental resorptive lesions in dogs. In: Proceedings of the Third European

Congress of Veterinary Dentistry. Amsterdam, The Netherlands: European Veterinary
Dental Society; 1994.

[193] Kaplan B. Root resorption of the permanent teeth of a dog. JAVMA 1967;151:708–9.
[194] Kuntsi-Vaattovaara H. Clinical report: root resorption in a dog. In: Proceedings of the

Fifth World Veterinary Dental Congress. Birmingham, UK, 1997, p. 244.

[195] Levin J. Internal tooth resorption in 2 Basenjis and a Samoyed. In: Proceedings of the

Eighth Annual Veterinary Dental Forum. Philadelphia, PA, 1994, p. 113.

[196] Lyon KF. Cover photograph: eﬀects of root resorption on a dog molar. J Vet Dent

1993;10:1.

[197] Meyer R, Suter G, Triadan H. Epidemiologische und morphologische Untersuchungen

am Hundegebib II. Mitteilung: Morphologische Untersuchungen. Schweiz Arch
Tierheilkd 1980;122:503–17.

[198] Okuda A, Fujimoto K, Nishida T, et al. Histological aspects of resorptive lesions in two

dog cases. In: Proceedings of the Ninth Annual Veterinary Dental Forum. Vancouver,
British Columbia, Canada, 1995, p. 49–50.

[199] Pollok S, Turk MH. Gingival hyperplasia: a case study of an inﬂammatory response. Vet

Med Small Anim Clin 1979;74:691–4.

[200] Rathfon BL. Resorptive lesions in the dog. In: Proceedings of the 10th Annual Veterinary

Dental Forum. Houston, TX, 1996, p. 210–1.

[201] Roes F, Ro¨cken FE, Trautvetter E. Zahnhartsubstanzresorption bei einem Riesensch-

nauzer. Der Praktische Tierarzt 1996;77:694–6.

835

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

[202] Schreyer J, Plu¨mer E. Odontoklastische Zahnhartsubstanzresorptionen bei Hunden.

Kleintierpraxis 2001;46:465–74.

[203] Hamp SE, Olsson S.-E, Farso-Madsen K, et al. A macroscopic and radiologic

investigation of dental diseases of the dog. Vet Radiol Ultrasound 1984;25:86–92.

[204] Crossley DA, Dubielzig RR, Benson KG. Caries and odontoclastic resorptive lesions in a

chinchilla (Chinchilla lanigera). Vet Rec 1997;141:337–9.

[205] Okuda A, Kanehisa J, Heersche JN. The eﬀects of sodium ﬂuoride on the resorptive

activity of isolated osteoclasts. J Bone Miner Res 1990;5 (Suppl 1):S115–20.

[206] Kameyama Y, Nakane S, Maeda H, et al. Eﬀect of ﬂuoride on root resorption caused by

mechanical injuries of the periodontal soft tissues in rats. Endod Dent Traumatol 1994;10:210–4.

[207] Aller S. Case report: neck lesions in a cat. J Vet Dent 1987;4:9–10.
[208] Clarke D. Cervical neck lesions: restore or extract? Newsletter of the Australian

Veterinary Dental Society 1992;7:64–5.

[209] Cukjati JF. Cat neck lesions (academy neck lesion study and survey). In: Proceedings of

the Third Annual Veterinary Dentistry Forum. New Orleans, LA, 1989, p. 41–3.

[210] Dicks F, Zo¨llner W. Glas-Ionomer-Zement fu¨r einen praktikablen Einsatz in der

Tierzahnheilkunde. Der Praktische Tierarzt 1988;69:32–6.

[211] Emily P. Silver glass ionomer in the treatment of feline neck lesions. Br Vet Dent Assoc J

1989;2:2–3.

[212] Floyd M. External odontoclastic resorptive lesions in cats—another opinion. J Vet Dent

1991;8:22–3.

[213] Golden AL. Restoration of cat neck lesions using glass ionomers. In: Proceedings of the

Third Annual Veterinary Dental Forum. New Orleans, LA, 1989, p. 39–40.

[214] Golden AL, Marretta SM. The use of ESPE Ketac-Bond Aplicap for the restoration of

cervical lesions. J Vet Dent 1989;6:5–7.

[215] Hawkins BJ. Cervical line lesions (neck lesions) in the cat. In: Proceedings of the First

Annual Veterinary Dental Forum. Cincinnati, OH, 1987, p. 1–5.

[216] Lyon KF. Feline dental disease: treatment of subgingival resorption lesions. J Vet Dent

1990;7:13–4.

[217] Reichart P. Kariestherapie im Hunde- und Katzengebib. Tiera¨rztliche Praxis 1978;6:361–4.
[218] Zetner K, Steurer I. Long-term results of restoration of feline resorptive lesions with

micro-glass-composite. J Vet Dent 1995;12:15–7.

[219] Eberhard TL. Surgical treatment planning for feline odontoclastic resorptive lesions. In:

Proceedings of the 14th Annual Veterinary Dental Forum. Albuquerque, NM, 2000. p. 30–4.

[220] Harvey CE. External odontoclastic resorption lesions in cats—a treatable disease? J Vet

Dent 1991;8:16–7.

[221] van Foreest A. Veterinaire tandheelkunde (8). Extraheren van gebitselementen bij de kat.

Tijdschr Diergeneeskd 1995;120:106–12.

[222] Johnston N. Feline dental disease—the neck lesion. Br Vet Dent Assoc J 1993;6:1–4.
[223] Eisner E. Chronic subgingival tooth erosions in cats. Veterinary Medicine 1989;84:378–87.
[224] Zetner K. Diagnosis, pathogenesis, and treatment of neck lesions in cats. Vet Q 1994;

16 (Suppl 1):22S–4S.

[225] Glickman I, Pruzansky S, Ostrach M. The healing of extraction wounds in the presence of

retained root remnants and bone fragments. An experimental study. Am J Orthod Oral
Surg 1947;33:263–83.

[226] Dupont G. Crown amputation with intentional root retention for advanced feline

resorptive lesions—a clinical study. J Vet Dent 1995;12:9–13.

[227] Lommer MJ, Verstraete FJM. Prevalence of odontoclastic resorption lesions and peri-

apical radiographic lucencies in cats: 265 cases (1995–1998). JAVMA 2000;217:1866–9.

[228] Sunakawa M, Tokita Y, Suda H. Pulsed Nd:YAG laser irradiation of the tooth pulp in

the cat: II. Eﬀect of scanning lasing. Lasers Surg Med 2000;26:477–84.

[229] Tokita Y, Sunakawa M, Suda H. Pulsed Nd:YAG laser irradiation of tooth pulp in the

cat: I. Eﬀect of spot lasing. Lasers Surg Med 2000;24:398–404.

836

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

[230] Dorenkamp B. Zahnerkrankungen bei Katzen—Klassiﬁzierung und Behandlung.

Biologische Tiermedizin 1993;2:40–5.

[231] Pierce A, Lindskog S. The eﬀect of an antibiotic/corticosteroid paste on inﬂammatory

root resorption in vivo. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1987;64:
216–20.

[232] Sae-Lim V, Wang CY, Choi GW, et al. The eﬀect of systemic tetracycline on resorption of

dried replanted dogs’ teeth. Endod Dent Traumatol 1998;14:127–32.

[233] Levin L, Bryson EC, Caplan D, et al. Eﬀect of topical alendronate on root resorption of

dried replanted dog teeth. Dental Traumatology 2001;17:120–6.

[234] Igarashi K, Adachi H, Mitani H, et al. Inhibitory eﬀect of the topical administration of a

bisphosphonate (risedronate) on root resorption incident to orthodontic tooth movement
in rats. J Dent Res 1996;75:1644–9.

[235] Liewehr FR, Craft DW, Primack PD, et al. Eﬀect of bisphosphonates and gallium on

dentin resorption in vitro. Endod Dent Traumatol 1995;11:20–6.

[236] Alatli I, Hammarstro¨m L. Root surface defects in rat molar induced by 1-hydroxy-

ethylidene-1, 1-bisphosphonate. Acta Odontol Scand 1996;54:59–65.

[237] Gioso MA, Correa HL, Venturini MAFA, et al. Prevalence of feline odontoclastic

resorptive lesions in three diﬀerent cities in Brazil. In: Proceedings of the Sixth World
Veterinary Dental Congress. Hobart, Australia: Australian Veterinary Dental Society;
1999.

[238] Debowes LJ. Identiﬁcation, grading, and treatment of feline resorptive lesions. In: Pro-

ceedings of the Ninth Annual Veterinary Dental Forum. Vancouver, British Columbia,
Canada, 1995, p. 24–6.

[239] Pierce A, Berg J.-O, Lindskog S. Calcitonin as an alternative therapy in the treatment of

root resorption. J Endod 1988;14:459–64.

[240] Trope M, Moshonov J, Nissan R, et al. Short vs long-term calcium hydroxide treatment

of established inﬂammatory root resorption in replanted dog teeth. Endod Dent
Traumatol 1995;11:124–8.

837

A.M. Reiter, K.A. Mendoza / Vet Clin Small Anim 32 (2002) 791–837

Feline nasopharyngeal polyps

Rebecca K. Muilenburg, DVM

Thomas R. Fry, DVM, MS

36569 Innisbrook Circle, Purcellville, VA 20132, USA

Cascade Veterinary Specialists, 660 NW Gilman Boulevard, Suite C-2,

Issaquah, WA 98027, USA

Feline nasopharyngeal polyps (FNPs) are benign, nonneoplastic, peduncu-

lated masses that arise from the mucosa of the nasopharynx, auditory (Eusta-
chian) tube, or middle ear and may extend into the nasopharynx, tympanic
cavity, or both [1]. A relatively uncommon disease, synonyms for these masses
include feline inﬂammatory polyps, pharyngeal polyps, aural polyps, middle
ear polyps, feline respiratory tract polyps, and otopharyngeal polyps [1,2].
FNPs are most often found in young cats, although they can occur in mid-
dle-aged or older cats. These polyps are the most common mass of the feline
external ear canal and are the second most common cause of nasopharyngeal
disease after lymphoma [3–5]. Similar polyps have been reported in the horse
and dog [6,7], and aural-pharyngeal polyps have been reported in iguanas [8].
A histologically similar mass has been described in the trachea of a cat [9].

Etiology

The etiology of FNPs is unknown. Because this disease occurs with

increased frequency in young cats, a congenital etiology in which these pol-
yps may arise as aberrant growths from the remnants of the branchial arches
was proposed by Baker in 1982 [10]. It has also been suggested that these
polyps may be secondary to an infection ascending from the nasopharynx
[11,12]. The typical inﬂammatory changes seen histopathologically com-
bined with clinical signs in these patients may imply a viral or bacterial con-
tribution. To potentially support this theory, feline calicivirus has been
recovered from the polyps of some aﬀected cats [1,11,13] as well as from the
nasopharynx of aﬀected cats. Because caliciviruses are ubiquitous, a cause
and eﬀect inference may not be appropriate. Whether inﬂammation and
infection are primary or secondary to FNP formation is unclear [13].

Vet Clin Small Anim 32 (2002) 839–849

* Corresponding author.

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 2 4 - 4

History and clinical signs

FNPs have been reported in cats ranging from a few weeks to 15 years of

age, although patients are often less than 2 years old [1,5,13]. No apparent
breed or sex predilection has been reported [13]. Although FNPs are most
often unilateral, they may be bilateral.

Clinical signs may be present for weeks to years before diagnosis, and

signs vary depending on the location of the polyp, because the larger portion
of the polyp may be pharyngeal, middle ear–based, or aural (external ear
canal) [1,5]. Less than half of cats have visible polyps [11]. Cats with FNPs
may have a chronic history, which may include signs consistent with upper
respiratory infection, including nasal discharge and sneezing. Although
many cats have no history of nasal or ocular discharge, those that do may
have rhinitis or sinusitis that is secondary to the obstruction of normal air-
ﬂow [6]. These cats may be partially responsive to the empiric treatment of
upper respiratory infection [1]. Other clinical signs reported include dys-
pnea, stridor, weight loss, epistaxis, voice change, and dysphagia [1,3,12,
13]. Less commonly, cyanosis and syncopal episodes may occur, particularly
in cats with large polyps obstructing the nasopharynx. Cats with FNPs may
present with signs of otitis externa, otitis media, or otitis interna. These most
commonly include otorrhea and head shaking, but they may also include
head tilt, Horner’s syndrome, nystagmus, and ataxia depending on which
portion of the ear is involved by the polyp, infection, or both [1,13].

Diagnosis

FNP should be considered as a diﬀerential diagnosis in any cat with evi-

dence of chronic upper respiratory disease. Diﬀerential diagnoses include
upper respiratory tract infections caused by feline calicivirus and feline rhi-
notracheitis virus, nasal foreign bodies, neoplasia, and mycotic disease, such
as cryptococcosis [1,13].

Diagnosis of nasopharyngeal masses can be made by oropharyngeal exami-

nation, digital palpation of the soft palate, skull radiographs, endoscopy, com-
puted tomography (CT) or magnetic resonance (MR) scans, nasopharyngeal
biopsy, or any combination of these [3]. For pharyngeal polyps, a diagnosis can
often be made on oral examination with digital palpation and the use of a Snook
spay hook to retract the soft palate. A more thorough examination is possible
with an endoscope using a retroﬂexed view, which allows for complete visualiza-
tion of the nasopharynx, oropharynx, openings to the auditory tubes, and cau-
dal nasal passages. Careful examination of the external ear canals should be
performed in all cases using either an otoscope or video-otoscope. A preopera-
tive biopsy should most deﬁnitely be considered if neoplasia is considered a likely
diﬀerential diagnosis or if there are any atypical features to the presentation.

Skull radiographs should be considered in all cases. Nasopharyngeal pol-

yps are best visualized on lateral radiographs as a soft tissue density in the

840

R.K. Muilenburg, T.R. Fry / Vet Clin Small Anim 32 (2002) 839–849

nasopharynx (Fig. 1) [1]. Ventrodorsal, oblique lateral views, and frontal
open-mouth projection skull radiographs should also be obtained to evalu-
ate the tympanic bullae. Radiographically detectable lesions in the tympanic
bullae usually require long-standing and sustained inﬂammatory stimuli,
and changes indicate chronicity rather than severity [14]. Changes may
include bulla enlargement, proliferative periosteal response, and increased
soft tissue density within the bullae (Fig. 2) [15]. Lack of radiographic
changes in the bullae does not eliminate middle ear disease just as the pres-
ence of an intact tympanic membrane does not preclude middle ear disease
[13,16]. Evaluation of bulla changes on standard radiographic views in cats
with known otitis media has a true-positive rate of only 75% [1]. This means
that as many as 25% of cats with known otitis media are untreated if radio-
graphs alone are used as the deciding factor to rule out middle ear disease.
MR and CT technology is becoming more accessible and more aﬀordable,
and they may be used in lieu of radiographs in the diagnosis of bulla disease.
Both modalities should aﬀord a higher true-positive diagnostic rate for the
detection of otitis media. If otitis interna or central neurologic disease is sus-
pected, MR has distinct advantages over CT. If middle ear changes are
present by radiography, CT, or MR, surgical intervention in the middle ear
(bulla osteotomy) can typically be recommended with conﬁdence.

Histopathology

FNPs typically present as grossly as ovoid- to elliptically shaped white to

pink pedunculated masses with variable ulceration. In the ear canal, these
polyps are usually smooth-surfaced red masses [1].

Fig. 1. Lateral radiograph of a feline skull. Note the nasopharyngeal polyp (arrow) occupying
the pharynx. (Courtesy of Norman Ackerman, DVM, Huntsville, AL.)

841

R.K. Muilenburg, T.R. Fry / Vet Clin Small Anim 32 (2002) 839–849

Histologically, FNPs typically have epithelium, which varies from strati-

ﬁed squamous to pseudostratiﬁed ciliated columnar, covering a core of ﬁbro-
vascular connective tissue containing scattered lymphocytes, plasma cells,
and macrophages [1,5,12,16,17]. Focal mucosal ulceration may be seen. The
polyp stalk may originate from the mucosa of the nasopharynx, auditory
(Eustachian) tube, or tympanic bulla [1,11–13]. Because the mucosal lining
between these regions is continuous and histologically similar, identifying the
exact anatomic origin of these polyps is diﬃcult [1,13]. The polyp may extend
into the nasopharynx. tympanic cavity, or external ear canal.

Treatment

Surgical anatomy of the middle ear

Surgery of the middle ear involves a number of vital anatomic structures

that must be preserved. Middle ear structures include the tympanic mem-
brane, the tympanic cavity, the auditory tube, the auditory ossicles (malleus,
incus, and stapes) with their associated muscles and ligaments, the tympanic

Fig. 2. Ventrodorsal radiograph of a feline skull. Note the asymmetry in density between the
osseus bullae. The right bulla (arrow) has an increased periosteal response, indicating chronic
inﬂammation. These changes do not indicate an active response, however. Note that each osseus
bulla lies just caudomedial to the two temporomandibular joints. (Courtesy of Norman
Ackerman, DVM, Huntsville, AL.)

842

R.K. Muilenburg, T.R. Fry / Vet Clin Small Anim 32 (2002) 839–849

nerve, and the sympathetic nerve supply to the ipsilateral eye [14]. The tym-
panic membrane separates the middle ear cavity from the horizontal ear
canal at the external auditory meatus.

The tympanic cavity constitutes the bulk of the middle ear. Most of this

air-ﬁlled cavity lies within the tympanic bulla. In the cat, it is divided into a
larger ventromedial and a smaller dorsolateral compartment by a nearly
complete thin bony septum [14]. The nasopharynx is connected to the tym-
panic cavity by the auditory (Eustachian) tube, which opens on the dorso-
medial aspect of the tympanic cavity [14]. The auditory ossicles are
located in the dorsomedial compartment and form a chain from the tym-
panic membrane to the inner ear. The tympanic nerve can also be located
in this area [14]. The sympathetic nerve ﬁbers to the eye course along a
promontory that lies medially on the dorsal aspect of the bulla. These ﬁbers
then pass through a narrow ﬁssure in the dorsal aspect of the septum into
the dorsolateral compartment and continue along the dorsomedial wall
before entering the petrous temporal bone medial to the auditory os of
the Eustachian tube at the rostral apex of the bulla [1]. The opening of the
auditory tube is located in the rostromedial aspect of the dorsolateral com-
partment [1].

Important structures located adjacent to the tympanic bulla include the

carotid artery, facial nerve, and hypoglossal nerve. The hypoglossal nerve
lies ventral to the tympanic bulla, the facial nerve exits the stylomastoid
foramen caudodorsal to the tympanic cavity and courses ventrolateral to
the osseous bulla, and the carotid artery is slightly ventromedial to the tym-
panic bulla [14]. The lingual artery and vein lie on the lateral border of the
hypoglossal muscle [1].

Objectives of surgery

Surgical objectives include removal of all inﬂamed and infected tissue,

because failure to do so may predispose the patient to recurrence of the polyp.
Described surgical techniques used in the treatment of FNPs include
some combination of traction-avulsion, ventral bulla osteotomy, myringo-
tomy, lateral ear resection, ear canal ablation combined with lateral bulla
osteotomy, and, in a limited role, laser debulking and sterilization of the
middle ear. Surgical exposure of the tympanic cavity, when done, also allows
access to material for culture and sensitivity or histopathologic testing.

The surgical technique(s) chosen depend on the location of the polyp and

presence or absence of radiographic changes to the middle ear as well as on
operator preference. Many surgeons advocate ventral bulla osteotomy in all
cases regardless of radiographic evidence of middle ear pathologic ﬁndings
[13,18], although others perform ventral bulla osteotomy only in cases where
radiographic changes consistent with otitis media exist [17]. This same study
advised that owners be warned of the potential for polyp recurrence if no
bulla osteotomy is done in the face of normal radiographs.

843

R.K. Muilenburg, T.R. Fry / Vet Clin Small Anim 32 (2002) 839–849

Preoperative considerations

Because many patients with FNPs demonstrate some signs of respiratory

compromise before surgical intervention, these cats should be considered at
risk for complications during anesthetic induction. Preoxygenation for 5 to
10 minutes before induction may increase the margin of safety and allow for
a smooth and rapid induction that allows immediate access for intubation
[1]. Preoperative antibiotic prophylaxis should be avoided so as to maximize
a yield if culture and sensitivity testing is to be performed, particularly on
tympanic bulla contents.

Surgical technique

Traction-avulsion

The nasopharynx can be evaluated by retracting the soft palate ros-

trally using a spay hook or stay sutures. The polyp can then usually be
displaced from the nasopharynx to the oropharynx by digital pressure.
The polyp may be grasped with right-angle forceps or alligator forceps,
with traction applied to avulse the entire polyp from its attachment in the
auditory tube or middle ear [13]. Hemorrhage is usually minimal and can
be controlled with digital pressure. In rare instances, the soft palate may
need to be split to facilitate exposure of the polyp [13]. Soft palate split-
ting is accomplished via a midline incision from the tip of the soft palate
and can extend rostral to the level of the hard palate. A midline incision
avoids the paired palatine vessels that lie lateral to the incision. Bleeding
is typically minimal. This approach allows for increased exposure and
visualization of the nasopharynx; however, in most cases, this is an un-
necessary step. Closure is accomplished via a two- or three-layer simple con-
tinuous pattern using synthetic monoﬁlament absorbable suture (ie, 4-0
poliglecaprone).

Polyps in the ear canal may be removed similarly by grasping with a for-

ceps and applying gentle traction. Hemorrhage is typically easily controlled
by pressure with cotton-tipped applicators or cotton. A lateral ear canal
resection can be performed if exposure is inadequate or the polyp cannot
be grasped in the intact ear canal [1,5]. Lateral ear resection has been well
described in numerous standard surgery textbooks.

Myringotomy

Myringotomy is incision of the intact tympanic membrane. Although not

a deﬁnitive treatment for FNPs, it may have a limited role for acquisition of
material for culture or biopsy when otoscopic examination reveals an intact
tympanum in a patient in which middle ear disease is suspected. A small
incision may also allow for direct visualization of polyps in these cases. This
procedure has a low likelihood of complications.

844

R.K. Muilenburg, T.R. Fry / Vet Clin Small Anim 32 (2002) 839–849

Bulla osteotomy

The middle ear can be approached surgically by means of a lateral or ven-

tral bulla osteotomy [9]. A lateral bulla osteotomy allows exposure for dis-
ease processes involving the external, middle, and inner ear compartments,
whereas the ventral approach is useful for middle and inner ear disease. A
lateral bulla osteotomy in combination with ear canal ablation has been
described as a treatment for FNPs, but the author believes this aggressive
approach is seldom indicated because it obviously leaves the patient without
an external ear canal. Numerous studies indicate that either traction-avul-
sion or traction-avulsion combined with ventral bulla osteotomy has a high
success rate and still allows for preservation of the ear canal [4,11,17,18].
Improved exposure of the tympanic cavity and the provision of gravity-
assisted ventral drainage are advantages of the ventral approach [9]. Because
the osseus septum divides the feline bullae into two compartments, the lat-
eral bulla osteotomy approach makes evacuation of the medial compart-
ment problematic. Another undesirable feature of the lateral approach is
the need for repositioning if both bullae are aﬀected.

One aspect of ventral bulla osteotomy that is disconcerting, particularly

to inexperienced surgeons, is the surgical approach. Reference to well-posi-
tioned ventrodorsal radiographs is useful, but in most instances, the bullae
are easily palpated in the anesthetized patient in dorsal recumbency. The
bullae are located just caudal and medial to a line connecting the temporo-
mandibular joints (see Fig. 2). A paramedian incision is made directly over
the bulla through the skin and platysma muscle. Sharp and blunt dissection
is used to separate the digastricus muscle laterally from the more medial sty-
loglossal and hypoglossal muscles. This plane of dissection is directly ventral
to the bulla. A Freer periosteal elevator is useful to elevate periosteum and
overlying soft tissues. As an anatomic landmark, the hypoglossal nerve lies
directly ventral to the bulla and should be preserved. The facial nerve lies
lateral to the bulla and the internal carotid artery medial to the bulla, and
both structures should be avoided. Exposure is easily maintained with either
Weitlander or small Gelpi self-retaining retractors. Entry into both compart-
ments of the bulla is easily accomplished with a Steinmann pin and chuck or
a neurosurgical air drill with appropriate burr. Widening of the initial entry
defect is done with either rongeurs or the air drill, taking care to avoid the
adjacent neurovascular structures. The ventral portion of the osseus trans-
verse septum should be removed to evaluate the dorsolateral compartment,
which is the most common site of FNP origin [2]. All polyp-associated soft
tissue and attachments should be removed using small curettes and eleva-
tors. Extensive lavage of the tympanic bulla should be performed.

We have used the carbon dioxide laser in approximately a dozen cases to

aid in the process of extirpation of FNP remnants from the middle ear. The
laser is useful in that it vaporizes cellular remnants and sterilizes the surfaces
that are lased. Further studies are indicated before its routine use can be rec-
ommended.

845

R.K. Muilenburg, T.R. Fry / Vet Clin Small Anim 32 (2002) 839–849

Anatomic closure is performed. Many surgeons recommend the use of

drains in the middle ear, exiting adjacent to the incision. Use of drains in mid-
dle ear surgery, although having a long tradition, has recently been ques-
tioned, speciﬁcally with regard to their necessity in studies involving ear
canal ablation and lateral bulla osteotomy for treatment of end-stage otitis [19].

Operative and postoperative complications

Overall, complications are more common with bulla osteotomy than with

traction-avulsion. Potential complications include polyp regrowth, wound
drainage, otitis media, Horner’s syndrome, hypoglossal nerve damage, facial
nerve paralysis, vestibular disturbances, and damage to the auditory ossicles
and vascular structures [5,11,14–16]. Signiﬁcant intraoperative hemorrhage
is also possible but rare.

Owners should be warned that polyp regrowth is a possibility in all

patients, especially in those treated by traction-avulsion alone. Traction-
avulsion may not adequately remove all inﬂammatory tissue within the tym-
panic bulla, and remnants may develop into new polyps within months to
years [11,13]. In a study of 31 cats with pharyngeal masses, 17% (5/29) of
patients available for long-term follow-up experienced regrowth [17]. Of
these 5 patients, only 1 had undergone ventral bulla osteotomy at the time
of initial surgery. Recurrence rates of 50% for aural polyps compared with
11% for pharyngeal polyps were seen in another study [11].

Horner’s syndrome, characterized by the clinical signs of ptosis, miosis,

enophthalmos, and elevation of the third eyelid, is most likely caused by
direct damage to the oculosympathetic trunk as it passes through the tym-
panic cavity. With this in mind, aggressive curettage or direct suctioning
of the dorsomedial aspect of the bulla should be avoided to reduce the risk
of damaging these nerve ﬁbers [13]. Horner’s syndrome can occur with
either traction-avulsion or ventral bulla osteotomy but is much more likely
with ventral bulla osteotomy. In two previous retrospective studies, Horn-
er’s syndrome occurred in 83% (10/12) and 81% (22/27) of cases, respectively
[17,19]. In the ﬁrst study, 17% (2/12) had long-term deﬁcits, and in the sec-
ond study, 4% (1/22) of Horner’s syndrome patients had permanent aﬀecta-
tions. In another recent study of 37 cats with inﬂammatory polyps, 43% of
cats treated by traction alone and 57% of cats that underwent a surgical pro-
cedure (ventral bulla osteotomy or total ear canal ablation with lateral bulla
osteotomy) demonstrated signs of Horner’s syndrome after surgery [11].
In most cases, Horner’s syndrome resolves in weeks to months, although
owners should be warned that some cases might be permanent.

Another potential sequela to bulla osteotomy, facial nerve paralysis,

occurs less commonly. One study reported that 33% (3/9) of cats treated sur-
gically demonstrated facial nerve paralysis [11]. In another study, 26% (5/19)
of cats (2 that had a lateral bulla osteotomy combined with total ear canal
ablation and 3 treated by ventral bulla osteotomy) developed facial nerve

846

R.K. Muilenburg, T.R. Fry / Vet Clin Small Anim 32 (2002) 839–849

paralysis. In the latter study, signs resolved within 3 days in 2 cats but per-
sisted for up to 12 weeks in the other 3 cats. All patients had eventual res-
olution of facial nerve paralysis [18].

Otitis interna or otitis media is another reported complication when

FNPs involve the middle ear. Reports on the incidence of infection have var-
ied widely, ranging from 13% to 83% [4,17,18]. Faulkner and Budsberg’s
study [4] included 12 cats with external ear canal and middle ear polyps,
although all cats were aﬀected with pharyngeal masses in Kapatkin et al’s
report [17]. Another recent study involving cats with middle ear disease
included 7 cats with FNPs, 57% (4/7) of which had bacterial otitis media
[18]. It is not surprising that rates of infection may be increased in cats with
evidence of external ear canal and middle ear involvement compared with
patients that have only pharyngeal masses. The inference is that the auditory
tube is a better barrier to acquired middle ear infection than the ruptured
tympanic membrane.

Signs of otitis interna include head tilt, ataxia, and nystagmus and may be

a result of overzealous curettage of the dorsomedial compartment of the
tympanic cavity [1,13]. Nystagmus often resolves within 24 hours of surgery,
but head tilt or ataxia may persist [4,13]. Bulla osteotomy resulted in
improvement but never in complete resolution of head tilt in cats that devel-
oped this sign secondary to middle ear disease [18].

Damage to the hypoglossal nerve or vascular structures adjacent to the

tympanic bulla can occur during the surgical approach and may be mini-
mized by careful surgical technique. Signs of hypoglossal nerve damage
include swallowing, prehension, and mastication deﬁcits [13].

Adjunct treatment

Most sources advocate the use of antibiotics after bulla osteotomy, with

antibiotic selection based on culture and sensitivity test results [1,9,11]. Use
of systemic corticosteroids after surgery is controversial. One study found
that postoperative prednisolone therapy signiﬁcantly reduced the rate of
recurrence of polyps after traction, although the location of the polyp (pha-
ryngeal versus aural) in patients that received prednisolone was not clearly
deﬁned [11]. Other authors state that topical or systemic corticosteroid
administration is not beneﬁcial and is contraindicated [12]. Routine use of
analgesics for purposes of preemptive analgesia and postoperative analgesia
is recommended, particularly when bulla osteotomy is performed.

Summary

In summary, an ideal diagnostic plan for cats with suspected FNPs

should include a thorough anesthetized oropharyngeal examination, oto-
scopic examination, and imaging studies, which may consist of a bulla

847

R.K. Muilenburg, T.R. Fry / Vet Clin Small Anim 32 (2002) 839–849

radiographic series or specialized imaging studies such as CT or MR scans.
In general, if signs indicative of otitis media are present, ventral bulla osteo-
tomy should be advised. As a result of the distinct appearance of FNPs, a
preoperative biopsy is not indicated in all instances but should be considered
if there are atypical features to the history or presentation. Traction-avul-
sion of polyps through the external ear canal or auditory tube from the oro-
pharynx may have a lower success rate than traction-avulsion combined
with ventral bulla osteotomy. If the veterinarian opts to treat FNPs by trac-
tion-avulsion alone as a result of ﬁnancial constraints imposed by the client,
the client should be strongly cautioned regarding potential recurrence.

Postoperative complications are possible with any treatment option, but

neurologic impairment, including Horner’s syndrome, facial neuropathy,
and hypoglossal neuropathy, is signiﬁcantly more likely after surgical inter-
vention by ventral bulla osteotomy than after traction-avulsion alone. A
high percentage of these complications are self-limiting. In all instances,
appropriate culture and sensitivity and biopsy specimens should be collected
so as to enable provision of appropriate postoperative care. Antibiotic ther-
apy should be provided based on culture and sensitivity test results. The use
of postoperative steroids to prevent recurrence is controversial.

References

[1] Pope ER. Feline inﬂammatory polyps. Semin Vet Med Surg (Small Anim) 1995;10:87–93.
[2] Pope ER, Constantinescu GM. Feline respiratory tract polyps. In: Bonagura, JD, editor.

Kirk’s Current Veterinary Therapy XIII. 1st edition. Philadelphia: WB Saunders; 2000.
p. 794–6.

[3] Allen H, Broussard J, Noone K. Nasopharyngeal diseases in cats: a retrospective study of

53 cases (1991–1998). J Am Anim Hosp Assoc 1999;35:457–61.

[4] Faulkner JE, Budsberg SC. Results of ventral bulla osteotomy for treatment of middle ear

polyps in cats. J Am Anim Hosp Assoc 1990;26:496–9.

[5] Harvey CE, Goldschmidt MH. Inﬂammatory polypoid growths in the ear canal of cats.

J Small Anim Pract 1978;19:669–77.

[6] Bradley RL: Selected oral, pharyngeal, and upper respiratory conditions in the cat. Vet

Clin North Am Small Anim Pract 1984;14:1173–84.

[7] Fingland RB, Gratzek A, Vorhies MW, et al. A nasopharyngeal polyp in a dog. J Am

Anim Hosp Assoc 1993;29:311–4.

[8] Uhl EW, Jacobson E, Bartick TE, et al. Aural-pharyngeal polyps associated with

Cryptosporidium infection in three iguanas (Iguana iguana). Vet Pathol 2001;38:239–42.

[9] Sheaﬀer KA, Dillon AR. Obstructive tracheal mass due to an inﬂammatory polyp in a cat.

J Am Anim Hosp Assoc 1996;32:431–4.

[10] Baker G. Nasopharyngeal polyps in cats [letter]. Vet Rec 1982;111:43.
[11] Anderson DM, Robinson RK, White RAS. Management of inﬂammatory polyps in 37

cats. Vet Rec 2000;147:684–7.

[12] Kirpensteijn J. Aural neoplasms. Semin Vet Med Surg (Small Anim) 1993;8:17–23.
[13] Davidson JR. Otopharyngeal polyps. In: Bojrab MJ, editor. Current techniques in small

animal surgery. 4th edition. Philadelphia: Lea & Febiger; 1998. p. 147–50.

[14] Boothe H. Surgery of the tympanic bulla (otitis media and nasopharyngeal polyps). Probl

Vet Med 1991;3:254–69.

848

R.K. Muilenburg, T.R. Fry / Vet Clin Small Anim 32 (2002) 839–849

[15] Lane JG, Orr CM, Lucke VM, et al. Nasopharyngeal polyps arising in the middle ear of the

cat. J Small Anim Pract 1981;22:511–22.

[16] Remedios AM, Fowler JD, Pharr JW. A comparison of radiographic versus surgical

diagnosis of otitis media. J Am Anim Hosp Assoc 1991;27:183–8.

[17] Kapatkin AS, Matthiesen DT, Noone KE. Results of surgery and long-term follow-up in

31 cats with nasopharyngeal polyps. J Am Anim Hosp Assoc 1990;26:387–92.

[18] Trevor PB, Martin RA. Tympanic bulla osteotomy for treatment of middle ear disease in

cats: 19 cases (1984–1991). JAVMA 1993;202:123–8.

[19] Devitt CM, Seim HB, Willer R, et al. Passive drainage versus primary closure after total

ear canal ablation-lateral bulla osteotomy in dogs: 59 dogs (1985–1995). Vet Surg 1997;26:
210–6.

849

R.K. Muilenburg, T.R. Fry / Vet Clin Small Anim 32 (2002) 839–849

Feline thyroid surgery

Sheldon Padgett, DVM, MS

Metropolitan Veterinary Hospital, 1053 South Cleveland-Massillon Road,

Akron, OH 44321, USA

Thyroid disease in the cat presents a unique challenge to the surgeon, be-

cause the patients commonly have concurrent secondary metabolic abnormal-
ities that can predispose them to signiﬁcant and sometimes life-threatening
complications. It is necessary not only to consider the best method of treat-
ment but the eﬀects that thyroid disease can have on other organ systems.

Surgical indications

Thyroid surgery in the cat is generally limited to removal of the thyroid

(thyroidectomy), although a thyroid biopsy may occasionally be indicated.
Thyroidectomy is most commonly performed as therapy for hyperthyroidism.
Additionally, because of the intimate anatomy of thyroid and parathyroid
glands, the thyroid is usually removed as part of a parathyroidectomy per-
formed as therapy for functional parathyroid disease.

Hyperthyroidism

Since the ﬁrst report of the disease in 1979 by Peterson et al [1], hyper-

thyroidism has become the most commonly diagnosed endocrine disease
in cats. The disease is caused by functional benign adenomatous hyperplasia
in approximately 98% of cases. Functional thyroid tumors (carcinomas)
account for the remaining 2% of hyperthyroid cats [2].

Feline hyperthyroidism is usually treated by one of the following methods:

surgical removal of abnormal tissue (thyroidectomy), daily oral medication
(methimazole [Tapazole]), or radioactive iodine therapy. Each treatment has
pros and cons depending on the patient’s medical condition, availability of
the therapy, and client’s tolerance for complications or ability to medicate.

Vet Clin Small Anim 32 (2002) 851–859

E-mail address: spadgett@bright.net (S. Padgett).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 2 3 - 2

Medical therapy options for hyperthyroidism

Long-term oral antithyroid drug administration is widely available to all

owners and requires no anesthesia or specialized equipment. The initial cost
of therapy is lower, and no hospitalization is necessary. Methimazole is not
a curative therapy but controls hyperthyroidism by blocking thyroid hor-
mone synthesis. This medication may not be appropriate for animals that
are diﬃcult to medicate or experience side eﬀects associated with the drug.
Approximately 18% of cats have been reported to have mild to severe reac-
tions to the drug (self-excoriation, hematologic abnormalities, vomiting,
anorexia, and hepatopathy) [3,4]. Long-term thyroid hormone level moni-
toring and methimazole dosage adjustment are necessary, which adds to the
expense of medical therapy.

Radioactive iodine therapy is a safe and eﬀective method of curing hyper-

thyroidism. Only one treatment is necessary, no systemic side eﬀects are
seen, and no anesthesia risks are necessary. Although it is an excellent pro-
cedure, it is still not widely available. The cost is signiﬁcant, and there is a
chance that up to 8% of cats may need a second treatment to achieve euthy-
roidism [3]. Depending on regulatory mandates in the area, posttreatment
isolation is necessary, often for a number of weeks.

Surgical considerations

Surgical anatomy

The thyroid glands are paired and located on either side of the trachea in

the paratracheal fascia, normally just distal to the caudal larynx (thyroid
cartilage). The normal feline thyroid gland is approximately 1 cm long, 3
to 5 mm wide, and 1 mm thick [5]. When a thyroid gland is pathologically
enlarged, the thyroid can be found further caudal than normal because of
gravitational migration; therefore, the entire cervical area should be avail-
able for exploration (Fig. 1). Each thyroid gland has two associated para-
thyroid glands. The external parathyroid gland is found lying just beneath
the thyroid capsule, usually at the cranial pole of the thyroid gland,
although there is tremendous variation in this location. The internal para-
thyroid gland is found embedded in the thyroid tissue at the middle to cau-
dal aspect of the thyroid gland.

The carotid sheath is also found in the paratracheal fascia near the thy-

roid glands. Each thyroid gland is supplied by a cranial thyroid artery,
which is a branch of the carotid artery. This enters the thyroid at the cranial
pole and gives rise to a small vascular branch supplying the external para-
thyroid gland. The caudal thyroid artery seen in the dog is usually absent
in the cat.

The recurrent laryngeal nerves are ﬁne in the cat and run in the paratra-

cheal fascia near the thyroid glands. The left recurrent laryngeal nerve is

852

S. Padgett / Vet Clin Small Anim 32 (2002) 851–859

located dorsolateral to the trachea, and ventral to the esophagus. The right
recurrent laryngeal nerve is located lateral to the trachea and dorsomedial to
the sternothyroideus muscle [5].

Preoperative evaluation

Elevated circulating thyroid hormone levels should be conﬁrmed by lab-

oratory analysis. Another method of documenting hyperthyroidism is tech-
netium 99m radionuclide imaging [6]. This test documents hyperthyroidism,
the distribution of the abnormal tissue (one gland or both), the presence of
functional ectopic thyroid tissue, and metastatic thyroid carcinoma.
Unfortunately, the equipment necessary for this imaging modality limits its
availability.

Assessment of the systemic health of the patient is important. Preoperative

blood work and urinalysis should be performed to reveal any concurrent met-
abolic abnormalities. More than 90% of hyperthyroid cats have an increase in
liver-associated enzymes, and more than 20% have a high blood urea nitro-
gen or creatinine level [7]. Because the typical feline hyperthyroid patient is
geriatric, concurrent metabolic or neoplastic abnormalities are common.

Electrocardiograms and thoracic radiographs are recommended before

anesthesia. Hyperthyroidism has been found to lead to tachycardia, gallop
rhythm, cardiomegaly, and heart murmurs in many cats [7,8]. This has been
attributed to increased sympathetic tone, hypertrophic cardiomyopathy in
response to increased thyroid hormones, and cardiac changes that compen-
sate for altered peripheral tissue function [7,8]. Ventricular arrhythmias and

Fig. 1. Intraoperative view of a unilateral thyroid tumor in a mature cat. The histologic
diagnosis was multinodular adenomatous hyperplasia. The external parathyroid gland (PT) is
located cranially (to the left).

853

S. Padgett / Vet Clin Small Anim 32 (2002) 851–859

heart failure secondary to cardiac disease are not common but are seen more
frequently in older hyperthyroid cats. Thoracic radiographs are also useful
to demonstrate obvious metastatic disease from thyroid carcinoma.

Both tachycardia and ventricular arrhythmias should be controlled before

anesthesia. This can usually be achieved by controlling the hyperthyroidism
with methimazole (10–15 mg/kg of body weight administered orally) [3] for 2
to 4 weeks before surgery. This is not appropriate for cats that have demon-
strated adverse eﬀects to this medication. Propranolol therapy (0.5 mg/kg
administered orally 2–3 times a day) for 3 to 5 days before surgery has also
been suggested to decrease anesthetic risk associated with cardiac changes
[5]. Propranolol not only decreases heart rate but has also been shown to
lower serum T3 levels [9]. This drug should be used with extreme caution
in cats with congestive heart failure because of the negative inotropic eﬀects
of propranolol.

Anesthetic considerations

Premedications containing acepromazine (0.1 mg/kg administered intra-

muscularly) are commonly recommended to decrease autonomic tone in the
hyperthyroid patient, thereby decreasing the potential for arrhythmias [10].
Drugs that potentiate arrhythmias should be avoided because of the fre-
quency of hypertrophic cardiomyopathy, tachycardia, and arrhythmias.
Atropine should be avoided in the hyperthyroid patient because it may
induce arrhythmias. Ketamine should not be used because it sensitizes the
heart to catecholamine-induced arrhythmias [11]. Isoﬂurane via mask and
propofol are suggested for use as induction agents.

An electrocardiogram should be used during anesthesia to monitor for

arrhythmias. Support of circulating volume and renal function via careful
administration of a balanced electrolyte solution is prudent.

Surgical technique

Surgical preparation of the entire ventral cervical area to the level of the

thoracic inlet should be performed, because the enlarged thyroid can travel
caudally as a result of gravity. The patient is positioned in dorsal recumbency
with the forelimbs pulled caudally. The neck should be slightly extended. The
approach is via a ventral midline incision caudal to the larynx. Blunt dissec-
tion in the midline between the strap muscles of the neck (sternohyoideus and
sternothyroideus) is performed. The trachea is visualized, and blunt dis-
section is undertaken in the paratracheal fascia lateral to the proximal
trachea. Care is taken not to damage the tracheal vascularity, which occurs
as a pedicle on the lateral aspects close to the trachea. The recurrent laryngeal
nerve and the carotid sheath should be identiﬁed and preserved. If the thy-
roid glands are not found immediately caudal to the larynx, the paratracheal
area should be explored to the level of the thoracic inlet.

