Document Outline

Vet Clin Small Anim 38 (2008) xi–xii

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

deal with the clinical situations where down-regulation of the ocular immune
system fails and leads to ocular pathology.

One of my constant aggravations is the extent to which veterinary science

lags behind its medical and basic science cousins. All too often, the basic tools
that are used to investigate the immune system (be they monoclonal antibodies,
nucleic acid primers or microarrays) just are not available to veterinary scien-
tists to the same extent as medical researchers or rodent biologists. Therefore,
just to use one example, we have difficulties in evaluating the lymphocytic pop-
ulations infiltrating the lacrimal gland in a way that the clinicians for human
patients or the basic scientists working on rodent models simply do not.

For that reason I have asked the authors in this issue to seek links between

the veterinary, medical, and basic science perspectives of the diseases on which
they write. That is, of course, in addition to providing a thorough and up-
to-date approach to the clinical diagnosis and treatment of the diseases they cover.
I hope you will agree with me that all of them have excelled in this respect. We
have excellent reviews of conditions commonly seen in practice, such as keratitis
and uveitis, while contributions on more rarely seen conditions, such as extraoc-
ular myositis and sudden acquired retinal degeneration, provide completely new
information that will be of interest to those in referral ophthalmology clinics and
those in general practice. It is always difficult to juggle clinical work, teaching, and
research, so to be asked to provide a cutting edge summary of an area of oph-
thalmic disease in addition to all of that is a considerable burden. I thank all
the contributors most profoundly for their contributions to this issue. I hope
you will find that it aids you in your clinical work and stimulates you to see
the research that has been undertaken in ocular immunology and how much
more there remains to do!

David L. Williams, MA, VetMB, PhD, CertVOphthal, FRCVS

Associate Lecturer

Veterinary Ophthalmology

Department of Veterinary Medicine

University of Cambridge

Madingley Road

Cambridge CB3 OES, England, UK

Fellow

College Lecturer

Director of Studies

Veterinary Medicine and Pathology

St. John’s College

Cambridge CB2 1TP, England, UK

E-mail address:

doctordlwilliams@aol.com

xii

PREFACE

Immunology of the Ocular Surface

Brian C. Gilger, DVM

a,b,

Department of Clinical Sciences, College of Veterinary Medicine, North Carolina State

University, 4700 Hillsborough Street, Raleigh, NC 27606, USA

Center for Comparative and Translational Medicine, College of Veterinary Medicine, North

Carolina State University, 4700 Hillsborough Street, Raleigh, NC 27606, USA

he immunology of the ocular surface is a remarkable interplay between
the body’s innate and adaptive immune systems: a diverse array of
defense mechanisms that act to prevent microbial colonization. The innate

immune system is the first line of defense against invading organisms and
consists of anatomical barriers of the ocular surface (eg, mucins, epithelium)
and antimicrobial peptides present in the tear film. The innate immune
response is not antigen specific, but instead reacts well to a variety of organ-
isms. The adaptive immune system acts as a second line of defense and is
antigen specific, thus normally reacting only with the organism that induced
the response. It demonstrates immunologic memory and reacts more rapidly
on subsequent exposure to the same organism. In the unique ocular surface
microenvironment, these two arms of the immune response may have distinct
functions, but there is interplay between these systems to balance tolerance of
normal flora, the exposure of environmental irritants, the limited blood and
lymph supply of the cornea, and ocular tissues that are exquisitely sensitive
to overt inflammatory response. The immune system uses an effective but
complex series of interrelated processes to prevent microbial invasion: barrier
functions of the surface mucins and epithelium, tear film proteins, antigen
presenting cells, special antigen-recognition receptors called the Toll-like recep-
tors, and a reactive lacrimal gland that liberates white blood cells and IgA.
Breakdown of any of these functions may lead to infectious keratitis or,
conversely, immune-mediated disease.

OCULAR SURFACE INNATE IMMUNOLOGY
Epithelium and Mucins

The physical action of eyelid closure and tear washout is the first barrier to
ocular microorganism invasion. The ocular surface is also lined with nonkerati-
nized stratified epithelial cells, whose superficial layers are in contact with

*College of Veterinary Medicine, North Carolina State University, 4700 Hillsborough
Street, Raleigh, NC 27606. E-mail address: bgilger@ncsu.edu

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.11.004

Vet Clin Small Anim 38 (2008) 223–231

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

microorganisms (ie, the ocular flora). The epithelium itself acts as a barrier to
invasion of microorganisms because of the presence of epithelial intercellular
tight junctions and by the rapid renewal of epithelial cells with frequent
shedding of superficial layers of potentially invaded epithelium.