854

S. Padgett / Vet Clin Small Anim 32 (2002) 851–859

If no preoperative radionuclide scan has been performed, careful inspec-

tion of both thyroid glands is necessary to determine whether the disease is
unilateral or bilateral. Bilateral disease is manifested as two enlarged glands.
Unilateral disease should lead to atrophy of the normal gland as a result of
negative feedback induced by the autonomous hormone production from
the abnormal gland. If the thyroid gland opposite an enlarged gland seems
to be normal, it is almost certainly hyperplastic as well. In this case, the
normal-appearing gland should at least be biopsied if not removed. Ap-
proximately 70% of hyperthyroid cats have bilateral disease [7,8]. Diﬀerent
methods of removing the thyroid glands are reviewed below. Resected thy-
roid tissue should always be submitted for histopathologic examination.
Routine closure of the neck muscles and skin is then performed.

Surgical thyroidectomy techniques diﬀer primarily according to the meth-

od by which parathyroid function is preserved. No matter which of the fol-
lowing techniques is used, the underlying principles of meticulous dissection,
careful attention to preserving vascularity, and avoiding other vital struc-
tures in the area are crucial.

Extracapsular thyroidectomy technique

This technique involves no dissection of the thyroid parenchyma. The

vascularity to the thyroid gland is ligated, the parathyroid gland is sharply
dissected from the capsule of the thyroid, and the entire thyroid gland (with
capsule) is removed [12]. This technique is not recommended because of the
high incidence of postoperative complications.

Modiﬁed extracapsular thyroidectomy technique

This technique is similar to the extracapsular technique except that the

external parathyroid gland vascularity is preserved, thereby decreasing post-
operative morbidity. Using a ﬁne-tipped electrode, the thyroid capsule is
cauterized around the external parathyroid gland and its vascularity, with
at least 2-mm margins. A number 11 scalpel blade or ﬁne scissors are used
to cut through the cauterized area. The rim of thyroid capsule containing the
external parathyroid gland and vessel is dissected free from the thyroid
parenchyma, which is still encompassed by the rest of the thyroid capsule.

Intracapsular thyroidectomy technique

This technique minimizes damage to the blood supply of the external para-

thyroid gland by leaving the thyroid capsule in place. A small incision is
made in an avascular area of the thyroid capsule. Using a cotton-tipped
applicator or hemostat, meticulous blunt dissection is performed to remove
the thyroid gland from its capsule. It is particularly important not to dam-
age the area of the capsule associated with the external parathyroid gland
and its associated vascularity. Any pieces of the thyroid remaining in the
capsule should be carefully removed, because adenomatous tissue left within
the capsule has been associated with hyperthyroid recurrence [12–14].

855

S. Padgett / Vet Clin Small Anim 32 (2002) 851–859

Modiﬁed intracapsular thyroidectomy technique

This technique diﬀers from the routine intracapsular method by removing

more of the thyroid capsule, leaving only the capsule associated with the
external parathyroid gland and its blood supply. After the thyroid gland
is carefully removed from the capsule, a number 15 scalpel blade or ﬁne scis-
sors are used to create a peninsula of capsular tissue containing only the
gland and its blood vessel. The remainder of the capsule is excised, decreas-
ing the likelihood of adenomatous tissue remaining in situ.

Staged bilateral removal technique

To avoid damage to the two remaining external parathyroid glands during

a bilateral thyroidectomy, some authors suggest a staged unilateral removal
[5]. The time between surgeries allows resolution of transient damage to the
parathyroid gland or its vascularity. The surgeries are staged by at least 3
weeks [5]. Although this technique does decrease the incidence of postoper-
ative hypocalcemia, there is added risk as a result of two anesthetic events.

Parathyroid autotransplantation

In the event there is inadvertent removal or complete devascularization of

the external parathyroid gland during surgery, autotransplantation can save
the function of the parathyroid gland. This is achieved by mincing the para-
thyroid gland into approximately 1-mm cubes and inserting them into a stab
incision in one of the local neck muscles [15]. This allows revascularization
and minimizes the duration and severity of postoperative hypocalcemia
[15,16]. This has been recommended as routine procedure [17], which may
not be prudent, because there is a chance that diseased thyroid may be trans-
planted with the parathyroid gland [16].

Postoperative complications and management

Hypocalcemia

The most serious complication of bilateral thyroidectomy is postopera-

tive hypocalcemia as a result of acute hypoparathyroidism, which is
reported in 11% to 82% of cats depending on the method of thyroidectomy
(Table 1) [12,14,18]. Hypoparathyroidism usually develops because of inad-
vertent removal of all parathyroid glands or disruption of the vascular sup-
ply to the remaining parathyroid glands.

Hypocalcemia usually occurs 24 to 72 hours after surgery. Not all hypo-

calcemic patients warrant therapy, and the decision to treat is based on the
severity of clinical signs. Clinical signs of hypocalcemia can be mild (irritabil-
ity and decrease in appetite) or severe (ear and facial twitching, seizure-like
muscular tetany). Clinical signs do not usually occur until total serum calcium
is less than 6.5 mg/dL [5]. It is recommended that serum calcium concentra-
tions be measured at least once 24 hours after bilateral thyroidectomy.

Mild signs of hypocalcemia can be treated with oral supplementation

alone. Cats showing moderate to severe signs of hypocalcemia should have

856

S. Padgett / Vet Clin Small Anim 32 (2002) 851–859

their calcium levels conﬁrmed by laboratory analysis if possible. If immedi-
ate results are not possible, clinical signs alone warrant treatment, because
severe hypocalcemia can be fatal. Moderate to severe hypocalcemia is trea-
ted by slow intravenous infusion of calcium gluconate (1 mL of a 10% cal-
cium gluconate solution containing 9.3 mg of ionized calcium administered
over 10 minutes). This should be followed by intravenous infusion of a cal-
cium gluconate solution (10 mL of a 10% calcium gluconate solution in 250
mL of 0.9% NaCl at 2.5 mL/kg/h for 8 to 12 hours [5,11,15]. If serum cal-
cium concentrations decrease after discontinuing the slow intravenous calci-
um infusion, another 8 to 12 hours of therapy is indicated.

Oral supplementation of calcium (500–700 mg/kg of calcium gluconate

administered orally and divided into 3 doses per day) [5] should be started
as well as a vitamin D analogue (dihydrotachysterol solution at a rate of
0.03 mg/kg/d for 3 days and then decreased to 0.01–0.02 mg/kg/d) [10].
Dihydrotachysterol often does not show an eﬀect until 48 to 72 hours after
starting therapy. As the serum calcium concentrations increase over a period
of days to weeks, the oral medications can be tapered. Although some
patients require lifelong oral calcium therapy, most do not need oral supple-
mentation for more than 3 to 6 weeks.

Azotemia

Many patients with hyperthyroidism have concurrent renal insuﬃciency.

It is possible that treatment of hyperthyroidism may lead to clinical evidence
of a previously masked chronic renal insuﬃciency, because treatment has
been shown to decrease mean glomerular ﬁltration rate and to increase
serum creatinine and blood urea nitrogen concentrations [19,20]. It may
be prudent to achieve temporary euthyroidism via medical management and
assess the degree of azotemia before achieving permanent euthyroidism with
surgical therapy.

Nerve damage

The recurrent laryngeal nerve can be damaged on exploration of the para-

tracheal fascia, leading to voice change or laryngeal paralysis. For this rea-
son, it is important to know the normal anatomy of the area and strive to

Table 1
Incidence of hypocalcemia after bilateral thyroidectomy

Bilateral Thyroidectomy
Technique

Incidence of Hypocalcemia

Reference

Intracapsular

15% (8/53)

22% (11/50)

36% (7/19)

Modiﬁed intracapsular

33% (10/30)

Extracapsular

82% (9/11)

Modiﬁed extracapsular

23% (6/26)

Staged intracapsular

11% (1/11)

857

S. Padgett / Vet Clin Small Anim 32 (2002) 851–859

avoid undue tension or trauma to the recurrent laryngeal nerve. Laryngeal
paralysis can be temporary if the nerve is mildly damaged. If clinical signs in
this area are caused by permanent laryngeal paralysis, surgical correction is
required.

Horner’s syndrome can also occur if the sympathetic trunk is damaged.

Careful retraction of the cervical musculature should minimize this risk.

Recurrent hyperthyroidism

Patients undergoing bilateral thyroidectomy can experience recurrence of

disease months to years after surgical treatment [13,14]. This has been re-
ported to occur more commonly with the intracapsular technique [14].
Recurrence is attributed to leaving nests of abnormal thyroid tissue at the
surgical sight, which become functional after hypertrophy [13,14], and to
adenomatous ectopic thyroid tissue [13]. Alternatively, metastatic thyroid
carcinoma could cause recurrence of disease. The best method of diagnosing
the location of the abnormal tissue is a technetium 99m thyroid scan.

Hypothyroidism

Although it has been suggested that cats undergoing bilateral thyroidec-

tomy should receive thyroid hormone replacement, most patients show no
clinical signs referable to hypothyroidism. Hormone replacement therapy
is not necessary in most patients.

References

[1] Peterson ME, Johnson GF, Andrews LK. Spontaneous hyperthyroidism in the cat. In:

Scientiﬁc Proceedings of the American College of Veterinary Internal Medicine [abstract].
Guelph (ON): Aqim College; 1979, p. 108.

[2] Turrel JM, Feldman EC, Nelson RW, et al. Thyroid carcinoma causing hyperthyroidism in

cats: 14 cases (1981–1986). JAVMA 1988;193:359–64.

[3] Kintzer PP. Considerations in the treatment of feline hyperthyroidism. Vet Clin North Am

Small Anim Pract 1994;24:577–85.

[4] Peterson ME, Kintzer PP, Hurvitz AI. Methimazole treatment of 262 cats with hyper-

thyroidism. J Vet Intern Med 1988;2:150–7.

[5] Flanders JA. Surgical therapy of the thyroid. Vet Clin North Am Small Anim Pract

1994;24:607–21.

[6] Peterson ME, Becker DV. Radionuclide thyroid imaging in 135 cats with hyperthyroidism.

Vet Radiol Ultrasound 1984;25:23–7.

[7] Liu S, Peterson ME, Fox PR. Hypertrophic cardiomyopathy and hyperthyroidism in the

cat. JAVMA 1984;185:52–7.

[8] Peterson ME, Kintzer PP, Cavanagh PG, et al. Feline hyperthyroidism: pretreatment

clinical and laboratory evaluation of 131 cases. JAVMA 1983;183:103–10.

[9] Foster DJ, Thoday KL. Use of propranolol and potassium iodate in the presurgical

management of hyperthyroid cats. J Small Anim Pract 1999;40:307–15.

[10] Feldman EC. Feline hyperthyroidism (thyrotoxicosis). In: Feldman EC, Nelson RW,

editors. Canine and Feline Endocrinology and Reproduction. 2nd edition. Philadelphia:
WB Saunders; 1996. p. 118–66.

858

S. Padgett / Vet Clin Small Anim 32 (2002) 851–859

[11] Muir W, Hubbell JAE. Handbook of veterinary anesthesia. St. Louis: CV Mosby; 1989.

p. 74–86.

[12] Flanders JA, Harvey HJ, Erb HN. Feline thyroidectomy: a comparison of postoperative

hypocalcemia associated with three diﬀerent surgical techniques. Vet Surg 1987;16:362–6.

[13] Swalec KM, Birchard SJ. Recurrence of hyperthyroidism after thyroidectomy in cats. J Am

Anim Hosp Assoc 1990;26:433–7.

[14] Welches CD, Scavelli TD, Matthiesen DT, et al. Occurrence of problems after three

techniques of bilateral thyroidectomy in cats. Vet Surg 1989;18:392–6.

[15] Flanders JA. Surgical treatment of hyperthyroid cats. Mod Vet Pract 1986;67:711–5.
[16] Padgett SL, Tobias KM, Leathers CW, et al. Eﬃcacy of parathyroid gland autotrans-

plantation in maintaining serum calcium concentrations after bilateral thyroparathy-
roidectomy in cats. J Am Anim Hosp Assoc 1998;34:219–24.

[17] Norsworthy GD. Feline thyroidectomy: a simpliﬁed technique that preserves parathyroid

function. Vet Med 1995;90:1055–63.

[18] Birchard SJ, Peterson ME, Jacobson A. Surgical treatment of feline hyperthyroidism:

results of 85 cases. J Am Anim Hosp Assoc 1984;20:705–9.

[19] Dibartola SP, Broome MR, Stein BS, et al. Eﬀect of treatment of hyperthyroidism on renal

function in cats. JAVMA 1996;208:875–8.

[20] Graves TK, Olivier NB, Nachreiner RF, et al. Changes in renal function associated with

treatment of hyperthyroidism in cats. Am J Vet Res 1994;55:1745–9.

859

S. Padgett / Vet Clin Small Anim 32 (2002) 851–859

Feline gastrointestinal foreign bodies

Trevor N. Bebchuk, DVM

Department of Small Animal Clinical Sciences, Western College of Veterinary Medicine,

University of Saskatchewan, 52 Campus Drive, Saskatoon, Saskatchewan, S7N 5B4 Canada

Gastrointestinal foreign bodies are responsible for a wide range of clinical

presentations in veterinary practice. Cats may demonstrate vague mild signs
of a chronic nature, they may have acute or severe vomiting and diarrhea, or
they may be in hypovolemic or septic shock. The clinical signs vary depend-
ing on the location of the obstruction, the degree of obstruction, the foreign
body causing the obstruction, and the chronicity of obstruction. The
required treatment depends on numerous clinical and laboratory variables,
which must be evaluated before developing a treatment plan. The goal of
treatment is always relief of the obstruction with minimum morbidity. This
article focuses on some of the diﬀerent types of obstructions reported in cats,
how they aﬀect the animal at diﬀerent levels of the gastrointestinal tract,
diagnostic techniques, and the recommended treatments.

The vast range of clinical presentations possible with gastrointestinal for-

eign bodies precludes any speciﬁc generalizations regarding the appearance
of a cat with this problem. Clinical signs that may be present include vomiting,
diarrhea, regurgitation, ptyalism, inappetence, anorexia, depression, dehy-
dration, abdominal pain, abdominal distention, palpable ﬁrm segments of the
intestines, palpable intestinal dilation, and many more. It is imperative that a
complete physical examination, including an oral examination, be performed.
It is recommended that laboratory assessment, including routine hematology
tests, a serum biochemical proﬁle, and urinalysis, be performed. Radiographs
are essential for the diagnosis of most gastrointestinal foreign bodies, and a
contrast study is necessary in some cases. Obtaining both left and right lateral
recumbent radiographs of the abdomen may prove beneﬁcial in some animals.
Fluid and gas contents are extremely mobile and tend to move to the depen-
dent portion of the stomach during postural changes. The redistribution of
gas can act as a negative contrast medium to highlight foreign objects or dis-
orders only visible on one lateral projection [1]. The same principle applies to

Vet Clin Small Anim 32 (2002) 861–880

E-mail address: trevor.bebchuk@usask.ca (T.N. Bebchuk).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 3 0 - X

the intestinal tract. Ultrasonography is useful for the diagnosis of certain for-
eign body obstructions. Ultimately, either endoscopy or surgery is required
for deﬁnitive diagnosis and treatment of foreign body obstruction.

The most common hematologic abnormalities in cats with foreign body

obstruction range from a leukocytosis with or without a mild left shift to
a degenerative left shift in cases with peritonitis from intestinal perforation.
Dehydration and electrolyte imbalance are expected in animals that are
vomiting. These animals may have a normal pH or primary metabolic
acidosis. A normal pH results from equal loss of gastric acid secretions and
base secretions from the proximal duodenal, bile, and pancreatic juices. Pri-
mary metabolic acidosis is caused by a relatively greater loss of base secre-
tions from the upper intestinal tract, and lactic acidosis results from
dehydration and inadequate perfusion of splanchnic viscera, skin, and
muscle [2]. Cats with pyloric obstruction may demonstrate a hypokalemic,
hypochloremic, metabolic alkalosis. The dehydration and laboratory abnor-
malities should be corrected by appropriate intravenous ﬂuid therapy, which
should be initiated before surgical intervention. Deﬁnitive correction
requires removal of the inciting cause of the vomiting and inﬂammation.

Foreign body location

Esophagus

The esophagus is responsible for the transport of food, water, and saliva

from the pharynx to the stomach. It is frequently under tension during the act
of swallowing and bolus formation. Unlike the intestines, the esophagus does
not have a serosal surface, and this may delay early ﬁbrin sealing of entero-
tomy sites compared with the rest of the intestines. Vascular supply to the
esophagus is segmental, with the cervical portion supplied by branches of the
thyroid and subclavian arteries and the thoracic portion supplied by the bron-
choesophageal arteries and segmental branches of the aorta [3].

Esophageal obstruction is less common than other gastrointestinal

obstructions. When it occurs, it can be complete or partial, each of which
demonstrates diﬀerent clinical signs and sequelae. The most common clini-
cal signs are dysphagia and regurgitation depending on the level of obstruc-
tion. When the obstruction is incomplete, signs of chronic wasting, such as
emaciation, may be observed. If aspiration has occurred as a result of the ob-
struction, abnormal pulmonary sounds, such as crackles, may be auscultable,
and clinical signs, including coughing, mucopurulent nasal discharge, and
fever, may be observed. If a foreign body has perforated the esophagus, a
secondary mediastinitis or pyothorax may develop [4]. Esophagobronchial
ﬁstula formation secondary to a perforating foreign body can occur, leading
to secondary pulmonary pathologic ﬁndings [5].

There are three narrowed anatomic regions in the esophagus where a for-

eign body is likely to become lodged. These include the cricopharyngeal

862

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

sphincter, the base of the heart, and the esophageal hiatus in the diaphragm.
Additionally, foreign bodies may lodge at the thoracic inlet, where advance-
ment may be impeded by surrounding soft tissues [4].

Diagnosis of esophageal foreign body obstruction is based on clinical

signs, a normal pharyngeal examination, and radiographs. In most cats, the
esophagus is not normally visible on radiographs. When distended with air
or ﬂuid, it becomes a visible structure. Esophageal dilatation can be evi-
dence of an esophageal foreign body if the cat has no previous history of
megaesophagus. The presence of a radiopaque object (Fig. 1) in the region
of the esophageal lumen provides a deﬁnitive diagnosis of foreign body
obstruction, but not all foreign objects are radiographically visible. An
esophagram can be performed to visualize radiolucent objects (Fig. 2).
When an esophagram is performed, there is a risk of aspiration, and ﬂuid
barium rather than barium paste can be used to reduce the risk [3]. A ﬂuo-
roscopic examination, in conjunction with a contrast esophagram, should be
used when possible because it allows assessment of swallowing, esophageal
motility, and gastroesophageal sphincter function. If there is evidence of
periesophageal gas or ﬂuid accumulation, mediastinal eﬀusion, or pleural
eﬀusion, there may be an esophageal rupture, and barium should not be
used. In these cases, aqueous iodine or iohexol should be used [6].

Removal of esophageal foreign bodies should be considered an emer-

gency procedure, because the longer an object remains in the esophagus, the
greater is the risk of aspiration and esophageal wall injury by pressure
necrosis. Most esophageal foreign bodies can be deﬁnitively diagnosed and
removed endoscopically [4,7]. This includes ﬁshhooks, some of which may
be large or embedded in the esophageal wall [8]. The equipment needed to
be successful at endoscopic removal of esophageal foreign bodies includes
but is not limited to a rigid proctoscope or ﬂexible ﬁberoptic endoscope, a
light source, long blunt-ended grasping forceps, ﬂexible alligator forceps,
and Foley catheters. In the event a foreign body is located in the caudal
thoracic esophagus and cannot be removed endoscopically, an attempt
should be made to gently push it into the stomach. This should only be per-
formed if it is possible without causing further injury to the esophagus.
When successful, this technique allows for a laparotomy and gastrotomy
to remove the foreign body rather than the higher morbidity thoracotomy
and esophagotomy. When perforations are present or when removal of a
foreign body endoscopically carries a high risk of perforation, surgical
removal via esophagotomy is recommended. With some foreign bodies, such
as embedded ﬁshhooks, a combined surgical and endoscopic approach may
be used. A surgical approach is made, and without an esophagotomy, the
portion of the hook protruding through the esophageal wall is cut and
removed. The endoscope is then used to retrieve the rest of the hook that
remains within the lumen [8].

The surgical approach to the cervical esophagus is via a ventral midline in-

cision, and the thoracic esophagus is approached via a left lateral intercostal

863

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

thoracotomy at the appropriate intercostal space for the level of the obstruc-
tion. At the level of the heart base, where the aorta pushes it to the right, the
esophagus can be approached via a right lateral intercostal thoracotomy
at the fourth or ﬁfth intercostal space. The esophagus is then packed oﬀ from
the rest of the neck or thorax using moistened laparotomy sponges. Ideally,

Fig. 1. Right lateral (A) and ventrodorsal (B) projections of the neck and thorax of a 1-year-old
female domestic shorthair cat. A 5-cm needle was ingested, punctured the pharynx, and is now
lying ventral to the larynx and trachea extending to the level of the ﬁfth cervical vertebral body.

864

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

Fig. 2. Right lateral (A) and ventrodorsal (B) projections of an esophagram in a 3-year-old
castrated male domestic longhair cat. There is a large ﬁlling defect in the thoracic esophagus
extending from the fourth thoracic vertebral body to the diaphragm. This is best viewed on the
lateral projection. This foreign body was not visible on survey radiographs. It was determined
to be clumps of hair and vegetation and was removed endoscopically.

865

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

the cranial esophagus can be suctioned of contents before the esophagotomy
incision, but if this is not possible, an assistant can hold the esophagus
closed cranial and caudal to the proposed esophagotomy site to prevent con-
tamination from leakage. If the esophageal wall over the foreign body looks
healthy, the incision can be made over the object and should be made large
enough to remove the foreign body with no additional damage to the esoph-
ageal wall. If the esophagus seems compromised, the incision should be
made aboral to the foreign body, beginning at the foreign body but large
enough to remove the object with minimal manipulation. Once the foreign
object has been removed, the esophageal mucosa is examined for any evi-
dence of perforation. The esophagotomy incision can be closed in one or
two layers. The ﬁrst layer is an appositional pattern in the mucosa and sub-
mucosa, and the second layer is an appositional pattern in the muscularis.
The holding layer is the submucosa. If the esophageal wall does not look
healthy, a resection and anastomosis should be performed on the damaged
region. Anastomosis of the esophagus can be performed with a one- or two-
layer closure. Suture material for esophageal surgery in cats should be a syn-
thetic, monoﬁlament, absorbable suture, such as polydioxanone, in a 4-0
size and with a swaged on reverse cutting needle. If there is still concern rel-
ative to leakage of the esophagus at the site of the anastomosis or esopha-
gotomy, an omental patch can be brought through the diaphragm and
wrapped over the anastomosis, a pericardial reinforcement can be applied,
or a pedicle intercostal graft can be used [9,10]. Anastomosis of the esoph-
agus should not be under tension, because the esophagus is constantly in
motion as a result of swallowing and respiration. The anastomosis may
dehisce if the tension is excessive. The major complication of surgery for
esophageal perforation is infection, and this occurred in 57% of cases in one
study with and without dehiscence [5]. In the postoperative period, a phar-
yngostomy tube can be used to provide caloric and ﬂuid requirements; how-
ever, this is controversial, because the presence of the intraluminal tube may
impair esophageal healing [5].

Stomach

Foreign bodies of the stomach are common and may be an incidental

ﬁnding in some cases (Fig. 3). Clinical signs of gastric foreign bodies range
from asymptomatic to intermittent or persistent vomiting as a result of out-
ﬂow obstruction, gastric distention, and mucosal irritation. Vomiting is
more common with foreign bodies in the pyloric antrum, because distention
or noxious stimulation of the duodenum and/or pyloric antrum stimulates
vomiting, whereas similar distention of the fundus does not [3]. Cats com-
monly ingest string, yarn, and other string-like material when they play.
This can result in a linear foreign body, which is frequently anchored under-
neath the tongue or at the pylorus, causing intestinal plication. A gastric for-
eign body is generally not an emergency unless it is a linear foreign body or

866

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

Fig. 3. Right lateral (A) and ventrodorsal (B) projections of a gastric foreign body (vintage
Canadian nickel) discovered as an incidental ﬁnding when obtaining radiographs of a 10-year-
old cat after trauma.

867

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

the foreign object is lodged in the pylorus and causing obstruction and
severe vomiting.

Young animals are presented with gastric foreign bodies more frequently

than older animals, and this should be a diﬀerential diagnosis for any kitten
that has clinical signs of vomiting. The cats are presented with a history of
intermittent or persistent vomiting or a more chronic history of inappetence/
anorexia and depression. Physical examination of these cats may be unre-
markable; however, many have some level of dehydration or abdominal
pain, and a gastric foreign body is palpable in some cats. The stomach is cra-
nially located in the abdomen and is shielded by the caudal costal arches,
making routine palpation of gastric foreign bodies diﬃcult. If a linear for-
eign body is present, plicated intestines may be palpable.

Laboratory abnormalities most commonly include evidence of dehydra-

tion. For example, an elevated hematocrit as well as elevated blood urea
nitrogen, creatinine, and total protein levels can be expected. In cases of
severe vomiting as a result of pyloric obstruction, a hypochloremic and
hypokalemic metabolic alkalosis may be present. In cases of vomiting with-
out pyloric obstruction, a metabolic acidosis would be expected as a result
of losses of base-rich duodenal and pancreatic secretions as would dehydra-
tion and lactic acidosis.

Radiopaque foreign bodies may be diagnosed radiographically; however,

this is not always the case. Radiolucent foreign bodies may require a con-
trast gastrogram for diagnosis. This can be performed using barium; how-
ever, if an esophageal, gastric, or intestinal rupture is suspected, aqueous
iodine or iohexol should be used [6]. As in the case of the esophagus, many
gastric foreign bodies can be diagnosed and removed endoscopically. Con-
trast radiographs and endoscopy allow the clinician to discern between for-
eign bodies and other causes of vomiting, such as gastric neoplasia and
gastric ulceration. An additional tool for the diagnosis of a gastrointestinal
foreign body is ultrasonography. Using ultrasound, foreign bodies of the
stomach and intestines may be identiﬁed, and many foreign bodies have a
characteristic ultrasonographic appearance depending on the tendency to
transmit or attenuate the ultrasound beam [11]. If the foreign body is
smooth and rounded, vomiting can be induced with xylazine (Rompun) at
a dose of 1 mg/kg of body weight [12]. This should only be attempted with
objects that can be expulsed with no harm to the esophagus and no risk of
lodging in the esophagus.

Foreign bodies that have sharp edges or are large should not be removed

by endoscopy because of the risks of esophageal laceration and lodging the
foreign body in the esophagus. These foreign bodies are best removed by
gastrotomy [12]. Radiographs should be taken just before surgery to ensure
that the object has not moved from the stomach. The surgical approach to
the stomach for gastrotomy is via a cranial ventral midline laparotomy.
Because the intestines and stomach can contain foreign bodies concurrently,
a thorough exploratory laparotomy should be performed. The stomach is

868

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

packed oﬀ from the rest of the abdomen with moistened laparotomy
sponges, and stay sutures are placed between the greater and lesser curva-
tures, near the cardia, and in the pyloric antrum. The gastrotomy incision
is created between these stay sutures in the hypovascular region between the
greater and lesser curvatures on the ventral surface of the stomach. The inci-
sion should not be made too close to the pylorus to prevent excessive nar-
rowing of the gastric lumen. Closure of the gastrotomy can be performed
in one or two layers. If a two-layer closure is used, the ﬁrst layer is an appo-
sitional pattern that must incorporate the submucosa, and the second layer
is an inverting pattern in the serosa and muscularis [12]. Suture material
should be synthetic, monoﬁlament, absorbable, and size 3-0 or 4-0.

In the postoperative period, hydration status and electrolyte levels should

be monitored. Intravenous ﬂuid therapy should be continued and adjusted to
address any abnormalities. If the animal has been anorexic and vomiting for a
sustained period, hypokalemia is expected. This can be treated with intrave-
nous ﬂuids containing 20 to 40 mEq/L of potassium chloride. The animal
should not receive more than 0.5 mEq/kg/h of potassium chloride. If vomiting
continues, treatment with an antiemetic may be necessary. If vomiting has
ceased, the cat should be started on a bland diet 12 to 24 hours after surgery.

The prognosis for removal of gastric foreign bodies via gastrotomy is

good. Recovery can be complicated by local or generalized peritonitis if
there is gastric perforation present or if spillage of gastric contents occurs
during gastrotomy. The latter is uncommon if moist laparotomy sponges are
used eﬀectively to isolate the stomach from the rest of the abdomen before
making the gastrotomy incision.

Small intestine

Small intestinal obstruction by a foreign body is a common condition in

cats. This obstruction can be complete or partial. The clinical signs vary in
severity as a result of the level and degree of obstruction. The location of the
obstruction also contributes to variability in the presenting clinical signs.
Common clinical abnormalities include vomiting, anorexia, depression, and
abdominal tenderness. Many intestinal foreign bodies can be detected with
careful abdominal palpation. Results of abdominal palpation include intes-
tinal distention, a palpable object, and abdominal pain.

The diagnosis of foreign body obstruction is usually made radiographi-

cally. The classic radiographic sign of mechanical obstruction is the presence
of multiple loops of gas-ﬁlled small intestine of various diameters. A small
intestinal diameter greater than 1.6 times the depth of the midcentrum of the
ﬁfth vertebra has been used as a predictor of intestinal obstruction in dogs
[13]. A similar ratio may be useful in cats. The dilation may not be present to
the same degree in cases of partial intestinal obstruction. If the object is
radiopaque, it can be identiﬁed on plain radiographs (Fig. 4); however,
many foreign objects are radiolucent and require contrast radiography for

869

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

identiﬁcation (Fig. 5) [14]. Some foreign objects, such as various fruit seeds
and corn cobs, although nonopaque, can be identiﬁed on survey radio-
graphs because of their characteristic shape and contained gas lucencies
[14]. Liquid barium or barium paste in food can be used if there is no sus-
picion of gastrointestinal perforation. If perforation is suspected, aqueous
iodine or iohexol should be used [6]. Most proximal small intestinal obstruc-
tions are visible within 6 hours, whereas a 24-hour study may be required for
more distal obstructions [15,16]. The presence of a nonopaque foreign body
is seen as a ﬁlling defect in the intestinal lumen or as a complete obstruction
to the ﬂow of barium in some cases. Another imaging modality that can be
used to diagnose and characterize intestinal foreign bodies is ultrasono-
graphy [11,17]. Using ultrasonography, gastrointestinal motility can be
assessed; when increased, it often signals the location of a mechanical
obstruction. The presence of ﬂuid/gas distention also indicates the location
of the foreign object, and some objects may have characteristic acoustic sig-
nals. Ultrasonography may not accurately predict the presence of intestinal
perforation even when wall thickness is measured [11].

The treatment of intestinal foreign body obstruction is surgical. The sur-

gical approach is via a ventral midline laparotomy extending from the
xiphoid to the pubis. The entire intestinal tract should be explored to deter-
mine if there are multiple foreign bodies and to assess if the object caused
any intestinal trauma in transit. If the bowel segment containing the foreign
body is healthy, the foreign object can be removed through an antimesen-
teric enterotomy incision just aboral to the obstruction. This placement
ensures that the intestine excised is healthy. There is potential for the intes-
tine immediately overlying the foreign body to be compromised as a result
of pressure necrosis of the intestinal wall contacting the object. Enterotomy
proximal to the obstruction is not recommended, because distention with
gas and ﬂuid and the passage of the foreign body may have caused some
degree of vascular compromise [2,18]. The enterotomy incision is made large
enough to manipulate the foreign material out of the intestinal lumen with-
out causing further intestinal trauma. This usually requires an incision the
length of the diameter of the obstructing object [18]. The enterotomy inci-
sion is then closed with size 4-0, synthetic, monoﬁlament, absorbable suture
material, such as polydioxanone, in a simple continuous or simple interrup-
ted appositional pattern [19,20]. If the bowel segment demonstrates evidence
of necrosis, such as a thin intestinal wall and dark discoloration, a resection
and anastomosis should be performed. End-to-end anastomosis can be
accomplished using a simple interrupted appositional pattern or a modiﬁed
simple continuous appositional pattern with the same type of suture material
used for enterotomy closure [20,21]. The modiﬁed simple continuous pattern
is performed by ﬁrst placing two separate sutures at the mesenteric and anti-
mesenteric borders. They are tied, leaving a 3- to 4-cm end for a stay suture,
to which mosquito forceps are attached. The needle end of the suture is then
used to complete the anastomosis. One strand advances the perimeter of the

870

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

Fig. 4. Right lateral (A) and ventrodorsal (B) radiographic projections of the abdomen of a 2-
year-old male domestic shorthair cat after ingestion of ﬁshing tackle. Three radiopaque foreign
bodies are visible, including two hooks (one with leader attached) and a lead ﬁshing weight.

871

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

Fig. 5. Right lateral (A) and ventrodorsal (B) survey radiograph projections of a 3-year-old cat
with a 4-day history of vomiting and depression. The stomach is distended with gas, and there is
a large and primarily ﬂuid-ﬁlled bowel loop in the caudal right abdomen. (C,D) A barium
contrast gastrointestinal study in the same cat. This contrast study conﬁrms the presence of
foreign body obstruction of the proximal jejunum, and the distended bowel seen on the survey
radiographs is proximal to the obstruction. Complete obstruction is conﬁrmed by the fact that
the barium does not ﬂow past the lesion. There is mild gastroesophageal reﬂux.

872

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

intestine with bites 2 to 3 mm apart from mesenteric to antimesenteric knots
and a square knot tied to the tagged end of the knot at the antimesenteric
border. The other strand is advanced on the other side in the opposite di-
rection and tied in the same fashion [20]. After the anastomosis has been

Fig. 5 (continued).

873

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

completed, an omental or serosal patch can be used to aid in sealing the
anastomosis [10,19,22].

A unique form of intestinal obstruction that is seen commonly in cats is

linear foreign body obstruction. This is caused by foreign bodies, such as
string, thread, a nylon stocking, or carpet ﬁbers. One of the most common
causes is sewing thread alone or in combination with sewing needles [23].
The linear object becomes ﬁxed around the base of the tongue or in the pylo-
rus. The intestinal peristaltic waves attempt to move the object aborally, and
the intestine gradually gathers up in a pleated fashion like an accordion on
the foreign object. The linear object becomes imbedded in the mesenteric
border of the small intestine and can erode through the intestinal wall, lead-
ing to leakage of intestinal contents and a local or generalized peritonitis
[24]. Clinical signs with these foreign bodies are generally not severe, because
the obstruction is not complete. Vomiting tends to be less frequent and
severe than with other foreign bodies. If peritonitis develops, a rapid deteri-
oration in the cat’s status may be observed. The linear object is generally not
palpable even with careful abdominal palpation; however, plication and
clumping of the intestine often are observed.

The radiographic signs of a linear foreign body include small intestinal

accordion-like pleating, shortening or gathering of the intestine, increased
luminal gas bubbles, and peritonitis secondary to bowel lacerations [23].
When clumping and plication of the intestine are visible on survey radio-
graphs, this is strong circumstantial evidence for the presence of a linear for-
eign body (Fig. 6). Another common radiographic sign of a linear foreign
body is a pattern of small, eccentrically located, luminal gas bubbles that are
tapered at one or both ends [23]. In a review of 64 cats with linear foreign
bodies, if three or more of these bubbles were seen, a deﬁnitive diagnosis
of a linear foreign body could be made [23]. Both ventrodorsal and lateral
projections should be obtained. The lateral view is preferred to determine
whether intestinal clumping is present, because the intestines are often on
the right side in a ventrodorsal view of a normal cat. The purpose of the ven-
trodorsal projection is to conﬁrm that suspicious gas bubbles are located in
the small bowel. If a barium contrast study is used to conﬁrm the diagnosis,
the abnormal plication of the intestines should be more evident, and the for-
eign object may appear as a linear ﬁlling defect. Once the contrast material
has passed through to the colon, a small amount may remain in the foreign
body, making it look like a linear opaque structure in the intestinal lumen.

On ultrasonographic examination, the bowel may have a ‘‘ribbon candy’’

appearance. Ultrasonography of linear foreign bodies reveals plication or
corrugation of the intestinal tract that can be similar to the ultrasonographic
appearance of intussusception. The two disorders can be diﬀerentiated by
the presence of a hyperechoic structure within the lumen and lack of the wall
layers to form a complete concentric ring [11].

Linear foreign bodies may cause only mild chronic intermittent signs

because of the partial nature of the intestinal obstruction [25]. Continued

874

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

Fig. 6. Right lateral (A) and ventrodorsal (B) survey radiographic projections of a 2-year-old
spayed female domestic shorthair cat with a linear foreign body (sewing thread). There is
clumping of the intestines on the lateral projection, and several eccentric tapered gas bubbles
can be seen in the small intestine. There is generalized decreased detail suggesting some free
peritoneal ﬂuid.

875

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

peristaltic activity may lead to erosion of the mesenteric border intestinal
wall with subsequent local or generalized peritonitis and vascular compro-
mise of the intestinal wall. For this reason, linear foreign body ingestion
should be treated as an emergency condition. Surgical management is the
traditional approach to this disorder. It has been reported that cats with
mild clinical signs, no pyrexia or severe abdominal pain, a linear foreign
body located around the base of the tongue, and evidence of only a mild left
shift on a complete blood cell count can be managed conservatively by cut-
ting free the sublingual attachment of the foreign body and maintaining the
cat in the hospital for observation and intravenous ﬂuid therapy. Contrain-
dications to this conservative management approach include obvious pyloric
anchoring of the linear foreign body, the presence of severe abdominal pain
and pyrexia, radiographic evidence of peritonitis, and a degenerative left
shift on routine hematology testing [24].

Unlike most other intestinal foreign bodies, linear foreign bodies are not

easily removed from a single enterotomy incision. Pulling the object out
through a single proximal enterotomy can cause friction of the object against
the intestinal mesenteric border, and occult perforations may develop. It is
preferable to remove the object in short segments via multiple enterotomies
on the antimesenteric border. To begin, the anchor point, which is usually sub-
lingual or pyloric, must be released. Incising the string under the tongue or
performing a gastrotomy to release a pyloric anchor point accomplishes this.
The linear object can then be removed by one or more intestinal enterotomies.
A technique has been described for removal of linear foreign bodies from a
single enterotomy incision. The procedure is performed by creating a single
enterotomy in the antimesenteric border of the proximal duodenum. The lin-
ear foreign body is tied or sewn to a red rubber catheter, which is advanced
into the duodenum aborally, and the enterotomy is closed. The red rubber
catheter is then milked aborally along the intestine, relieving the plication
as it is advanced, until it is advanced the length of the colon; an assistant can
then retrieve it and the linear foreign body from the anus [26]. This technique
may not be eﬀective for certain linear objects that have caused more severe pli-
cation or are knotted or matted and cannot be advanced aborally [27]. It is dif-
ﬁcult to detect perforations on the mesenteric border should they occur,
because they are shielded by the attachment of the mesentery and mesenteric
vasculature. Where laceration has occurred, inﬂammation and infection at the
site are occasionally walled oﬀ, making removal of the string diﬃcult. In these
cases, the intestine may not resume normal functioning after surgery [12].

At our teaching hospital, we see a combination of local clients and refer-

ral cases. Over the past 10 years, we have treated 113 cats for gastrointesti-
nal foreign bodies, with 2 of these cats presented on two diﬀerent occasions.
Linear foreign bodies caused almost 50% of these cases. Over half of these
cases were caused by needles and thread or thread alone. Other foreign
objects identiﬁed were coins, earplugs, small toys, trichobezoars, an almond,
ﬁshing tackle, and a cork. Cats are susceptible to intestinal obstruction by

876

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

many diﬀerent types of foreign bodies, but linear foreign body obstruction is
clearly the most common type.

Although most feline intestinal foreign bodies can be removed with a

good prognosis, intestinal surgery is not without morbidity. Intestinal dehis-
cence is the most signiﬁcant complication and has been associated with an
80% mortality rate in cats and dogs after intestinal surgery. This compares
with a mortality rate of only 7.2% in animals without dehiscence [28]. In a
series of 121 dogs undergoing small intestinal anastomosis or enterotomy,
dehiscence was identiﬁed in 15.7% of the animals, with a mortality rate of
73.7% [29]. It has also been shown that the survival rate is negatively corre-
lated with multiple intestinal procedures [28]. For these reasons, it is impor-
tant to ensure that only a single enterotomy or resection and anastomosis is
performed whenever possible, only healthy intestine is sutured, and intesti-
nal closure is meticulous.

When extensive intestinal injury results from the passage of a foreign

body or the presence of a linear foreign body, there may be a large or multi-
ple areas of intestinal necrosis. This would require the resection of a large or
multiple segments of the small intestine. This can result in various clinical
signs known as short bowel syndrome. The pathophysiology of this syn-
drome is a result of decreased secretin and cholecystokinin in the proximal
duodenum, thereby decreasing pancreatic and biliary secretions. The loss of
intestinal brush border enzymes also contributes to the syndrome, with the
changes resulting in maldigestion. Decreased intestinal transit time and
decreased mucosal surface area may also contribute to malabsorption. The
decreased transit time, increased lumenal osmotic pressure, bacterial over-
growth, and gastric hypersecretion ultimately result in diarrhea, dehydra-
tion, electrolyte imbalances, and malnutrition. The precise percentage of
small intestinal length that can be removed in cats without causing short
bowel syndrome is not known. In people, 40% to 50% can be removed
safely, but when greater then 75% is removed, nutritional status cannot be
maintained on enteral nutrition alone [30]. In four of ﬁve experimental dogs,
resection of 85% of the small intestine did not aﬀect their ability to live for
11 to 24 months with no special therapy [31]. If massive resection of the
small intestine is required, it must be anticipated that the cat is going to need
nutritional support until adaptive changes in the intestine can occur. These
adaptive changes include increased bowel diameter, crypt and villus mucosal
cell hyperplasia, an increase in villus height and crypt depth, and an increase
in the number of epithelial cells per unit length of the villus. In the interim,
antidiarrheal therapy may be necessary as well as antibacterial therapy to
limit bacterial overgrowth [30].

Large intestinal foreign bodies

Large intestinal foreign body obstruction is exceedingly rare in cats.

Once passed into the colon, most objects are passed in the feces without

877

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

complication. In the rare case where colonic obstruction is identiﬁed, it can
be treated in a similar fashion as small intestinal obstruction. It should be
recognized that the large intestine heals similar to the stomach and small
intestine but that healing is delayed. Wound tensile strength lags behind that
of the small intestine, and suture line failure is more likely. The blood supply
is segmental, there is a large population of bacteria, and solid feces place
more tension on the suture line than the liquid ingesta in the small intestine.
All these factors theoretically lead to greater morbidity and mortality with
colonic surgery [19]. One study, however, did not detect a diﬀerence in dehis-
cence between animals undergoing large or small intestinal surgery [28].

Conclusion

Cats can be aﬀected by foreign body obstruction at all levels of their gas-

trointestinal tract. Foreign bodies can be diagnosed and identiﬁed by a com-
bination of physical examination and palpation techniques, medical
imaging, and endoscopy. Medical imaging can consist of survey radio-
graphs, contrast radiographs, or any combination of the three. Endoscopy
is useful for the identiﬁcation and removal of both esophageal and gastric
foreign bodies. If possible, esophageal surgery should be avoided. Small
intestinal foreign bodies are most commonly linear foreign bodies in cats,
and these constitute an emergency presentation. Traditionally, these have
been addressed as a surgical emergency, but in selected cases, conservative
management is appropriate. When surgery is performed for intestinal for-
eign bodies, the object should be removed using the fewest number of enter-
otomies necessary for removal with minimal intestinal trauma. If there is
evidence of intestinal necrosis, a resection and anastomosis is performed.
To prevent leakage and to aid in intestinal healing, an omental or serosal
patch can be placed over the enterotomy and anastomotic sites after suture
closure. If large intestinal segments must be removed, the cat may need to be
managed for short bowel syndrome with nutritional and pharmacologic
support.

The successful management of cats with gastrointestinal obstruction is

based on a knowledge of the relevant anatomy, proper use of diagnostic and
therapeutic techniques, an understanding of the physiologic eﬀects of
obstruction, and an appreciation for intestinal tract healing at the aﬀected
location. If all these factors are considered, most cats with gastrointestinal
foreign bodies can be managed with a good prognosis.

References

[1] Armbrust LJ, Biller DS, Hoskinson JJ. Case examples demonstrating the clinical utility of

obtaining both right and left lateral abdominal radiographs in small animals. J Am Anim
Hosp Assoc 2000;36:531–6.

878

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

[2] Orsher RJ, Rosin E. Small intestine. In: Slatter D, editor. Textbook of small animal surgery,

vol. 1. 2nd edition. Philadelphia: WB Saunders; 1933. p. 593–611.

[3] Hedlund CS. Surgery of the digestive system. In: Fossum TW, editor. Small animal surgery.

St. Louis: Mosby; 1997. p. 232–308.

[4] Ryan WW, Greene RW. The conservative management of esophageal foreign bodies and

their complications: a review of 66 cases in dogs and cats. J Am Anim Hosp Assoc 1975;
11:243–9.

[5] Parker NR, Walter PA, Gay J. Diagnosis and surgical management of esophageal per-

foration. J Am Anim Hosp Assoc 1989;25:587–94.

[6] Agut A, Sanchez-Valverde MA, Torrecillas FE, et al. Iohexol as a gastrointestinal contrast

medium in the cat. Vet Radiol Ultrasound 1994;35:164–8.

[7] Houlton JEF, Herrtage ME, Taylor PM, et al. Thoracic oesophageal foreign bodies in the

dog: a review of ninety cases. J Small Anim Pract 1985;26:521–36.

[8] Michels GM, Jones BD, Huss BT, et al. Endoscopic and surgical retrieval of ﬁshhooks

from the stomach and esophagus in dogs and cats: 75 cases (1977–1993). JAVMA 1995;
207:1194–7.

[9] Fingeroth JM. Surgical diseases of the esophagus. In: Slatter D, editor. Textbook of small

animal surgery, vol. 1. 2nd edition. Philadelphia: WB Saunders; 1993. p. 534–59.

[10] Hosgood G. The omentum—the forgotten organ: physiology and potential surgical

applications in dogs and cats. Compend Contin Educ Pract Vet 1990;12:45–51.

[11] Tidwell AS, Penninck DG. Ultrasonography of gastrointestinal foreign bodies. Vet Radiol

Ultrasound 1992;33:160–9.

[12] Van Sluys FJ. Gastric foreign bodies. In: Slatter D, editor. Textbook of small animal

surgery, vol. 1. 2nd edition. Philadelphia: WB Saunders; 1993. p. 568–71.

[13] Lamb CR. Recent developments in diagnostic imaging of the gastrointestinal tract of the

dog and cat. Vet Clin North Am Small Anim Pract 1999;29:307–41.

[14] Lamb CR, Hansson K. Radiological identiﬁcation of nonopaque intestinal foreign bodies.

Vet Radiol Ultrasound 1994;35:87–8.

[15] Aronson LR, Brockman DJ, Cimino Brown D. Gastrointestinal emergencies. Vet Clin

North Am Small Anim Pract 2000;30:555–79.

[16] Sparkes AH, Papasouliotis K, Barr FJ, et al. Reference ranges for gastrointestinal transit

of barium-impregnated polyethylene spheres in healthy cats. J Small Anim Pract 1997;
38:340–3.

[17] Newell SM, Graham JP, Roberts GD, et al. Sonography of the normal feline gastro-

intestinal tract. Vet Radiol Ultrasound 1999;40:40–3.

[18] Ellison GW. Intestines. In: Bojrab MJ, Ellison GW, Slocum B, editors. Current techniques

in small animal surgery. 4th edition. Baltimore: Williams & Wilkins; 1998. p. 245–54.

[19] Ellison GW. Wound healing in the gastrointestinal tract. Semin Vet Med Surg 1989;4:

287–93.

[20] Weisman DL, Smeak DD, Birchard SJ, et al. Comparison of a continuous suture pattern

with a simple interrupted pattern for enteric closure in dogs and cats: 83 cases (1991–1997).
JAVMA 1999;214:1507–10.

[21] DeHoﬀ WD, Nelson W, Lumb WV. Simple interrupted approximating technique for

intestinal anastomosis. J Am Anim Hosp Assoc 1973;9:483–9.

[22] Crowe DT. The serosal patch: clinical use in 12 animals. Vet Surg 1984;13:29–38.
[23] Felts JF, Fox PR, Burk RL. Thread and sewing needles as gastrointestinal foreign bodies in

the cat: a review of 64 cases. JAVMA 1984;184:56–9.

[24] Basher AWP, Fowler JD. Conservative versus surgical management of gastrointestinal

linear foreign bodies in the cat. Vet Surg 1987;16:135–8.

[25] Willis SE, Farrow CS. Partial gastrointestinal obstruction for one month due to a linear

foreign body in a cat. Can Vet J 1991;32:689–91.

[26] Anderson S, Lippincott CL, Gill PJ. Single enterotomy removal of gastrointestinal linear

foreign bodies. J Am Anim Hosp Assoc 1992;28:487–90.

879

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

[27] Muir P, Rosin E. Failure of the single enterotomy technique to remove a linear intestinal

foreign body from a cat [abstract]. Vet Rec 1995;136:75.

[28] Wylie KB, Hosgood G. Mortality and morbidity of small and large intestinal surgery in

dogs and cats: 74 cases (1980–1992). J Am Anim Hosp Assoc 1994;30:469–74.

[29] Allen DA, Smeak DD, Schertel ER. Prevalence of small intestinal dehiscence and asso-

ciated clinical factors: a retrospective study of 121 dogs. J Am Anim Hosp Assoc 1992;
28:70–6.

[30] Yanoﬀ SR, Willard MD. Short bowel syndrome in dogs and cats. Semin Vet Med Surg

1989;4:226–31.

[31] Guthbertson EM, Gilﬁllan RS, Burhenne HJ, et al. Massive small bowel resection in the

beagle, including laboratory data in severe undernutrition. Surgery 1970;68:698–705.

880

T.N. Bebchuk / Vet Clin Small Anim 32 (2002) 861–880

Diagnosis and treatment of portosystemic

shunts in the cat

D. Michael Tillson, DVM, MS*,

James T. Winkler, DVM

Department of Clinical Sciences, College of Veterinary Medicine,

Auburn University, Auburn, AL 36849, USA

Portosystemic shunts (PSSs) are abnormal venous communications per-

mitting blood in the portal system to enter the systemic venous system
directly without passing through the liver. The shunted blood contains
abnormally high levels of compounds absorbed from the intestinal tract that
are normally removed, detoxiﬁed, or metabolized by the liver. Cats with
PSSs present with a wide variety and intensity of physical and clinical signs,
and they should have a complete evaluation, because medical stabilization
and surgical intervention oﬀer the best chance of managing this condition.

PSSs are generally categorized based on whether they were present from

birth or develop over time (congenital or acquired), the shunt location in re-
lation to the hepatic parenchyma (extrahepatic or intrahepatic), and whether
the shunt was a large single vessel or smaller multiple vessels. Most feline
congenital PSSs consist of a large single vessel representing a developmental
error in which a vascular communication is maintained between abdominal
veins derived from the cardinal and vitelline veins [1]. Extrahepatic shunts
are located outside the liver parenchyma and most commonly empty directly
into the prehepatic vena cava; however, cats may have atypical PSS connec-
tions. Two cats had a PSS arising from a remnant of the umbilical vein [2],
and a shunt can ﬂow into any systemic vessel, including other caudal vena
caval tributaries (eg, renal vein, phrenicoabdominal vein), the azygous vein,
or the internal thoracic vein [1,3,4]. Intrahepatic shunts course through the
left hepatic division, signifying a patent ductus venosus, but they may also
be found in the central and right hepatic divisions [1,5]. Intrahepatic shunts
drain into the hepatic veins or posthepatic vena cava and comprise approx-
imately 10% of feline PSSs [6–10]. Multiple acquired PSSs occur secondary

Vet Clin Small Anim 32 (2002) 881–899

* Corresponding author.

E-mail address: tillsdm@vetmed.auburn.edu (D.M. Tillson).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 1 9 - 0

to sustained portal hypertension as seen with advancing hepatic disease and
ﬁbrosis or after surgical attenuation of a shunt. Acquired shunts arise from
the opening of previously nonfunctional vessels between the portal and sys-
temic circulation [1,11]. Portal vein atresia, failure of the portal vein to
develop normally, requires the presence of a PSS for the animal to survive.

Hepatic circulation in the cat is derived from the hepatic artery and the

portal circulation, with the portal vein supplying from 65% to 75% of hep-
atic blood. The portal vein comprises a fusion of the mesenteric veins and
the gastrosplenic vein and is joined by the gastroduodenal and cranial pan-
creaticoduodenal veins [12]. Cats normally have a portal vein pressure of 10
to 13 cm of water [6,7]. The presence of a PSS oﬀers less vascular resistance
than the hepatic vasculature; thus, up to 80% of the portal blood volume
may be shunted through a PSS into the systemic circulation.

Historical ﬁndings and clinical signs

Having an increased level of suspicion based on historical and physical

ﬁndings is essential in making the diagnosis of a feline PSS. Most cats with
a PSS are less than 1 year of age at presentation, but signs may have been
present for many months before presentation [6,12–17]. Although a bias
toward male cats was initially suspected, a more equal distribution between
the sexes is reported now [7,10,12,14,18]. Many aﬀected cats are purebred
(Persian, Siamese, Himalayan, and Burmese); however, domestic short-
haired cats make up the single largest ‘‘breed group’’ [4,6,14–16,18]. Aﬀected
cats may seem normal or may be small with an unkempt appearance. The
liver is typically not palpable, but the kidneys may be more easily palpated
[15]. Some cats with a PSS seem to have golden- or copper-colored irises
[15,16]. Congenital cardiac murmurs, a peritoneopericardial hernia, and a
high rate of cryptorchism (24%) have been reported in cats with a PSS
[6,9,15,19].

Historical ﬁndings or presenting signs are often indicative of abnormal-

ities of the nervous, gastrointestinal, or urinary system. Central nervous sys-
tem signs are combined under the umbrella term of hepatic encephalopathy
(HE). Clinical signs associated with HE include altered consciousness
(stuporous or comatose), behavioral changes (lethargy or aggression in
cats), intermittent seizures of varying severity, visual deﬁcits or amaurosis,
ataxia, circling, head pressing, disorientation, hyperexcitability, tremors,
and apparent hallucinations [6,7,12,13,15,16,18,21]. The etiology of HE is
not clear and may diﬀer from one animal to the next; however, it is thought
to result from high levels of toxic compounds in the systemic circulation that
normally are transformed or removed by the liver. Whereas ammonia is the
most commonly cited of these compounds, aromatic amino acids, mercap-
tans, and benzodiazepine-like substances are also implicated. HE has been
reviewed in the veterinary literature [20,22–24].

882

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

Gastrointestinal signs include intermittent anorexia, polyphagia, pica,

vomiting, diarrhea, constipation, and weight loss [2,6,11–13,15,16,20,21].
Excessive salivation (ptyalism) in a cat is a common ﬁnding, and its presence
in a cat should heighten suspicion of a PSS [2,6,12,15,18,20,21]. Signs refer-
able to the urinary system include polyuria, polydipsia, urinary crystals and
calculi (ammonium biurate or urate crystals), hematuria, proteinuria, stran-
guria, and pollakiuria [4,6,7,12,15,16,18,21]. The reasons for polyuria and
polydipsia are not clearly understood but may include a decreased renal
medullary gradient secondary to limited blood urea nitrogen production, psy-
chogenic polydipsia, altered function of portal vein osmoreceptors, and stim-
ulation of central thirst centers [15]. Polyuria and polydipsia generally resolve
after shunt ligation. Identifying ammonium biurate crystals or uroliths indi-
cates signiﬁcant hepatic dysfunction. The presence of urinary crystals, calculi,
or obstruction in any young cat should place a PSS on the diﬀerential list.
Other clinical signs include intermittent fever, recurrent upper respiratory
infections, and intolerance of anesthetic agents [6,7,12,15,21].

Clinicopathologic testing

A complete biochemical proﬁle and complete blood cell count are per-

formed on any cat presenting with historical and clinical signs consistent
with PSS. Classic changes on a biochemical proﬁle include increased values
for aminotransferase and serum alkaline phosphatase as well as decreased
levels of blood urea nitrogen, albumin, and glucose [15]. Although the
changes are consistent in dogs, they are less so in cats, requiring a higher
degree of reliance on clinical judgment to further the diagnosis [12]. Poiki-
locytosis was the most common abnormality on a complete blood cell count,
although some reports include mild anemia, microcytosis, and spherocytosis
as potential abnormalities [12,15]. Subclinical sepsis from translocated
enteric bacteria normally ﬁltered out by the liver has been proposed as the
reason for an increased white blood cell count in dogs with a PSS, although
one study failed to support this hypothesis [25].

On urinalysis, urine speciﬁc gravity can vary, but it is frequently hypo-

sthenuric or isosthenuric. High levels of ammonia and uric acid excreted
into the urine, which are normally converted to water-soluble urea and allan-
toin by the liver, create an environment supportive of ammonium biurate
crystal precipitation [15,26]. These crystals can be detected on microscopic
examination.

When results from biochemical tests, hematology, and urinalysis support

the diagnosis of PSS, further supportive diagnostics are performed. Serum
bile acid (SBA) measurement has widely replaced sulfobromophthalein or
iodcyanine green clearance testing [17]. SBAs are biliary salts released into
the duodenum to digest ingested fats. They are then recovered by reabsorp-
tion in the distal ileum and enter the portal circulation. Normally, hepatic

883

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

extraction removes more than 95% of biliary salts from portal blood, per-
mitting the liver to recycle the bile salts into the biliary system. Cats and
dogs with a PSS have increased circulating bile acids because of shunting
of portal blood around the liver. In animals with normal hepatic function,
SBA concentrations are low in the fasted state (2.3–3.1 lmol/L) and only
moderately increased after feeding (5.3–9.9 lmol/L), although reference
ranges vary between laboratories. Cats with a PSS may have SBA concen-
trations reaching more than 10 times the high end of normal [17,27,28]. Ele-
vated SBA concentrations are not, however, pathognomonic for a PSS.

Measuring SBA levels is more sensitive than measuring liver enzymes or

BSP retention testing and is equal to ammonia tolerance testing (ATT) in
detecting hepatic dysfunction [17,28]. Furthermore, SBA measurements are
more convenient than BSP or ATT. SBAs are stable at room temperature for
48 hours and longer at

ÿ20°C [27]. Postprandial SBA levels are normally

higher than fasting levels, although spontaneous gallbladder contractions
during fasting can cause fasting SBA levels to exceed postprandial levels [28].

Determining blood ammonia levels was considered to be the best method

of evaluating hepatic function before the wide spread acceptance of SBA
measurement. Circulating ammonia levels are increased in cats with a PSS,
because hepatocyte conversion of ammonia to urea through the Krebs-
Henseleit cycle is drastically impaired, and only a small amount of ammonia
is consumed in the conversion of glutamate to glutamine by other tissues
[18,26]. Obtaining accurate ammonia levels has limited the usefulness of this
test, because samples must be drawn in cold heparinized tubes and trans-
ferred on ice to the laboratory for immediate testing. Red blood cells have
two to three times the ammonia concentration of serum, and delays in sep-
arating red cells from plasma or hemolysis during venopuncture spuriously
increase ammonia levels [26]. Ammonia levels of greater than 100 lg/dL are
considered to be abnormal by our laboratory.

ATT is a challenge test used to determine the liver’s ability to handle an

increased ammonia load. Initially, a baseline ammonia concentration is
established, and exogenous ammonia is then diluted in 30–50 mL of H

and administered at a rate of 100 mg/kg of body weight not to exceed a dose
of 3 g orally or rectally. Plasma ammonia levels are determined 30 minutes
later [20,26]. Animals with normal liver function should have less than a
twofold increase in plasma ammonia; increases of threefold to 10-fold indi-
cate hepatic insuﬃciency [26]. Although clinicians must exercise judgment in
patient selection for ATT, the creation of iatrogenic HE is reportedly
uncommon [26]. Despite this fact, the equally accurate and more convenient
SBA testing has generally replaced ATT in diagnosing PSSs.

Diagnostic imaging

Many diﬀerent imaging modalities are available to aid in diagnosing a

PSS. Initially, abdominal radiographs are taken on suspect cats. Although

884

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

a PSS is not directly visualized, abdominal ﬁlms can provide supportive evi-
dence of a PSS. Cats with a PSS normally have a small liver (microhepatica)
with poor abdominal detail [29]; however, lack of abdominal detail is also
common in young healthy cats with limited body fat and must not be over-
interpreted. Abdominal ﬁlms may aid in ruling out abdominal conditions
that could cause the cat’s clinical signs.

Ultrasound has become the diagnostic tool of choice for imaging PSSs

[16]. It is noninvasive, can provide a guide to the location of the shunting ves-
sel, and does not require isolation of the cat after imaging. The primary dis-
advantage is the technical skill required to successfully ﬁnd a PSS in a cat or
small dog. Signiﬁcant skill is required, and operator experience is a major fac-
tor for success [7,14]. Sedation is often needed to keep the cat comfortable
and still during a ‘‘shunt hunt,’’ but caution must be used when sedatives are
given to cats with compromised hepatic function. Intrahepatic shunts were
more easily demonstrated on ultrasonography [29], but as veterinary ultra-
sound has continued to advance and skills have increased, extrahepatic
shunts are routinely found. Ultrasonographic equipment with Doppler capa-
bilities help in ﬁnding shunts by identifying areas of turbulence in the vena
cava, which indicate where the shunt merges with it [14,16]. Suspected shunt
vessels that run dorsally are considered to be portoazygous shunts [14]. The
negative predictive value of ultrasonography for PSS identiﬁcation is
reported to be 33% [14]; thus, an inconclusive ultrasound examination does
not rule out the presence of a PSS. When historical ﬁndings and clinical signs
are consistent with a PSS, further diagnostics must be undertaken.

Portal scintigraphy involves using a radiolabeled compound, generally

technetium (

Tc), to outline portal blood ﬂow [29–32]. With the cat under

sedation,

Tc is placed into the colon, where it is rapidly absorbed into the

portal system. A gamma camera records the path of the

Tc, and still

images are captured. In a cat with normal portal circulation, the area occu-
pied by the liver mass is ﬁrst illuminated by the

Tc [30,31]. In cats with a

PSS, the area represented by the liver is a void, whereas the heart is the ﬁrst
organ to show signiﬁcant illumination. Computer software can calculate a
shunt fraction, the percentage of portal blood shunting around the liver.
Normal dogs maintain a shunt fraction of approximately 5%, whereas dogs
with a PSS have an average shunt fraction of 84% [31]. Cats with a PSS seem
to have a lower shunt fraction, with an average of 52% [32]. The reason for
this variation is unclear, although calculation errors based on an expected
heart-to-liver interval of 16 seconds have been suggested as the cause.
Researchers propose the appropriate heart-to-liver interval for small dogs,
cats, and puppies to be less than 8 seconds [32,33]. Determining a shunt frac-
tion is particularly helpful in monitoring the progress of postsurgical PSS
patients receiving a partial ligation or the application of a slow occlusion
technique.

Whereas scintigraphy has a high positive predictive value for detecting a

PSS, it has several disadvantages [32]. Scintigraphy is not speciﬁc for shunt

885

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

location, because the resolution is generally not good enough to discern the
actual shunt vessel [31,32]. There is poor interoperator correlation for deriv-
ing the same shunt fraction for the same patient [34]. Scintigraphy also
requires speciﬁc licensing for ordering and handling the radioisotopes, and
the imaging equipment is expensive and usually only available through
referral institutions. State regulations vary, but isolation of radioactive pets
is generally required for some period. This often delays surgical intervention
and can be diﬃcult in critical patients.

Contrast radiology, or portography, which experienced a decline as other

imaging modalities gained favor, is still the gold standard for imaging PSSs.
Although there are several techniques used to perform portography (cranial
mesenteric artery injection, transsplenic portography, and retrograde porto-
graphy), the most widely used is mesenteric vein portography [7,8,12,29].
Mesenteric portography is a moderately invasive surgical procedure per-
formed under general anesthesia. After aseptic surgical preparation, a small
incision is created along the ventral midline in the area of the umbilicus, and
a loop of jejunum is gently exteriorized. An over-the-needle intravenous
catheter (20–24 gauge) is placed in a mesenteric vein and secured with 4/0
silk (Fig. 1). A sterile extension set is ﬂushed with saline and attached to
the catheter. After checking to ensure that all connections are secure, an

Fig. 1. An intravenous catheter is placed into a mesenteric vein and secured. This setup was
used for operative mesenteric portography and for measuring portal pressures during
portosystemic shunt surgery. In surgery, the catheter is connected to a water manometer that
is allowed to equalize with the pressures. As the shunt vessel is occluded, the level in the
manometer should rise to reﬂect the increasing pressure.

886

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

iodinated contrast agent is injected through the catheter into the portal
system [7]. The injection is recorded on videotape or captured with a rapid
ﬁlm changer. When only a single ﬁlm can be taken, the image is obtained
near the end of the injection procedure [29]. Using the largest catheter
possible and injection with adequate pressure ensure the best possible study.
A dorsoventral view is obtained, the cat is rolled into lateral recumbency,
and the study is repeated. A recent study in dogs found that when only a
single radiographic view is taken, the left lateral position is optimal [35].

A normal portogram shows vascular arborization throughout the liver. A

portogram of a PSS shows contrast in the portal vein, which then enters the
caudal vena cava (portocaval shunt) or the azygous vein (portoazygous
shunt) without any contrast reaching the liver. There is minimal to no con-
trast within the hepatic silhouette (Fig. 2A). A torturous vessel leading from
the portal system usually represents the shunt (Fig. 2B,C). Some intrahe-
patic shunts have partial hepatic ﬁlling. Multiple extrahepatic shunts are
generally visualized near the kidneys in dogs and cats with connections to
the caudal vena cava. Many subtle variations exist, and consultation with
a veterinary radiologist or surgeon is recommended for clinicians with
limited experience in interpreting portograms. When all liver lobes show
normal vascular pattens with no premature contrast ﬁlling of the caudal
vena cava or the azygous vein, a macroscopic PSS can be ruled out.

Because moving from the operating room suite to the radiology area is

often problematic; we perform the entire procedure in the radiology area,
which reduces anesthesia time and the risk of catheter displacement resulting
in a failed study. With experience, mesenteric portography is a fairly brief
procedure, and the cat can either be recovered before performing the deﬁn-
itive surgical procedure or, if all anesthetic parameters are within acceptable
limits, the abdominal incision can be temporarily closed and the cat moved
to the operating room for immediate surgical correction. In general, we make
this decision based on anesthesia time, the cat’s preanesthesia status, and
anticipated time for shunt identiﬁcation and ligation.

Medical management

Presurgical or medical management of a PSS strives to control the cat’s

clinical signs that cause morbidity. These signs, primarily signs associated
with HE, were previously described in this article. Dietary management is
the mainstay of PSS patient management [6,11,22,36]. Low-protein diets
(eg, Hill’s l/d or k/d [Hill’s Pet Nutrition, Inc., Topeka, KS]) are recommen-
ded to decrease the production of nitrogenous wastes. In many patients, the
restricted diet results in a drastic reduction in the severity of their clinical
signs. Animals with PSS often have gastric mucosa ulceration. The resulting
gastric hemorrhage is the equivalent of a high-protein meal; thus, some clini-
cians routinely use gastric protectants (eg, 250–500 mg of sucralfate oral

887

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

suspension administered orally every 8–12 hours) in dogs [7,36]. This recom-
mendation can be applied to cats with suspected gastrointestinal hemor-
rhage.

Lactulose is a synthetic disaccharide that is not digested in the small

intestinal tract but is degraded by colonic bacteria. It decreases colonic
transit time and acidiﬁes colonic contents [22,36]. These actions result in
reduced time for bacterial generation of ammonia, and the acidic environ-
ment encourages the conversion of ammonia to NH

, which is poorly

absorbed from the colon [22]. A recommended starting dose ranges from
1 to 5 mL administered orally every 8 hours [9,36]. This dose can be modi-
ﬁed in both frequency and volume to eliminate clinical signs and to obtain
one to two soft stools without causing diarrhea [7]. Lactulose can also be
administered in the form of an enema to cats that are comatose or when oral
administration is contraindicated using a ratio of two parts of warm water
to one part of lactulose [6,7,9,21,22,36].

Oral antibiotics are used in the medical management of PSS patients to

reduce or alter intestinal bacterial populations, particularly in the colon.
Neomycin and metronidazole are most commonly referenced for use in
animals with HE [8,9,22,36]. Metronidazole is eﬀective in controlling the

Fig. 2. Portography in a cat using a rapid ﬁlm changer. (A) A dorsoventral portogram shows
portal vein ﬁlling. An abnormal vessel is obscured by the spinal column, and there is no contrast
within the hepatic parenchyma. (B) The lateral view more clearly demonstrates the presence of
a portocaval shunt. Contrast ﬁlls the portal vein and bypasses the liver to ﬂow into the caudal
vena cava. A pool of contrast is forming in the urinary bladder from the previous dorsoventral
portogram. (C) A later ﬁlm in the series shows some contrast ﬁlling the more ventral hepatic
lobes.

888

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

microbial population of the colon; however, it is absorbed systemically, and
toxicity from overdosage or decreased hepatic metabolism can cause neuro-
logic signs. The recommended dosage is 7.5 mg/kg administered orally every
8 hours or 10 mg/kg administered orally every 12 hours [9,36]. Neomycin is
an aminoglycoside with a high potential for renal toxicity but with minimal
absorption after oral administration [9]. This may give neomycin an edge
over metronidazole. Neomycin is dosed at a rate of 20 mg/kg administered
orally every 8 to12 hours [9,36]. We ﬁnd that other antibiotics, including
amoxicillin (15–20 mg/kg administered orally every 8 hours) and amoxicil-
lin/clavulanic acid (10–15 mg/kg administered orally every 12 hours), are
eﬀective in reducing or eliminating signs of HE.

Anesthesia concerns

Anesthesia can be diﬃcult in cats with a PSS. Almost all anesthetic

agents, with the exception of isoﬂurane and sevoﬂurane, require some hep-
atic metabolism. We use reversible opioids for premedication and analgesia,
allowing for a lesser dependence on the use of higher concentrations of inha-
lation anesthetic agents. Some authors recommend performing a mask
induction of PSS patients [7,21], but we ﬁnd that this can be stressful on cats
and small dogs. A substantially reduced dose of a short-acting agent such as
propofol given intravenously allows for rapid intubation and has been well
tolerated by our patients. Other authors have recommended intravenous
ketamine for induction [6,14].

While under general anesthesia, cats with a PSS are closely monitored for

hypothermia, hypoglycemia, and hypotension [6,21]. Their small body size
and limited fat deposits, together with the long surgical incision, invite
hypothermia. Creating a nest of warm blankets and using supplemental
heating devices such as warm-water and warm-air blankets help to maintain
the cat’s body temperature. Warm intravenous ﬂuids may be helpful, but
unless the ﬂuids are warmed close to the patient, the slow ﬂuid rate used for
cats results in ﬂuids reaching room temperature long before they reach
the cat. Abdominal lavage with warm saline with prolonged contact with
abdominal viscera for several minutes may help to increase core body tem-
perature before closing. However, minimizing anesthesia time is the most
important measure that can be taken to preserve body temperature. Surgical
ﬂuids should contain 2.5% to 5% dextrose to help prevent hypoglycemia
[37]. Blood glucose levels are checked after moving into the operating room
and every 30 minutes. We continue the same ﬂuids after surgery until the cat
resumes eating. Cats under inhalation anesthesia are often hypotensive,
emphasizing the importance of indirect or direct blood pressure monitoring
during surgery. Adequate perfusion reduces postsurgical complications
and ensures that intraoperative portal pressure measurements are reason-
ably accurate [21]. Care must also be taken to prevent overhydration from

889

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

intravenous ﬂuids and ﬂuids ﬂushed through the mesenteric catheter when
portal pressure is measured.

Surgical management

Shunt vessel occlusion is the primary goal when performing surgery on a

cat with a PSS and is dependent on the surgeon accurately identifying the
PSS. The most common PSS in cats arises from the left gastric vein or the
portal vein and drains into the caudal vena cava (portocaval shunt) between
the phrenicoabdominal vein and the liver [8]. Although portoazygous shunts
are the second most common presentation, in the authors’ experience, cats
have a greater variety of shunt presentations.

A PSS is approached through a ventral midline incision from the xiphoid

to the pubis. The caudal portion of the thoracic cavity is prepared and
draped, because the locations of some shunts, especially intrahepatic shunts,
are more easily dissected by extending the abdominal incision to include a
caudal sternotomy. Perioperative antibiotics (eg, a ﬁrst-generation cephalo-
sporin at a dose of 22 mg/kg administered intravenously) are given after
induction and every 2 hours during the procedure. We do not normally con-
tinue antibiotics after surgery.

A thorough abdominal exploration is performed before attempting to

locate the shunt. Shunt identiﬁcation requires the surgeon to have excellent
familiarity with the portal system in normal animals. Presurgical imaging
may be able to direct the surgeon to the location of the shunt; however, the
accuracy of diﬀerent imaging techniques can vary. Begin by looking for a
typical portocaval shunt. By retracting the duodenum toward midline, the
surgeon can inspect the caudal vena cava between the phrenicoabdominal
vein and the liver. It is at this location where most portocaval shunts enter
the caudal vena cava. If the shunt is not visualized here, the omental bursa is
opened, allowing inspection of the medial aspect of the portal vein. Large
anomalous vessels that could potentially be the shunt are identiﬁed and fol-
lowed to termination. Next, the greater and lesser curvatures of the stomach
as well as the dorsal aspect of the stomach are examined to identify any
abnormal vessels coursing dorsally toward the azygous vein. If the shunt
is not found in any of these locations, the surgeon continues to examine
individual portal contributaries until the shunt is discovered (see Fig. 3A).
If a shunt is not found after all these eﬀorts, either there is no shunt, the
shunt is intrahepatic, the shunting is microvascular in nature, or the surgeon
is having a really bad day. If one was not performed before surgery, a porto-
gram is obtained to determine if a shunt is really present.

Portocaval shunt dissection can be performed with duodenal retraction or

through the omental bursa depending on surgeon preference. After identify-
ing the PSS vessel, right-angle or ligature-passing forceps are used to care-
fully dissect it free from surrounding tissue. Forceps are placed beside the
shunt and gently opened, bluntly separating the shunt from the surrounding

890

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

Fig. 3. Intraoperative pictures of a feline portosystemic shunt. (A) A large torturous vein is
found associated with the colon in a cat. This shunt arose from the colonic vein and emptied
into a renal vein. This is typical of the variation in shunt location that is encountered in cats.
(B) A close-up view of the previous shunt. This shunt was completely ligated with the caudal
strand of 2/0 silk.

891

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

tissue. Dissection is continued until the forceps can be cleanly passed around
the shunt. PSSs are large, thin walled, and torturous in nature, and approach-
ing them with careless or aggressive dissection is a recipe for disaster. Dissect-
ing forceps should never be closed when they are behind the vessel for fear of
grasping the back wall of the vein. The exception to this rule is when the for-
ceps are used to grasp the ligating suture as it is cautiously pulled around the
vessel. If there is any resistance when the forceps are removed, they are
opened and replaced. Never pull on forceps that have resistance, because
such action could easily tear the vein. Ligatures are placed as close as possible
to the termination of the shunt vessel. We use 2/0 silk suture for shunt liga-
tion (see Fig. 3B). After the suture completely encircles the shunt, it is pro-
tected from inadvertent handling by placing it back within the abdominal
cavity during the preparation to monitor portal pressures.

A mesenteric catheter is placed as previously described, and saline-ﬁlled

extension tubing with a stopcock and water manometer or pressure trans-
ducer is attached (see Fig. 1) The stopcock or transducer is placed level with
the heart and maintained in this position. Once this setup is in place, the mano-
meter is ﬁlled with saline and the stopcock is opened between the catheter
and manometer, allowing the saline column within it to fall until the level
indicates the portal pressure in centimeters of water. Temporary shunt
occlusion is accomplished by crossing the preplaced silk suture until blood
ﬂow through the shunt ceases. Portal pressures should rise in response to
this action. If pressures do not increase, several possibilities should be inves-
tigated: the vessel being occluded is not the PSS, there is a second shunt ves-
sel, the anesthesia depth is excessive, the patient is hypotensive, the
mesenteric catheter is occluded, or the stopcock is not opened correctly.
These factors must be pondered before continuing with shunt ligation. After
the shunt has been ligated, the abdominal viscera are observed for signs of
congestion and portal hypertension, and a liver biopsy is obtained for histo-
pathologic examination. The abdomen is then closed in a routine manner.

The increased pressure created within the hepatic portal system (portal

pressure) by occluding the shunt is the limiting factor as to whether or not
a PSS can be completely occluded during surgery. With shunt occlusion, por-
tal pressures may rise moderately (<10 cm of water) or dramatically (>20 cm
of water). Guidelines for acceptable intraoperative portal pressures have
been established (Table 1). During shunt occlusion, the abdominal viscera
are carefully examined for signs of portal hypertension. These signs include
increased intestinal peristaltic activity and blanching, mesenteric vein disten-
tion, and congestion or cyanosis of the pancreas and intestinal tract [7,21,37].
When there is a discrepancy between portal pressures and physical indicators
of portal hypertension, the physical changes must take precedence. If the
physical parameters and portal pressures remain within an acceptable range
(see Table 1), the shunt is ligated.

When portal pressures indicate that it would not be prudent to com-

pletely ligate the PSS, several options exist: partial PSS ligation or one of the

892

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

techniques that slowly occludes the PSS, such as an ameroid constrictor or
cellophane band [7,11,38]. A review found that only 34% of cats undergoing
PSS surgery had the PSS completely ligated, with the remainder having par-
tial attenuation of the PSS vessel [7]. Partial ligation is performed by carefully
tying the ﬁrst throw of the silk ligature while closely monitoring the rising por-
tal pressures with the goal of obtaining as much occlusion as possible before
the portal pressures rise above established guidelines. Although estimating
the percentage of shunt occlusion is subjective, 50% to 75% occlusion is often
obtained with this technique. Additional throws are carefully placed so as not
to tighten the initial throw any further. Cats receiving partial ligation of a PSS
are at risk of postsurgical shunt thrombosis and subsequent portal hyper-
tension. The use of systemic anticoagulants in animals after partial shunt
ligation has been suggested, but it is not something that we generally do [37].

After partial ligation, we recommend a second abdominal operation to

attempt complete shunt vessel occlusion 1 to 2 months after the initial pro-
cedure. In anticipation of a second procedure, we place a loop of 2/0 poly-
propylene suture around the shunt at the conclusion of the ﬁrst operation.
The suture is placed in a protected position and is retrieved during the sec-
ond procedure, eliminating the need to dissect the shunt vessel from the scar
tissue induced by the original surgery. While measuring portal pressures, the
polypropylene suture is used to completely occlude the shunt. Three cats
with partial ligation of their PSS had successful shunt occlusion during a
second procedure, and a fourth cat required a third procedure before com-
plete occlusion was achieved [13]. If preexisting portal hypertension or mul-
tiple acquired shunts are found during the second operation, no further
occlusion is attempted. We have had success in completing shunt ligation
in roughly 30% of our canine patients, but the one cat receiving a partial
ligation had multiple shunts at the time of the second procedure (Winkler
JT et al, submitted for publication).

Ameroid constrictors have gained great popularity for surgical manage-

ment of portosystemic shunts [11]. The ameroid constrictor that we use
(Research Instruments and Manufacturing; Corvallis, OR) is composed of

Table 1
Guidelines for shunt vessel attenuation

1. Baseline portal pressure for cats is 10–13 cm of water [6].
2. Portal pressures should not exceed 18 mm Hg [13].
3. Increases in portal pressure of 10 cm of water appear to be well tolerated [6].
4. Portal pressures should not exceed 20 cm of water and should not exceed 10 cm of water over

pre-ligation baseline [7].

5. Pressures should not exceed 20 cm of water, but the increase over baseline was routinely

greater than 10 cm of water [10].

6. Recommend partial ligation when portal pressures (during temporary occlusion) rise to

greater than 17 cm of water, or portal pressures increase to greater than 9 cm of water above
baseline, or central venous pressure decreases by 1 cm of water or more. This study was
performed in dogs [47].

893

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

hydroscopic compressed casein material inside a metal band with a central
channel and an access slot that allows it to be passed around the shunt
[11]. After placement, a key slides into the access slot, preventing the con-
strictor from sliding oﬀ the shunt. The constrictor absorbs abdominal ﬂuid,
causing the dehydrated casein to expand; because the metallic band prevents
outward expansion, the material expands inwardly, causing gradual shunt
occlusion. The presence of a foreign protein should also incite an inﬂamma-
tory reaction that may help with the gradual occlusion of the shunt. It is
hypothesized that gradual shunt occlusion allows time for hepatic adaption
and regeneration [11].

It may not be necessary to measure portal pressures during PSS surgery

when an ameroid constrictor is used [11], but we believe that portal pressure
measurement is important. The time required for the ameroid constrictor
occlusion is variable, representing a potential risk [11,39]. If shunt occlusion
is more rapid than expected, or if the liver cannot accommodate the in-
creased portal ﬂow, the constrictor may simply force the opening of quies-
cent portocaval channels that are later identiﬁed as multiple extrahepatic
shunts. We make a judgment call about using the ameroid constrictor based
on the severity of portal hypertension during shunt vessel occlusion and the
gross appearance of the liver (size, texture, and color). We use ameroid con-
strictors cautiously in patients with high (>25–28 cm of water) portal pres-
sures, often opting for partial shunt vessel occlusion instead.

After surgery, an ameroid constrictor shifted, causing acute postsurgical

portal hypertension in one patient and resulting in the recommendation that
a ﬁne suture be placed through the central channel to secure the constrictor
to adjacent tissues [11]. Other surgeons simply remove the metallic band to
decrease the weight of the ameroid constrictor, rationalizing that the inﬂam-
matory reaction created by foreign protein is responsible for shunt vessel
occlusion [39]. Patients receiving ameroid constrictor for occlusion of a PSS
improve after surgery, but long-term follow-up studies are still pending. One
of two cats receiving an ameroid constrictor was documented as developing
multiple extrahepatic shunts [11]. (Note: Since this chapter was submitted
for publication, two articles have evaluated the long-term outcome of cats
having a single extrahepatic portosystemic shunt treated by the placement
of an ameroid constrictor [40,41]. One paper found a poor long-term
outcome in cats treated with ameroid ring placement [40]. The other paper
found minimal surgical complications but a substantial risk of post-surgical
complications [41]. Despite these ﬁndings, the second paper suggested a
better long-term outcome. This was in spite of 5/11 cats undergoing scinti-
graphy after surgery having evidence of continued portosystemic shunting
of blood.)

Other techniques have been reported for occlusion of PSSs. These include

placing a cellophane band around the shunt vessel to incite local inﬂamma-
tion, thus encouraging shunt occlusion, and intravascular placement of
thrombogenic coils. Placing a cellophane band around the shunt has been

894

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

successful in causing vessel occlusion in dogs, but this technique is not re-
ported in the cat; although mixed results were reported at a surgical confer-
ence [38,40,41]. Thrombogenic coils have been used for patent ductus
venosus occlusion in the dog [42], but use in the cat has not been reported.
Complications with thrombogenic coils include the risk of portal hyper-
tension caused by rapid shunt vessel occlusion, coil migration to nonshunt
locations, and recanalization of the shunt vessel. These factors as well as the
need to perform multiple procedures, cost, technical expertise, and equip-
ment availability limit the applicability of this technique.

The surgical technique for intrahepatic shunts is initially similar to that

for extrahepatic shunts. After abdominal exploration, the portal vein is fol-
lowed to the liver. Each liver lobe is gently and carefully palpated for a ‘‘soft
spot’’ indicating the shunt’s path through that lobe. In most cases, intra-
hepatic shunts are considered to represent a patent ductus venosus, but this
technically only applies to shunts within the left hepatic division, whereas
shunts traversing the central and right hepatic divisions have an unknown
etiology [5,10]. If an intrahepatic shunt is suspected, we ﬁnd that passing
a red rubber catheter down a splenic vein and through the portal system
until it enters the caudal vena cava makes shunt identiﬁcation easier and
allows the surgeon to decide whether intrahepatic or posthepatic dissection
and ligation or attenuation would be best. Once the shunt is dissected and
the encircling ligature is passed, the catheter can be withdrawn into the por-
tal vein and used to measure portal pressures during shunt occlusion. Oper-
ative ultrasound can also be used for identifying intrahepatic shunts.

Surgical management of multiple extrahepatic PSSs is controversial, with

the most widely accepted technique being vena cava banding [37,43]. Vena
cava banding involves placing catheters in both the portal system and the
caudal vena cava and then encircling the vena cava (with a silk ligature or
umbilical tape) cranially to the conﬂuence of the shunt vessel and vena cava.
The vena cava is attenuated until its pressure is 1 to 2 cm of water higher
than the measured portal pressure. Although 60% of animals with multiple
PSSs undergoing caval banding improve, the authors ﬁnd the rationale for
this technique inconsistent. Expecting the liver to respond to increased caval
pressure by increasing hepatic blood ﬂow through a liver that could not
handle the portal pressure before surgery fails to make sense. For animals
with multiple acquired shunts, we generally obtain a liver biopsy and pursue
aggressive medical management, although some surgeons report positive
results and recommend using this technique [6,43].

Complications

Mortality in cats undergoing PSS surgery ranges dramatically from 10%

to 71% [6,10,12,13,37]. The high end of this range is from an early review and
is much higher than our experience suggests. A better understanding of the

895

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

presurgical, surgical, and postsurgical management of animals with porto-
systemic shunts has improved survival over the years. Although Portosyste-
mic shunts surgery is relatively ‘‘bloodless,’’ there is a potential for
catastrophic hemorrhage if the shunt vessel is torn or incised during dissec-
tion. The small body size of cats with a PSS increases the risk of fatal hem-
orrhage with shunt vessel damage. Furthermore, there is a substantial risk of
acute portal hypertension if shunt vessel ligation is required to control hem-
orrhage. Primary shunt vessel repair or the creation of a secondary shunt
using either a vein graft or another vessel from the portal system to the sys-
temic circulation may prevent death from acute hypertension [44]. Avoiding
hemorrhage by means of careful surgical technique by or under the guidance
of an experienced surgeon is preferred.

Fatal portal hypertension can occur during surgery or in the postopera-

tive period and is attributed to acute shunt occlusion. Whereas measuring
intraoperative portal pressures helps to gauge the potential for postsurgical
hypertension, several factors, including depth of anesthesia, patient posi-
tion, thrombosis of a partial attenuation, and movement of an ameroid con-
strictor, can cause this extremely serious complication. Postsurgical PSS
patients are closely monitored for signs of portal hypertension (Table 2).
Acute shunt vessel thrombosis can result in the rapid onset of fatal portal
hypertension. Our only experience with this complication was in a dog that
collapsed acutely 48 hours after a partial ligation; we have not encountered
it in a cat.

Some cases of complete or partial shunt ligation or ameroid constrictor

occlusion develop ascites. Postsurgical ascites is generally a temporary

Table 2
Clinical signs of postsurgical portal hypertension

1. Poor peripheral perfusion

5. Intestinal ileus

a. Delayed capillary reﬁll time

6. Hemorrhagic diarrhea

b. Poor mucus membrane color

7. Ascites

2. Tachycardia

8. Endotoxic or hypovolemic shock

3. Severe abdominal pain

9. Death

4. Abdominal distention

Comments

a. If signs of postsurgical portal hypertension occur, an immediate return to surgery is

recommended. Ideally, the occluding ligature would be removed, relieving the
hypertension. In two cases of dogs that developed postsurgical hypertension, rapid
surgical intervention was not helpful.

b. Care must be taken to avoid confusing normal postoperative pain with the signs of portal

hypertension. Opinions on the use of pain medications in the postoperative period diﬀer.
Because adequate pain relief is important, however, postsurgical opioid use is routine at
our hospital.

c. Frequent monitoring is required during the postsurgical recovery period and for a

minimum of 72 hours after surgery.

Data from references [7,8,13,21, and 37].

896

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

condition that resolves without intervention as surgically induced portal
hypertension is normalized, although this resolution might be from the
development of multiple extrahepatic shunts. In our experience, postsurgical
ascites resolves in 2 to 3 weeks without treatment. If therapy is desired, a
salt-restricted diet and diuretics have been recommended [36]. We have
found that in the preoperative patient, ascites is most consistent with signiﬁ-
cant preexisting liver pathologic ﬁndings. As such, the existence of presurgi-
cal ascites carries a poor prognosis, and these animals are seldomly surgical
candidates for shunt ligation or attenuation.

Seizures are reported to occur in dogs after PSS occlusion, with the time

frame for seizure development ranging from 0 to 3 days after surgery [45,46].
The etiology of postsurgical seizures is not clearly understood, but the lead-
ing theories include alteration of neurotransmitter levels in the circulation or
the sudden decrease in benzodiazepine type compounds in the circulation.
These seizures are considered to be diﬀerent from those caused by hepatic
encephalopathy [45,46]. Seizure management involves standard pharmaco-
logic agents: diazepam, pentobarbital, phenobarbital, and potassium bro-
mide. A continuous rate infusion of propofol was the only eﬀective control
for seizures in the few patients that we have encountered. The emergence of
postsurgical seizures is a grave development with a reported mortality of
between 50% and 80% [45,46]. Our experience supports a poor prognosis,
because patients are severely debilitated and often neurologically compro-
mised after seizure cessation, and owners often discontinue supportive care
and request euthanasia. Anecdotally, surgeons suggest that the incidence of
seizures in cats is greater than in dogs, but we could ﬁnd only one reference
documenting the course of postsurgical seizures in a cat [2].

The long-term prognosis for cats with a PSS is generally related to the

degree of shunt vessel occlusion. Cats with complete shunt vessel occlusion
are expected to have a relatively normal life. Cats undergoing partial shunt
occlusion may remain stagnant with some continued shunting, progress to
complete occlusion, or stimulate the development of multiple shunts. SBA
measurements indicate that there is still substantial shunting in these cats.
Cats with partial ligation are reported to have a less favorable prognosis
than dogs: 58% good to excellent for cats compared with 94% for dogs
[7]. Many cats experience the return of clinical signs in less than a year
[13]. The development of multiple extrahepatic shunts decreases a cat’s life
span, but diligent medical management may permit them to have good-
quality lives for many years.

References

[1] Payne J, Martin RA, Constantinescu GM. The anatomy and embryology of portosystemic

shunts in dogs and cats. Semin Vet Med Surg 1990;5:76–82.

[2] Brockman DJ, Brown D, Holt D. Unusual congenital portosystemic communication re-

sulting in persistence of the extrahepatic umbilical vein. J Small Anim Pract 1998;39:244–8.

897

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

[3] Burger B, Whiting P, Breznok EM. Congenital feline portosystemic shunts. JAVMA

1986;188:515–21.

[4] Ware W, Montavon P, DiBartola SP, et al. Atypical portosystemic shunt in a cat. JAVMA

1986;188:187–8.

[5] Lamb C, White R. Morphology of congenital intrahepatic portacaval shunts in dogs and

cats. Vet Rec 1998;142:55–60.

[6] Schunk C. Feline portosystemic shunts. Semin Vet Med Surg 1997;12:45–50.
[7] Birchard SJ, Sherding RG. Feline portosystemic shunts. Compend Contin Educ Pract Vet

1992;14:1295–300.

[8] Martin RA, Freeman L. Identiﬁcation and surgical management of portosystemic shunts in

the dog and cat. Semin Vet Med Surg 1987;2:302–6.

[9] Levy J, Bunch S, Komtebedde J. Feline portosystemic vascular shunts. In: Bonagura JD,

editor. Current Veterinary Therapy. Philadelphia: WB Saunders; 1995. p. 743–9.

[10] White R, Forrester-van Hijfte M, Petrie G, et al. Surgical treatment of intrahepatic

portosystemic shunts in six cats. Vet Rec 1996;139:314–7.

[11] Vogt J, Krahwinkel D, Bright R, et al. Gradual occlusion of extrahepatic portosystemic

shunts in dogs and cats using the ameroid constrictor. Vet Surg 1996;25:495–502.

[12] Scavelli TD, Hornbuckle W, Roth L, et al. Portosystemic shunts in cats: seven cases (1976–

1984). JAVMA 1986;189:317–25.

[13] VanGundy T, Booth HW, Wolf AM. Results of surgical management of feline porto-

systemic shunts. J Am Anim Hosp Assoc 1990;26:55–62.

[14] Holt D, Schelling C, Saunders M, et al. Correlation of ultrasonographic ﬁndings with

surgical, portographic and necropsy ﬁndings in dogs and cats with portosystemic shunts: 63
cases (1987–1993). JAVMA 1995;207:1190–3.

[15] Center S, Magne ML. Historical, physical examination and clinicopathologic features of

portosystemic vascular anomalies in the dog and cat. Semin Vet Med Surg 1990;5:83–93.

[16] Lamb C, Forrester-van Hijfte M, White R, et al. Ultrasonographic diagnosis of congenital

portosystemic shunt in 14 cats. J Small Anim Pract 1996;37:205–9.

[17] Center S, Baldwin B, Erb H, et al. Bile acid concentrations in the diagnosis of hepatobiliary

disease in the cat. JAVMA 1986;189:891–6.

[18] Blaxter A, Holt P, Pearson G, et al. Congenital portosystemic shunts in the cats: a report of

nine cases. J Small Anim Pract 1988;29:631–45.

[19] Lunney J. Congenital peritoneal pericardial diaphragmatic hernia and portocaval shunt in

a cat. J Am Anim Hosp Assoc 1992;28:163–6.

[20] Tyler J. Hepatoencephalopathy. Part I. Clinical signs and diagnosis. Compend Contin

Educ Pract Vet 1990;12:1096–73.

[21] Holt D. Critical care management of the portosystemic shunt patient. Compend Contin

Educ Pract Vet 1994;16:879–92.

[22] Tyler J. Hepatoencephalopathy. Part II. Pathophysiology and treatment. Compend Contin

Educ Pract Vet 1990;12:1260–70.

[23] Maddison J. Hepatic encephalopathy. J Vet Intern Med 1992;6:341–53.
[24] Hardy R. Pathophysiology of hepatic encephalopathy. Semin Vet Med Surg 1990;5:100–6.
[25] Tobias KS, Besser T. Evaluation of leukocytosis, bacteremia and portal vein oxygen

tension in clinically normal dogs and dogs with portosystemic shunts. JAVMA 1997;211:
715–8.

[26] Leveille-Webster C. Laboratory diagnosis of hepatobiliary disease. In: Ettinger SJ, Feldman

EC, editors. Textbook of Veterinary Internal Medicine. Philadelphia: WB Saunders; 2000.
p. 1277–93.

[27] Center S, Baldwin B, et al. Evaluation of serum bile acid concentrations for the diagnosis of

portosystemic venous abnormalities in the dog and cat. JAVMA 1985;186:1090–4.

[28] Center S, Erb H, et al. Measurement of serum bile acid concentrations for diagnosis of

hepatobiliary disease in cats. JAVMA 1995;207:1048–10.

[29] Moon M. Diagnostic imaging of portosystemic shunts. Semin Vet Med Surg 1990;5:120–6.

898

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

[30] Newell S, Selcer B, et al. Use of hepatobiliary scintigraphy in clinically normal cats. Am J

Vet Res 1994;55:762–8.

[31] Daniel G, Bright RM, et al. Per rectal portal scintigraphy using 99m technetium per-

technetate to diagnose portosystemic shunts in dogs and cats. J Vet Intern Med 1991;5:23–7.

[32] Forrester-van Hijfte M, McEvoy F, et al. Per rectal portal scintigraphy in the diagnosis and

management of feline congenital portosystemic shunts. J Small Anim Pract 1996;37:7–11.

[33] Koblik P, Hornof W. Transcolonic sodium pertechnetate Tc 99m scintigraphy for

diagnosis of macrovascular portosystemic shunts in dogs, cats and potbellied pigs: 176
cases (1988–1992). JAVMA 1995;207:729–33.

[34] Samii V, Kyles AE, et al. Evaluation of interoperator variance in shunt fraction calculation

after transcolonic scintigraphy for diagnosis of portosystemic shunts in dogs and cats.
JAVMA 2001;218:1116–9.

[35] Scrivani P, Yeager A, et al. Inﬂuence of patient positioning on sensitivity of mesenteric

portography for detecting an anomalous portosystemic blood vessel in dogs: 34 cases
(1997–2000). JAVMA 2001;219:1251–3.

[36] Taboada J. Medical management of animals with portosystemic shunts. Semin Vet Med

Surg 1990;5:107–19.

[37] Butler L, Fossum TW, et al. Surgical management of extrahepatic portosystemic shunts in

the dog and cat. Semin Vet Med Surg 1990;5:127–33.

[38] Youmans K, Hunt G. Cellophane banding for the gradual attenuation of single extra-

hepatic portosystemic shunts in eleven dogs. Aust Vet J 1998;76:531–7.

[39] Youmans K, Hunt G. Experimental evaluation of four methods of progressive venous

attenuation in dogs. Vet Surg 1999;28:38–47.

[40] Hunt GB. Portosystemic shunts: constrict, ligate or band. In: Proceedings of the 10th

Annual Symposium of the American College of Veterinary Surgeons [abstract]. Chicago,
IL, 2000.

[41] Harari J, Lincoln J, et al. Lateral thoracotomy and cellophane banding of a congenital

portoazygous shunt in a dog. J Small Anim Pract 1990;31:571–3.

[42] Partington B, Partington C, et al. Transvenous coil embolization for treatment of patent

ductus venosus in a dog. JAVMA 1993;202:281–4.

[43] Whiting P, Peterson S. Liver and biliary system—portosystemic shunts. In: Slatter D,

editor. Textbook of Small Animal Surgery. Philadelphia: WB Saunders; 1993. p. 660–77.

[44] Poy N, Degner D, et al. Splenocaval shunting for alleviation of portal hypertension in a

dog: a case report. Vet Surg 2001;998:348–53.

[45] Hardie EM, Kornegay J, et al. Status epilepticus after ligation of portosystemic shunts. Vet

Surg 1990;19:412–7.

[46] Matushek K, Bjorling DE, et al. Generalized motor seizures after portosystemic shunt

ligation in dogs: ﬁve cases (1981–1988). JAVMA 1990;196:2014–7.

[47] Swalec K, Smeak DD. Partial vs complete attenuation of single portosystemic shunts. Vet

Surg 1990;19:406–11.

899

D.M. Tillson, J.T. Winkler / Vet Clin Small Anim 32 (2002) 881–899

Megacolon in the cat

Robert W. Bertoy, DVM, MS

Veterinary Surgical Service, El Dorado Hills, CA 95762, USA

Megacolon has been deﬁned as a condition of persistent increased bowel

diameter always associated with chronic constipation [1]; however; megaco-
lon is not a speciﬁc disease entity. It is merely a subjective evaluation of the
diameter of the colon, usually based on radiographic assessment.

Megacolon is the end result of an obstructive or pseudo-obstructive con-

dition that causes constipation, which is progressive, often resulting in inca-
pacitating and debilitating obstipation. Megacolon may be thought of as the
most advanced stage in the spectrum of chronic constipation.

Because of the severity of the clinical signs associated with megacolon, sur-

gical therapy may be indicated. The procedure used depends on the cause. In
some cases, surgical intervention before the development of severe megaco-
lon may alleviate the clinical problem before physical debilitation develops.

Classiﬁcation

There are many reported causes of megacolon in human beings; fewer

have been documented in the cat. As our awareness of this disorder and our
diagnostic abilities increase, more speciﬁc causes may be uncovered. This is
important, because a successful treatment plan requires accurate diagnosis
of the primary problem. Megacolon has been categorized as congenital or
acquired, primary or secondary, intrinsic or extrinsic, functional or mechan-
ical, and dilated or hypertrophic. The terms dilated and hypertrophic
recently appeared in the literature and are confusing [2]; the colonic muscula-
ture is histologically hypertrophied in idiopathic megacolon, yet, grossly, the
colon is severely dilated [3,4]. The surgeon should conceptualize the causes of meg-
acolon as either colonic inertia or outlet obstruction (Table 1), because these
categories may dictate the surgical treatment. In the cat, idiopathic megaco-
lon is the most common diagnosis, occurring in approximately 60% to 70%
of the cases reported in the literature (this percentage may be much higher in

Vet Clin Small Anim 32 (2002) 901–915

E-mail address: bertoydvm@aol.com (R.W. Bertoy).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 2 0 - 7

clinical practice). Neurologic disorders constitute approximately 11% of the
cases, and those under the heading of outlet obstruction occur in 24% of the
cases, with most obstruction cases (96%) occurring as the result of pelvic
fracture malunion [2].

Anatomy

The colon begins at the ileocolic sphincter. The cecum is a diverticulum of

the proximal colon in the dog and cat. These species have a cecocolic oriﬁce
that lies just aboral to the ileocolic junction. The colon is divided into
ascending, transverse, and descending portions. The ascending colon begins
at the ileocolic sphincter and runs cranially, ending at the right colic ﬂexure.
The colon continues at this point as the transverse colon, which runs from
right to left to its termination at the left colic ﬂexure. The descending colon,
the longest portion, begins at this point and ends near the level of the pelvic
inlet, where the rectum begins [5]. The colorectal junction is diﬃcult to iden-
tify morphologically; landmarks that have been used include the pubic brim,
pelvic inlet, and seventh lumbar vertebrae. It may be more accurate to deﬁne
it as the point where the cranial rectal artery penetrates the seromuscular
layer of the large intestine, which it does just cranial to the pubic brim. This
landmark is consistent, readily identiﬁable during celiotomy, and related to
the vascular supply and not only to a nearby osseous structure. This should
eliminate confusion between total and subtotal colectomy in which diﬀeren-
tiation is often the result of one’s deﬁnition of the colorectal junction. This
deﬁnition is used in this article.

The proximal portion of the colon receives its vascular supply from anas-

tomosing arcades via branches of the ileocolic artery, a branch of the cranial
mesenteric artery. The colic and right colic branches of the ileocolic artery

Table 1
Possible causes of megacolon in the cat

I. Colonic inertia

A. Idiopathic megacolon
B. Secondary to neurologic disease

1. Trauma to colonic innervation
2. Associated with congenital abnormalities of the caudal spine
3. Chagas disease
4. Dysautonomia

C. Secondary to a variety of medical conditions
D. Secondary to prolonged colonic distention (eg, outlet obstruction)

II. Outlet obstruction

A. Pelvic fracture malunion
B. Colonic, rectal, or anal stricture or tumor
C. Intrapelvic extraluminal mass
D. Foreign body or improper diet
E. Anal or rectal atresia

902

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

supply the ascending colon, and the middle colic branch supplies the trans-
verse colon. The descending colon is supplied primarily by the left colic
artery, the cranial branch of the caudal mesenteric artery. The left colic
artery forms an anastomosis with the middle colic artery and supplies the
descending colon by numerous vasa recti rather than by anastomosing
arcades as in the small intestine and proximal portion of the colon. The cra-
nial rectal artery is the caudal branch of the caudal mesenteric artery and
primarily supplies the cranial rectum. It also supplies a short segment of the
terminal colon via several vasa recti. The venous drainage mirrors that of
the arterial supply, except that the cranial rectal vein continues cranially
as the left colic vein, joining the ileocolic vein to form the caudal mesenteric
vein in the area of the left colic ﬂexure. The caudal mesenteric vein is short
and empties directly into the portal vein [5].

Innervation of the colon is supplied by an intrinsic component and an ex-

trinsic component. The intrinsic portion consists of groups of neurons that
form the submucosal (Meissner’s) plexus and the myenteric (Auerbach’s)
plexus located between the outer longitudinal and inner circular
smooth muscle layers. The sensory component of the plexuses receives infor-
mation on the composition of the intestinal contents and the state of the
muscular wall from nerve endings near the epithelial cell layer and smooth
muscle layers. Eﬀector nerve endings innervate secretory cells and smooth
muscle motor units. The intrinsic innervation is responsible for intestinal
contractions, which occur even in the total absence of extrinsic innervation.
The extrinsic innervation is provided by preganglionic parasympathetic
(cholinergic) ﬁbers that are responsible for stimulating smooth muscle cell
activity and by postganglionic sympathetic (adrenergic) ﬁbers that suppress
smooth muscle activity [6].

Physiology

Motility

Muscle contractions facilitate mixing of ingesta and propulsion of fecal

matter toward the anus. Slow-wave contractions, which originate in the
middle of the colon, spread orally and aborally and account for antiperistal-
tic and peristaltic movement of ingesta. Segmentation contractions ensure
adequate mixing of ingesta. Mass contractile propulsive movements that
evacuate the colon and propel feces into the rectum and anal canal originate
in the distal segment of the colon [6].

Movements in the proximal portion of the colon can occur independent

of the extrinsic nerve supply. Distention of the colon usually initiates these
intrinsic contractions. Entry of food into the stomach or duodenum is
responsible for a reﬂex contraction of the colon, the gastrocolic and duode-
nocolic reﬂex [6]. The eﬀects of extrinsic parasympathetic and sympathetic
stimulation on the colon were discussed previously.

903

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

The distal colon, rectum, and anus are more dependent on the extrinsic

nerve supply for contractile and reﬂex activity. Disruption of this
innervation can result in decreased motility or incontinence. Distention of
the distal colon and proximal rectum initiates general visceral aﬀerent
impulses that pass to the sacral spinal cord. General visceral eﬀerent ﬁbers in
the sacral spinal cord complete a reﬂex arc that results in evacuation of the
large bowel. Enhanced coordinated reﬂex contraction of the smooth muscle
of the colon and rectum, reﬂex relaxation of the internal anal sphincter,
reﬂex relaxation of the paraspinal muscles, voluntary relaxation of the
external anal sphincter, and an increase in intra-abdominal pressure during
the Valsalva maneuver result in defecation. Disruption of any of the steps of
this process can result in constipation. Voluntary contraction of the external
anal sphincter muscle can interrupt the defecation reﬂex, causing fecal
material to move back into the distal colon and proximal rectum (accom-
modation) to be stored until the next defecation reﬂex is initiated [6]. An
abnormally large rectal capacity or increased rectal compliance may result in
retention of stool at this location. If this occurs, the rectoinhibitory and
contraction reﬂexes may not be initiated, resulting in diﬃculty with
evacuation as with megarectum or perineal hernia. Normal continence
requires not only an intact defecatory reﬂex and sphincter mechanism but an
adequate reservoir for storage of stool during accommodation. If stool
cannot be adequately stored after conscious contraction of the external anal
sphincter, incontinence results [6].

Absorption/secretion

Most large intestinal ﬂuid absorption occurs in the proximal half of the

colon. The mucosa of the colon has a high capacity for active sodium
absorption. Chloride is absorbed passively by an electrochemical gradient
created by this active sodium absorption. The colonic mucosa also actively
absorbs chloride ions while simultaneously secreting bicarbonate ions in the
same transport process. The absorption of sodium and chloride creates a
large osmotic gradient, which is responsible for water absorption. Although
only 20% of gastrointestinal water absorption occurs in the colon, this
amount is critical for normal homeostasis. The bicarbonate ions help to neu-
tralize the acidic byproducts of bacterial metabolism [6,7].

Potassium is lost in the feces by the addition of potassium-rich mucus and

desquamated epithelial cells and by active transport into the colonic lumen
by mucosal cells [8]. Mucus, the primary secretory product of the colon,
lubricates and facilitates passage of fecal material and protects the mucosa
from mechanical and chemical injury [6].

Synthesis of numerous nutrients occurs secondary to the metabolic pro-

cesses of colonic bacteria. This is of little signiﬁcance in the dog and cat,
except for vitamin K synthesis, but plays a much greater role in herbivorous
species [7].

904

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

Pathophysiology

General

Normal dogs and cats can retain feces in the colon for several days with-

out suﬀering any demonstrable adverse eﬀects. With prolonged retention of
feces, the eﬃcient absorptive processes of the colon further dehydrate and
solidify the feces to the point that fecal concretions are produced. These are
diﬃcult and painful to eliminate and can become so large that passage
through the pelvic canal is physically impossible [8]. This intractable con-
stipation, or obstipation, can produce severe distention of the colon if not
treated. If this distention is prolonged, irreversible changes can occur that
aﬀect normal colonic motility, and secondary colonic inertia can result. The
degree of distention and duration of time needed to produce these irrevers-
ible changes are unknown; however, it seems that both are important factors
in the development of secondary colonic inertia. Based on empiric observa-
tions, some have suggested that 6 months is a dependable period [9,10]; how-
ever, there are no hard data available to back up these assertions.

Severely constipated animals may exhibit central nervous system depres-

sion, anorexia, and weakness, which are signs that have been attributed to
the absorption of unidentiﬁed toxins produced by bacterial metabolism in
the stagnant colonic lumen. Vomiting can occur secondary to prolonged
intestinal obstruction, the eﬀect of the absorbed toxins of colonic bacterial
metabolism on the chemoreceptor trigger zone, or vagal aﬀerent stimulation
of the vomiting center triggered by bowel distention. Paradoxically, diarrhea
may be present. Liquid feces may pass around the fecal concretions. Addi-
tionally, these concretions can irritate the colonic mucosa and cause secre-
tion of mucus and ﬂuid as well as exudation of blood, resulting in watery
and sometimes blood-tinged diarrhea. Severely constipated patients may
present with many complaints, including tenesmus, depression, lethargy,
anorexia, weight loss, and vomiting as well as occasional watery, mucoid,
or bloody diarrhea solely related to severe fecal impaction [8,11].

Idiopathic megacolon

Idiopathic megacolon, a condition of unknown cause and the primary

condition recognized in association with progressive intractable constipation
in the cat, is seen almost exclusively in middle-aged and older cats. To date,
there is no recognized speciﬁc histologic abnormality as there is for Hirsch-
sprung’s disease in human beings. Despite this, it seems that an abnormality
associated with either the intrinsic or extrinsic innervation to the lower large
intestine, the myoneural junction, or the smooth muscle itself is the likely
cause for the progressive colonic inertia. In human beings, idiopathic consti-
pation of colonic origin seems to be caused by abnormal colonic motility,
resulting in slow colonic transport [11–13]. Two causes of this altered motil-
ity are atonicity of the colonic musculature and excessive muscle activity or

905

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

spasticity of the distal large intestine or associated structures, resulting in
failure of the defecation reﬂex [11,12,14]. Previous studies have shown that
smooth muscle from megacolonic cats develops less isometric stress than
normal feline colonic smooth muscle. It is entirely possible, however, that
colonic distention from chronic obstipation may disrupt smooth muscle
myoﬁlaments, resulting in these functional abnormalities (ie, what is cause
and what is eﬀect is not known) [15]. Other studies in human beings have
shown altered release of the inhibitory neurotransmitter nitric oxide in mega-
colonic smooth muscle. It is postulated that the loss of this inhibitory nerve
input induces constipation by increasing nonpropagating contractions in the
rectum (similar to that seen in opiate-induced constipation in human beings)
with slowed transit in the right colon [16]. Investigation of this problem is
necessary to elucidate further the mechanism of this disorder.

Megacolon secondary to neurologic or medical disease

Although trauma to the sacrocaudal spinal cord reportedly can cause sec-

ondary megacolon, there are few reports of this in the veterinary literature.
Disruption of the extrinsic nerve supply to the distal large intestine can
interrupt distal colonic motility and interfere with the complicated interac-
tion between the colon, rectum, and anus during the defecation reﬂex,
resulting in chronic constipation and megacolon. Theoretically, removal
of the nonfunctional dilated colon would be therapeutic only as long as the
rectoanal reﬂexes remain intact.

Experimental breeding studies in Manx cats produced several cats with

congenital megacolon. These cats showed partial or complete absence of the
sacral and caudal spinal cord in conjunction with sacral agenesis or dys-
genesis. This results in interference with normal defecation and obstipation that
shows up early in life. Most also had problems with urinary and fecal incon-
tinence. Treatment was not attempted in any of these cats because of the
presence of multiple defects [17]. Some Manx cats can exhibit clinical signs
of megacolon later in life. It is unclear whether a milder form of neurologic
dysfunction is responsible for this ‘‘late-onset’’ megacolon or if these cats
are neurologically normal and suﬀer from idiopathic megacolon. I have per-
formed surgery (colectomy) in several of these cats with excellent results.

Aganglionic megacolon (Hirschsprung’s disease), the most common

cause of megacolon in human infants, has not been documented in cats or
dogs. The aganglionic segment lacks the submucosal and myenteric plexuses
and typically is located in the rectum and sigmoid colon. This segment
remains tonically contracted, eﬀectively producing an obstruction [11,12].
Classic surgical therapy involves removal of the aganglionic segment and
any nonfunctional dilated colon.

Several other medical and neurologic diseases have been reported to

cause megacolon in human beings [1], but they have not been proven to exist
in the cat.

906

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

Outlet obstruction

There are many reported causes of megacolon secondary to outlet

obstruction (see Table 1). Pelvic fracture malunion is the most common.
Colonic or rectoanal tumors and strictures and intrapelvic extraluminal
tumors or masses (eg, enlarged prostate or perineal hernia) are reported
causes of megacolon in animals. Foreign bodies, improper diet, and anal
or rectal atresia have been reported only in isolated cases.

As stated previously, prolonged retention of feces in the colon as a result of

any cause can produce such severe distention that irreversible changes occur,
resulting in signiﬁcantly reduced colonic smooth muscle function. Reduced
motility or severe colonic inertia can result, further complicating treatment.

There are human patients who present because of chronic constipation

with a condition of paradoxic puborectalis contraction. During defecation,
the puborectalis (and other paraspinal muscles) normally relax. Paradoxic
contraction of any of these muscles during defecation results in an outlet
obstruction causing constipation; if severe, this can result in megarectum
or megacolon. Most patients are ﬁrst treated conservatively, with surgical
treatment (partial division or resection of the puborectalis muscle) reserved
for those with persistent problems [18]. It is unknown if this condition oc-
curs in the cat, but the observation of a dilated rectum in some cats with idio-
pathic megacolon suggests that a distal obstructive or pseudo-obstructive
condition is possible.

Diagnosis

Because most owners are unaware of their animal’s bowel habits, it is dif-

ﬁcult to judge what is normal for an individual animal. Consequently, diag-
nosis and treatment are often initiated only when signs are severe and the
disease is long standing. Ruling out all other causes of intractable constipa-
tion results in a diagnosis of idiopathic megacolon. A complete physical
examination should be performed on all cats suspected of having megaco-
lon. A digital rectal examination should be performed to assess any evidence
of distal colonic or rectal stricture or tumor. The presence of perineal hernia
should also be noted, although this is likely secondary to the chronic tenes-
mus associated with chronic obstipation. A neurologic examination should
be performed with speciﬁc emphasis on the function of the sacrocaudal spi-
nal cord. Although laboratory data are usually normal in cases of megaco-
lon, these tests (complete blood cell count, serum chemistry, and urinalysis)
should be performed to rule out other causes of constipation and to help
identify any complicating conditions before pursuing other more invasive
diagnostic procedures. Abdominal radiography should be performed and
demonstrates a distended colon impacted with fecal material in all cases
of megacolon regardless of the primary cause. Radiography is used primar-
ily to rule out obstructive diseases, such as pelvic fracture malunions, sacro-
caudal spinal trauma or deformities, and intramural or mural colonic or

907

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

rectoanal obstructive lesions. Contrast radiography or endoscopy may also
be necessary to rule out an obstructive disease. Only when all known causes
of megacolon have been eliminated as a possibility can a diagnosis of idio-
pathic megacolon be made.

Treatment

Medical

Initial management involves correction of dehydration and acid-base and

electrolyte abnormalities if obstipation has been prolonged. This is followed
by evacuation of the colon using stool softeners and enemas. Manual removal
of softened fecaliths often is necessary and should be performed with caution
to minimize damage to the mucosal barrier and to prevent the absorption of
luminal bacteria and toxins into the systemic circulation. Prophylactic anti-
biotics should be administered before manual removal, because mucosal
trauma is inevitable. Long-term management involves high-ﬁber diets, stool
softeners, bulk laxatives, and periodic enemas in the hope that constipation
can be controlled. Long-term medical management has been reviewed else-
where and is not discussed in this article. If recurrent obstipation or debilita-
tion occurs, surgery should be considered as a therapeutic option.

Surgical

General

Surgical treatment involves identifying the underlying cause and surgi-

cally removing the obstruction when necessary. Partial pelvectomy or pelvic
reconstructive procedures can be used for pelvic fracture malunion [10].
These procedures have been described; however, the criteria for selection
of a speciﬁc orthopedic procedure are not well deﬁned [19]. Mass excision,
segmental proctectomy or colectomy, or bougienage can be performed to
excise obstructive mass lesions or strictures. Foreign bodies should be
removed and diet changes made when indicated. Anal or rectal atresia can
be treated with pull-through procedures. The major dilemma in the treat-
ment of outlet obstruction is whether the dilated colon can function prop-
erly after removal of the obstruction. In cases of pelvic fracture malunion,
it has been recommended that removal of the obstruction (pelvic resection)
be performed in cats that have had clinical signs of constipation of less than
6 months’ duration, with colectomy being reserved for those cats with a
longer history of clinical signs [9,10]. It seems reasonable to extend these
guidelines to all forms of outlet obstruction causing megacolon. Again, these
guidelines have been made empirically based on a small number of cases.

Colectomy for megacolon

This technique is the only surgical procedure presented in depth in this

article, because most cats with megacolon are candidates for this procedure.

908

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

In cats with concurrent perineal hernia, colectomy is performed before peri-
neal herniorrhaphy is considered. Colectomy alone often resolves the clini-
cal signs [2].

Preoperative preparation using enemas to evacuate the colon is diﬃcult

to perform and not recommended. It is much easier to prevent hard feces
from contaminating the peritoneal cavity than the liquid feces left in the
colon after the administration of enemas. Antimicrobial prophylaxis is war-
ranted, and cefazolin at the rate of 20 mg/kg of body weight given at the
time of anesthetic induction and continued for 24 hours is one rational
choice. The goal of surgery is to remove all the dilated colon, which means
that the resection sites are the most distal aspect of the ileum (total colec-
tomy) or 2 to 4 cm distal to the cecum (subtotal colectomy) and at a point
as far distally as possible that permits suturing of the anastomosis, generally
2 cm cranial to the pubic brim. I ﬁnd that a total colectomy is easier to per-
form and allows a more tension-free anastomosis than a subtotal colectomy,
and it is my procedure of choice.

Any hard stool in the rectum and anal canal is removed digitally before

surgery so that it does not cause excessive straining or diﬃcult defecation
after colectomy. The cat is placed in dorsal recumbency, and a ventral mid-
line celiotomy is performed from a few centimeters cranial to the umbilicus
to the pubic brim. The only abnormality in cats with idiopathic megacolon
is a colon greatly distended with fecaliths. In cats with pelvic fracture mal-
union, there may be scar tissue at the pelvic inlet, which may complicate the
distal colonic resection. The distal ileum, cecum, and colon are exteriorized
and packed oﬀ with moistened laparotomy sponges. Solid feces can be dis-
placed from the resection sites by squeezing the wall of the colon. Hard feca-
liths may have to be broken down manually so as to facilitate this maneuver.
Any impacted feces in the rectum not removed digitally before surgery should
be moved proximal to the resection site at this time. The terminal arcade of
the jejunal vessels supplying the most distal portion of the ileum is ligated if
a total colectomy is performed. If the decision has been made to perform a
subtotal colectomy, this step is eliminated and a site 2 to 4 cm distal to the
cecum that easily allows vessel ligation is chosen as the proximal resection
site. The ileocolic (total colectomy), right colic, and middle colic vessels
(total colectomy and subtotal colectomy) that supply the ascending (total
colectomy) and transverse colon (total and subtotal colectomy) are then
doubly ligated and transected (Fig. 1). Care must be taken when ligating the
colic and ileocolic vessels so that damage is not done to the cranial mesen-
teric artery. If this occurs, the blood supply to the small intestine can be
interrupted with catastrophic results. The caudal mesenteric vessels are then
doubly ligated and transected (Fig. 2), and the cranial rectal artery is ligated
2 cm cranial to the pubic brim approximately 1 cm caudal to the point where
the artery penetrates the seromuscular layer of the rectum (Fig. 3). Atrau-
matic intestinal forceps are placed proximal and distal to the planned resec-
tion sites. The distal ileum, cecum, and colon (total colectomy) (Fig. 4) or

909

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

the transverse and descending colon (subtotal colectomy) are then resected.
The bowel ends are anastomosed using a single layer of simple interrupted
3-0 or 4-0 polydioxanone suture. With a total colectomy, there is great diﬀer-
ence in bowel diameters, and a simple end-to-end anastomosis is impossible.
The ileal diameter is increased by making a diagonal cut, and the rectal diam-
eter is decreased by oversewing the antimesenteric side as needed to match the

Fig. 1. Ligation site of the right colic vessels (isolated over the hemostat). The blue suture to the
right of the hemostat is the ligation site for the terminal arcade of the jejunal vessels supplying
the distal ileum (when performing a total colectomy).

Fig. 2. Ligation site of the middle colic vessels (isolated over the hemostat).

910

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

ileal and rectal diameters (Figs. 5 and 6). The ileal mesentery is sutured with
3-0 or 4-0 absorbable suture in a simple continuous pattern. The abdominal
cavity is then lavaged with warm sterile saline, and the abdomen is closed
routinely. To speed closure, I use a simple continuous pattern of 3-0 poly-
dioxanone suture in the linea alba.

Although the exact pathophysiologic mechanism responsible for idio-

pathic megacolon is unknown, surgical therapy has been extremely rewarding

Fig. 4. The resected colon, cecum, and distal ileum (total colectomy). Note small amount of
ileum resected (at tip of the tissue forceps on the left).

Fig. 3. Ligation site of the caudal mesenteric vessels (isolated over the hemostat).

911

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

in cases refractory to medical management. Total colectomy with ileo-
rectal anastomosis or subtotal colectomy with preservation of the cecum and
ascending colon with colorectal anastomosis has been used successfully
without signiﬁcant alteration in either reservoir or sphincter continence
[20,21]. Although stool storage and water absorption are altered after total
colectomy, the distal small intestine apparently adapts by increasing its stool
capacity (by an increase in diameter) and its ability to absorb water [20]. In a

Fig. 5. Oversewing the antimesenteric side of the colon to match ileal diameter (total
colectomy).

Fig. 6. Completed ileorectal anastomosis (total colectomy).

912

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

small study of four colectomized cats, there seemed to be no alteration in
enteric function or general health [22]. When reviewing the available veteri-
nary literature, it appears that greater than 90% of cats have an uncomplicated
recovery and function ‘‘normally’’ and without long-term complications.
There seems to be an overall complication rate of less than 5%. This includes
stricture at the anastomotic site (1 of 150 cases), death caused by peritonitis
from leakage at the anastomosis (1 of 150 cases), death from unknown cause
(1 of 150 cases), persistent diarrhea (2 of 150 cases), and recurrent constipa-
tion (3 of 38 cases) [19].

Most cats begin eating within 24 to 48 hours after colectomy. Small-volume

watery diarrhea develops for approximately 5 days. Most cats have
semisolid nonformed stools at 1 week and soft formed stools by 6 to 12
weeks after colectomy [13,20,21]. Cats can be expected to have a softer than
normal stool and an increase in stool frequency (2–6 times per day). This
mirrors what is seen in human patients [1,11–13]. The occasional cat may
have an episode of constipation, but this seems to be relatively minor and
is easily managed by digital evacuation, diet changes, stool softeners, and
possibly enemas [13,21]. It is possible that these cats represent those with
a greatly dilated rectum before surgery. This enlarged rectum, or megarec-
tum, may be caused by chronic distention from large fecaliths or the result
of a distal obstructive or pseudo-obstructive condition that interrupts the
defecation reﬂex, such as paradoxic puborectalis contraction in human
beings.

There have been cats that are presented for chronic diarrhea or perineal

soiling (sometimes with perianal dermatitis if chronic) after colectomy. It
has been suggested (but not proven) that this may be the result of small
intestinal bacterial overgrowth [20]. Continence requires not only normal
anal sphincter function but an adequate reservoir for storage of feces.
Because reservoir capacity is reduced after colectomy, malabsorption diar-
rhea (steatorrhea) caused by bacterial overgrowth may appear as perineal
soiling. I have empirically treated some of these cats with antibiotics (typi-
cally metronidazole), which has resulted in elimination of clinical signs. In a
few of these cases, the cats required long-term antibiotic therapy (up to 6
weeks) before returning to a more normal postoperative course. If small
intestinal bacterial overgrowth is suspected, the diagnosis should be proven
by proper testing. A low serum cobalamin concentration, high serum folate
concentration, and elevated concentrations of breath hydrogen all help to
support this diagnosis.

In human patients with megacolon, total colectomy with ileorectal anas-

tomosis seems to be superior to subtotal colectomy, because recurrence of
constipation has been seen with less extensive resections. In the cat, it has
been reported that subtotal colectomy with preservation of the ileocolic
valve is a successful surgical treatment for idiopathic megacolon. It is rec-
ommended that the ileocolic valve be spared if at all possible so as to reduce
the risk of small intestinal bacterial overgrowth [21]. Whether preservation

913

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

of the ileocolic valve after removal of 80% to 90% of the colon actually pre-
vents colonic microorganisms from colonizing the ileum is unknown but
seems unlikely. It is my opinion that total colectomy is the technique of
choice; it is easier to perform, allows a tension-free anastomosis, and is more
likely to prevent recurrent constipation because all the colon is removed.

Summary

Megacolon is a condition that is not uncommon in the cat. Most cases are

idiopathic (a cause cannot be determined), and these seem to be a result of
colonic inertia. Pelvic fracture malunions are the next most common cause
and result in a pelvic outlet obstruction. Total or subtotal colectomy oﬀers
good long-term results in cases of idiopathic megacolon and chronic cases of
pelvic fracture malunion, and the technique is described in detail.

References

[1] McCready RA, Beart RW. The surgical treatment of incapacitating constipation associated

with idiopathic megacolon. Mayo Clin Proc 1979;54:779–83.

[2] Washabau RJ, Holt D. Pathogenesis, diagnosis, and therapy of feline idiopathic

megacolon. Vet Clin North Am Small Anim Pract 1999;29:589–603.

[3] Gattuso JM, Kamm MA, Talbot JC. Pathology of idiopathic megarectum and megacolon.

Gut 1997;41:252–7.

[4] Gattuso JM, Smith VV, Kamm MA. Altered contractile proteins and neural innervation in

idiopathic megarectum and megacolon. Histopathology 1998;33:34–8.

[5] Evans HE, Christensen GC. Miller’s anatomy of the dog. 2nd edition. Philadelphia: WB

Saunders; 1979.

[6] Banks WJ. Applied veterinary histology. Baltimore: Williams & Wilkins; 1981. p. 373–408.
[7] Pass MA. Large intestine: physiology. In: Slatter DH, editor. Textbook of small animal

surgery. Philadelphia: WB Saunders; 1985. p. 753–6.

[8] Ford RB. Constipation and dyschezia. In: Ford RB, editor. Clinical signs and diagnosis in

small animal practice. New York: Churchill Livingstone; 1988. p. 491–504.

[9] Matthiesen DT, Scavelli TD, Whitney WO. Subtotal colectomy for the treatment of

obstipation secondary to pelvic fracture malunion in cats. Vet Surg 1991;20:113–7.

[10] Schrader SC. Pelvic osteotomy as a treatment for obstipation in cats with acquired stenosis

of the pelvic canal: Six cases (1978–1989). JAVMA 1992;208–13.

[11] Beck DE, Jagelman DG, Fazio VW. The surgery of idiopathic constipation. Gastroenterol

Clin North Am 1987;16:143–56.

[12] Poisson J, Devroede G. Severe constipation as a surgical problem. Surg Clin North Am

1983;63:193–217.

[13] Preston DM, Hawley PR, Lennard-Jones JE, Todd IP. Results of colectomy for severe

idiopathic constipation in women (Arbuthnot Lane’s disease). Br J Surg 1984;71:547–52.

[14] Von der Ohe MR, Camilleri M, Carryer PW. A patient with localized megacolon and

intractable constipation: evidence for impairment of colonic muscle tone. Am J Gastro-
enterol 1994;89:1867–70.

[15] Washabau RJ, Stalis IS. Alterations in colonic smooth muscle function in cats with

idiopathic megacolon. Am J Vet Res 1996;57:580–7.

[16] Koch TR, Otterson MF, Telford GL. Nitric oxide production is diminished in colonic

circular muscle from acquired megacolon. Dis Colon Rectum 2000;43:821–8.

914

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

[17] Deforest ME, Basrur PK. Malformations and the Manx syndrome in cats. Can Vet J 1979;

20:304–14.

[18] Pfeifer J, Afachan F, Wexner SD. Surgery for constipation: a review. Dis Colon Rectum

1996;39:444–60.

[19] Rosin E. Megacolon in cats: the role of colectomy. Vet Clin North Am Small Anim Pract

1993;23:587–94.

[20] Bertoy RW, MacCoy DM, Wheaton LG, Gelbeng HB. Total colectomy with ileorectal

anastomosis in the cat. Vet Surg 1989;18:204–10.

[21] Bright RM, Burrows CF, Goring R, Fox S, Tilmant L. Subtotal colectomy for treatment of

acquired megacolon in the dog and cat. JAVMA 1986;188:1412–16.

[22] Gregory CR, et al. Enteric function in cats after subtotal colectomy for treatment of

megacolon. Vet Surg 1990;19:216–20.

915

R.W. Bertoy / Vet Clin Small Anim 32 (2002) 901–915

Perineal urethrostomy

Charles W. Smith, DVM, MS

Small Animal Surgery Section, Department of Veterinary Clinical Medicine,

College of Veterinary Medicine, 38 Small Animal Clinic, 1008 Hazelwood Drive,

University of Illinois, Urbana, IL 61801, USA

Indications

Perineal urethrostomy is a surgical procedure used in male cats to create

a permanent opening between the pelvic urethra and the skin in the perineal
region. Indications for perineal urethrostomy include recurrent urethral
obstruction, urethral obstruction that cannot be relieved by catheterization
and reverse ﬂushing, urethral strictures, and urethral trauma or neoplasia. A
component of feline lower urinary tract disease is urethral plug formation
with obstruction, especially in male cats. Without proper medical and die-
tary management, urethral obstruction may ensue. Obstructions that cannot
be relieved by catheterization or recurrent urethral obstructions can be man-
aged by perineal urethrostomy. It must be emphasized that perineal ure-
throstomy is only an adjunct to medical management of cats with recurrent
or persistent urethral plug formation. Fortunately, with medical and dietary
management and with a high level of owner compliance, the need for perineal
urethrostomy in patients with lower urinary tract disease has decreased. Ure-
thral strictures may occur with chronic urethritis or repeated trauma from
catheterization. Likewise, perineal urethrostomy is indicated if the stricture
is within the penile urethra. If the location of the obstruction is unknown,
contrast radiography may be necessary. Other urethral trauma or neoplasia
may damage the urethra, requiring a perineal urethrostomy.

Surgical technique

Several techniques and modiﬁcations of perineal urethrostomy have been

described since the early 1960s [1–5]. The goal of these techniques is to
remove the small penile urethra and take advantage of the wider pelvic or

Vet Clin Small Anim 32 (2002) 917–925

E-mail address: cwsmith@uiuc.edu (C.W. Smith).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 3 2 - 3

membranous portion of the urethra, which is three to four times larger than
the penile urethra, to produce a widened tube for urine passage. Another
technique originally introduced in 1968 [6] and modiﬁed in 2000 [7] involves
connecting the pelvic urethra to the preputial mucosa to maintain a natural
and more cosmetic opening. The major concern with this technique is hav-
ing the anastomosis buried beneath the skin. Perineal urethrostomy as ﬁrst
described by Wilson and Harrison [5] is the author’s choice of a surgical
technique.

The perineal area, including 4 to 5 cm at the ventral base of the tail, is

clipped and prepared for aseptic surgery. The patient is positioned in either
dorsal or ventral recumbency. In ventral recumbency (author’s preference),
the patient is gently positioned over a rolled towel, with light tension on the
hindlimbs and with the tail secured over the back. A pursestring suture of
3-0 nylon is placed around the anus, carefully avoiding the anal sacs. If the
cat is intact, castration is performed ﬁrst. If possible, the urethra is catheterized
to allow identiﬁcation of the urethra. An elliptic skin incision is made
around the prepuce and scrotum, leaving at least a centimeter of intact skin
between the anus and the incision (Fig. 1A). Enough skin is removed at the
base of the scrotum and prepuce to permit slight tension on the skin-urethra
anastomosis such that the skin edges do not roll inward and make contact.
The penis and the ischiocavernous and ischiourethralis muscles are isolated
by careful blunt dissection within the subcutaneous tissues (see Fig. 1B). The
cranial and caudal scrotal arteries can be electrocoagulated but rarely bleed
enough to cause concern. The ischiocavernous and ischiourethralis muscles
are incised using scissors at their ischial attachments to minimize hemor-
rhage. Retracting the penis laterally to tense these muscles facilitates incision
at the ischial attachments (see Fig. 1C). The ventral pubic attachment is
incised carefully with scissors (see Fig. 1D). Careful blunt dissection ven-
trally and digital elevation of the penis and pelvic urethra from the pelvic
ﬂoor permit mobilization and posterior displacement of the penis and pelvic
urethra. Adequate mobilization is accomplished when there is minimal to no
tension on the incision line. The retractor penis muscle, bulbocavernous
muscle, and bulbourethral glands are identiﬁed on the dorsal aspect of the
penis (see Fig. 1E). The retractor penis muscle is dissected from the urethra,
transected near the external anal sphincter muscle, and excised (see Fig. 1F).
Care is taken in this dissection to prevent damage to the rectum and pelvic
nerves (Fig. 2). The penile urethra is incised on its dorsal surface from the tip
of the penis to the bulbourethral glands (see Fig. 1G) using iris scissors. At
the level of the bulbourethral glands, the pelvic urethra is about 4 mm in
diameter. Closed mosquito hemostats can be introduced into the pelvic ure-
thra to its box-locks when the proper level is reached. The incised pelvic ure-
thra and approximately two thirds of the penile urethra are sutured to the
skin using 4-0 monoﬁlament nylon or polypropylene or 5-0 synthetic
absorbable suture material in an interrupted pattern (see Fig. 1H). The
remaining urethra and penile tissue distal to the urethrostomy site are

918

C.W. Smith / Vet Clin Small Anim 32 (2002) 917–925

removed (see Fig. 1I). An absorbable mattress suture may be placed through
the body of the remaining penile shaft to control hemorrhage but is often
unnecessary. The remaining skin incision is closed. The pursestring suture
in the anus is removed. An Elizabethan collar is used to prevent licking until
the sutures are removed at 7 to 10 days. Indwelling urinary catheters are

Fig. 1. (A) Elliptic incision incorporating the scrotum and prepuce (S

¼ scrotum; P ¼ prepuce;

¼ skin incision). (B) Removal of prepuce and scrotum (A ¼ caudal scrotal artery; B ¼ cranial

scrotal artery; C

¼ dorsal artery and vein of penis, prostatic artery). (C) Dissection of the penis

from the surrounding tissue to its pelvic attachments on the ischium (ICM

¼ ischiocavernous

muscle; IUM

¼ ischiourethral muscle; PeC ¼ crus of penis; Pe ¼ penis; PU ¼ pelvic urethra).

(D) Incision of the ligament of the penis (PeL

¼ ligament of penis). (E) Exposure of the retractor

penis muscle (BUG

¼ bulbourethral glands; BSM ¼ bulbocavernous muscle; RPM ¼ retractor

muscle of penis). (F) Insertion of a probe into penile urethra (PeU

¼ penile urethra).

(G) Incision of the penile urethra through the glans penis to the pelvic urethra. (H) Suture of
the pelvic and penile urethral mucosa to the perineal skin (S-1

¼ initial sutures). (I) Placement of

through-and-through suture through the body of the penis (S-1

¼ initial sutures; S-2 ¼ through-

and-through mattress suture; S-3

¼ mucosa-to-skin sutures). (From Wilson GP, Kusba JK.

Perineal urethrostomy in the cat. In: Bojrab MJ, editor. Current Techniques in Small Animal
Surgery. Philadelphia: Lea & Febiger; 1983. p. 328–30; with permission. Adapted from Wilson
GP, Harrison JW. Perineal urethrostomy in cats. JAVMA 1971;159:1789; with permission.)

919

C.W. Smith / Vet Clin Small Anim 32 (2002) 917–925

avoided after surgery. Shredded paper should be used instead of litter during
the healing period.

Postoperative complications

Perineal urethrostomy in the cat, when indicated and when properly per-

formed, is beneﬁcial to the patient, rewarding to the surgeon, and plagued
with few complications. Complications reported after perineal urethrostomy

Fig. 1 (continued).

920

C.W. Smith / Vet Clin Small Anim 32 (2002) 917–925

include hemorrhage from erectile tissue, wound dehiscence, cystitis or
ascending urinary tract infection, urethral stricture, urinary and fecal incon-
tinence, perineal hernia, and rectourethral ﬁstula.

Hemorrhage from cavernous tissue can occur but rarely becomes a seri-

ous problem. Because the urethra is surrounded by cavernous tissue, some
hemorrhage can be expected. Generally, accurate apposition of the urethra
to the skin prevents most bleeding from the penile cavernous tissue. Care-
fully incising the ischiocavernosus and ischiourethralis muscles at their
ischial attachment avoids hemorrhage from these muscles. Some hemor-
rhage occurs when the bulbospongiosus muscle is incised but rarely becomes
a problem. Licking of the surgical site can cause excessive bleeding when
Elizabethan collars are not used or when they are removed by the patient.
When bleeding occurs, careful cleaning of the urethral stoma is indicated
to maintain urine outﬂow.

Urine leakage into the perineal tissue is an infrequent complication but

may lead to cellulitis and wound dehiscence. Severe urine leakage into the
perineum and caudal thigh areas requires soft tissue urine drainage and a

Fig. 2. Pelvic nerves and rectum in relation to the penis and pelvic musculature. (From Wilson
GP, Kusba JK. Perineal urethrostomy in the cat. In: Bojrab MJ, editor. Current Techniques in
Small Animal Surgery. 2nd edition. Philadelphia: Lea & Febiger; 1983. p. 326; with permission.)

921

C.W. Smith / Vet Clin Small Anim 32 (2002) 917–925

closed indwelling catheter system [8]. Avoiding laceration of the urethra and
accurately apposing the urethra to the skin prevent this problem. Necrosis
of the urethra and skin is prevented by gentle tissue handling and avoiding
tension on the suture line. When the mucosa is traumatized by repeated
catheterization, an indwelling catheter, or uroliths, it is more prone to tear-
ing by tissue forceps and suture material. An indwelling soft Foley catheter
may be necessary for a few days. Long-term indwelling catheters should be
avoided after surgery, however, because of the risk of stricture formation
and ascending urinary tract infection [9–12].

Postoperative bacterial cystitis is a common complication in patients with

perineal urethrostomy [11,13–15]. The explanation for this complication is
not completely known. Several reasons have been suggested, and some have
been investigated. The use and misuse of indwelling catheters have been
shown to increase the risk for ascending urinary tract infection [9,12,16].
The urethral stoma, with the larger urethral opening and its closer proximity
to the anus, may predispose the cat to bacterial contamination of the lower
urinary tract. With the removal of the penile urethra, a part of the urethral
mucosal barrier is removed. It has been reported that perineal urethrostomy
using the Wilson and Harrison technique may cause impairment of striated
muscle urethral sphincter function as measured by urethral pressure pro-
ﬁlometry and electromyography, thereby increasing the frequency of ascend-
ing urinary tract infection [17]. In a subsequent study, Griﬃn et al [18]
recommended sharp dissection and incising the ischiocavernosus and
ischiourethralis muscles with a scalpel blade to avoid damage to the urethral
branches of the pudendal nerve coursing ventral and medial to the insertions
of these muscles. A later study reported that neither sharp nor blunt intra-
pelvic dissection signiﬁcantly alters the postoperative urodynamic status in
male cats based on their sphincter electromyography studies. Furthermore,
because the pelvic plexus and pudendal nerve lie dorsal to the urethra,
aggressive dissection may impair lower urinary tract function. By preserving
the dorsal aspect of the urethra’s attachment in the extensive dissection tech-
nique, normal lower urinary tract function is maintained [19]. Because lower
urinary tract contamination and infection occur after perineal urethrostomy
and may be subclinical, it is important to monitor patients using urinalysis
and culture testing and to treat with the appropriate antimicrobial drugs.

Stricture of the urethral stoma (Fig. 3) is the most diﬃcult postoperative

complication to manage. The stricture develops because of excessive granu-
lation and scar tissue formation around the opening. Several factors have
been proposed as causes, including inﬂammation, suture tension, poor skin-
to-urethra apposition, inadequate mobilization of the penis and pelvic ure-
thra, self-trauma, and indwelling urinary catheters [8,11,12]. Strictures are
best prevented by using good surgical technique, avoiding postoperative
catheters, and using collars to prevent excessive licking and self-trauma.
Extensive surgical experience with a good urethrostomy technique is the best
way to minimize stricture formation. Once stricture occurs, corrective surgery

922

C.W. Smith / Vet Clin Small Anim 32 (2002) 917–925

is necessary. Simple dilation of the urethral stoma yields only temporary
relief. Using an elliptic incision, the scarred terminal portion of the urethra
and surrounding skin must be excised, the pelvic urethra must be gently
mobilized further from the surrounding tissues, and the urethra must be
incised dorsally to permit suturing of the pelvic urethra to the skin. If the
urethra cannot be suﬃciently mobilized to permit satisfactory repair, a pre-
pubic urethrostomy [20,21] or subpubic urethrostomy [22] may be per-
formed. A more recent study on prepubic urethrostomy reported a high
level of complications with the technique [20]. It seems that the prognosis
is better with a subpubic urethrostomy.

Urinary incontinence is an infrequent occurrence after perineal urethro-

stomy. This may occur as a sequel to overdistention of the urinary bladder
during the obstruction episodes or may be the result of disruption of the pel-
vic plexus and the pudendal nerve during surgery. Overdistention of the urinary
bladder should be recognized and treated early to minimize incontinence,
and frequent emptying should be performed until continence returns. Care-
ful and minimal dissection to the dorsal pelvic urethra is the most important
measure to prevent damage to the pelvic innervation and to maintain

Fig. 3. The perineal area of a cat with a urethral stricture after perineal urethrostomy. Note the
pinpoint urethral opening.

923

C.W. Smith / Vet Clin Small Anim 32 (2002) 917–925

continence. Fecal incontinence has been reported as an infrequent compli-
cation of perineal urethrostomy [23]. This complication can develop with the
same aggressive dissection around the pelvic urethra, with resultant damage
to the pelvic innervation.

Iatrogenic perineal hernia has been reported as a complication of perineal

urethrostomy [10,24,25]. This complication occurs when the pelvic dia-
phragm is damaged during isolation and mobilization of the pelvic urethra.
Again, careful surgical dissection during mobilization of the pelvic urethra
should prevent this complication. Surgical repair may be necessary to man-
age this complication.

A rare but serious complication that can occur is laceration of the rec-

tum, leading to a rectourethral ﬁstula. I have seen this complication once.
It occurs when the surgeon is dissecting dorsal to the pelvic urethra to mobi-
lize the pelvic urethra or dissecting the retractor penis muscle from its
attachment to the anal sphnicter and rectum and penetrates the rectum. Air
bubbles and feces are seen coming from the urethral stoma. Surgical closure
of the ﬁstula and treatment of the ascending urinary tract contamination/
infection are necessary. Careful dissection and knowledge of anatomic struc-
tures should prevent this complication.

Prognosis

Perineal urethrostomy is an excellent surgical procedure to help manage

acute and recurrent urethral obstruction, penile urethral stricture, and neo-
plasia and trauma to the penis. Many cats with recurrent lower urinary tract
disease with infection, urethral plugs, and uroliths in the bladder can be suc-
cessfully managed with a proper diet and antibiotics. In cats in which a peri-
neal urethrostomy is indicated, the urethral stoma heals rapidly and is often
eﬀective in preventing urethral obstruction. With experience, the procedure
can be performed successfully with minimal complications. When complica-
tions occur, many can be managed successfully.

References

[1] Blake JA. Perineal urethrostomy in cats. JAVMA 1968;153:1499–506.
[2] Carbone MG. A modiﬁed technique for perineal urethrostomy in the male cat. JAVMA

1967;151:301–5.

[3] Johnston DE. Feline urethrostomy: a critique and new method. J Small Anim Pract 1974;

15:421–35.

[4] Richards DA, Hinko PJ, Morse EM, Jr. Feline perineal urethrostomy: a new technique for

an old problem. J Am Anim Hosp Assoc 1972;8:66–73.

[5] Wilson GP, Harrison JW. Perineal urethrostomy in cats. JAVMA 1971;169:1789–93.
[6] Christensen NR. Preputial urethrostomy in the male cat. JAVMA 1964;145:903–8.
[7] Lih-Seng Y, Shih-Chien C. Modiﬁed perineal urethrostomy using preputial mucosa in cats.

JAVMA 2000;216:1092–5.

[8] Scavelli TD. Complications associated with perineal urethrostomy in the cat. Probl Vet

Med 1989;1:111–9.

924

C.W. Smith / Vet Clin Small Anim 32 (2002) 917–925

[9] Lees GE, Osborne CA, Stevens JB, et al. Adverse eﬀects of open indwelling urethral

catheterization in clinical normal male cats. Am J Vet Res 1981;42:825–33.

[10] Leighton RL. Perineal hernia in a cat [abstract]. Feline Pract 1979;9:44.
[11] Smith CW, Schiller AG. Perineal urethrostomy in the cat: a retrospective study of

complications. J Am Anim Hosp Assoc 1978;14:225–8.

[12] Smith CW, Schiller AG, Smith AR, et al. Eﬀects of indwelling urinary catheters in male

cats. J Am Anim Hosp Assoc 1981;17:427–33.

[13] Gregory CR, Vasseur PB. Long-term examination of cats with perineal urethrostomy. Vet

Surg 1983;12:210–2.

[14] Osborne CA, Caywood DD, Johnston GR, et al. Perineal urethrostomy versus dietary

management in prevention of recurrent lower urinary tract disease. J Small Anim Pract
1991;32:296–305.

[15] Smith CW, Weigel RM, Smith AR. Perineal urethrostomy in the cat—an update. Feline

Pract 1991;19:20–6.

[16] Lees GE, Osborne CA. Use and misuse of indwelling urinary catheters in cats. Vet Clin

North Am Small Anim Pract 1984;14:599–608.

[17] Gregory CR, Holliday TA, Vasseur PB, et al. Electromyographic and urethral pressure

prolifometry: assessment of urethral function before and after perineal urothrostomy in
cats. Am J Vet Res 1984;45:2062–5.

[18] Griﬃn DW, Gregory CR, Kitchell RL. Preservation of striated muscle urethral sphincter

function with use of a surgical technique for perineal urethrostomy in cats. JAVMA
1989;194:1057–60.

[19] Sackman JE, Sims MH, Krahwinkel DJ. Urodynamic evaluation of lower urinary tract

function in cats after perineal urethrostomy with minimal and extensive dissection. Vet
Surg 1991;20:55–60.

[20] Baines SJ, Rennie S, White RAS. Prepubic urethrostomy: a long-term study in 16 cats. Vet

Surg 2001;30:107–13.

[21] Bradley RL. Prepubic urethrostomy: an acceptable urinary diversion technique. Probl Vet

Med 1989;1:120–7.

[22] Ellison GW, Lewis DD, Boren FC. Subpubic urethrostomy to salvage a failed perineal

urethrostomy in a cat. Comp Contin Educ Pract Vet 1989;11:946–51.

[23] Wilson GP, Kusba JK. Perineal urethrostomy in the cat. In: Bojrad MJ, editor.

Current Techniques in Small Animal Surgery. 2nd edition. Philadelphia: Lea & Febiger;
1983. p. 325–33.

[24] Johnson MS, Gourley IM. Perineal hernia in a cat: a possible complication of perineal

urethrostomy. Vet Med Small Anim Clin 1980;75:241–2.

[25] Welches CS, Scavelli TD, Aronsohn MG, et al. Perineal hernia in the cat: a retrospective

study of 40 cases. J Am Anim Hosp Assoc 1992;28:431–8.

925

C.W. Smith / Vet Clin Small Anim 32 (2002) 917–925

Treatments for feline long bone fractures

Joseph Harari, MS, DVM

Veterinary Referral Services, 21 East Mission Avenue, Spokane, WA 99202, USA

With an increasing feline pet population (73 million) that currently

exceeds the canine pet population (68 million), veterinary practitioners and
specialists are expected to treat a growing variety and number of fractures
involving the humerus, radius/ulna, femur, and tibia/ﬁbula. Although feline
orthopedics involves principles similar to those used in treating canine bone
fractures, cats are not small dogs, and diﬀerences exist with regard to patient
and fracture management. Furthermore, the adage that feline fractures
always heal if ‘‘fragments are placed in the same room’’ is not supported
by clinical and experimental reports of complications after surgical treat-
ments [1–7]. In this article, treatment options for feline long bone fractures
are reviewed, and case examples are presented.

Perioperative considerations

Orthopedic injuries in cats may be open or closed and are characterized

by simple (single) or complex (comminuted) fractures. Causes are based on
the environment of the patient and include various forms of trauma, such as
vehicular and ﬁrearm accidents, falls, ﬁghts, and, unfortunately, human
abuse [4,8–12]. As with all trauma cases, a general physical examination
should be performed initially to identify and treat cardiopulmonary instabil-
ity quickly; bone fractures are rarely life threatening but are indicative of
possibly serious concomitant soft tissue lesions. Pulmonary contusions,
pneumothorax, and ruptured urinary bladder are common derangements
identiﬁed in feline trauma patients [10,12,13]. Cachectic animals (ie, cats
missing for several days) require ﬂuid and caloric resuscitation before any
anesthetic, surgical, or stressful diagnostic procedures. Establishing normo-
thermia is also a critical goal in the initial treatment of lost or abandoned
animals.

Vet Clin Small Anim 32 (2002) 927–947

E-mail address: jharari103@aol.com (J. Harari).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 2 5 - 6

Preoperative evaluations of fracture patients are based on the type of

injury, signalment, and patient health status. A complete blood cell count
and serum chemistry analysis, thoracic/abdominal radiography/ultra-
sonography, electrocardiography, limb motor and sensory testing, and blad-
der and bowel integrity checks are all important examinations. Preliminary
survey radiographs of aﬀected limbs are helpful in guiding the clinician (and
owner) in determining treatment choices; selection of a type of repair is
often determined by detailed preoperative radiographs of the anesthetized
cat. Experienced surgeons also have alternate options when unusual or
unexpected intraoperative ﬁndings are encountered. Postoperative radio-
graphs are mandatory for evaluation of the orthopedic repair(s) and to
document the placement of implants. All perioperative views should be
orthogonal projections (ie, at right angles to each other).

Perioperative analgesics (fentanyl patches, opioids, and sedatives) are

useful in reducing general anesthetic requirements and in stabilizing frac-
tious or painful cats and are necessary from a humane standpoint for ortho-
pedic patients; options are reviewed in another article in this issue.

Bone infections have not been well documented in cats compared with

dogs [4,5,10,14]. Antibiotics are used on a therapeutic basis if infection is
present as a result of open contaminated wounds or a break in surgical asep-
tic technique [5,15]. Prophylactic antibiotics are administered intravenously
during anesthetic induction of the patient and repeated during surgery to
prevent infection during prolonged procedures, with the use of metal im-
plants, and in the presence of damaged devitalized tissues or hematomas. Com-
monly used antimicrobial agents include amoxicillin, ampicillin, amoxicillin/
clavulanate, ﬁrst-generation cephalosporins, and clindamycin. In small
animals, the most frequent isolates in cases of osteomyelitis are various
Staphylococcus, Streptococcus, Proteus, Klebsiella, and Pseudomonas species
as well as Escherichia coli. Anaerobes identiﬁed include Bacteroides, Fuso-
bacterium, and Clostridium species. Chronic infections may require implant
removal, bone debridement, lavage, and drainage. Deep cultures of infected
tissues or implants are helpful in identifying causative agents and antimicrobial
sensitivity patterns.

Postoperative rehabilitation in the cat is often limited by the animal’s

independent nature, dislike for handling by strangers, and apprehension
when placed in a noisy and unfamiliar hospital environment. Recovery can
also be mitigated by proximity to dogs. In quiet isolated clinic rooms or at
home, controlled indoor activity and games are encouraged to maintain
joint mobility and tissue health, muscle tone, and skeletal weight bearing.
Passive ﬂexion and extension exercises performed by owners and nurses
on agreeable or sedated cats are also a useful adjunct in reducing patient
morbidity. Free roaming by the patient outdoors is usually permitted after
radiographic and physical evaluations conﬁrming bone union and after
removal of external appliances.

928

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

Operative considerations

Fracture biology

According to most authors, nearly 50% of all feline fractures involve the

long bones, and the femur is most frequently injured (Table 1). Supraphysio-
logic forces (compression, tension, bending, and rotation) applied individu-
ally to these long bone columns produce distinct fracture patterns (Fig. 1)
[15,16]. In clinical cases, a combination of these forces often exists and
causes complex patterns characterized by comminutions and ﬁssures. Addi-
tionally, internal bone stresses and strains can induce further collapse and
displacement of comminuted, oblique, or transverse fractures. Selection
of an implant or external splintage is therefore based on the ability of the

Fig. 1. Types of loading forces applied to bone columns and resultant fractures. (Courtesy of
Joseph Harari, MS, DVM)

Table 1
Percentage of long bone fractures compared with all fractures in cats

Source

Bone

Hill

1977 [42]

Phillips

1977 [11]

Knecht

1978 [43]

Schrader

1994 [5]

Smith

1994 [39]

Griﬀon

1994 [13]

Humerus

—

Radius/ulna

Femur

Tibia/ﬁbula

Private practice, England.

Academic hospital, midwestern United States.

Medical center, New York City.

929

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

ﬁxation to eﬀectively neutralize distractive forces at the fracture site and
thereby promote bone union.

Bone healing depends on fracture site stability and vascularization [15–

17]. Movement of fragments at the fracture site is described as strain—a ratio
in which the change in length (ie, fragmentary displacement) of the gap is
divided by the original gap length [15,16,18]. Fibrous tissue can tolerate up
to 20% strain, cartilage can tolerate up to 10%, and bone can tolerate only
2% [16]. Excessive strain limits the development of lamellar bone. Clinically,
the consequences of this interfragmentary strain theory are that prolonged
motion produces nonunion and small fracture gaps between major fragments
are associated with higher strain compared with large gaps and comminu-
tions. In operative procedures, therefore, a mechanical approach is used
when applying a plate to obtain cortical contact (rigid stability) during com-
pression of a transverse fracture, whereas a comminuted fracture is treated
biologically with fewer invasive buttressing or neutralizing procedures using
an external ﬁxator or plate possibly in combination with an intramedullary
(IM) pin [4,15,16,18–20]. Recently, interlocking nails have also been used
in the latter fashion to treat comminuted fractures in cats [3]. In Germany,
a newly developed IM nail with three lamellae (Trilam nail) extending down
the pin shaft has been used to treat feline long bone fractures biologically [21].

Sources of blood supply to the fracture site include intramedullary, peri-

osteal, and extraosseous soft tissues. Transformation of the fracture hema-
toma into healing bone follows an orderly sequence of inﬂammation, repair,
and remodeling phases [15–17]. These stages begin immediately after trauma
to the bone occurs and exist in overlapping temporal relations: inﬂamma-
tion (days) provides cellular elements and growth factors, whereas angiogen-
esis and ﬁbroplasia predominate in the repair (weeks) phase.

Undiﬀerentiated pluripotential cells in the fracture develop into ﬁbro-

blasts and then into chondroblasts or osteoblasts based on oxygen tension
(low for cartilage and high for bone cells). During remodeling (months), a
mineralized callus is transformed into woven and then lamellar bone via
osteoclastic and osteoblastic cellular activities. In cases of rigid ﬁxation
(compression plate) and cortical apposition of fragments, direct bone union
occurs without cartilage precursors. Gap healing is a type of direct union
characterized by stable ﬁxation (neutralization plate, external ﬁxation) and
fracture gaps of less than 1 mm. With primary bone union, radiographic evi-
dence of healing is identiﬁed by the absence of a fracture line and a contin-
uous medullary cavity with cortical outlines. Fracture healing and bone
column reconstruction are thus considered ‘‘anatomic.’’ Secondary or in-
direct bone union is associated with fracture gaps, comminuted segments,
micromotion, and implants, such as pins, wires, and external ﬁxators. In this
situation, transformation of the ﬁbrocartilaginous callus into bone tissue
reduces strain during healing. Radiographically, a combined endosteal and
periosteal osseous callus is visualized [8,15,22]. Subsequent remodeling in
the years to follow alters the size and shape of the bone callus. Assuming

930

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

that joint planes above and below the fracture(s) are parallel, this type of
fracture healing is considered ‘‘functional,’’ because anatomic reconstruc-
tion of the damaged bone column has not been attempted [19,20,23].

Implant selection

External and internal implants used in feline orthopedics include cerclage

and Kirschner (K) wires, IM pins, bone plates, screws, interlocking nails,
and external skeletal ﬁxation (ESF) systems (Tables 2 and 3). The ESF
approach uses metal clamps and connecting bars (IMEX Veterinary Inc.,

Table 3
Characteristics of implants for fracture repair

Implant

Characteristics

Cerclage wire

Inexpensive; requires periosseous dissection and at least two

wires per oblique fracture; often loosen or break and delay
healing; provides fragmentary apposition; placed adjacent to
Kirschner wires to prevent slippage; ancillary to plate or pin
ﬁxation; rarely used alone

External skeletal ﬁxation

Moderate expense but reusable bars and clamps; limited or no

surgical exposure; useful for open wounds; variable frames
can neutralize most distractive forces; staged disassembly to
load bone during healing; can be combined with
intramedullary pin; rigorous postoperative care

Kirschner wire

Inexpensive; requires periosseous dissection; useful for

fragmentary apposition and to reduce shear; ancillary with
cerclage wires to plate or pin ﬁxation; rarely used alone

Intramedullary pin

Inexpensive; requires periosseous dissection and eventual

retrieval; controls bending and provides axial alignment;
usually combined with wires, external skeletal ﬁxation, or
plate; rarely used alone

Plate and screws

Expensive; requires extensive periosseous dissection; neutralizes

all distractive forces and provides early weight bearing; can
be combined with intramedullary pin, stacked, or cut to
length

Interlocking nail

Moderate expense; requires periosseous dissection; neutralizes

all distractive forces

Table 2
Implants for feline orthopedic patients

Fixation

Sizes/Types

Cerclage wires

20–22 gauge; loop or twist tie

External ﬁxation pins

0.0625, 0.078, or 0.044 inch in diameter; smooth or threaded;

attached directly to acrylic column or via small clamps to
metal or carbon connecting bars

Kirschner wires (pins)

0.035, 0.045, 0.054, or 0.062 inch in diameter

Intramedullary pins

0.0625, 0.078, or 0.094 inch in diameter; smooth; double trochar

Plate and screws

2.0–2.7-mm diameter screws; dynamic compression,

reconstruction, or cuttable plate

Interlocking nails

2.0-mm diameter screws; 4.0–4.7-mm diameter nails

931

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

Longview, TX; Securos Inc., Charlton, MA) or acrylic tubes (Innovative
Animal Products, Inc., Rochester, MN) attached to percutaneous pins.
Threaded percutaneous pins enhance ESF frame rigidity compared with
smooth pins; however, the raised thread design requires predrilling for
placement. The ﬁxator pinhole diameter should be approximately 25% to
30% of the bone diameter to avoid creating stress risers and iatrogenic frac-
tures; this may limit the location of positive-proﬁle pins to metaphyseal and
condylar regions of bone [2,24,25].

It is impossible to identify the ‘‘best’’ ﬁxation device because of the var-

iability in patients, lesions, expertise, equipment, and ﬁnances. Each implant
has advantages that should be maximized and disadvantages that need to be
minimized. Combinations of ﬁxations should be considered: an IM pin (con-
trols bending) with ESF (fewer percutaneous pins needed, controls rotation
and collapse), an IM pin with bone plate (minimal fracture dissection, plate
protected across defects), cerclage wires (fragment apposition) alongside K
wires (fragment apposition, reduced cerclage slippage), and an IM pin with
K/cerclage wires. Partially threaded IM pins are less desirable than smooth
pins because their insertion is more diﬃcult, the thread-shaft junction is
predisposed to failure, and they fail to provide increased bone-holding
strength [26].

The interlocking nail system (Innovative Animal Products, Inc.) is com-

posed of an IM pin or nail with prefabricated holes to receive proximal and
distal transcortical screws. This combined approach controls compressive,
tensile, bending, and rotational distractive forces. A recently described
lamellar IM pin provided support against axial, bending, and rotational
forces; source, size, and cost of equipment for treating feline patients were
not delineated [21].

In addition to these implants, fresh or frozen (Veterinary Transplant

Services, Inc., Kent, WA) cancellous grafting should be considered during
open repair involving major bone defects or in cases of anticipated poor heal-
ing as a result of or disease [27]. Fresh autogenous tissue can be harvested
from the greater tubercle of the humerus, wing of the ilium (along with cort-
ical bone), or medial tibial tuberosity. Bone union is enhanced by placing this
material at the main fracture site before wound closure. It provides osteogen-
esis from donor site cells, osteoinduction of recipient site pluripotential cells
by bone morphogenetic proteins in the graft, and osteoconduction as a trellis
for ingrowing fracture ﬁbrovascular elements. Commercially available graft
material provides a lattice (cancellous bone chips) and bone morphogenetic
proteins (demineralized cortical bone powder); it can be ‘‘extended’’ by mix-
ing with the patient’s blood or with minimal autogenous graft tissue. Bioac-
tive glass, a ceramic composed of calcium, sodium, phosphates, and silicon
dioxide, has also been used for osteoconduction and possible osteoinduction
[3,27]. Cortical chip autografts and banked allografts have been used to
promote fracture healing in experimentally created feline tibial nonunion
models stabilized with a bone plate and screws [6,7].

932

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

Light-weight cast ﬁxation is a useful alternative to surgical repair for

tibial or ulnar fractures; ideal scenarios involve healthy, indoor, nonsenile
patients with minimally (50% or more major fragment overlap) displaced
closed fractures [5,10,28]. Casts can be initially applied proximal to the
elbow or stiﬂe to immobilize joints above and below the lesion. As clinical
and radiographic evidence of bone union is noted, replacement casts can
be placed just below these joints to facilitate mobility and reduce soft tissue
irritation and patient morbidity.

Casts decrease but do not prevent bending and rotational instabilities if

interdigitation of fragment ends (transverse fractures) and muscular con-
traction help to lock in the fracture (Fig. 2). Over-the-shoulder or hip joint
spica casts can be used for humeral or femoral fractures in some cases as a
patient ‘‘salvage’’ alternative to euthanasia when surgical repair or amputa-
tion cannot be performed. Muscular contraction, bone displacement, and
ineﬃcient stabilizing capabilities by these casts limit their usefulness com-
pared with surgical treatments. Forelimb amputation is best performed by
removal of the scapula and using regional muscle to protect the thoracic
inlet and major axillary neurovascular elements [5]. For hindlimb amputa-
tion, a midshaft or proximal third femoral osteotomy provides satisfactory

Fig. 2. Healed midshaft tibial/ﬁbular fractures and valgus angulation after application of a
synthetic cast.

933

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

cosmetic results and protection of the inguinal region. Cage rest can permit
bone healing in immature patients with incomplete or mildly displaced frac-
tures; however, its eﬃcacy is unpredictable in most other cases.

Surgical approaches

Based on the concepts originally described in human orthopedics, atten-

tion has been directed in veterinary surgery to treat comminuted fractures in
a more ‘‘tissue-friendly’’ approach via biologic osteosynthesis in contrast to
precise anatomic reconstruction after major surgical dissection [18–20]. This
biologic approach is characterized by reduced surgical invasiveness at the
fracture site to preserve perosseous vascularity while still obtaining fracture
stability by using an IM pin combined with a bone plate or external ﬁxator
or an interlocking nail (Figs. 3 and 4) [3,15,18,20,29,30]. In the combined
plate/rod or ESF/pin procedure, an IM pin approximately one third of the
narrowest diaphyseal diameter is initially applied to a humeral, femoral, or
tibial fracture to provide axial alignment and regain normal bone length.
Bone fragments are handled gently, periosteal stripping is avoided, and the
fracture hematoma is not aspirated. Central comminutions can be encircled
with suture not for fracture stability but for enhanced regional callus forma-
tion along the bone axis. External ﬁxator pins or a bone plate and screws are
then applied proximally and distally to the main bone fragments in an alter-
nating manner ensuring joint parallelism above and below the injured bone.
The IM pin protects the plate from strains associated with central bone
defects and open screw holes and does not need to be removed. If ESF is
used for femoral fractures, the IM pin can protrude proximally through the
gluteal muscles and is connected (‘‘tied-in’’) to the ﬁxator frame; this
increases stability of the combined internal/external ﬁxation and avoids sub-
cutaneous soft tissue/sciatic nerve irritation compared with an unconnected
cut IM pin and external frame (Fig. 5) [2,23]. A tie-in conﬁguration can be
used for humeral fractures if the pin exits cranially to the greater tubercle
and does not compromise limb function (Fig. 6) [23,25]. An externally con-
nected IM pin is usually the ﬁnal implant removed during staged disassem-
bly of the ESF conﬁguration during the bone healing process. In addition to
ESF/pin and plate/rod combinations, an interlocking nail can be used to
treat complex fractures in a tissue-friendly biologic manner [3]. With this
system, an IM nail is used initially to obtain axial alignment and length, and
subsequent proximal and distal transcortical screws engage the nail to stabi-
lize the main fragments; centrally located comminutions and the main frac-
ture hematoma can be avoided to preserve inherent vascularity. Critical to
any operative procedure is adherence to Halsted’s principles for surgical dis-
section: asepsis, preservation of vascular elements, closure of dead space,
gentle tissue handling, and removal of necrotic tissue.

Closed repair (the most traditional biologic approach) and ESF applica-

tion are more easily performed with radial or tibial fractures compared with

934

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

humeral or femoral lesions, because the fewer muscles below the elbow and
stiﬂe joints cause less displacement and ease reduction of bone fragments
[4,18,31]. Additionally, a hanging limb preparation used during radial or
tibial fracture repairs aids in obtaining axial realignment [18,31]. In this
maneuver, the paw is attached proximally to a ceiling hook or intravenous
stand, and the surgery table is dropped a few centimeters away from the
patient so that body weight produces a straight hanging limb. Postoperative
radiographic assessment helps to guide manipulations of the external
clamps, pins, and bones to avoid severe angular and rotational discrepan-
cies. Successful closed reduction and blind IM pinning of femoral and tibial
fractures has also been described [32]. Young cats with transverse fractures
treated within 72 hours of injury had the best prognosis.

Fig. 3. (A) Preoperative radiograph of a comminuted midshaft femoral fracture in a mature
cat. (B) Postoperative view after application of an intramedullary pin and a 2.7-mm dynamic
compression plate using a biologic approach for fracture ﬁxation.

935

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

Implant removal

General indications for removal of ﬁxation devices include recurrent pain

or lameness associated with regional soft tissue irritation, implant migra-
tion, obstruction to healing by loose or infected implants, and a loose or un-
necessary implant associated with a healed fracture. Speciﬁc complications
have been described for certain ﬁxations. Fracture or implant-associated
neoplasia and stress protection of bone by plates have not been well doc-
umented in cats compared with dogs [33,34]. IM pins penetrating joint sur-
faces and aﬀecting movement are removed. Pins and interlocking nails used
for femoral fractures and causing sciatic neuropraxia warrant removal [3].

Fig. 4. (A) Comminuted proximal tibial shaft fractures in a senile cat. (B) Postoperative
radiographic views after application of an intramedullary pin and a type II ﬁxator.

936

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

Fig. 5. (A) Preoperative view of an open, comminuted, distal femoral gunshot injury.
(B) Postoperative view after placement of an intramedullary pin ‘‘tied-in’’ to a modiﬁed type
Ia external skeletal ﬁxation (ESF) frame. The distal centrally threaded condyle pin is connected
laterally (two bars) and medially (cranial, bent bar). (C) Radiographic view 3 months after
surgery depicting fracture healing and a reduced ESF frame.

937

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

Fig. 6. (A) Oblique, comminuted, midshaft humeral fracture in a young cat. (B) Postoperative
radiograph using a limited approach and application of an intramedullary pin externally
connected (‘‘tied-in’’) to a type Ia external ﬁxator. (C) Follow-up radiograph 10 weeks after
surgery depicting bone union and a ﬁxator frame previously reduced by removal of two pins at
5 weeks.

938

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

K wires often migrate into extraosseous soft tissue regions and do not cause
clinical problems. Cerclage wires are removed because of looseness or break-
age and subsequent delaying of healing as a result of motion-induced trau-
ma to ﬁbrovascular elements at the fracture site. Bone plates and screws are
removed because of premature failure or association with chronic osteo-
myelitis [33]. Percutaneous external ﬁxation pins are removed sequentially
during healing to load the bone and enhance fracture callus remodeling or
if they are loose and causing lameness from perosseous reactions (see Figs. 5
and 6) [18].

Fracture healing

Determination of feline fracture healing rates and guidelines for radio-

graphic assessment are diﬃcult to standardize because of variables, such
as patient age, type of injury, and method of ﬁxation, including cancellous
grafting [2,22,25,27,32,35]. Immature (<1 year of age) cats should be eval-
uated every 2 to 4 weeks, whereas young (1–5 years of age) and middle-aged
(6–10 years of age) patients can be examined at 4- to 6-week intervals. Older
patients warrant re-examination at 6- to 8-week intervals.

In a retrospective multi-institutional review (J. Harari, unpublished data)

of feline fractures treated with ESF, the only consistent variable aﬀecting
healing rates was patient age—immature cats healed faster than older cats re-
gardless of the injury or method of ﬁxation. Furthermore, the healing rates
for cats were generally longer than in other studies involving ESF-treated
canine fractures. This may be caused by excessive ﬁxator rigidity when using
frames relatively similar to those being used for canine fracture patients [22].
Other reported complications of feline fracture healing include nonunions of
single IM pin ﬁxation of long bones as a result of rotational instability and
delayed union of open distal tibial injuries [36].

Humeral fractures

The feline humerus is more slender and straight than its canine counter-

part, thus permitting ease in application of IM pins and plates. The brachial
groove also has a less distinct curvature [37,38]. On the distal medial aspect
of the bone, a supracondylar foramen contains the median nerve and bra-
chial artery. Neurovascular damage secondary to bone fragmentation or
implant impalement (medial bone plating) is a potential complication [5].
There is no supratrochlear foramen as in other carnivores; a coronoid fossa
craniodistally receives the medial coronoid of the ulna during ﬂexion of the
elbow joint [37,38]. Cats have a relatively straighter and wider condyle than
dogs; hence, condylar fractures are less frequent [8].

Most fractures are comminuted and involve the middle and distal por-

tions of the diaphysis [5,8,10,39]. Although radial nerve injury with these

939

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

fractures is not common, evidence of nerve dysfunction (lack of sensation on
dorsum of the paw, lack of weight bearing or limb extension) should be
assessed before and after surgery. A lateral approach to the shaft is most
often used to apply stabilizing implants, such as an IM pin, cerclage/K
wires, ESF, bone plate/screws, and interlocking nails (see Fig. 6). The ten-
sion side of the bone is the craniolateral surface, although the technical ease
of medial bone plating has been documented in dogs [31]. External ﬁxation
devices are applied laterally as a result of the proximity of the thoracic wall
on the medial side of the limb. A distal percutaneous pin can be placed
safely through the humeral condyle; it exits medial to the bone and should
be connected with a cranial bar attached to lateral pins. An IM pin should
exit laterally by the greater tubercle to avoid the shoulder joint proximally
and is placed into the medial aspect of the humeral condyle or epicondyle
distally to prevent trauma to the elbow joint. Proximal bone injuries involv-
ing the physis can be repaired via small pins or large K wires placed normo-
grade into the shaft using a lateral approach. Distal fractures to the condyle
can be approached laterally or medially and warrant initial reconstruction
and stabilization of the articular lesion with small pins or a screw; the con-
dyle is then stabilized to the humeral shaft by IM pins or wires [5,10].

Radial and ulnar fractures

The radius is craniolateral to the ulnar proximally and medial to it dis-

tally; along with muscular attachments, this orientation permits supination
and pronation of the limb [37,38]. The relatively straight ulna is larger than
the radius, and the two are joined in the middiaphysis by an interosseous
membrane. A caudal process (radial tuberosity) below the radial head serves
as an attachment for the biceps brachia muscle. The saucer-shaped appear-
ance and slight curvature of the radial shaft make external ﬁxation challeng-
ing in a medial-to-lateral plane.

Because of its weight-bearing status, fractures of the radius are primarily

treated, although olecranon and ulnar styloid process lesions can produce
joint instabilities and should also be addressed. Furthermore, severely com-
minuted proximal radial fractures may sometimes require external ﬁxation
of the proximal ulna or distal humerus (temporarily) to permit en bloc bone
union. IM pinning of radial fractures is not recommended because of carpal
joint trauma during pin placement and subsequent protrusion as well as lack
of implant stability.

Most fractures involve the middle and distal aspects of the radial shaft

[5,10]. Small plates are applied on the cranial (tension side) aspect of the
bone using a craniomedial approach (Fig. 7). The cephalic vein medially and
the extensor muscles/tendons cranially are identiﬁed and protected dur-
ing the procedure. For ESF, small external ﬁxation pins can be applied
in a medial-to-lateral plane (type Ia or II) or angled from the midline in a

940

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

craniocaudal plane (type Ib) to obtain more secure bone purchase. Casting
is a reasonable nonsurgical option for treatment of a minimally displaced
radial shaft fracture.

Fig. 7. (A) Three views of a distal radius/ulna nonunion treated 6 months earlier with external
skeletal ﬁxation (ESF). (B) Postoperative views after a second surgery involving placement of
two stacked cuttable plates and 2.0-mm screws.

941

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

Femoral fractures

The feline femur is a pronounced straight tubular bone protected by mus-

culature on all sides. A large medullary cavity and abundant perosseous vas-
cular supply contribute to rapid healing. Sciatic nerve injuries are not often
identiﬁed before surgery; the proximal location of the nerve near an IM pin
or nail placed by the greater trochanter makes it susceptible to iatrogenic
trauma, however. Most fractures involve the femoral diaphysis or condyles
[5,10]. Proximal head and neck injuries are approached from a lateral aspect
and stabilized with K wires, or a femoral head and neck resection is per-
formed. Salvage resections produce excellent clinical results in cats because
of their agility and low body weights [5,10,39].

A lateral approach to the shaft is useful for IM pinning, nailing, bone

plating, or external ﬁxation (Fig. 8) [2,5,10,23,40]. The lateral aspect of the
bone is the tension side. As with the humerus, lateral ESF devices are used
to avoid trauma to the body wall; distally placed condylar pins can safely
exit medially and connect via a cranial bar to lateral pins (see Fig. 5). Tie-
in conﬁgurations and double-connecting bars enhance ESF frame stabil-
ity and reduce the need for excessive percutaneous pinning through lateral

Fig. 8. Seven-week postoperative radiograph of a comminuted femoral shaft fracture repaired
with a 4.7-mm interlocking nail (Courtesy of Robert Parker, DVM, New York, NY).

942

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

Fig. 9. (A) Preoperative radiograph of open, transverse, distal tibial/ﬁbular shaft fractures
secondary to a road traﬃc accident in a young cat. (B) Postoperative view of the external
skeletal ﬁxation (ESF) frame and medial degloving injuries. (C) Radiographic view 2 months
after surgery depicting bone union and a type II ESF frame. (D) Healed soft tissue wounds 2
months after surgery.

943

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

musculature. Distal fractures involving supracondylar or condylar (physeal)
regions can be approached laterally, and the patella can be deﬂected medi-
ally to permit retrograde pinning from the fragments proximally into the
shaft [4,25]. Most physeal injuries are Salter-Harris type I or II lesions [8].

Small (0.045–0.062 in) pins are placed abaxially at the articular condylar

surface to avoid impingement of joint motion and can be crossed dynami-
cally (bend up the canal) or statically (penetrate contralateral cortex) prox-
imal to the fracture; another option is uncrossed, slightly convergent, larger
pins. Intercondylar fractures require reconstruction and stabilization with
pins or bone screw (2.0–2.7 mm) to obtain articular congruity. The condyle
is then reattached to the femoral shaft via IM pins. Closed reduction of
physeal fractures and ﬂexion bandaging have also been described [5].

Tibial and ﬁbular fractures

In the crus, the uniformly tubular tibia is the primary weight-bearing

structure; ﬁbular fractures are not stabilized unless proximal and distal
lesions produce joint instabilities. Avulsion injuries involving collateral lig-
aments are usually stabilized by small pins or wires. Distal tibial shaft frac-
tures are usually open injuries because of the paucity of overlying soft
tissues; reduced extraosseous vascularity may also contribute to delayed
unions (Fig. 9) [4,5]. Most fractures involve the diaphysis, and a medial
approach provides direct access to the bone. Medial saphenous neurovascu-
lar elements in the middiaphyseal region should be preserved. Repairs of

Fig. 10. Preoperative (right) and 6-week postoperative (left) views of a comminuted proximal
tibial shaft fracture in an old cat stabilized with a 2.7-mm dynamic compression plate (Courtesy
of Howard Lawrence, DVM, Spokane, WA).

944

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

fractures include IM pinning or nailing, plating, and external ﬁxation [3–
5,17,35,39].

IM rods are placed normograde from the tibial plateau, medial to the

patellar tendon, and distally into the shaft [32,41]. Although the tension side
of the bone may be along the craniolateral surface, medial plating is per-
formed because it oﬀers an uncomplicated approach to a ﬂat bone surface
(Fig. 10). External ﬁxation is easily accomplished in a medial-to-lateral
plane (type Ia or Ib) or in convergent craniocaudal planes (type Ib). ESF
is used most frequently in tibial injuries because of the tibia’s size and shape,
reduced overlying musculature permitting direct access to the bone and
allowing ease of bone realignment, and need for open wound management
[4,18,31]. As with radial fractures, minimally displaced tibial fractures can
be treated with synthetic casts in patients with good healing potential (see
Fig. 2) [28].

Summary

Orthopedic injuries in cats occur frequently and are amenable to a variety

of surgical and nonsurgical treatment options. Complications and delayed
healing have been reported and can be attributed to improper ﬁxation.
Clinicians have numerous options ranging from external to internal ﬁxation,
casting, cage rest, and limb amputation. The goals of reducing patient mor-
bidity and obtaining a return to normal function warrant the selection of an
appropriate treatment based on the nature of the lesion, available expertise,
and directives of the client.

References

[1] Emery MA, Murakami H. The features of fracture healing in cats after immediate and

delayed open reduction. J Bone Joint Surg Br 1967;49:571–9.

[2] Langley-Hobbs SJ, Carmichael S, McCartney W. Use of external ﬁxators in the repair of

femoral fractures in cats. J Small Anim Pract 1996;37:95–101.

[3] Larin A, Eich CS, Parker RB. Repair of diaphyseal fractures in cats using interlocking

intramedullary nails:12 cases (1996–2000). JAVMA 2001;219:1098–104.

[4] Richardson ER, Thacher CW. Tibial fractures in cats. Compend Contin Educ Pract Vet

1993;15:383–94.

[5] Schrader SC. Orthopedic surgery. In: Sherding RG, editor. The cat: clinical diseases and

clinical management. 2nd edition. New York: Churchill Livingstone; 1994. p. 1649–710.

[6] Toombs JP, Wallace LJ, Bjorling DE. Evaluation of Key’s hypothesis in the feline tibia: an

experimental model for augmented bone healing studies. Am J Vet Res 1985;46:513–7.

[7] Toombs JP, Wallace LJ. Evaluation of autogeneic and allogeneic cortical chip grafting in a

feline tibial nonunion model. Am J Vet Res 1985;46:519–28.

[8] Farrow CS. Radiology of the Cat. St. Louis: Mosby; 1994. p. 255.
[9] Kolata RJ, Kraut NH, Johnston DE. Patterns of trauma in urban dogs and cats: a study of

1000 cases. JAVMA 1974;164:499–502.

[10] Leighton RL, Robinson G. Orthopedic surgery. In: Holzworth J, editor. Diseases of the

cat. Philadelphia: WB Saunders; 1987. p. 100–45.

945

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

[11] Phillips IR. A survey of bone fractures in the dog and cat. J Small Anim Pract 1979;20:

661–74.

[12] Whitney WO, Mehlhaﬀ CJ. High-rise syndrome in cats. JAVMA 1987;191:1399–403.
[13] Griﬀon DJ, Walter PA, Wallace LJ. Thoracic injuries in cats with traumatic fractures. Vet

Comp Orthop Traumatol 1999;12:74–7.

[14] Johnson KA. Osteomyelitis in dogs and cats. JAVMA 1994;205:1882–7.
[15] Hulse DA, Johnson AL. Fundamentals of orthopedic surgery and fracture management.

In: Fossum TW, editor. Small animal surgery. St. Louis: Mosby; 1997. p. 705–33.

[16] Radasch RM. Biomechanics of bone and fractures. Vet Clin North Am Small Anim Pract

1999;29:1045–82.

[17] Remedios AR. Bone and bone healing. Vet Clin North Am Small Anim Pract 1999;

29:1029–44.

[18] Aron DN, Palmer RH, Johnson AL. Biological strategies and a balanced concept for repair

of highly comminuted long bone fractures. Compend Contin Educ Pract Vet 1995;17:35–49.

[19] Gerber C, Mast JW, Ganz R. Biological internal ﬁxation of fractures. Arch Orthop

Trauma Surg 1990;109:295–303.

[20] Palmer RH. Biological osteosynthesis. Vet Clin North Am Small Anim Pract 1999;

29:1171–86.

[21] Hach V. Initial experience with a newly developed medullary stabilization nail (Trilam

nail). Vet Comp Orthop Traumatol 2000;13:109–14.

[22] Johnson AL. Bone healing with external skeletal ﬁxation. In: Proceedings of the Ninth

Annual Complete Course in External Skeletal Fixation. Athens (GA): University of
Georgia; 2002. p. 99.

[23] Aron DN, Foutz TL, Keller WG. Experimental and clinical experience with an IM pin/

external skeletal ﬁxator tie-in conﬁguration. Vet Comp Orthop Traumatol 1991;4:86–94.

[24] Harari J, Seguin B, Bebchuk T. Closed repair of tibial and radial fractures with external

skeletal ﬁxation. Compend Contin Educ Pract Vet 1996;6:651–64.

[25] Langley-Hobbs SJ, Carmichael S, McCartney W. External skeletal ﬁxation for stabilisation

of comminuted humeral fractures in cats. J Small Anim Pract 1997;38:280–5.

[26] Howard PE, Brusewitz GH. An in vitro comparison of the holding strength of partially

threaded vs nonthreaded intramedullary pins. Vet Surg 1983;12:119–22.

[27] Fitch R, Kerwin S, Newman H. Bone autografts and allografts in dogs. Compend Contin

Educ Pract Vet 1997;19:558–75.

[28] Oakley RE. External coaptation. Vet Clin North Am Small Anim Pract 1999;29:1083–96.
[29] Hulse DA, Hyman W, Nori M. Reduction in plate strain by addition of an intramedullary

pin. Vet Surg 1967;26:451–9.

[30] Hulse DA, Kerry K, Fawcett A. Eﬀect of intramedullary pin size on reducing bone plate

strain. Vet Comp Orthop Traumatol 2000;13:185–90.

[31] Harari J, Roe SC, Johnson AL. Medial plating for repair of middle and distal diaphyseal

fractures of the humerus in dogs. Vet Surg 1986;15:45–8.

[32] Newman ME, Milton JL. Closed reduction and blind pinning of 29 femoral and tibial

fractures in 27 dogs and cats. J Am Anim Hosp Assoc 1989;25:61–8.

[33] Emmerson TD, Muir P. Bone plate removal in dogs and cats. Vet Comp Orthop

Traumatol 1999;12:74–7.

[34] Fry PD, Jukes HF. Fracture associated sarcoma in the cat. J Small Anim Pract 1995;

36:124–6.

[35] Ross JT, Matthiesen DT. Use of multiple pin and methylmethacrylate external skeletal

ﬁxation for the treatment of orthopaedic injuries in the dog and cat. Vet Comp Orthop
Traumatol 1993;6:115–21.

[36] Peck JN. Feline fracture complications. In: Proceedings of the 11th Annual American

College of Veterinary Surgeons Symposium. Chicago (IL): Amer Coll Vet Sx; 2001. p. 378.

[37] Crouch JE. Text-Atlas of Cat Anatomy. Philadelphia: Lea & Febiger; 1969. p. 43–8.

946

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

[38] Hudson LC, Hamilton WP. Atlas of feline anatomy for veterinarians. Philadelphia:

WB Saunders; 1993. p. 34–7.

[39] Smith CW. Feline orthopedics. Feline Pract 1994;22:16–27.
[40] Foland MA, Shwarz PD, Salman MD. The adjunctive use of half-pin (Type I) external

skeletal ﬁxators in combination with intramedullary pins for femoral fracture ﬁxation. Vet
Comp Orthop Traumatol 1991;4:77–85.

[41] McLaughlin R. Internal ﬁxation: intramedullary pins, cerclage wires, and interlocking

nails. Vet Clin North Am Small Anim Pract 1999;29:1097–116.

[42] Hill FWG. A survey of bone fractures in the cat. J Small Anim Pract 1977;18:457–63.
[43] Knecht CD. Fractures in cats: A survey of 100 cases. Feline Pract 1978;8:43–6.

947

J. Harari / Vet Clin Small Anim 32 (2002) 927–947

Lumbosacral and pelvic injuries

Otto I. Lanz, DVM

Department of Small Animal Clinical Sciences, Virginia-Maryland Regional

College of Veterinary Medicine, Duckpond Drive, Blacksburg, VA 24061, USA

Fractures of the pelvis have been reported to account for approximately

20% to 22% of fractures that occur in cats [1–6]. The incidence of pelvic frac-
tures is second only to fractures of the femur, which account for 33% to 38%
of all appendicular skeleton fractures in the cat [6]. The age distribution for
cats with pelvic fractures in one study was found to be less than 12 months
of age. One large retrospective study concluded that traumatic injuries of all
types occurred in cats with a mean age of 15.6 months [7], whereas another
retrospective study found that fractures of the pelvis occurred in cats with a
mean age of 16.8 months [8].

The sex distribution reported in one study of cats with pelvic fractures

was found to be virtually identical [6–8]. Some authors have reported that
trauma generally tends to occur more often in male cats than in female cats;
however, a study of fractures of all types in cats found a male-to-female
ratio of 51:49 [9]. The etiology of pelvic fractures in cats remains controver-
sial in the veterinary literature. One retrospective study reported that falls
from great heights are the most commonly identiﬁed cause of pelvic frac-
tures in cats [10], and other retrospective studies name motor vehicle trauma
as the most commonly identiﬁed cause of pelvic fractures in cats [3,4,6–8].

The force of the traumatic insult required to fracture the pelvic bones is

such that simultaneous multiple injuries can be expected to occur. In a large
study evaluating pelvic fractures in cats, a 58.6% incidence of additional
injuries was found [10]. The musculoskeletal system was involved in
56.4% of these cases with additional injuries, the abdomen and thorax were
involved in 23.2% (mainly consisting of hemothorax and pneumothorax),
and the nervous system was involved in 20.4% [10]. Another large retrospec-
tive study found that 71.8% of cats suﬀered serious extrapelvic injury after
pelvic fractures [6]. The most common injury was fracture of the sacrum,
which was seen in 19.4% of the cases [6]. Coxofemoral luxation, Salter-Harris

Vet Clin Small Anim 32 (2002) 949–962

E-mail address: olanz@vt.edu (O.I. Lanz).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 2 9 - 3

type I fracture of the femoral head, or fracture of the femoral neck occurred
in 13.6% of the cases (Fig. 1) [6]. In this same study, cats less than 12 months
of age had Salter-Harris type I fractures of the proximal femur [6]. Older
animals had either coxofemoral luxations or comminuted femoral neck frac-
tures. Femoral fractures not involving the coxofemoral joint occurred in
12.6% and 10.6% of cats, and all cats displayed various degrees of ischiatic
nerve dysfunction [6]. Surprisingly, only 10.6% of cats had radiographic
conﬁrmation of pneumothorax or pulmonary contusion [6].

Soft tissue trauma

Various other soft tissue injuries are associated with pelvic fractures. Uri-

nary tract trauma is commonly associated with pelvic fractures, although
only 0.5% of cats with pelvic trauma are diagnosed with urinary tract inju-
ries [10]. Rupture of the bladder is reported to be the most common soft tis-
sue injury associated with pelvic fractures (Fig. 2). Three mechanisms of
injury have been described to induce bladder trauma. First, intravesicular
pressure increases quickly in a full bladder, which is caused by rapid eleva-

Fig. 1. Ventrodorsal radiograph of a cat with a chronic coxofemoral dislocation and healed
femoral fracture.

950

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

tion of abdominal pressure after blunt trauma of the caudal abdomen.
Second, penetrating bone fragments from the fractured pelvis cause direct
trauma to the bladder. Finally, bruising of the bladder wall may result in
localized areas of necrosis. Other injuries to the urinary system associated
with pelvic fractures include avulsion of the bladder neck from the urethra,
torn urethra, avulsion of the ureters, and renal trauma.

Peripheral nerve damage is commonly found in pelvic fracture cases. One

study reported an incidence of 13.9% of lumbosacral plexus damage in cats
with pelvic fractures [10]. Avulsion of the lumbosacral roots is rare, but sa-
croiliac luxations with cranial displacement of the ilium have been reported
to cause damage to the L6 and L7 nerve roots [11]. Lumbosacral trunk
injury is reported to occur with ilial fractures with craniomedial displace-
ment of bone fragments. Sciatic nerve entrapment may occur secondary
to ischial or acetabular fractures. Sciatic deﬁcits in cats have been reported
to be a result of intimate contact with bone fragments or callus formation
during the healing process.

Gastrointestinal injury directly related to pelvic fractures is considered a

rare occurrence. Lacerations of the rectum most commonly involve the cau-
dal 4 to 6 cm of the organ [12]. The most common long-term gastrointestinal
complication of pelvic fractures is megacolon. Other soft tissue injuries that
can be associated with pelvic fractures include abdominal wall herniation,
diaphragmatic herniation, and rupture of the prepubic tendon [1,4,6,10].

Fig. 2. Lateral abdominal radiograph of a retrograde cystourethrogram demonstrating a
ruptured bladder.

951

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

Initial assessment

Because pelvic trauma in the cat can cause injury in other areas of the

body, one must completely evaluate the entire patient and not limit the diag-
nostic evaluation to the pelvis and immediate surrounding area. The follow-
ing information should be obtained during the initial history and physical
examination: cause of the trauma; duration of time between the traumatic
incident and the examination; ability of the animal to ambulate, urinate, and
defecate; and whether the animal is hemorrhaging.

Physical examination should start with a visual assessment of the animal.

The thorax should be inspected for visible abnormalities along with the rate
and quality of chest wall movement. The body wall, inguinal area, and peri-
neum should be evaluated for bruising, distention, and asymmetry. The
thorax should be auscultated for cardiac arrhythmias, abnormal lung
sounds, shifted cardiac sounds, and borborygmi. This is important, because
these signs are associated with diaphragmatic hernias. Abdominal palpation
should be done systematically. An abdomen that feels empty on palpation is
also suggestive of a diaphragmatic hernia.

If possible, a rectal examination should be performed. The pelvic exami-

nation can provide information on the nature of the pelvic fractures, the
degree of displacement, and the width of the pelvic canal. Assessment of
peripheral nerve damage is essential in all cases of pelvic fractures. It has
been reported that cutaneous, sensory, and voluntary motor deﬁcits are the
most common peripheral nerve deﬁcits associated with pelvic fractures [1].
Pelvic limb reﬂex abnormalities and severe pain may also be present. Anal
tone should be assessed during the rectal examination as well as tail sensa-
tion and voluntary tail movement. Serial neurologic examinations may be
necessary because of their prognostic signiﬁcance.

With the high incidence of extrapelvic trauma, routine thoracic and

abdominal survey radiographs are warranted (Fig. 3). Early radiographic

Fig. 3. Lateral thoracic radiograph of a cat with a diaphragmatic hernia.

952

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

evaluation of the thorax and abdomen is most beneﬁcial, but this should not
interfere with initial resuscitation of the animal or cause severe stress to the
patient. It is important to note that not all thoracic injury is radiographically
visible immediately preceding the traumatic event. Pneumothorax, air bron-
chograms, pleural eﬀusion, and diaphragmatic hernias are other conditions
that should be ruled out after radiographs of the thorax are taken.

Abdominal survey radiographs are assessed for peritoneal ﬂuid, such as

blood, urine, bile, or exudates as well as the presence of free abdominal air,
which may be indicative of a ruptured hollow viscus organ. The retroperito-
neal space should be closely examined on abdominal radiographs. Increased
density and loss of the renal outline may be indicative of urine extravasation
or hemorrhage. An excretory urogram should be performed if one suspects
injury to a ureter or kidney. A positive-contrast retrograde cystourethro-
gram is indicated if one suspects a rupture of the bladder or urethral dam-
age. Abdominal centesis or a diagnostic peritoneal lavage is indicated if
there is evidence of free abdominal ﬂuid. Fluid samples should be submitted
for cytology testing and analysis. It is important to remember that needle
paracentesis is only 47% diagnostic, whereas a diagnostic peritoneal lavage
is reported to be 95% diagnostic [13,14].

Conservative versus surgical management

Indications for surgical versus conservative management of pelvic frac-

tures in cats are not clearly deﬁned in the veterinary literature [3,4,15,16].
Many veterinarians believe that most pelvic fractures in the cat can be man-
aged conservatively; however, no long-term studies have been performed in
cats to support this statement (Fig. 4). Guidelines for surgical reduction and
stabilization have been purported in cats. Surgical intervention should be
performed when there is narrowing of the pelvic canal, extreme pain, neuro-
logic deﬁcits, acetabular involvement, and inability to walk after 3 days of
conservative management (Fig. 5) [2–4,16]. Cats with extreme pain, inability
to walk, or altered status often have damage to the sciatic nerve or lumbo-
sacral trunk caused by cranial and medial displacement of the ilium or
fracture/dislocation of the sacroiliac joint.

Early surgical reduction and stabilization should be performed when the

pelvic canal is narrowed on pelvic radiographs and rectal palpation. The
prevalence of megacolon in cats after pelvic trauma remains uncertain. A
delay in surgical intervention has been reported to adversely aﬀect the clin-
ical outcome. Medical treatment consists of oral administration of stool
softeners and laxatives. Chronic partial obstruction of feces through the
colon results in progressive colonic distention and neuromuscular deteriora-
tion [2,16]. Pelvic narrowing may not immediately result in clinical signs
of colonic dysfunction (Fig. 6). In a retrospective study evaluating six cats
with obstipation secondary to stenosis of the pelvic canal, it was found that

953

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

obstipation developed a variable amount of time after pelvic trauma [2].
Delayed versus immediate onset of clinical signs of intestinal obstruction did
not positively contribute to the prognosis; however, a delay in surgical inter-
vention did seem to aﬀect clinical outcome adversely [2] because of chronic
partial obstruction of the colon, resulting in megacolon and intramural neu-
romuscular deterioration. This study did report excellent clinical signs when
the obstruction was relieved within the ﬁrst 6 weeks of pelvic injury and that
medical management of the megacolon did not prevent megacolon from
occurring [2]. Continued administration of laxatives, enemas, and manual
evacuation of the colon should be anticipated in the postoperative period
if surgical correction is attempted.

Surgical procedures that are reported for pelvic reconstruction include

partial pelvectomy, corrective osteotomy with redirection of impinging
bone, triple pelvic osteotomy and symphyseal separation, and distraction
by use of an autogenous or allogenic corticocancellous graft or metallic
implant [2,5]. Partial pelvectomy or repositioning of bones in cats with mega-
colon has the advantage of allowing future evacuation of the colon. If sur-
gery is to be performed, the location of the obstruction is best determined by
rectal palpation. Repositioning of the pelvis is diﬃcult if the injury is long
standing as a result of bony callus or large amounts of dense ﬁbrous tissue.
When pelvic fractures are being managed without surgery, cats should be

Fig. 4. Ventrodorsal radiograph of a cat with an ilial body fracture that was managed with
conservative management. This fracture went on to heal without complications.

954

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

conﬁned for 3 to 4 weeks until pelvic fragments are stable. During this time,
rectal palpation should be performed weekly to ensure early detection of
obstructing callus or bone fragments [2].

Surgical techniques have been directed at re-establishing an adequate

diameter of the pelvic canal. The disadvantages associated with pelvic recon-
struction include the substantial amount of soft tissue dissection necessary
and the need for performing pelvic osteotomies (Fig. 7). Complications
reported with pelvic osteotomies include sciatic nerve injury, laceration of
the urethra, and possible injury to the rectum. It is recommended that a ure-
thral catheter be placed before surgery to help facilitate identiﬁcation of the
urethra and prevent iatrogenic trauma of the urethra [2]. Because of the dis-
advantages and potential complications associated with corrective osteoto-
mies and partial pelvectomy, subtotal colectomies may oﬀer an alternative.

Subtotal colectomies oﬀer an alternative surgical procedure for cats with

obstipation or constipation secondary to pelvic trauma. After surgical
removal of the ileocecal valve and dilated colon, the feces become semi-
formed and can pass easily through the narrowed or collapsed pelvic canal.
A study evaluating subtotal colectomies for the treatment of obstipation sec-
ondary to pelvic fracture malunions in cats recommended that subtotal

Fig. 5. Ventrodorsal radiograph of a cat with an acetabular fracture and ilial body fracture.
This is an example of a case that should be managed surgically and not conservatively.

955

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

colectomy be performed after 6 months of constipation associated with pel-
vic fracture malunion or in cats developing refractory constipation after
pelvic reconstruction [16]. This same study recommended partial pelvic
resection for cats with a history of pelvic trauma within the past 6 months
that have early problems associated with constipation.

Fracture patterns

In a large retrospective study evaluating 103 cats with pelvic fractures, it

was found that the most common fracture involved the pelvic ﬂoor, which
occurred in 90% of the cats [6]. Sacroiliac luxation was the second most
common injury, seen in approximately 60% of the cases, followed by ilial
body fractures, seen in 48.5% of the cases [6]. The sacroiliac luxations were
bilateral in 27% of the cases, and, surprisingly, only 2 cats had sacral wing
fractures [6].

A classiﬁcation scheme for sacral fractures in dogs and cats has been

described as follows [17]:

Type I alar: alar oblique fracture line on the ventrodorsal radiograph

originating on or immediately adjacent to the juxta-articular notch
and terminating on the articular surface of the wing of the sacrum

Fig. 6. (A,B) Pelvic radiographs of a cat with megacolon secondary to past pelvic fractures.

956

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

Type II foraminal: foraminal longitudinal, or oblique fracture line on the

ventrodorsal radiograph through the ﬁrst or ﬁrst and second sacral
foramina

Type III transverse: transverse fracture
Type IV avulsion: avulsion fracture of the origin of the sacrotuberous

ligament

Type V comminuted: comminuted fracture

In a retrospective study evaluating 17 cats with sacral fractures, 53% of

the cats had a type III (transverse) sacral fracture, which was postulated
to result from traction injury or after blunt trauma to the tail base [17]. Most
the cats with a type III fracture of the sacrum were neurologically normal;
however, two cats had tail paralysis, and one cat had tail paralysis and uri-
nary and fecal incontinence [17]. These neurologic deﬁcits were attributable
to damage of the coccygeal nerve roots. Type I fractures occurred in 41% of
the cats, and most of these cats had normal neurologic signs [17]. One cat
(6%) sustained a type II sacral fracture. Type IV or V sacral fractures were
not observed in any of the feline cases [17]. In this study, most cats had bilat-
eral sacroiliac subluxations, which were present without concurrent pelvic
injuries. The suggestion has been made that this may be attributable to a rel-
atively weak attachment between the pelvis and sacrum [17].

Fig. 7. (A,B) Pelvic radiographs of a cat with megacolon in which pelvic restoration was
attempted. The cat in these radiographs eventually had a subtotal colectomy for its megacolon.

957

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

In 48.5% of cats with ilial body fractures, 36% fractures were oblique and

14% were comminuted, with no cats having bilateral ilial fractures [6].
Ischial body fractures or avulsions of the tuber ischium were seen in 26%
of the cats, with only one animal having bilateral involvement [6]. Ischial
body fractures were signiﬁcantly associated with sacral fractures. The prac-
tical signiﬁcance of this ﬁnding suggests that cats with a fracture of the
ischium must be evaluated for the presence of concurrent sacral fractures.
Acetabular fractures occurred in 17.5% of the cases and were unilateral in
all cases, with most (77%) being two-piece fractures [6].

The most common combination of pelvic fractures reported in cats is pel-

vic ﬂoor fractures with unilateral ilial body fracture and contralateral sac-
roiliac luxation [6,10]. This was seen in 16.5% of the cases [6]. Pelvic ﬂoor
fractures in combination with unilateral ilial body fracture or pelvic ﬂoor
fractures in combination with unilateral sacroiliac luxation were the next
most common combinations of pelvic fractures reported.

Eleven cats (10.6%) had various degrees of ischiatic nerve dysfunction,

ranging from mild paresis to absence of pain perception in response to a
noxious stimulus [6]. There was a strong correlation between ischiatic nerve
function and ipsilateral ilial body fracture. The severity of ilial fracture
seems not to be associated with the presence of ischiatic nerve injury [6].
Deﬁcits in pudendal or pelvic nerve function are noted in only a small per-
centage of the cases, and signs ranged from inability to fully empty the uri-
nary bladder to loss of anal tone, voluntary urination, and perineal
sensation [3,6,17]. Studies have found sacral body fracture, coxofemoral
joint injury, femoral diaphyseal fracture, peripheral nerve injury, and pul-
monary injury to be the most common injuries in cats with fractures of the
pelvis [1,6,10,17]. Injury to the abdominal wall, diaphragm, and lower uri-
nary system were uncommon [1,6,10,17].

Coxofemoral dislocation, Salter-Harris type I fracture of the femoral

head, or fracture of the femoral neck occurs at low frequency in cats with
pelvic fracture [6,9,10,18]. Animals less than 12 months of age have Salter-
Harris type I fractures involving the femoral head, and older animals have
either coxofemoral dislocations or comminuted femoral neck fractures
[6,9,10,18]. The incidence of coxofemoral injury decreases in cats with ipsi-
lateral sacroiliac luxations [17]. Craniodorsal dislocation is the most com-
mon direction of coxofemoral dislocations. Dislocation causes tearing of
the ligament of the head of the femur, joint capsule, and, occasionally, a
small piece of cartilage and bone from the femoral head. In patients without
fractures of the acetabulum or femoral head, closed reduction with or with-
out ﬂexion or an Ehmer sling is indicated [4]. Closed reduction of coxofem-
oral dislocation is associated with a high failure rate of 64.8% in cats and
dogs, whereas the success rate with open reduction approaches 82% [15].
Because there are better functional results with orthopedic salvage proce-
dures, excision arthroplasty is an option for the treatment of cats with recur-
ring coxofemoral dislocation.

958

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

Sacrocaudal fractures

Fracture/luxation of the sacrocaudal area in cats is quite frequently asso-

ciated with neurologic deﬁcits of the nerve roots of the sacral and coccygeal
nerve segments [4,14,19]. Lesions of the sacral and caudal nerve roots often
produce urinary incontinence with a ﬂaccid and distended bladder, fecal
incontinence with a dilated anus, and a paralyzed atonic tail unassociated
with pelvic limb paralysis. The S1 to S3 spinal cord segments and nerve
roots contribute to the pelvic nerve, which transmits sensory information
from and parasympathetic motor innervation to the detrusor muscle. The
pelvic nerve is also responsible for sensory information from and parasym-
pathetic motor innervation to smooth muscle of the descending colon.

The S1 to S3 spinal cord segments and nerve roots contribute to the

pudendal nerve, which transmits sensory information from the external ure-
thral sphincter, anal sphincter, and perineal area. The pudendal nerve also
provides motor innervation to the external urethral sphincter and the stri-
ated muscle of the anal sphincter. Lesions involving the S1 to S3 spinal cord
segments and nerve roots result in an absence of reﬂex urination or defeca-
tion and persistently dilated sphincters.

The bladder is easy to express manually, and feces spontaneously fall out

of the rectum. Smooth muscle contracts even when denervated, resulting in
some autonomous urination and defecation; however, the contractions are
not suﬃcient to empty the entire bladder, and feces must be manually
removed from the rectum. The location of the sacrocaudal fracture/luxation
may not correlate well with the severity of clinical signs or neurologic deﬁ-
cits seen in cats. The inciting traumatic event can produce complete or par-
tial neurologic dysfunction depending on whether there is complete
transection or stretching of the nerve and nerve roots.

The most common causes of sacrocaudal fracture/luxation reported in

cats include automobile trauma and bite wounds [4,14,19]. Most of the
lesions reported in one large retrospective study were at the third sacral and
ﬁrst caudal vertebrae [19]. Concomitant injuries are usually present with
these injuries, and cats present with rear limb weakness or diﬃculty in walk-
ing. On physical examination, hyperpathia is usually elicited on palpation of
the tail head, and the perineal region may be soiled with feces and urine.
Other physical examination ﬁndings include absence of the bulbourethral
reﬂex; decreased sensation to the perineal region; and a large, ﬂaccid, dis-
tended bladder. The tail may have lost its natural curl, and there may not
be voluntary or reﬂex tail movement. Deep pain sensation from the caudal
vertebrae may also be depressed or absent.

Multiple injuries, especially pelvic fractures or femoral fractures, are

common in cats with sacrocaudal fracture/luxation [3,4,6,11,19]. Ocular
trauma is also common, as reported in a previous study of cats with sacro-
caudal fracture/luxation [19]. Rear limb weakness is commonly seen on pre-
sentation of these animals and is attributable to concurrent pelvic or femoral

959

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

fractures. In cats having no evidence of orthopedic trauma other than in the
sacrocaudal area, the rear limb weakness is thought to be secondary to dis-
traction of the cauda equina nerve roots. If the nerve roots are not severed,
recovery of rear limb function is expected and thought to be a transient neu-
rapraxia.

In one study, cats presented with sacrocaudal fracture/luxation were

divided into ﬁve groups depending on their initial neurologic status [19]. Cats
with tail head hyperesthesia only were managed conservatively. In these
cases, chronic sacrocaudal pain may be caused by pressure on peripheral
nerves caused by scar tissue or callus formation. Because cats in this group
function well, surgical decompression or stabilization is not advocated.

Cats initially presented with loss of motor and sensory function to the tail

only have an excellent prognosis for return of tail function without surgical
intervention [19]. Tail amputation is not recommended in these cases,
because these cats have impairment of the caudal nerves with normal puden-
dal and pelvic nerve function. Cats initially presented with a ﬂaccid tail and
urinary retention only can be expected to recover fully. These cats have nor-
mal pudendal nerve function but have urinary retention secondary to pelvic
nerve damage. One cat was reported not to recover completely in this group
because of irreversible damage of the pelvic nerves without clinically evident
pudendal nerve damage [19]. Cats initially presented with a ﬂaccid tail and
diminished anal tone only have a recovery rate of 75% [19]. Cats in this
group have damage to the pudendal, pelvic, and caudal nerves. Some anal
tone, perineal sensation, and innervation to the striated muscle and urethra
remain. Cats initially presented with ﬂaccid tails and absent anal tone have
severe damage to the pudendal, pelvic, and caudal nerves, which may not be
reversible. Most cats presented with sacrocaudal fracture/luxation are cate-
gorized into this group [19]. The recovery rate for this group is 50%. It has
been reported that some cats recover fully even with initial signs that indicate
severe neurologic impairment. Needle electromyography of the anal sphinc-
ter and caudal muscles can be performed 5 to 7 days after an insult to the
sacral and caudal nerve roots. If any motor unit action potentials are seen,
some integrity of the nerves is preserved and may be of prognostic value.

It is still controversial as to whether or not surgical intervention quickens

recovery time and decreases complications. Tail amputation has been sug-
gested to decrease soilage of the tail and to decrease pain associated with
traction of the nerves of the cauda equina secondary to the weight of the tail.
This has not been seen in one large retrospective study of cats with sacrocau-
dal fracture/luxation. Tail amputation or stabilization in conjunction with
routine dorsal laminectomy has also been advocated for the treatment of
cats with sacrocaudal fracture/luxation. The reported success rates in the
veterinary literature after conservative management of cats with sacrocaudal
fracture/luxation are similar to the success rates with surgical intervention.

Medical treatment of cats with sacrocaudal fracture/luxation is aimed at

preventing bladder distention leading to detrusor atony. Often, the bladder

960

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

cannot be manually expressed because of severe spasticity in the urethra and
external and internal sphincter. Attempts at manual expression may injure
the detrusor muscle or rupture the bladder. The cat should be aseptically
catheterized every 6 to 8 hours. In female cats, an indwelling catheter is
more practical and may need to be left in place for up to 7 days. Oral admin-
istration of diazepam at a rate of 2 to 5 mg/kg of body weight per cat every 8
hours may produce enough relaxation to make manual expression of the
bladder possible. Phenoxybenzamine, an a-blocking agent, has also been
reported to reduce urethral spasms. It is important to remember that phe-
noxybenzamine may cause severe hypotensive eﬀects in some animals. After
approximately 1 week, the urethral and anal sphincters relax, and reﬂex uri-
nation and defecation occur. Reﬂex contraction is hyperactive, and the blad-
der empties when ﬁlled with small amounts of urine. A urinalysis should be
performed routinely to monitor cystitis. If bacterial cystitis occurs, culture
and sensitivity testing should be performed so as to administer the appropri-
ate antibiotics. Constipation resulting from pelvic nerve injury can be man-
aged with dietary changes. Laxatives and enemas can be used to manage
occasional episodes of constipation.

In one study, rear limb weakness, urination, and tail problems resolved in

1 week in cats with sacrocaudal fracture/luxation. This same study found
that cats not able to urinate normally within 1 month remained incontinent.
It is important to note that tail function has been reported to take several
months to improve.

Conclusion

A good return to function can generally be expected for cats undergoing

surgical repair of pelvic fractures. The outcome is dependent on thorough
preoperative evaluation, proper diagnostic evaluation, and patient manage-
ment. One must be cognizant of the concomitant injuries that can occur in
cats with pelvic fractures and perform a comprehensive physical examination.
Feline patients with extensive pelvic fractures may require multiple surgical
procedures and intensive postoperative care; despite these eﬀorts, they may
still have long-term orthopedic problems and neurologic dysfunction. Timely
and precise surgical intervention is necessary to impart a favorable prognosis.

References

[1] Verstraete F, Lambrechts N. Diagnosis of soft tissue injuries associated with pelvic

fractures. Compend Contin Educ Pract Vet 1992;14:921–30.

[2] Schrader SC. Pelvic osteotomy as a treatment for obstipation in cats with acquired stenosis

of the pelvic canal: six cases (1978–1989). JAVMA 1992;200:208–13.

[3] Payne J. Selecting a method for managing pelvic fractures in dogs and cats. Vet Med 1993;

969–73.

961

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

[4] DeCamp CE. Principles of pelvic fracture management. Semin Vet Med Surg 1992;7:63–70.
[5] Ferguson J. Triple pelvic osteotomy for the treatment of pelvic canal stenosis in a cat.

J Small Anim Pract 1996;37:495–8.

[6] Bookbinder PF, Flanders JA. Characteristics of pelvic fracture in the cat. Vet Comp

Orthop Traumatol 1992;5:122–7.

[7] Kolata R. Pattern of trauma in urban dogs and cats: a study of 1,000 cases. JAVMA 1974;

5:499–502.

[8] Bennet D. Orthopedic disease aﬀecting the pelvic region of the cat. J Small Anim Pract 1975;

16:723–38.

[9] Hill F. A survey of bone fractures in the cat. J Small Anim Pract 1977;18:457–63.

[10] Bohmer E. Beckenfrakturen beim Hund in den Jahren 1970–1977 [inaugural dissertation].

Munich: Ludwig Maximilians-Universitat Munchen; 1985.

[11] Jacobson A, Schrader SC. Peripheral nerve injury associated with fracture or fracture-

dislocation of the pelvis in dogs and cats: 34 cases (1978–1982). JAVMA 1987;190:569–72.

[12] Muir P. Rectal perforation associated with pelvic fracture in a cat. Vet Rec 1998;142:371–2.
[13] Crowe DT, Crane SW. Diagnostic abdominal paracentesis and lavage in the evaluation of

abdominal injuries in dogs and cats: clinical and experimental investigations. JAVMA
1976;168:700–5.

[14] Bjorling DE, Latimer KS, Rawlings CA, et al. Diagnostic peritoneal lavage before and

after abdominal surgery in dogs. Am J Vet Res 1983;44:816–20.

[15] Ablin L, Gambardella P. Orthopedics of the feline hip. Compend Contin Educ Pract Vet

1991;13:1379–86.

[16] Matthiesen DT, Scavelli TD, Whitney WO. Subtotal colectomy for the treatment of

obstipation secondary to pelvic fracture malunion in cats. Vet Surg 1991;20:113–7.

[17] Anderson A, Coughlan AR. Sacral fractures in dogs and cats: a classiﬁcation scheme and

review of 51 cases. J Small Anim Pract 1997;38:404–9.

[18] Smith CW. Feline orthopedics. Feline Pract 1994;22:16–27.
[19] Smeak D, Olmstead M. Fracture/luxations of the sacrococcygeal area in the cat: a retro-

spective study of 51 cases. Vet Surg 1985;14:319–24.

962

O.I. Lanz / Vet Clin Small Anim 32 (2002) 949–962

Surgical diseases of the feline stiﬂe joint

Ron M. McLaughlin DVM, DVSc

Department of Clinical Sciences, College of Veterinary Medicine,

Mississippi State University, Mississippi State, MS 39762, USA

Abnormalities of the feline stiﬂe joint typically cause hind limb lameness,

joint swelling, and pain. They occur less frequently in cats than in dogs and
are often the result of trauma, such as vehicular trauma, a fall from height,
or an altercation with another animal. Cruciate ligament rupture, patellar
luxation, stiﬂe dislocation, fractures of the patella and distal femur, and sep-
tic arthritis caused by bite wounds are common traumatic causes of stiﬂe
lameness in cats. Fortunately, stiﬂe injuries in cats can often be managed
more conservatively than in dogs, and the prognosis for adequate return
of joint function is generally good.

Nontraumatic causes of stiﬂe lameness in the cat include congenital

patellar luxation and immune-mediated arthropathies. Although immune-
mediated joint diseases are not treated surgically, they are important in the
diﬀerential diagnosis for stiﬂe lameness in cats. They typically cause poly-
arthritis with an acute or insidious onset of lameness and shifting leg lameness.
Feline chronic progressive polyarthritis and caliciviral arthritis are two types
of immune-mediated joint disease seen in cats. Feline chronic progressive
polyarthritis usually occurs in male cats between the ages of 1 and 5 years
of age. It is an erosive arthritis that most commonly aﬀects the carpi and
tarsi, although other joints may be involved. The antigen responsible for the
immune response is unconﬁrmed, but the disease is linked to serologic evi-
dence of exposure to feline syncytium-forming virus and feline leukemia
virus [1–3]. Caliciviral arthritis is a transient self-limiting polyarthritis asso-
ciated with calicivirus, and it is acquired by natural infection or vaccination.
It is typically seen in kittens and usually resolves within 1 week without
treatment [2,4]. Mycoplasma species and Lyme disease are also possible
causes of polyarthritis in cats [2].

A thorough physical examination, complete understanding of the

patient’s history, and knowledge of common feline diseases are needed to

Vet Clin Small Anim 32 (2002) 963–982

E-mail address: McLaughlin@cvm.msstate.edu (R.M. McLaughlin).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 2 1 - 9

conﬁrm a diagnosis and recommend the appropriate treatment for cats with
stiﬂe abnormalities. In cases of traumatic injury, the patient must be care-
fully evaluated for diaphragmatic hernia, traumatic myocarditis, pneumo-
thorax, hemothorax, pulmonary contusions, abdominal bleeding, bladder
rupture, neurologic deﬁcits, and concurrent orthopedic injuries before sur-
gery is considered. This article describes common causes of feline stiﬂe lame-
ness that are amenable to surgical treatment. The clinical signs, radiographic
ﬁndings, treatment options, and prognosis are reviewed.

Cranial cruciate ligament rupture

The cranial cruciate ligament (CrCL) prevents stiﬂe hyperextension,

excessive internal rotation of the tibia, and cranial translation of the tibia
relative to the femoral condyles (drawer motion). Injury and rupture of the
CrCL causes joint pain, eﬀusion, instability, and, eventually, osteoarthritis.
The incidence of CrCL rupture is much lower in cats than in dogs. The rea-
son for the lower incidence of CrCL rupture in cats is unclear but may be
partially a result of the fact that the CrCl is larger than the caudal cruciate
ligament in cats, whereas it is smaller than the caudal cruciate ligament in
dogs and human beings [5]. Interestingly, partial rupture of the CrCL is
reported less frequently in cats than in dogs. It is unclear if this is because
cats show minimal clinical signs with partial CrCL tears, recover more
quickly when partial tears occur, or simply do not get partial tears of the
CrCL as often as dogs.

The cause of CrCL rupture is not always apparent in cats, although trauma

is considered the most common etiology. In many cases, however, the
traumatic event goes unnoticed by the owner. The application of a supra-
physiologic load to the ligament, a sudden hyperextension of the stiﬂe, or
a severe internal rotation of the tibia when the stiﬂe is ﬂexed can all lead
to traumatic rupture of the CrCL. Conformational, hormonal, inﬂamma-
tory, and hereditary factors are thought to play a role in CrCL rupture in
dogs but have not been evaluated in cats. It has been noted, however, that
many cats presented for CrCL instability are overweight [6].

Clinical signs and diagnosis

Clinical signs of acute CrCL rupture include a sudden onset of non-

weight-bearing lameness, stiﬂe pain, and palpable joint eﬀusion. The lame-
ness may gradually improve with rest but often redevelops with activity. The
lameness may also be progressive and persistent and is often worse after
exercise. Most cats have a positive cranial drawer sign, positive tibial com-
pression test, joint pain, and stiﬂe eﬀusion when the rupture is acute. In cats
with chronic rupture, crepitus may be palpable with manipulation of the
joint. In some cases, a ‘‘click’’ or ‘‘popping’’ sensation may be noted during
palpation of the stiﬂe and can be indicative of a meniscal injury. Unlike the

964

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

case in dogs, joint laxity is generally palpable even in chronic cases of CrCL
rupture in cats, and the medial buttress (the periarticular ﬁbrous tissue over
the region of the medial collateral ligament) is not usually as prominent.

The diagnosis of a CrCL injury is based on history and clinical signs.

CrCL injury should be strongly suspected in cats with stiﬂe pain, joint eﬀu-
sion, and lameness. A positive cranial drawer sign or a positive tibial com-
pression test is used to conﬁrm the presence of joint instability. Radiography
of the stiﬂe may reveal joint eﬀusion in acute cases and signs of osteoarthri-
tis in more chronic cases. Meniscal calciﬁcation or ossiﬁcation is also re-
ported in cats with CrCL rupture and can be observed radiographically as
a mineral density in the area of the medial or lateral meniscus [7,8].

Treatment

Conservative medical therapy is an eﬀective treatment for CrCL rupture

in cats and is usually the method of choice for most feline patients. Normal
limb function usually returns within 5 weeks after treatment with restricted
activity and weight loss alone. In a study of 18 cats treated conservatively
for 1 month after rupture of the CrCL, all the cats had a normal gait and
a normal pain-free range of motion. Muscle atrophy was minimal [9]. Al-
though these cats demonstrated normal clinical joint function, many had
periarticular thickening over the medial aspect of the stiﬂe joint, radio-
graphic evidence of osteoarthritis, and persistent joint laxity. Longer-term
studies are needed to conﬁrm that progressive osteoarthritis does not aﬀect
future clinical function in cats treated conservatively for CrCL rupture.
Currently, the use of nonsteroidal anti-inﬂammatory medications is not
routinely recommended because of the success of rest and exercise restric-
tion alone and because of potential complications associated with their use
in cats.

Surgical exploration and stabilization of the stiﬂe joint are recommended

only in cats with persistent lameness unresponsive to conservative therapy.
Obese cats may also beneﬁt from early surgery. The presence of meniscal
calciﬁcation is not necessarily an indication for surgical intervention,
because some cats with this condition were without clinical signs [9]. If pain
or lameness persists after medical management, however, removal of the
mineralized meniscal tissue may provide relief of clinical signs [7]. It is
important that cats be evaluated for cardiomyopathy before anesthesia is
considered to allow surgical treatment of a CrCL rupture [10].

Both intracapsular and extracapsular stabilization techniques have been

advocated for treatment of CrCL rupture in cats [11]; however most sur-
geons prefer extracapsular techniques. Extracapsular techniques imbricate
the lateral joint tissues to prevent cranial drawer motion and minimize inter-
nal rotation of the tibia. Most extracapsular techniques use a combination
of capsular imbrication (sutures placed in the lateral ﬁbrous joint capsule)
and lateral retinacular imbrication (sutures placed from the fabella to the

965

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

tibial crest). Many variations in technique exist using diﬀerent numbers of
sutures and various suture placements.

The technique recommended for extracapsular stabilization of the feline

stiﬂe is similar to that used in dogs. The limb is shaved and prepared from
the hip to the metatarsus. A hanging leg preparation is used to allow manip-
ulation of the stiﬂe joint during surgery. A lateral parapatellar arthrotomy is
performed, and the stiﬂe is explored. A small Hohmann retractor or mos-
quito hemostat can be placed between the condyles (intercondylar notch)
and over the caudal edge of the tibial plateau to drawer the tibia cranially
and improve visualization of the joint. Torn remnants of the CrCL are
excised along with damaged portions of the menisci. The joint is lavaged,
and the joint capsule is closed with small absorbable suture material in a
continuous pattern. An extracapsular suture is then placed from the lateral
fabella to tibial crest (Fig. 1). A hole is drilled across the tibial crest to serve
as an anchor point for the suture. Alternatively, the suture can be placed
through the distal portion of the straight patellar tendon. Nonabsorbable
monoﬁlament suture is used in most cases. The appropriate size is based

Fig. 1. Model showing the placement of an extracapsular suture for stabilization of the stiﬂe
joint after rupture of the cranial cruciate ligament. The suture is placed through a hole drilled in
the tibial crest and around the lateral fabella (arrow).

966

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

on the size of the patient and surgeon’s preference, although 2-0 and 0
sutures are used commonly. Once the extracapsular suture is tightened, the
lateral fascia is imbricated over the extracapsular suture using mattress
sutures (3-0 monoﬁlament absorbable suture). Subcutaneous tissues and
skin are closed routinely.

After surgery, the cat’s activity is restricted for 6 weeks. Running and

jumping should be avoided. External coaptation bandages are not necessary
during the postoperative recovery period.

Collateral ligament rupture

The collateral ligaments provide the primary medial and lateral stability to

the stiﬂe joint. Collateral ligament injuries in the cat are usually a result of vio-
lent trauma, such as that occurring when the animal is hit by an automobile or
falls from a signiﬁcant height. Injury to the medial collateral ligament occurs
more frequently than injury to the lateral collateral ligament [12]. Meniscal
injury is commonly associated with rupture of the collateral ligaments.

Clinical signs and diagnosis

Patients with rupture of a collateral ligament often present with a history

of trauma. Most patients are nonweight-bearing lame immediately after the
injury. Palpation of the limb reveals pain in the stiﬂe joint, swelling, and joint
instability. The aﬀected limb and the entire patient must be carefully exam-
ined for other ligamentous, orthopedic, neurologic, or soft tissue injuries.

The diagnosis of collateral ligament injury is based on physical examina-

tion ﬁndings and radiography. The joint should be examined with the cat
under heavy sedation or anesthesia to assess joint stability. Stabilization
of respiratory and cardiovascular signs associated with trauma should be
undertaken ﬁrst if necessary. The stability of the stiﬂe is assessed by the
application of varus and valgus stress to the joint. It is important to apply
pressure with the stiﬂe in extension and in ﬂexion. In a normal joint, both
collateral ligaments are taut in extension, but the lateral collateral ligament
is slightly lax when the stiﬂe is ﬂexed. The medial collateral ligament remains
taut during stiﬂe ﬂexion. Disruption of the lateral collateral ligament results
in varus instability, whereas disruption of the medial collateral ligament
results in valgus instability.

Stress radiographs are helpful in some cases to conﬁrm collateral liga-

ment disruption. When varus or valgus stress is applied to the joint (with the
cat under anesthesia), the joint space appears widened radiographically if
the collateral ligament is ruptured.

Treatment

First-degree ligamentous sprains are mild and consist of damage to only a

few collagen ﬁbers [13]. Although hemorrhage and edema occur within the

967

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

parenchyma of the ligament, the ligament’s function remains intact, and the
joint remains stable. First-degree sprains are generally treated conservatively
with rest, external coaptation, and, in rare cases, the administration of non-
steroidal anti-inﬂammatory drugs [13].

Second- and third-degree sprains are more extensive, however. With a

second-degree sprain, the ligament is grossly intact, but its function is dis-
rupted. With third-degree sprains, the parenchyma of the ligament is disrup-
ted or avulsed from its bony attachment. Both second- and third-degree
sprains result in joint instability [13]. Surgical repair is indicated, because
conservative therapy often results in persistent joint instability, pain, lame-
ness, and osteoarthritis [6]. No studies describing the results of conservative
therapy for treatment of second- and third-degree collateral ligaments of the
stiﬂe in cats have been reported.

Surgical repair of a stiﬂe collateral ligament rupture begins with explora-

tion of the joint to conﬁrm the ligament damage and to identify other poten-
tial problems, including cruciate ligament rupture and meniscal injuries.
Once the collateral ligament is examined, three general options exist for
repair: primary repair of the ligament, reattachment of the ligament to the
bone if an avulsion injury is present, or replacement of the ligament with
a prosthetic ligament using suture material.

Primary repair of a collateral ligament injury is usually feasible only in

cases of ligament laceration. With traumatic disruption of the ligament, the
parenchymal damage present in most second- and third-degree sprains pre-
cludes primary repair. Primary repair consists of placing suture to approx-
imate the traumatized ends of the ligament. Several suture patterns have
been designed for use in ligaments, including the locking loop, Bunnel-
Mayer pattern, and three-loop pulley. Nonabsorbable monoﬁlament suture
is used (2-0 or 3-0).

If the ligament has avulsed from its bony origin or insertion, reattach-

ment of the bone fragment may restore function. The bone fragment is
attached with a lag screw if it is large enough or with several divergent
Kirschner wires. Unfortunately, in many cases, the avulsed fragment is too
small to allow placement of the implants, or the ligament itself is too dam-
aged to be functional.

In most cases of stiﬂe collateral ligament disruption, replacement with a

prosthetic ligament is necessary. Even in cases where the damaged ligament
is reattached or sutured, a prosthetic ligament is usually placed to protect
the ligament during healing. This involves placing monoﬁlament suture
material (2-0 or 0) in a ﬁgure-of-eight pattern to restore collateral ligament
function. The suture is attached proximally and distally at the normal ana-
tomic locations of the ligament’s origin and insertion. It is attached to the
bone using one of several methods. Bone tunnels can be drilled to allow pas-
sage of the suture. A screw and washer may be inserted into the proper sites
on the femoral condyle and the proximal tibia (Fig. 2). The suture is then
wrapped around the screw heads (beneath the washers). Alternatively,

968

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

specially designed bone anchors (BoneBiter Suture Anchor System; Androcles,
Inc., Warsaw, IN) can be inserted to attach the suture to the bone (Fig. 3).

After surgery, the cat’s activity is restricted for 4 to 6 weeks to allow heal-

ing. External coaptation may be applied, although bandages are not well
tolerated by some cats.

Stiﬂe dislocation

Stiﬂe dislocation (stiﬂe derangement, stiﬂe luxation) occurs infrequently

in cats and is the result of signiﬁcant trauma. Stiﬂe dislocation occurs when

Fig. 3. Prosthetic ligament replacement using bone anchors for repair of a medial collateral
ligament rupture. (A) Model of the stiﬂe joint showing placement of bone anchors and suture
material. (B) Craniocaudal radiographic view of the stiﬂe joint after surgery. (C) Lateral
radiographic view of the stiﬂe joint after surgery.

Fig. 2. Craniocaudal (A) and lateral (B) postoperative radiographic views of the stiﬂe joint
showing prosthetic ligament replacement of a ruptured lateral collateral ligament using screws
and washers.

969

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

multiple ligamentous injuries lead to gross instability of the joint. The pri-
mary stabilizing structures of the stiﬂe are the cruciate ligaments and the col-
lateral ligaments. Secondary stabilizers of the joint are the joint capsule,
menisci, and muscles and tendons that traverse the joint [12,14].

Clinical signs and diagnosis

Cats with stiﬂe dislocations are typically presented after a signiﬁcant

trauma (Fig. 4). Although numerous combinations of ligament injury have
been described, the most common combination is rupture of both cruciate
ligaments and one collateral ligament. The medial collateral ligament is
most common [12]. Medial injury may be more common when a cat is hit
by a car on the lateral side, causing tension on the medial joint restraints
[15]. Meniscal damage also occurs, and the joint capsule is torn or stretched.
Fortunately, the vascular and neurologic damage seen in human beings with
stiﬂe dislocations is rare in cats [12].

The diagnosis of stiﬂe dislocation is based on examination of the stiﬂe joint

and radiography. Typically, the stiﬂe region is swollen, and pain is elicited
with palpation or manipulation of the limb. Open wounds may be present
depending on the type of trauma that occurred, but the luxation is closed in
most cases. Although it is usually obvious that signiﬁcant injury to the stiﬂe
is present, it can be diﬃcult to accurately determine which ligaments (and
combinations of ligaments) are ruptured on the basis of physical examination
and radiography alone. Surgical exploration and careful examination of each
ligament are needed to conﬁrm the extent of the ligamentous injury [12].

Treatment

The limb is initially placed in a Robert Jones bandage for 48 hours to

reduce swelling and limit motion while the patient is stabilized. The joint

Fig. 4. Lateral (A) and craniocaudal (B) radiographic views of a dislocated stiﬂe in a cat.

970

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

is then surgically explored via a lateral parapatellar approach to assess
the cruciate ligaments and menisci. Ruptured cruciate ligaments and
damaged portions of the menisci are excised. The medial and lateral collat-
eral ligaments are then assessed through separate surgical approaches.
Each ligament is carefully examined and palpated while applying stress to
the limb. When the integrity of each ligament is determined, the joint is
debrided, lavaged, and then stabilized using one of several described surgical
methods.

Extra-articular stabilization is one method commonly used for repair of

stiﬂe dislocations [12,15]. The arthrotomy incision and any other tears in the
joint capsule are ﬁrst sutured with absorbable monoﬁlament suture material
in a simple continuous pattern. Large (0 or 1), monoﬁlament, nonabsorbable
sutures are then placed outside the joint capsule to replace the function of the
various ruptured or damaged ligaments. For rupture of the CrCL, one suture
is placed from the lateral fabella to the tibial crest, another from the medial
fabella to the tibial crest, and a third from the lateral fabella to the patellar
ligament distal to the patella. For rupture of the caudal cruciate ligament,
one suture is placed from the ﬁbular head to the patellar ligament just distal
to the patella, and another is placed from the caudomedial tibial plateau to
the patellar ligament just distal to the patella. The suture can be passed
through bone tunnels drilled in the ﬁbular head and caudomedial tibial pla-
teau. For rupture of the medial and lateral collateral ligaments, suture mate-
rial is placed in a ﬁgure-of-eight pattern from the distal femur to the tibial
plateau using the normal origin and insertion points of the ligament as land-
marks. The suture is anchored to the bone using bone tunnels, screws, and
washers or bone anchors (see section on collateral ligament rupture).

The sutures are preplaced, and the joint is reduced and held in a function-

al standing angle. This is approximately 150

° of extension in most patients,

although the normal standing angle of the contralateral stiﬂe can be mea-
sured before surgery to determine the ideal angle in each patient. The sutures
are then tightened to provide stability. The collateral ligament sutures are
tightened ﬁrst, followed by the sutures augmenting the CrCL and those aug-
menting the caudal cruciate ligament [6]. The joint is manipulated to assess
stability before closing the incision.

After surgery, the limb can be bandaged for 10 days, followed by 10 weeks

of restricted activity [6]. Unfortunately, many cats do not tolerate long-term
bandaging. Alternatively, a transarticular external ﬁxator can be placed to
provide stability [12,14]. The ﬁxator remains in place for 4 weeks [12]. The
limb is placed in a soft padded bandage for 4 weeks after removal of the ﬁx-
ator. Aron [12] described the use of transarticular ﬁxators in 12 dogs and one
cat after stabilization of a stiﬂe dislocation and reported few complications.
All 13 animals maintained nearly normal extension of the joint but lost 10

° to

° of ﬂexion. Despite the reduced range of motion and development of mod-

erate osteoarthritis, good clinical function resulted in all 13 patients. Never-
theless, in a study of four cats in which an external ﬁxator was applied for

971

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

6 weeks after surgical stabilization of the stiﬂe, complications occurred, such
as pin loosening, disruption of the ﬁxator, and fractures of the tibia and
femur [14]. These complications were thought to result from poor compliance
with the postoperative instructions to strictly conﬁne the cats for 10 to 16
weeks. All four cats in this study reportedly had good function once the ﬁx-
ator was removed and complications created by the external ﬁxator were
addressed. These ﬁndings suggest that an external ﬁxator can provide rigid
ﬁxation after surgical stabilization of the stiﬂe in cats and allow early weight
bearing and acceptable long-term function; however, conﬁnement during the
recovery period is necessary to reduce complications.

Another method of stabilizing stiﬂe dislocations is the placement of a

transarticular pin (Fig. 5). The joint is explored and debrided as previously
described. The stiﬂe joint is reduced and held in a functional weight-bearing
position (30

°–40° of ﬂexion) while a 3.0- or 3.5-mm Steinmann pin is in-

serted across the joint. The pin can be inserted in several ways. In one method,
the pin is driven from the distal aspect of the tibial crest in a proximal direc-
tion until it enters the stiﬂe joint through the nonarticular intercondylar area
cranial to the intercondyloid eminence. The joint is reduced, and the pin is
advanced across the joint and into the intercondylar fossa of the distal
femur. The pin is then advanced until it exits the cranial femoral cortex
proximal to the patella, and it is then cut ﬂush with the tibia [6]. In another
insertion method, the pin positioning is the same, but the pin is ﬁrst retro-

Fig. 5. Lateral radiographic view of a dislocated stiﬂe in a cat repaired with a transarticular pin.

972

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

graded from the nonarticular intercondylar eminence of the tibial plateau to
exit at the distal aspect of the tibial plateau. The direction of pin insertion is
then reversed, and it is normograded into the distal femur while holding the
stiﬂe in 30

° of extension [16,17]. A third method of transarticular pinning

involves placing two pins. The ﬁrst is positioned as just described. The sec-
ond is placed from the craniomedial aspect of the tibial crest, through the
intercondylar area, and into the lateral femoral condyle [17]. All three meth-
ods provide adequate stability.

After pin placement, the joint capsule is imbricated with absorbable

monoﬁlament sutures, and the subcutaneous tissues and skin are closed rou-
tinely. The limb is splinted to provide additional support during the healing
period, and strict conﬁnement is encouraged for 1 month. The transarticular
pin is removed in 3 to 4 weeks, and the cat is keep conﬁned for an additional 8
to 10 weeks [17]. One retrospective study reported excellent results in ﬁve of
six cats in which transarticular pins and external coaptation were used to sta-
bilize stiﬂe dislocations [17]. This study did not recommend the use of trans-
articular pinning without postoperative coaptation, because complications
were likely. Complications reported with the use of transarticular pinning
include pin bending, migration and loosening of the pins, and iatrogenic car-
tilage damage during pin insertion. Overall, good results have been achieved
with the use of transarticular pinning to stabilize stiﬂe dislocations in cats
[16,17]. Limb function is good, although reduced range of motion and osteo-
arthritis occur in most cases.

Patellar luxation

Patellar luxations in cats are generally medial and are thought to result

from trauma (although this trauma often goes unnoticed by the owner)
[6]. Unilateral or bilateral luxation may occur, and both male and female
cats can be aﬀected [18]. Most reported cases of patellar luxation occur in
young cats. In one report, 66% of cats with patellar luxation were less than
1 year of age [19]. In another report on eight cats, all were less than 3 years
of age [18]. Patellar luxation may also occur after surgical repair of other
stiﬂe injuries or after femoral fracture repair. Lateral luxation of the patella
is less common in cats and usually results from trauma. It is unilateral and
may occur concurrently with other orthopedic injuries.

Congenital medial patellar luxations are described in Devon Rex and

Abyssinian breeds and are usually bilateral [20,21]. Cats with congenital
patellar luxation may also have a shallower trochlear groove and slight medi-
al deviation of the tibial tuberosity in some cases. The coxa vara and lateral
bowing of the distal femur that are commonly seen in dogs with patellar lux-
ations are not typically seen in cats, however [18,19]. A weak association
between medial patellar luxation and hip dysplasia has also been reported
in cats [22].

973

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

Clinical signs and diagnosis

In many cases, congenital medial patellar luxation causes no clinical signs

and is detected only as an incidental ﬁnding during physical examination.
Some patellar laxity is present even in normal cats, and mild subluxation
of the patella can often be elicited during palpation [22]. It is possible for
some minor trauma to produce clinical signs in a cat with previously sub-
clinical patellar laxity.

In some cats, however, displacement of the patella from the trochlear

groove causes pain, acute lameness, locking of the limb in extension, or a
crouching gait. In a review of 21 cats with patellar luxation, 66% were mark-
edly lame and others had noticeable gait abnormalities [19]. The lameness
associated with patellar luxation is intermittent in some cases. In dogs,
patellar luxation may predispose the joint to rupture of the CrCL, although
this is rarely described in cats.

The diagnosis of patellar luxation is made by palpation of the joint and

evaluation of patellar stability. The grading system used to describe the
severity of the patellar luxation is the same as that used for dogs. In grade
I, the patella can be manually luxated using digital pressure but reduces
spontaneously when pressure is released. In grade II, the patella can be lux-
ated manually by digital pressure or with rotation of the tibia. The patella
can then be reduced manually and remains in place until the limb is ma-
nipulated. In grade III, the patella is luxated at the time of the physical
examination. It can be manipulated back into the trochlear groove but
immediately reluxates. The degree of lameness is variable with grade III lux-
ations but is usually persistent. In grade IV, the patella is permanently lux-
ated and cannot be reduced. Cats with grade I luxation rarely show clinical
signs, and only some of the cats with grade II or III patella luxation have
clinical signs [22]. Grade IV luxations are uncommon in cats.

The displaced patella may also be observed on craniocaudal radiographic

views of the stiﬂe joint. Although radiographs are not necessary to conﬁrm
the diagnosis of patellar luxation, they may help to identify other joint ab-
normalities.

Treatment

Restricted activity alone is often recommended to treat cats with patellar

luxation, particularly if there are few clinical signs. Surgery is not recom-
mended in cats without clinical signs. Surgical treatment is generally recom-
mended for cats with more severe luxation and for those with persistent
clinical signs, however. A combination of soft tissue procedures and bone
reconstructive procedures is used to restore normal anatomy and improve
the function of the patella mechanism. Several surgical techniques, including
imbrication, trochlear recession, and tibial crest transposition, may be used
together depending on the severity of the luxation.

974

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

Imbrication

Soft tissue imbrication techniques are used to enhance stability of the

patella. In many cases, imbrication of the retinaculum alone is adequate
to maintain reduction. The lateral retinaculum is imbricated to stabilize
medial patellar luxations, and the medial retinaculum is imbricated to stabi-
lize lateral patellar luxations. Nonabsorbable monoﬁlament suture (2-0) is
placed in a Lembert or mattress pattern to achieve imbrication. Alterna-
tively, a narrow strip of the retinacular tissue can be excised, and the resul-
tant edges are sutured. Another form of imbrication involves placing a
mattress suture from the fabella to the patella to prevent luxation.

Trochlear recession

Trochlear recession techniques deepen the trochlear groove to improve

patellar stability. The procedure must be performed carefully to avoid dam-
aging the articular cartilage, particularly in smaller cats. Either a triangular
wedge recession technique or a rectangular recession technique may be used.
In the triangular recession technique, a small hand saw is used to excise a
wedge-shaped osteochondral fragment, preserving the articular surface. A
small amount of bone is removed from the resultant defect, and the osteo-
chondral fragment is positioned back into the defect. The rectangular reces-
sion technique also preserves the articular cartilage. A small osteotome is
used to remove a rectangular osteochondral fragment from the trochlear
groove. A small section of bone from the resultant defect is removed with
a rongeur before replacing the fragment. Ideally, the trochlear is recessed
such that half of the thickness of the patella is seated within the trochlear
groove. Trochleoplasty techniques that do not preserve the articular carti-
lage are not recommended.

Tibial crest transposition

Tibial crest transposition is rarely needed to stabilize the patella in cats.

In cases of severe or recurrent luxation, however, transposing the insertion
of the straight patellar tendon can realign the quadriceps mechanism (and
thus the patella) over the trochlear groove and improve patellar stability.
Because the patella is encased in the quadriceps at the origin of the straight
patellar tendon, realigning the quadriceps mechanism allows the patella to
remain in the trochlear groove during joint movement. To correct a medial
patellar luxation, the tibial crest is moved to a more lateral position and
reattached using two small Kirschner wires. It is transposed medially in
cases of lateral patellar luxation.

The prognosis after surgical treatment of patellar luxation in cats is good.

The limb may be bandaged for 1 week after surgery, although many cats
resent bandages, and immobilization is not necessary for a satisfactory out-
come. Exercise is restricted for 4 weeks to allow healing. One report found
that 12 of 13 cats undergoing surgical treatment of patellar luxation had

975

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

excellent or fair-to-good function [19]. Other authors report similar positive
results with surgical correction [18].

Fractures of the distal femur

Distal femoral fractures in cats are a result of trauma and often in-

volve the stiﬂe joint. Supracondylar fractures of the distal metaphysis
may involve the femoropatellar joint. Fractures of the distal femoral growth
plate (Salter-Harris type I and II fractures) occur frequently in immature
cats (Fig. 6). Fractures involving the growth plate region are also relatively
common in young adult cats and may occur as a result of late closure of
the growth plate after early neutering [23]. Intra-articular fractures of
the femoral condyles or trochlea also occur in cats [24]. Each of these
distal femoral fractures requires internal ﬁxation to preserve stiﬂe joint
function.

Clinical signs and diagnosis

Cats with fractures of the distal femoral physis are typically presented

with a history of trauma and clinical signs of swelling, crepitus, and pain
in the distal femur. Most are initially not weight bearing, although some cats
begin to place weight on the limb if the injury is several days old. The diag-
nosis of distal femoral physeal fracture is based on palpation and radio-
graphy of the femur. The epiphyseal segment is usually displaced caudally
and proximally. Lateral, craniocaudal, and occasionally oblique views are
helpful to conﬁrm the presence of an articular fracture.

Fig. 6. Lateral (A) and craniocaudal (B) radiographic views of a Salter-Harris type II distal
femoral fracture in a cat.

976

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

Treatment

External coaptation is ineﬀective in stabilizing most distal femoral frac-

tures, because the epiphysis is displaced by its muscular attachments. Immo-
bilization of the fracture site is extremely diﬃcult with external coaptation,
and cats may not tolerate prolonged immobilization. Internal ﬁxation allows
better anatomic reduction and implant placement to provide stability to the
fracture site. Numerous methods have been described for stabilization of the
various types of distal femoral fractures, including a single intramedullary
pin, Rush pinning, cross-pins, smooth pins placed in Rush fashion, parallel
smooth pins, lag screw ﬁxation, external ﬁxators, and bone plating [25].

Repair of Salter-Harris type I and II fractures is usually achieved using

one of several described pinning methods. Smooth pins placed across the
growth plate are thought to allow continued growth and avoid complications
associated with altered growth or premature closure. Studies have found that
some growth disturbance is common after repair of Salter-Harris type I and
II fractures of the distal femur, however. Parker and Bloomberg [26] found
that the physis closed within 4 weeks of intramedullary pinning in all 17 dogs
and cats that they evaluated. Another study found that 82% of dogs eval-
uated after repair of Salter-Harris type I and II fractures had some growth
disturbance [27]. Fortunately, obvious lameness is uncommon, and cats
adapt well to any minor limb length discrepancy that develops after fracture
ﬁxation [28].

A common pinning technique used to stabilize fractures of the distal fem-

oral growth plate is placement of a single intramedullary pin [28]. A lateral
parapatellar approach is made to the stiﬂe joint and distal femur. The patella
is reﬂected medially to allow visualization of the fracture site and the inter-
condylar notch. A Steinmann pin is inserted into the distal femoral epiphysis
starting just cranial to the origin of the caudal cruciate ligament. The pin
should be 50% to 70% of the diameter of the femoral medullary cavity. The
fracture is reduced, and the pin is advanced across the fracture line and into
the medullary cavity of the metaphysis. It should be advanced proximally
and seated into the cancellous bone at the proximal femur. The pin is then
cut and countersunk below the articular surface at the intercondylar notch.
The joint is lavaged and closed routinely. The cat’s activity is restricted for
3 weeks. Complications with this technique are rare but can include instabil-
ity, infection, pin migration into the joint, and iatrogenic fracture of the fem-
oral epiphysis. A study of 52 cats with supracondylar fractures treated with
intramedullary pinning reported that 82% were free of lameness [28].

A modiﬁcation of the intramedullary pinning technique was reported in

nine cats [29]. In this report, a 2.4-mm Rush pin was placed in the medullary
cavity instead of a Steinmann pin. The hook on the end of the Rush pin was
positioned in the intercondylar notch and allowed easy removal of the
implant if needed. Because the Rush pins are blunted on the ends, holes must
be predrilled in the epiphysis and metaphysis to allow insertion.

977

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

Another pinning technique commonly used to stabilize Salter-Harris type

I and II fractures of the distal femur is the placement of two small pins
(Kirschner wires) across the fracture site (Figs. 7 and 8) [25,30]. A lateral
parapatellar approach is used to access the fracture site and stiﬂe joint. The
fracture is reduced and held in place with pointed reduction forceps. A lat-
eral Kirschner wire (0.045 or 0.062 in) is inserted through the nonweight-
bearing cartilage lateral to the trochlear ridge and 3 mm cranial to the origin

Fig. 7. Lateral (A) and craniocaudal (B) radiographic views 1 month after repair of a Salter-
Harris type II fracture with two cross-pins. Bony union is complete.

Fig. 8. Lateral (A) and craniocaudal (B) radiographic views of a supracondylar fracture in a cat.

978

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

of the long digital extensor tendon. A second Kirschner wire is inserted at a
corresponding point just medial to the medial trochlear ridge. The pins may
be placed either parallel to each other or in a crossed fashion and are
inserted such that they exit the cranial cortex of the femur approximately
2 cm proximal to the fracture line [25,30]. In some cases, the pins can be
positioned in Rush fashion. With this technique, the pins bounce oﬀ the
inner cortex of the femoral metaphysis and continue up the medullary cavity
during insertion. Placing the two Kirschner wires in this manner is also an
acceptable method of stabilizing growth plate fractures of the distal femur.
Once the pins are positioned, they are then cut and countersunk below the
cartilage surface using a nail set. The joint is lavaged and closed routinely.
The cat is conﬁned, and exercise is restricted for 3 weeks. The prognosis
after internal ﬁxation of distal femoral growth plate fractures in cats
using two Kirschner wires is excellent, although complications have been
reported, such as malalignment, implant failure, nonunion, and patellar
luxation [31]. Postoperative epiphyseal alignment has been shown to be an
important factor inﬂuencing the outcome after surgical repair [31].

Supracondylar fractures in adult cats are typically transverse or short obli-

que (Fig. 9). Repair is often achieved using the pinning techniques described
previously for Salter-Harris type fractures. In some cases, however, the frac-
ture is slightly more proximal or comminution is present, and ﬁxation using
plates and screws or an external ﬁxator is preferred. External ﬁxators may be
applied in a closed fashion if acceptable reduction can be achieved (Fig. 10).
If open reduction is indicated, a lateral approach to the femur is combined
with a parapatellar approach to the stiﬂe. The fracture is reduced, and trans-
ﬁxation pins are placed through the femur in a lateral-to-medial direction.
Ideally, three to four pins should be placed proximally and distally to the

Fig. 9. Lateral (A) and craniocaudal (B) radiographic views of a supracondylar fracture
stabilized with a type 1 acrylic external ﬁxator. The ﬁxator was applied in a closed fashion.

979

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

fracture site. The transﬁxation pins are then attached using acrylic or poly-
methylmethacrylate. Acrylic connecting bars are easily contoured to accom-
modate the curved distal femur, are lightweight, and eliminate the need to
ensure that all transﬁxation pins are inserted in the same plane. They also
allow the use of diﬀerently sized transﬁxation pins in a single frame. A type
I external ﬁxator is suﬃciently strong to stabilize distal femoral fractures in
cats, particularly if the fracture is well reduced. An intramedullary pin may
be inserted to augment the stability, although this is rarely necessary in cats.

Supracondylar fractures may also be stabilized with bone plates and

screws if the epiphyseal segment is large enough [24,32]. Reconstruction
plates (2.7 and 3.5 mm) have been used for repair of distal femoral fractures
in cats and provide suﬃcient stability [32]. Because the notches are located
between the screw holes, reconstruction plates are more easily contoured to
ﬁt the curved distal femur than standard bone plates. Proper contouring per-
mits the insertion of three screws in the small epiphyseal fragment of most
fractures. Miniplates can also be used to repair distal femoral fractures in
cats [24]. Various shapes are available (straight, T-plates, L-plates) to ensure
adequate ﬁxation of the distal fracture segment. Miniplates also allow the
insertion of small 2.0 mm screws into the epiphyseal fragment.

Intra-articular fractures involving the femoral trochlea or condyles

require internal ﬁxation to realign the articular surface and preserve joint

Fig. 10. Photograph of a cat with a type I external ﬁxator applied to the lateral aspect of the left
distal femur.

980

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

function. Screws placed in lag fashion (2.0 or 2.7 mm) are used to stabilize
condylar fractures and enhance healing by compressing bone fragments and
eliminating gaps or steps in the articular surface. Kirschner wires may be
used to anchor loose osteochondral fragments to the underlying bone.
Restricted activity, controlled exercise, and physical therapy are important
after stabilization of articular fractures to preserve joint function and min-
imize osteoarthritis.

Summary

Although stiﬂe disease is seen less frequently in cats than in dogs, accu-

rate diagnosis and proper treatment are important to restoring joint func-
tion. The prognosis for return of function after all but the most severe
stiﬂe injuries in cats is good if proper surgical technique is used and
adequate postoperative restrictions are enforced.

References

[1] Carro T. Polyarthritis in cats. Compend Contin Educ Pract Vet 1994;16:57–67.
[2] Hardie EM. Management of osteoarthritis in cats. Vet Clin North Am Small Anim Pract

1997;27:945–53.

[3] Pederson NC. Joint diseases of dogs and cats. In: Ettinger SJ, editor. Textbook of Veter-

inary Internal Medicine. 3rd edition, Philadelphia: WB Saunders; 1989. p. 2329–77.

[4] Bennett D, Gaskell RM, Mills A, Knowles JO, Carter SD, et al. Detection of feline calici-

virus antigens in the joints of infected cats. Vet Rec 1989;12:12–16.

[5] Alexander J, Shumway J, Lau R. Anterior cruciate ligament rupture. Feline Pract 1977;

6:38–9.

[6] Umphlet RC. Feline stiﬂe disease. Vet Clin North Am Small Anim Pract 1993;23:897–913.
[7] Reinke J, Mughannam A. Meniscal calciﬁcation and ossiﬁcation in six cats and two dogs.

J Am Anim Hosp Assoc 1994;30:145–52.

[8] Whiting PG, Pool RR. Intermeniscal calciﬁcation and ossiﬁcation in the stiﬂe joints of

three domestic cats. J Am Anim Hosp Assoc 1985;21:579–84.

[9] Scavelli TD, Schrader SC. Non-surgical management of rupture of the cranial cruciate

ligament in 18 cats. J Am Anim Hosp Assoc 1987;23:337–40.

[10] Janssens LAA, Janssens GO, Janssens DL. Anterior cruciate ligament rupture associated

with cardiomyopathy in three cats. Vet Comp Orthop Traumatol 1991;4:35–7.

[11] Denny HR. The hindlimb. In: Denny HR, editor. A guide to canine and feline orthopaedic

surgery. London: Blackwell Scientiﬁc Publications; 1993;3:327–56.

[12] Aron DN. Traumatic dislocation of the stiﬂe joint: treatment in 12 dogs and one cat. J Am

Anim Hosp Assoc 1988;24:333–40.

[13] Brinker WO, Piermattei DL, Flo GL. Ligamentous injuries. In: Handbook of small animal

orthopedics and fracture management. 2nd edition. Philadelphia: WB Saunders; 1990.
p. 314–23.

[14] Bruce WJ. Stiﬂe joint luxation in the cat: treatment using transarticular external skeletal

ﬁxation. J Small Anim Pract 1999;40:482–8.

[15] Hulse DA, Shires P. Multiple ligament injury of the stiﬂe joint in the dog. J Am Anim Hosp

Assoc 1986;22:105–10.

[16] Connery NA, Rackard S. The surgical treatment of traumatic stiﬂe disruption in a cat. Vet

Comp Orthop Traumatol 2000;13:208–11.

981

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

[17] Welches CD, Scavelli TD. Transarticular pinning to repair luxation of the stiﬂe joint in

dogs and cats: a retrospective study of 10 cases. J Am Anim Hosp Assoc 1990;26:207–14.

[18] Houlton JEF, Meynink SE. Medial patellar luxation in the cat. J Small Anim Pract

1989;30:349–52.

[19] Johnson ME. Feline patellar luxation: a retrospective case study. J Am Anim Hosp Assoc

1986;22:835–8.

[20] Engrall E, Bushnell N. Patellar luxation in Abyssinian cats. Feline Pract 1990;18:20–2.
[21] Flecknell M, Gruﬀydd-Jones T. Congenital luxation of the patellae in the cat. Feline Pract

1979;9:18–20.

[22] Smith GK, Langenbach A, Green PA, et al. Evaluation of the association between medial

patellar luxation and hip dysplasia in cats. JAVMA 1999;215:40–5.

[23] Houlton JEF, McGlennon NJ. Castration and physeal closure in the cat. Vet Rec 1992;

131:466–7.

[24] Chico AC, Font J, Marti JM. Trochlear femoral fractures in cats: results of seven cases. Vet

Comp Orthop Traumatol 2001;14:51–5.

[25] Brinker WO, Piermattei DL, Flo GL. Distal femoral fractures. In: Handbook of small

animal orthopedics and fracture management. 2nd edition. Philadelphia: WB Saunders;
1990. p. 129–34.

[26] Parker RB, Bloomberg MS. Modiﬁed intramedullary pin technique for repair of distal

femoral physeal fractures in the dog and cat. J Am Anim Hosp Assoc 1984;184:1259–65.

[27] Berg RJ, Egger EL, Konde LJ, McCurnin DM. Evaluation of prognostic factors for

growth following distal femoral physeal injuries in 17 dogs. Vet Surg 1984;13:172–80.

[28] Stigen O. Supracondylar femoral fractures in 159 dogs and cats treated using a normograde

intramedullary pinning technique. J Small Anim Pract 1999;40:519–23.

[29] Robinson A. Use of a Rush pin to repair fractures of the distal femur in cats. Vet Rec

2000;146:429–32.

[30] Franczuszki D, Chalman JA, Butler HC. The use of paired pins in the ﬁxation of distal

femur fractures in the dog and cat. J Am Anim Hosp Assoc 1986;22:173–8.

[31] Hardie EM, Chambers JN. Factors inﬂuencing the outcome of distal femoral physeal

fracture ﬁxation: a retrospective study. J Am Anim Hosp Assoc 1984;20:927–31.

[32] Lewis DD, van Ee RT, Oakes MG, Elkins AD. Use of reconstruction plates for stabi-

lization of fractures and osteotomies involving the supracondylar region of the femur. J Am
Anim Hosp Assoc 1993;29:171–8.

982

R.M. McLaughlin / Vet Clin Small Anim 32 (2002) 963–982

Feline injection site sarcomas

Bernard Se´guin, DVM, MS

Department of Surgical and Radiological Sciences, 2112 Tupper Hall,

University of California, Davis, CA 95616, USA

Sarcomas developing at the injection sites of vaccines have been recog-

nized for 10 years. Although there is strong epidemiologic evidence that
links the administration of inactivated feline vaccines and the subsequent
development of sarcomas at the site of injection, sarcomas developing at
sites of antibiotic and lufenuron administration, for example, have also been
reported occasionally [1–3]. It is now believed that vaccines are not the only
agents capable of causing sarcomas at the injection site; rather, virtually
anything that produces local inﬂammation has the potential to cause the
development of injection site sarcomas in susceptible cats [3]. Because the
incidence is relatively low and vaccines are the compounds that are given
to most cats in the population with suﬃcient frequency to make any corre-
lation [3], vaccines have received the most attention as potential causative
agents, and most of the research has focused on vaccines.

Feline injection site sarcomas are usually highly aggressive and invasive

locally, making them more challenging to treat than noninjection site sarco-
mas. They are also more likely to recur after surgical excision than noninjec-
tion site sarcomas [4]. Because of the many issues surrounding this problem,
the American Veterinary Medical Association, American Animal Hospital
Association, American Association of Feline Practitioners, and Veterinary
Cancer Society jointly formed the Vaccine-Associated Feline Sarcoma Task
Force (VAFSTF) in November of 1996, with the mission to plan and exe-
cute a coordinated response of research and education to this problem [5].

Epidemiology and etiopathogenesis

In 1985, a killed aluminum adjuvant rabies vaccine and a killed aluminum

adjuvant feline leukemia virus (FeLV) vaccine were introduced on the veter-
inary market [6]. Both of these vaccines became popular with veterinarians

Vet Clin Small Anim 32 (2002) 983–995

E-mail address: bseguin@ucdavis.edu (B. Se´guin).

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 2 2 - 0

for vaccinating cats. In 1986, focal cutaneous vasculitis and alopecia at sites
of rabies vaccination were reported in 13 dogs, 10 of which were Poodles
[7]. Rabies-speciﬁc ﬂuorescence was seen in the walls of dermal blood
vessels and in the epithelium of hair follicles in each of 3 dogs tested. By
1991, an increased number of injection site reactions received as biopsy
specimens at the Laboratory of Pathology of the University of Pennsylvania
beginning in September of 1988 were reported. The lesions were character-
ized as focal necrotizing granulomatous panniculitis associated with subcu-
taneous injection of rabies vaccine in cats and dogs [8]. Soon thereafter, the
possibility of a link between vaccination and the development of ﬁbrosarco-
mas in cats was raised [9].

Strong epidemiologic evidence has demonstrated an association between

the administration of inactivated FeLV and rabies vaccines and the subse-
quent development of soft tissue sarcomas [4,10–12]. Associations have also
been reported between feline panleukopenia and feline rhinotracheitis vac-
cines [4,13–15]. Tumor types that have been reported at the sites of vaccina-
tion are ﬁbrosarcoma (most common), malignant ﬁbrous histiocytoma,
osteosarcoma, rhabdomyosarcoma, undiﬀerentiated sarcoma, liposarcoma,
and chondrosarcoma [16]. The true incidence of sarcoma development after
vaccination is unknown. Some studies report an incidence of sarcomas per
vaccines administered [12], whereas others report a prevalence of sarcoma
per cat [10]. The incidence is estimated to be between 1/1000 and 1/10,000
vaccines administered [3,5,15]. It seems that the reaction to vaccines is addi-
tive and that the likelihood of sarcoma development increases with the num-
ber of vaccines given simultaneously at the vaccination site [12]. In a
retrospective study, it was determined that the risk of a cat developing a sar-
coma after administration of a single vaccine in the cervical-interscapular
region (a site not recommended anymore) was 50% higher than the risk of
a cat not receiving any vaccine at this site. The risk for a cat given two vac-
cines at the same site was approximately 127% higher, and the risk for a cat
given three to four vaccines was 175% higher [5,12]. Time to tumor develop-
ment in cats after vaccination has been reported to be between 2 months and
10 years [15].

A similar phenomenon of vaccine-associated sarcomas supports the

hypothesis that the sarcomas arise from inappropriate inﬂammatory or
immunologic reactions in cats [17,18]. This phenomenon relates to the devel-
opment of intraocular sarcomas after ocular trauma or chronic uveitis
[15,19–21]. The hypothesis is that the inﬂammatory reaction leads to uncon-
trolled proliferation of ﬁbroblasts and myoﬁbroblasts that undergo malig-
nant transformation in a subset of cats [15]. Transition zones from
inﬂammatory granuloma to sarcoma have been identiﬁed on histopatho-
logic examination as well as microscopic foci of sarcoma located in areas
of granulomatous inﬂammation, which, again, strongly supports the idea
that inﬂammation precedes the sarcoma development [5,22]. The unique
relation between trauma, inﬂammation, and tumorigenesis in the cat is still

984

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

not understood [15]. The component in the vaccine most commonly thought
to be associated with local postvaccination inﬂammation is the adjuvant
[23]. Aluminum is a common component of vaccine adjuvants in the form
of aluminum hydroxide or aluminum phosphate. Aluminum has been iden-
tiﬁed in postvaccinal granulomas and in some sarcomas at vaccination sites,
which may indicate a potential causative role in the development of these
tumors, but it may also be that the aluminum is only a marker of previous
vaccination [11].

The precise pathogenesis of vaccine-associated sarcomas is unknown.

The inﬂammation alone or in association with unidentiﬁed carcinogens or
oncogenes leads to neoplastic transformation and tumor development [5].
Growth factors or their receptors may be implicated or may be markers.
Vaccine-associated sarcomas have been found to be positive on immunohisto-
chemistry tests for platelet-derived growth factor and its receptor, epidermal
growth factor and its receptor as well as for transforming growth factor-b,
whereas sarcomas not associated with vaccines are negative or faintly
positive [18]. There is some evidence to suggest that lymphocytes or macro-
phages may play a role in the proliferation of ﬁbroblasts, perhaps through
platelet-derived growth factor [15,17,18].

Cancer is a disease with a genetic basis, and several studies have investi-

gated the presence of abnormalities at diﬀerent levels of the genetic code.
There seem to be major abnormalities in the chromosomes present in the
injection site sarcoma cells [24]. The role that diﬀerent genes may play has
also been researched. The gene c-jun is a proto-oncogene coding for a pro-
tein associated with cellular proliferation. Vaccine-associated sarcomas
express c-jun, whereas sarcomas not associated with vaccines do not [18].
The gene p53 is a tumor suppressor gene that plays a role in the regulation
of the cell cycle. When the p53 gene is absent or mutated, cells are able to
proceed through the cell cycle with damaged DNA, which can lead to malig-
nancies [15]. The role that p53 plays in the pathogenesis of vaccine-
associated sarcomas remains elusive but may have prognostic signiﬁcance
[15,25–28]. In one study, immunohistochemical detection of p53 protein in
a proportion of injection site–associated sarcomas suggested that mutation
of the p53 gene may play a role in the pathogenesis of these tumors [29].
Viral etiologies that have been evaluated thus far do not seem to play a role
in the pathogenesis of this tumor. Feline sarcoma virus, FeLV, and poly-
omavirus have all remained undetectable in feline vaccine-associated sarco-
mas [15,30,31].

Clinical approach

As with any tumor, ideally, these tumors should be treated early and

when small. Client education is imperative. First, clients should be warned
of the risk of vaccine/injection-associated sarcomas and the occurrence of

985

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

local vaccine/injection reactions. Owners should be taught how to examine
the injection site and to seek veterinary assistance if any of the following
scenarios occurs [5,15,32]:

1. A mass persists at the injection site for more than 3 months after in-

jection

2. A mass is present and is larger than 2 cm, regardless of time since in-

jection

3. A mass is still increasing in size 1 month after injection

Clients should not hesitate to seek veterinary assistance if there is any

doubt. After conﬁrmation of one of these three scenarios by the veterinar-
ian, a biopsy should be performed to ﬁnd out if the mass is a granuloma/
reactive mass or a sarcoma. A Tru-Cut (Travenol Laboratories, Dearﬁeld,
IL) needle biopsy, punch biopsy, or wedge biopsy is preferred. Cytologic
evaluation of ﬁne needle aspirates is considered unreliable for the diagnosis
of vaccine-associated sarcomas and is not recommended by the VAFSTF
[5,32]. Biopsies need to be performed by strict adherence to the following
principles of biopsy:

1. First, plan the site properly. It should be planned such that it is easy to

include the biopsy tract in the deﬁnitive resection or radiation ﬁeld.

2. Pay close attention to hemostasis and obliteration of dead space.

Avoidance of seromas and hematomas minimizes local contamination
of the biopsy site with cancer cells. Avoid using drains, because their
tract can also become contaminated with tumor cells.

3. If a biopsy is performed on the legs, the incision should be longitudinal

and not transverse.

4. Care should be taken not to traumatize the sample with forceps or

other handling instruments before ﬁxation.

5. It is best to use ‘‘uncontaminated’’ instruments (instruments that were

not used for acquiring the biopsy sample) when closing the biopsy tract.

These principles are aimed at minimizing the possibility of seeding tumor

cells beyond the already existing tumor.

If the mass is a vaccine reaction, there is currently no oﬃcial recommen-

dation from the VAFSTF on how to proceed (W.B. Morrison, personal
communication, 2001). The scientiﬁc evidence required to make a sound rec-
ommendation is lacking. It is unknown what proportion of granulomas may
become or induce the formation of a sarcoma, and the impact of excision of
a vaccine reaction on subsequent tumor development at the site has not been
elucidated. Fibrosarcomas have allegedly occurred in areas that were previ-
ously excised and determined on histopathologic examination to be injection
site reactions, raising questions about the utility of excising vaccine reac-
tions and the type of excision required (ie, marginal versus wide) [9,11,15].
Considering the existing evidence that inﬂammation can lead to sarcoma
development in some cats and the diﬃculty and frustration in treating these

986

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

tumors, it seems that the prudent thing to do may be to remove the
granuloma once it has met one or more of the three previously mentioned
criteria (W.M. Morrison, personal communication, 2001). The decision to
remove the granuloma or not must be an educated one on the part of the
owner. There is currently no scientiﬁc reason to justify wide excision of
granulomas; therefore, when the decision has been made to remove the
granuloma, marginal excision can be performed. These recommendations
regarding the management of granulomas may be shown to be erroneous
as we learn more about this clinical entity, and they may need to be revised
when newer information becomes available. If the mass is a sarcoma, proper
treatment planning becomes essential, which re-emphasizes the importance
of performing a biopsy ﬁrst before making any kind of treatment decision.

Cats with a biopsy-conﬁrmed injection site sarcoma need to be staged.

Cats with a suspected injection site sarcoma based on the history and loca-
tion of the tumor can have the staging process started either before or after
the biopsy procedure is performed but always before deﬁnitive treatment is
initiated. A complete blood cell count, chemistry panel, and urinalysis along
with feline leukemia virus (FeLV) and feline immunodeﬁciency virus (FIV)
tests determine the overall health status. Although no association is apparent
between viral status (FeLV, FIV) and tumor development, the course of the
disease may be altered because of compromise of the immune system [15].

Thoracic radiographs should be performed to look for evidence of meta-

stasis, which occurs in 10% to 24% of the cases [33,34]. Regional lymph
nodes should be evaluated through palpation, radiographs, or ultrasonogra-
phy and cytology where applicable. Metastasis occurs primarily in the lungs,
but other sites, such as the regional lymph nodes, mediastinum, pericardium,
liver, and pelvis, have also been reported [15,35–37]. Advanced imaging,
such as computed tomography (CT) or magnetic resonance imaging, is
extremely helpful and considered to be mandatory by many when planning
a surgical resection and potentially deciding to perform radiation therapy in
the neoadjuvant setting. In a study to evaluate the usefulness of CT imaging
of vaccine-associated sarcomas, the CT study revealed the tumor to be, on
average, twice as large as determined by physical examination and caliper
measurement, and recommendations regarding treatment were often altered
based on the CT results [15,38]. Even patients that have had a previous sur-
gical excision can beneﬁt from advanced imaging, because it can provide
information regarding the extent of the previous surgical ﬁeld and the area
that needs to be re-excised or included in the radiation treatment ﬁeld [15].

Treatment

Surgery still seems to be one of the most important components of the

treatment regimen for these tumors. Attempts at simple excision (ie, debulk-
ing or marginal excision) are rarely curative and ultimately lead to local

987

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

recurrence with a more diﬃcult second surgery [3,15]. Even attempts at
aggressive wide surgical excision are often incomplete and result in a 30%
to 70% recurrence rate [1,15]. Rear leg amputation has a higher rate of cure
than surgery in the interscapular space, but in some instances, amputation,
hemipelvectomy, or both still do not provide a cure [15,34]. Even when a his-
topathology report indicates no evidence of tumor cells at the surgical mar-
gins, there may be a 50% local recurrence rate [15]. Cats with aggressive
excision at the ﬁrst attempt have longer tumor-free intervals than those with
marginal excision (325 days versus 79 days), and cats with complete excision
have a longer tumor-free interval (>16 months versus 4 months) and sur-
vival time (>16 months versus 9 months) than those with incomplete excision
[34,39]. One study evaluating the prognosis after excision alone found a
median overall length of survival of 576 days, and based on the Kaplan-
Meier product limit method, approximately 10% of the cats can be consid-
ered cured of their tumor with excision alone [34]. From the Kaplan-Meier
product limit method, again, it was estimated that approximately 45% of the
cats with tumors located on their limbs treated by radical ﬁrst excision
(amputation) can be considered cured, whereas less than 10% of cats treated
by wide or marginal excision can be considered cured. It was also deter-
mined that there was a signiﬁcant diﬀerence in time to ﬁrst recurrence
between cats treated by wide excision versus cats treated by marginal exci-
sion (419 days versus 66 days, respectively) [34]. Median time to ﬁrst recur-
rence was 94 days [34] but can range from a couple of weeks to more than 6
months [15].

Aggressive surgical excision should be attempted. This means 3-cm mar-

gins laterally and one fascial plane deep to the tumor. If the tumor involves
or comes so close as to not be separated by a fascial plane from the scapula,
a spinous process, or the pelvis, a scapulectomy, spinous process resection,
or hemipelvectomy needs to be performed en bloc with the tumor excision.

After excision, the entire specimen needs to be submitted for histopatho-

logic examination. It is important to mark the surgical margins so as to help
the pathologist evaluate the margins for completeness. One of the best ways
to mark the margins is to use India ink or another marking dye [40].

Unfortunately, surgery alone often fails to provide a cure. Additional

local therapy, such as radiation therapy, is indicated so as to provide longer
control of the disease. If postoperative radiotherapy is part of the treatment
plan or the possibility of irradiation exists, the surgeon should place hemo-
clips in the surgical bed at the time of surgical excision; this allows identiﬁ-
cation of the surgical bed with conﬁdence in the near future so as to apply
the appropriate radiation treatment ﬁeld [15].

Radiation therapy alone is not considered an appropriate treatment for

these tumors. Irradiation alone should only be considered in the palliative
setting [15,41]. Radiotherapy combined with surgery seems to increase the
tumor control rate, however [15]. One study evaluated the eﬀectiveness of
radiation therapy in the preoperative setting with surgery in 33 cats [33].

988

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

Seventy-three percent of the cats had at least one surgery before irradiation,
and 10 cats only had microscopic disease at the time irradiation was initi-
ated. All cats underwent surgical excision 2 to 4 weeks after completion
of the radiation therapy. An attempt was made to excise all skin within the
irradiated ﬁeld whether gross tumor was present or not [33]. The median dis-
ease-free interval was 398 days, with a median overall length of survival of
600 days. Fifty-eight percent of the cats had treatment failure (local recur-
rence or metastasis). Forty-ﬁve percent had local recurrence. Based on the
Kaplan-Meier product limit method, approximately 30% of the cats can
be considered cured. Completeness of surgical margins was the only prog-
nostic factor for treatment success. The disease-free interval for cats with
incomplete surgical margins was 112 days versus 700 days for cats with com-
plete surgical margins [33].

It seems that complete surgical excision is of great importance; thus,

aggressive surgery should be performed irrespective of the addition of adju-
vant radiotherapy. Another study came to diﬀerent conclusions, however
[1]. In a study in which 76 cats received postoperative radiation therapy,
there was no diﬀerence in local recurrence rates between conservative exci-
sion and wide excision. Also, the status of the surgical margins was not a
prognostic factor for local recurrence or survival time, meaning that there
was no signiﬁcant diﬀerence between local recurrence rates and survival
times between cats with complete and incomplete surgical margins [1]. As
recognized by the authors, however, these data must be interpreted with
caution because of the low power of the analysis (power

¼ 0.06). In that

study, in cats that received conservative excision, radiotherapy was insti-
tuted immediately after or within 2 days of surgery. It was found that the
sooner the radiotherapy was started after surgery, the longer was the
disease-free interval and survival time [1]. It was not possible to determine
whether the longer disease-free interval and survival time were attributable
to the type of surgery (conservative versus wide excision), the shorter delay
between surgery and radiotherapy, or a combination of these factors. The
overall recurrence rate was 41%, median disease-free interval was 405 days,
and survival time from the start of radiation therapy was 469 days. Cats that
underwent more than one surgery before radiation therapy were more likely
to suﬀer from local recurrence and had a signiﬁcantly shorter disease-free
interval than those that underwent only one surgery [1]. Until more evidence
becomes available to the contrary, it seems prudent to recommend that vet-
erinarians perform as aggressive an excision as possible. Conversely, there is
some evidence to suggest that an extremely aggressive resection, meaning
5-cm margins laterally and two muscle planes deep, could be suﬃcient as the
sole treatment in most cats with injection site sarcomas, thereby avoiding
radiation therapy [42].

The debate as to which is the better approach between preoperative

and postoperative irradiation remains open. No clinical study has evaluated
this issue for feline injection site sarcomas. There are advantages and

989

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

disadvantages for each (Table 1) [43]. Postoperative radiation therapy refers
to the planned delivery of radiation after surgery. Typically, the radiation is
started 10 to 14 days after surgery, although it can be started as early as 24
hours after surgery [43]. Preoperative radiation refers to radiation delivered
before surgery. The surgery involves removing the tumor in toto at some

Table 1
Advantages and disadvantages of postoperative and preoperative radiation therapy

Postoperative radiation
therapy

Preoperative radiation
therapy

Delay in performing or

delivering therapy

A: No delay in performing

Sx because of acute side
eﬀects from xrt

A: No delay starting xrt

because of wound healing
problems

D: Delay in starting xrt

because of wound
healing complications
(eg, seroma); allows more
time for tumor cell
repopulation

D: Delay in performing Sx if

acute side aﬀects of xrt
persist

Wound healing

A: Not impaired from

previous xrt

D: Impaired from xrt

Size of radiation treatment

ﬁeld

D: Larger, because must

include all tissues
handled at Sx, including
entire incision line and
draining sites if present;
the larger the ﬁeld, the
greater the morbidity

A: Smaller, because no Sx

has been performed

Circulating cancer cells

released because of
manipulation of tumor
at surgery

D: Cancer cells released are

more likely to be viable;
may increase risk of
developing metastasis

A: Cancer cells entering

circulation less likely to be
viable because they were
killed as a result of xrt

Blood supply to tumor,

especially to cells at
periphery

D: Surgery disrupts blood

supply to tumor cells,
especially the ones at the
periphery of the tumor,
because these are the cells
that are mostly targeted
by xrt; hypoxic cells are
more radioresistant

A: Blood supply to cancer

cells at periphery of tumor
is at its best; therefore,
cells are most
radiosensitive from the
oxygenation stand point

Size of tumor

D: The tumor is as large as

can be and may be
diﬃcult to resect

A: Tumor may regress in size

as a result of xrt and make
it more amendable to
surgical resection;
however, making an
inoperable tumor
operable with xrt should
not be the main goal of xrt

¼ advantage; D ¼ disadvantage; xrt ¼ radiation therapy; Sx ¼ surgery.

Data from McLeod DA, Thrall DE. The combination of surgery and radiation in the

treatment of cancer. A review. Vet Surg 1989;18:1–6.

990

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

predetermined time rather than waiting to see if radiation is eﬀective or not
in controlling the tumor locally. Typically, surgery is performed 3 to 4 weeks
after the end of the radiation treatments [43]. Another potential way of pro-
viding radiation therapy is through the means of brachytherapy by implant-
ing iridium 192 [44].

Chemotherapy alone should not be considered for deﬁnitive therapy. As

for radiation therapy, chemotherapy alone may be attempted in the pallia-
tive setting. In the adjuvant or neoadjuvant setting, however, chemotherapy
may play an important role in the multimodality approach to treatment.
The use of various chemotherapy protocols has resulted in some partial
responses and, less frequently, in some complete responses [15]. Chemother-
apy in the preoperative setting may reduce tumor size, thereby facilitating
surgical resection, or chemotherapy can be used as a radiation sensitizer.
Agents that have been used include doxorubicin, cyclophosphamide, carbo-
platin, mitoxantrone, and vincristine [15]. In one study evaluating the use of
combination doxorubicin and cyclophosphamide in cats with nonresectable
tumors, 50% of the cats had a 50% or greater decrease in gross tumor bur-
den [45]. Seventeen percent had resolution of all clinically detectable tumors
[45]. Unfortunately, the responses were not durable, with median response
duration of 125 days. Median length of survival for responders was 242 days
versus 83 days for nonresponders [45]. Another study comparing surgery
and radiation therapy with or without doxorubicin found no signiﬁcant dif-
ference between the group receiving adjuvant chemotherapy and the group
that did not [46]. Median time to ﬁrst recurrence was 661 days for the group re-
ceiving adjuvant chemotherapy, whereas median time to ﬁrst recurrence
was not yet attained for the group not receiving adjuvant chemotherapy.
The median survival time was 701 days overall and 674 days and 842 days
for the group receiving adjuvant doxorubicin and the group not receiving
doxorubicin, respectively. The power of the study was 5% [46]. Another
study had similar results in which cats treated with surgery and irradiation
but with or without chemotherapy (doxorubicin and cyclophosphamide) did
not have a signiﬁcant diﬀerence in rate of local recurrence, rate of metasta-
sis, and survival time (power < 0.50) [1]. In another study, cats treated with
preoperative irradiation, concurrent chemotherapy with doxorubicin, and
surgical excision had a more prolonged disease-free interval (median of
360 days) than cats treated the same way but without doxorubicin (median
of 162 days). There was no diﬀerence in survival times between these two
groups of cats, however [47]. Although the last study may have conﬂicting
conclusions regarding the eﬀect of chemotherapy on disease-free interval,
one should be careful when comparing the results and conclusions of diﬀer-
ent studies, because the protocols were diﬀerent and there are other diﬀer-
ences that can aﬀect the results. It may be possible to use hyperthermia to
increase the local delivery of a liposome-encapsulated chemotherapeutic
agent in feline sarcomas [48]. Even though the metastatic rate is relatively
low (10%–24%), systemic chemotherapy may play a role in the delay or

991

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

prevention of the development of metastatic disease and needs to be further
investigated. Nevertheless, because of the aggressive nature of the injec-
tion site sarcomas in cats, for an owner who wants the best treatment, it
seems reasonable and justiﬁed to include chemotherapy as part of the treat-
ment regimen. Obviously, the best chemotherapy protocol remains to be
determined.

Immunotherapy has received comparatively little attention to date in the

treatment of injection site sarcomas. One study evaluated the role of xeno-
geneic cells that secrete human interleukin-2 [49]. The xenogeneic cells were
inﬁltrated at the time of surgical excision and implantation of iridium 192
for brachyradiotherapy. The inﬁltration was repeated several times during
a 2-month period. Of 16 cats treated with this protocol, 2 had local recur-
rence and 3 had metastases for a median length of survival of 16 months,
whereas 11 of 16 cats that did not receive the xenogeneic cell inﬁltrations
had tumor recurrence with a median length of survival of 8 months. One cat
did develop anaphylaxis [49].

Another interleukin, interleukin-12, is currently under investigation. Ace-

mannan has also been evaluated as an immunostimulant [50,51]. More stud-
ies are required before acemannan and other immunotherapies can be
recommended as part of the treatment regimen.

Patients that have been treated for an injection site sarcoma should be

rechecked: a physical examination should be performed monthly for 3
months and then at least every 3 months for the ﬁrst year and every 3 to
6 months thereafter. Additional diagnostic procedures should be performed
as indicated by the clinical signs and ﬁndings on the physical examination.

Conclusion

Injection site sarcomas are aggressive and thus are frustrating for the

attending veterinarians and the owners. They are a serious disease aﬄicting
cats and cause death more often than not. Signiﬁcant progress has been
made in the treatment of feline injection site sarcomas, and the best results
seem to be attained with a multimodality approach; at the present time, a
combination of aggressive surgery, irradiation, and chemotherapy seems
to be the treatment regimen that provides the best prognosis. Many ques-
tions remain unanswered, however, and too many patients still have treat-
ment failures. Because feline patients are presented to the veterinarian
with tumors of diﬀerent sizes and at diﬀerent locations, the best treatment
in each instance remains unknown. For example, it is not known whether
aggressive surgery alone is suﬃcient for long-term control in cats with rela-
tively small (<3 cm), discrete, and noninﬁltrative tumors that have not pre-
viously been resected [15]. Because factors that can predict the biologic and
clinical behavior have not been fully identiﬁed yet, recommending aggressive
multimodality therapy for all cats with injection site sarcomas seems neces-
sary at this time. Because vaccinations are the compounds that are most

992

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

commonly linked to these tumors, practitioners are encouraged to become
involved in the debate regarding vaccination protocols and to follow the
guidelines for vaccine administration put forward by the VAFSTF [5,15,52].

References

[1] Cohen M, Wright JC, Brawner WR, et al. Use of surgery and electron beam irradiation,

with or without chemotherapy, for treatment of vaccine-associated sarcomas in cats: 78
cases (1996–2000). JAVMA 2001;219:1582–9.

[2] Gagnon AC. Drug injection-associated ﬁbrosarcoma in a cat. Feline Pract 2000;28:18–21.
[3] MacEwen EG, Powers BE, Macy D, Withrow SJ. Soft tissue sarcomas. In: Withrow SJ,

MacEwen EG, editors. Small animal clinical veterinary oncology. 3rd edition. Philadelphia:
WB Saunders; 2001. p. 283–304.

[4] Hendrick MJ, Shofer FS, Goldschmidt MH, et al. Comparison of ﬁbrosarcomas that

developed at vaccination sites and at nonvaccination sites in cats: 239 cases (1991–1992).
JAVMA 1994;205:1425–9.

[5] Morrison WB, Starr RM, et al. Vaccine-associated feline sarcomas. JAVMA 2001;218:

697–702.

[6] Macy DW, Hendrick MJ. The potential role of inﬂammation in the development of

postvaccinal sarcomas in cats. Vet Clin North Am Small Anim Pract 1996;26:103–9.

[7] Wilcock BP, Yager JA. Focal cutaneous vasculitis and alopecia at sites of rabies

vaccination in dogs. JAVMA 1986;188:1174–7.

[8] Hendrick MJ, Dunagan CA. Focal necrotizing granulomatous panniculitis associated with

subcutaneous injection of rabies vaccine in cats and dogs: 10 cases (1988–1989). JAVMA
1991;198:304–5.

[9] Hendrick MJ, Goldschmidt MH. Do injection site reactions induce ﬁbrosarcomas in cats?

[letter]. JAVMA 1991;199:968.

[10] Coyne MJ, Reeves NCP, Rosen DK. Estimated prevalence of injection-site sarcomas in

cats during 1992. JAVMA 1997;210:249–51.

[11] Hendrick MJ, Goldschmidt MH, Shofer FS, et al. Postvaccinal sarcomas in the cat:

epidemiology and electron probe microanalytical identiﬁcation of aluminum. Cancer Res
1992;52:5391–4.

[12] Kass PH, Barnes WG, Spangler WL, et al. Epidemiologic evidence for a causal relation

between vaccination and ﬁbrosarcoma tumorigenesis in cats. JAVMA 1993;203:396–405.

[13] Burton G, Mason KV. Do postvaccinal sarcomas occur in Australian cats? Aust Vet J

1997;75:102–6.

[14] Lester S, Clemett T, Burt A. Vaccine site-associated sarcomas in cats: clinical experience

and laboratory review (1982–1993). J Am Anim Hosp Assoc 1996;32:91–5.

[15] McEntee MC, Page RL. Feline vaccine-associated sarcomas. J Vet Intern Med 2001;

15:176–82.

[16] Hendrick MJ, Brooks JJ. Postvaccinal sarcomas in the cat: histology and immunohisto-

chemistry. Vet Pathol 1994;31:126–9.

[17] Hendrick MJ. Feline vaccine-associated sarcomas: current studies on pathogenesis.

JAVMA 1998;213:1425–6.

[18] Hendrick MJ. Feline vaccine-associated sarcomas. Cancer Invest 1999;17:273–7.
[19] Dubielzig RR, Everitt J, Shadduck JA, et al. Clinical and morphologic features of post-

traumatic ocular sarcomas in cats. Vet Pathol 1990;27:62–5.

[20] Hakanson N, Shively JN, Reed RE, et al. Intraocular spindle cell sarcoma following ocular

trauma in a cat: case report and literature review. J Am Anim Hosp Assoc 1990;26:63–6.

[21] Pfeiﬀer RL, Monticello T, Bouldin TW. Primary ocular sarcomas in the cat. J Small Anim

Pract 1988;29:105–16.

993

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

[22] Hendrick MJ, Kass PH, McGill LD, et al. Postvaccinal sarcomas in cats. J Natl Cancer

Inst 1994;86:341–3.

[23] Vanselow BA. The application of adjuvants to veterinary medicine. Vet Bull 1987;57:881–96.
[24] McNiel EA, Hoots EA, LaRue SM. Characterization of chromosomal aberrations in feline

vaccine-associated sarcoma using comparative genomic hybridization. In: Proceedings of
the 19th Annual American College of Veterinary Internal Medicine Veterinary Medical
Forum, Denver, CO [abstract]. Lakewood (CO): American College of Veterinary Internal
Medicine; 2001. p. 857.

[25] Hershey AE, Dubielzig RR, Helfand SC. P53 expression in feline vaccinal sarcomas.

Proceedings of the 18th Annual Meeting of the American College of Veterinary Internal
Medicine, May 25–28, Seattle, WA; 2000. p. 390.

[26] Kanjilal S, Banerji N, Fifer A, et al. P53 tumor suppressor gene alterations in vaccine-

associated feline sarcoma. Presented at the 19th Annual Veterinary Cancer Society Con-
ference. Woods Hole, MA, November 13–16, 1999.

[27] Mayr B, Blauensteiner J, Edlinger A, et al. Presence of p53 mutations in feline neoplasms.

Res Vet Sci 2000;68:63–9.

[28] Mayr B, Reiﬁnger M, Alton K, Schaﬀner G. Novel p53 tumor suppressor mutations in

cases of spindle cell sarcoma, pleomorphic sarcoma and ﬁbrosarcoma in cats. Vet Res
Commun 1998;22:249–55.

[29] Nambiar PR, Jackson ML, Ellis JA, et al. Immunohistochemical detection of tumor

suppressor gene p53 protein in feline injection site-associated sarcomas. Vet Pathol 2001;
38:236–8.

[30] Ellis JA, Jackson ML, Bartsch RC, et al. Use of immunohistochemistry and polymerase

chain reaction for detection of oncornaviruses in formalin-ﬁxed, paraﬃn-embedded
ﬁbrosarcomas from cats. JAVMA 1996;209:767–71.

[31] Kidney BA, Haines DM, Ellis JA, et al. Evaluation of formalin-ﬁxed paraﬃn-embedded

tissues from vaccine site-sarcomas of cats for polyomavirus DNA and antigen. Am J Vet
Res 2001;62:828–32.

[32] Vaccine-Associated Feline Sarcoma Task Force. Diagnosis and treatment of suspected

sarcomas [abstract]. JAVMA 1999;214:1745.

[33] Cronin K, Page RL, Sponick G, Dodge R, Hardie EN, Price GS, et al. Radiation therapy

and surgery for ﬁbrosarcoma in 33 cats. Vet Radiol Ultrasound 1998;39:51–6.

[34] Hershey AE, Sorenmo KU, Hendrick MJ, Shofer FS, Vail DM. Prognosis for presumed

feline vaccine-associated sarcoma after excision: 61 cases (1986–1996). JAVMA 2000;
216:58–61.

[35] Esplin DG, Campbell R. Widespread metastasis of a ﬁbrosarcoma associated with vaccin-

ation site in a cat. Feline Pract 1995;23:13–6.

[36] Esplin DG, Jaﬀe MH, McGill LD. Metastasizing liposarcoma associated with a vaccin-

ation site in a cat. Feline Pract 1996;24:20–3.

[37] Rudmann DG, Van Alstine WG, Doddy F, et al. Pulmonary metastasis of a feline vaccine-

site ﬁbrosarcoma. J Vet Diagn Invest 1998;10:79–82.

[38] McEntee MC, Samii VF. The utility of contrast enhanced computed tomography in feline

vaccine-associated sarcomas: 35 cases [abstract]. Vet Radiol Ultrasound 2000;41:575.

[39] Davidson EB, Gregory CR, Kass PH. Surgical excision of soft tissue sarcomas in cats. Vet

Surg 1997;26:265–9.

[40] Seitz SE, Foley GL, Maretta SM. Evaluation of marking materials for cutaneous surgical

margins. Am J Vet Res 1995;65:826–33.

[41] Ogilvie GK. Recent advances in the treatment of vaccine-associated sarcomas. Vet Clin

North Am Small Anim Pract 2001;31:525–33.

[42] Kuntz CA. Modiﬁed wide local excision for vaccine associated soft tissue sarcomas in cats

[abstract]. Vet Surg 2000;29:481.

[43] McLeod DA, Thrall DE. The combination of surgery and radiation in the treatment of

cancer. A review. Vet Surg 1989;18:1–6.

994

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

[44] Rogers K, Walker M, Barton C, et al. Preliminary results of post-surgical iridium-192

brachytherapy for vaccine-associated sarcomas in 18 cats [abstract]. Vet Radiol Ultrasound
2000;41:576.

[45] Barber LG, Sorenmo KU, Cronin KL, Shofer FS. Combined doxorubicin and cyclo-

phosphamide chemotherapy for nonresectable feline ﬁbrosarcoma. J Am Anim Hosp Assoc
2000;36:416–21.

[46] Bregazzi VS, LaRue SM, McNiel E, Macy DW, Dernall WS, Powers BE, et al. Treatment

with combination of doxorubicin, surgery, and radiation versus surgery and radiation alone
for cats with vaccine-associated sarcomas: 25 cases (1995–2000). JAVMA 2001;218:547–50.

[47] King GK, Harris DF, Hottinger H. Prolonged remission and survival time in cats with

injection-site associated sarcomas using preoperative radiation and doxorubicin protocol:
61 cases (1998–2000) [abstract]. American College of Vet Radiol and Ultrasound 2001;
42:585.

[48] Matteucci ML, Thrall DE, Anyarambhatla G, et al. Hyperthermia eﬀect on uptake of

technetium-99M labeled liposomes in feline sarcomas [abstract]. Vet Radiol Ultrasound
1999;40:683.

[49] Quintin-Colonna F, Devauchelle P, Fradelizi D, et al. Gene therapy of spontaneous canine

melanoma and feline ﬁbrosarcoma by intratumoral administration of histoincompatible
cells expressing human interleukin-2. Gene Ther 1996;3:1104–6.

[50] Kent EM. Use of an immunostimulant as an aid in treatment and management of

ﬁbrosarcoma in three cats. Feline Pract 1993;21:13–7.

[51] King GK, Yates KM, Greenlace PG, et al. The eﬀect of acemannan immunostimulant in

combination with surgery and radiation therapy on spontaneous canine and feline
ﬁbrosarcomas. J Am Anim Hosp Assoc 1995;31:439–47.

[52] Richards J. Feline vaccination guidelines. Vet Clin North Am Small Anim Pract 2001;31:

455–72.

995

B. Se´guin / Vet Clin Small Anim 32 (2002) 983–995

Index

Note: Page numbers of article titles are in boldface type.

Acepromazine, 757

Alpha

-adrenoceptor agonists, 756–757

Ameroid constrictors, in portosystemic shunts, 893–894

Analgesia, multimodal, 749

preemptive, 749

Analgesic adjuvant agents, 756

perioperative, in cats, 758

Analgesic agents, pharmacology of, 750–757

Analgesic techniques, 757–761

Anesthesia, for treatment of hyperthyroidism, 854

in portosystemic shunts, 890–891

Anesthesia/analgesia, interpleural, 761

Anesthetics, local, in analgesic techniques, 755

perioperative, in cats, 756

Anti-inﬂammatory agents, nonsteroidal, 753–754

perioperative, in cats, 755

Ascites, following surgery in portosystemic shunt, 896

Azotemia, as complication of thyroidectomy, 857

Bones, long, feline, fractures of, treatments for, 927–947

Brachial plexus block, 759–760

Canthal V-plasty, medial, 767

Cartilage eversion, 774

Cat, gastrointestinal foreign bodies in, 861–880

long bones of, fractures of, treatments for, 927–947
lumbosacral/pelvic injuries, 949–962
megacolon in, 901–915

Vet Clin Small Anim 32 (2002) 997–1004

0195-5616/02/$ - see front matter

PII: S 0 1 9 5 - 5 6 1 6 ( 0 2 ) 0 0 0 4 0 - 2

Cat, gastrointestinal (continued )

nasopharyngeal polyps in, 839–849
ocular surgeries in, 765–790
odontoclastic resorptive lesions in, 791–837
pain recognition in, 750
perioperative pain management in, 747–763
portosystemic shunts in, diagnosis and treatment of, 881–899
stiﬂe joint of, surgical diseases of, 963–982
thyroid surgery in, 851–859

Cataracts, surgery for, 787

Ciliary body, pharmacologic ablation of, 783

Colectomy, for megacolon, 908–914

Collateral ligament rupture, clinical signs and diagnosis of, 967

treatment of, 967–969

Colon, absorption/secretion in, 904

anatomy of, 902–903
motility in, 903–904
pathophysiology of, 905–907
physiology of, 903–904

Conjunctiva, grafts of, in corneal ulcerations, 778–779, 780

Cornea, surgery of, 772–776

trauma to, management of, 772–776
ulcerations of, conjunctival grafts in, 778–779, 780

corneal-scleral-conjunctival transposition in, 780–782
linear grid keratotomy in, 776–778
superﬁcial keratectomy in, 779–780, 781

Corneal-scleral-conjunctival transposition, in corneal ulcerations, 780–782

Cranial cruciate ligament rupture, clinical signs and diagnosis of, 964–965

treatment of, 965–967

Cruciate ligament, cranial, rupture of, clinical signs and diagnosis of, 964–965

treatment of, 965–967

Cystitis, bacterial, following perineal urethrostomy, 922

Dental nerve blocks, 760–761

Ear, middle, surgical anatomy of, 842–843

Entropion, surgical techniques in, 766–769

Enucleation, transconjunctival, 783–785

transpalpebral, 785–786

Epidural anesthesia/analgesia, 757–759

998

Index / Vet Clin Small Anim 32 (2002) 997–1004

dosages for cats, 759

Esophagus, feline, foreign bodies in, 862–866

Exenteration, 788

Eye, feline, surgery of, 765–790

Eyelid(s), agenesis rotating/pedicle graft, 772, 773

defects of, sliding H-plasty for, 770–772
full-thickness resection of, 766
reconstructive procedures of, 769, 770
tacking of, 768–769
third, gland prolapse, cartilage eversion, 772, 773, 774

surgery of, 772, 774, 775

trauma to, management of, 765–766
wedge resection of, 766, 769–770

Femur, distal feline, fractures of, causes of, 976

clinical signs of, 976
treatment of, 977–981

feline, fractures of, 942–944

Fibia, feline, fractures of, 943, 944

Foreign bodies, gastrointestinal, in cat, 861–880

Fracture(s), femoral, feline, 942–944

ﬁbular, feline, 943, 944
humeral, feline, 939–940
long bone, feline, biology of, 929–931

healing of, 939
implant selection for, 931–934
treatments for, implant removal in, 936

operative considerations for, 929–939
perioperative considerations for, 927–928
surgical approaches for, 934–935, 936–938

patterns of, in lumbosacral/pelvic injuries, 956–958
radial, feline, 940–941
sacrocaudal, in lumbosacral/pelvic injuries, 959–961
tibial, feline, 943, 944–945
ulnar, feline, 940–941

Gastrointestinal foreign bodies, feline, 861–880

clinical presentations of, 861–862
hematologic abnormalities in, 862
in esophagus, 862–866
in large intestine, 877–878
in small intestine, 869–877

999

Index / Vet Clin Small Anim 32 (2002) 997–1004

Gastrointestinal (continued )

in stomach, 866–869
location of, 862–878

Globe(s), blind painful, surgery for, 782–786

proposed, replacement of, 789–790

Hemorrhage, following surgery in portosystemic shunt, 895

Hernia, iatrogenic perineal, following perineal urethrostomy, 924

Hotz-Celsus procedure, 769, 770

Humerus, feline, fractures of, 939–940

Hypertension, portal, following surgery in portosystemic shunt, 895–896

Hyperthyroidism, feline, causes of, 851

recurrent, following thyroidectomy, 858
treatment of, 851–852

anesthetic considerations for, 854
medical options for, 852
postoperative complications in, 856–858
preoperative evaluation for, 853–854
surgical anatomy and, 852–853
surgical techniques for, 854–856

Hypocalcemia, as complication of thyroidectomy, 856–857

Hypothyroidism, as complication of thyroidectomy, 858

Incontinence, urinary, following perineal urethrostomy, 923–924

Interpleural anesthesia/analgesia, 761

Intestine, large, feline, foreign bodies in, 877–878

small, feline, foreign bodies in, 869–877

Intraocular prosthesis, 786, 787

Keratectomy, superﬁcial, in corneal ulcerations, 779–780, 781

Keratotomy, linear grid, in corneal ulcerations, 776–778

Ketamine, 757

Laser surgery, ocular, 788

Lumbosacral/pelvic injuries, 949–962

conservative versus surgical management of, 953–956

1000

Index / Vet Clin Small Anim 32 (2002) 997–1004

fracture patterns in, 956–958
gastrointestinal injury related to, 951
initial assessment in, 952–953
peripheral nerve damage in, 951
physical examination in, 952
radiographic evaluation in, 952–953
sacrocaudal fractures in, 959–961
soft tissue trauma associated with, 950–951

Megacolon, classiﬁcation of, 901–902

colectomy for, 908–914
diagnosis of, 907–908
idiopathic, 905–906
in cat, 901–915

causes of, 901, 902

secondary to neurologic or medical disease, 906
secondary to outlet obstruction, 907
treatment of, 908–914

medical, 908
surgical, 908–914

Myringotomy, in feline nasopharyngeal polyps, 844

Nasopharyngeal polyps, feline, 839–849

bulla osteotomy in, 845–846
diagnosis of, 840–841, 842
etiology of, 839
histopathology of, 841–842
history and clinical signs of, 840
myringotomy in, 844
surgery in, and adjunct treatment, 847

objectives of, 843
operative and postoperative complications of, 846–847
preoperative considerations for, 844
techniques of, 844–846

traction-avulsion in, 844
treatment of, 842–846

Nerve blocks, dental, 760–761

intercostal, 761
radial, ulnar, and median, 760

Nerve damage, as complication of thyroidectomy, 857–858

Nociception, 748

Nonsteroidal anti-inﬂammatory agents, 753–754

perioperative, in cats, 755

1001

Index / Vet Clin Small Anim 32 (2002) 997–1004

Ocular proptosis, traumatic, 789–790

Ocular surgeries, in cats, 765–790

Odontoclastic resorptive lesions, feline, 791–837

adhesion molecules and, 818
and nonfelidae, compared, 823
classiﬁcation of, 801–812
clinical signs of, 819
compared with other dental hard substance defects, 822–823
crown amputation with root retention in, 825–826
diagnosis of, 819–822
etiopathogenesis of, 793–795
extraction of teeth in, 824–825
histopathologic appearance of, 812–818
in endocrine and metabolic imbalances, 798
in viral infections, 798–799
laser treatment of, 826
oral examination in, 819–820
pecularities of feline teeth and, 794–795
predisposition to, 800–801
prevalence of, 797, 800–801, 802–809
pulp response in, 818
radiographic ﬁndings in, 820–822
reparative phase of, 817–818
resoptive phase of, 813–817
restoration of, 824
risk factors for, 795–798
systemic and medical therapies in, 826–827
trauma and, 795
treatment of, 823–827
types of, 791, 792–793

OP3 (l) agonists, 751–752

OP3 (l) antagonists, 752–753

OP2 (K) agonists/OP3 (l) antagonists, 752

Opioids, 750

perioperative, in cats, 753

Orbit, surgery of, 787–788

Pain, pathophysiology of, 747–749

Pain management, feline, perioperative, 747–763

principles of, 749

Pain recognition, in cats, 750

1002

Index / Vet Clin Small Anim 32 (2002) 997–1004

Parathyroid glands, autotransplantation of, 856

staged bilateral removal of, 856

Patella, luxation of, in cat, causes of, 973

clinical signs and diagnosis of, 974
imbrication in, 975
tibial crest transposition in, 975–976
treatment of, 974–976
trochlear recession in, 975

Polyps, nasopharyngeal. See Nasopharyngeal polyps.

Portography, in portosystemic shunts in cat, 886–887, 888

Portosystemic shunt(s), acquired multiple, 881–882

ameroid constrictors in, 893–894
anesthesia in, 890–891
categorization of, 881
clinicopathologic testing in, 883–884
diagnostic imaging in, 884–887
dissection of, 891–892
extrahepatic, 881
historical ﬁndings in, clinical signs of, 882–883
in cat, diagnosis and treatment of, 881–899
intrahepatic, 881
medical management of, 887–890
mesenteric catheter in, 886, 892
partial ligation in, 892–893
surgical management of, 891–895

complications of, 895–897

Radius, feline, fractures of, 940–941

Sarcoma(s), injection site, feline, 983–995

clinical approach to, 985–987
epidemiology and etiopathogenesis of, 983–985
treatment of, 987–992

Scintigraphy, in portosystemic shunts in cat, 885–886

Seizures, following surgery in portosystemic shunt, 897

Sliding H-plasty, for eyelid defects, 770–772

Stiﬂe joint, feline, dislocation of, causes of, 969–970

clinical signs and diagnosis of, 970
treatment of, 970–973

lameness of, causes of, 963–964

in collateral ligament rupture, 967–969
in cranial cruciate ligament rupture, 964–967

1003

Index / Vet Clin Small Anim 32 (2002) 997–1004

Stiﬂe joint (continued )

in stiﬂe dislocation, 969–973

surgical diseases of, 963–982

Stomach, feline, foreign bodies in, 866–869

Thyroid surgery, feline, 851–859

surgical considerations for, 852–858
surgical indications for, 851

Thyroidectomy, extracapsular, 855

modiﬁed, 855

intracapsular, 855

modiﬁed, 856

Tibia, feline, fractures of, 943, 944–945

Transconjunctival enucleation, 783–785

Transpalpebral enucleation, 785–786

Ulna, feline, fractures of, 940–941

Ultrasound, in portosystemic shunts in cat, 885

Urethral stoma, stricture, following perineal urethrostomy, 922–923

Urethrostomy, perineal, 917–925

indications for, 917
postoperative complications of, 920–924
prognosis following, 924
surgical technique for, 917–920

Wedge resection, of eyelid, 766, 769–770

1004

Index / Vet Clin Small Anim 32 (2002) 997–1004

Document Outline

00 Preface
01 Feline perioperative pain management
02 Ocular surgeries in cats
03 Feline odontoclastic resorptive lesions_An unsolved enigma in veterinary dentistry
04 Feline nasopharyngeal polyps
05 Feline thyroid surgery
06 Feline gastrointestinal foreign bodies
07 Diagnosis and treatment of portosystemic shunts in the cat
08 Megacolon in the cat
09 Perineal urethrostomy
10 Treatments for feline long bone fractures
11 Lumbosacral and pelvic injuries
12 Surgical diseases of the feline stifle joint
13 Feline injection site sarcomas
14 Index

Wyszukiwarka

Podobne podstrony:
2010 4 JUL Topics in Cardiology
2002 3 MAY Lasers in Medicine and Surgery
2010 6 NOV Current Topics in Canine and Feline Infectious Diseases
2002 3 MAY Lasers in Medicine and Surgery
Instruments used in general surgery
Ian Hacking (1983) introductory topics in the philosophy of natural science
Hacking, Ian Introductory topics in the philosophy of natural science (1983)
Philippa Foot Moral Dilemmas and other topics in moral philosophy 2003
2008 4 JUL Emerging and Reemerging Viruses in Dogs and Cats
2011 4 JUL Organ Failure in Critical Illness
Clinical Advances in Cognitive Psychotherapy Theory and Application R Leahy, E Dowd (Springer, 200
Nugent 5ed 2002 The Government and Politics in the EU part 1
2009 4 JUL New Concepts in Diagnostic Radiology
Language Processing in Discourse A Key to Felicitous Translation M Doherty (2002)
2008 4 JUL Emerging and Reemerging Viruses in Dogs and Cats
Rindel MODELLING IN AUDITORIUM ACOUSTICS Sevilla 2002 Rindel 8p

więcej podobnych podstron

2002 4 JUL Topics in feline surgery

Document Outline