Mucins on the ocular surface serve several physiologic roles in the mainte-

nance of ocular health. Transmembrane mucins anchor the ocular tear film to
the external corneal epithelium, and thus act as a transition between the
hydrophobic epithelium and the aqueous tear film. Mucins also prevent bacterial
colonization and help eliminate surface foreign material

. There are several

types of mucins, and in human beings they commonly include the MUC1, -2,
-3, -4, and -5, MUC5AC, and MUC5B mucins. Mucin production and secretion
by goblet cells are stimulated by cytokines, such as interleukin (IL)-6 and gamma
interferon from dendritic cells in inflamed conjunctiva

. In addition, bacteria

may stimulate Toll-like receptors, which, through the nuclear factor-kappa B
(NF-jB) pathway, may induce mucin transcription

. Two secreted and two

membrane-bound mucins have been detected in the canine conjunctiva

Antimicrobial Peptides

Epithelial cells of the conjunctiva, lacrimal sac, and nasolacrimal surface consti-
tutively produce several antimicrobial peptides to assist in the prevention of
invasion of microorganisms. These peptides include lysozyme, lactoferrin, lip-
ocalin, angiogenin and secretory phospholipase A

. Lysozyme peptides

function by binding to the outer membrane of the bacteria, inserting full thick-
ness through cell membranes, and ultimately creating a pore that leads to cell
death

. Lactoferrin binds reversibly with iron, which is required for micro-

bial metabolism and growth, thus preventing its use by bacteria

. Lipocalin

(tear-specific prealbumin) scavenges bacterial products, and angiogenin has
general antimicrobial effects

. Inducible accessory pathways (APs), (ie, those

that become active during inflammation or infection), may be produced by res-
ident neutrophils (alpha defensins 1–3) or by epithelium (beta defensin 2)

The production of these APs are induced by interleukin-1b, tumor necrosis
factor a (TNF-a), and gram-negative bacteria lipopolysacharride (LPS), and
the mechanism likely through the Toll-like receptor complex (see below).
Induced APs also alert the adaptive immune system of a potential infection
by T-lymphocyte chemoattraction induced by alpha and beta defensins

OCULAR SURFACE ADAPTIVE IMMUNOLOGY

In systemic immunity, mucosal lymphocytes form a common mucosal immune
system called the mucosal associated lymphoid tissue (MALT). The lacrimal
gland, conjunctiva-associated lymphatic tissue (CALT) and the lacrimal
drainage associated lymphatic tissue (LDALT) also form a functional mucosal
immunologic group, a component of MALT, called the eye-associated lym-
phoid tissue (EALT)

. EALT is divided into two forms: an organized

lymphoid tissue, where lymphocytes are in follicles, and an extensive diffuse
lymphoid tissue. In lymphoid follicles (ie, organized lymphoid tissue), antigen

224

GILGER

is taken up from the environment by overlying follicle-associated epithelium
and presented to naive lymphocytes by antigen-presenting cells. This leads
to lymphocyte activation, proliferation, and eventual differentiation into
effector cells, (ie, B or T cells). These activated lymphocytes leave the follicle
and migrate to afferent lymphatic vessels, the bloodstream, and eventually to
effector organs: the lacrimal gland and conjunctiva (

The effector cells of mucosal immunity are comprised of the diffuse form of

lymphatic tissue that is interspersed along most mucus membranes and their
associated glands (including the lacrimal gland and conjunctiva). These consist
of intraepithelial lymphocytes and plasma cells. Lymphocytes are located in the
basal epithelial layer and lamina propria of the conjunctiva and lacrimal gland,
and are predominantly CD8þ suppressor/cytotoxic T cells and fewer CD4þ
T-helper cells

. The CD8þ cells likely promote an immunosuppressive envi-

ronment. Plasma cells in mucosal tissues contribute to secretory immunity, one
of the main immune effector mechanisms, by production of specific immuno-
globulins (eg, IgA), which pass through the overlying epithelium with help of
secretory component epithelial transporter molecules to form a surface layer
of IgA. IgA secreting plasma cells predominate in the lacrimal gland, and secre-
tory IgA are released into the tear film and ocular surface. The lacrimal gland
has a mixture of IgA-positive plasma cells and lymphocytes in the loose connec-
tive tissue between the secretory acini. Therefore, both the lacrimal gland and
conjunctiva are involved in the local production of secretory IgA

. Also, at

effector organs such as the lacrimal gland and conjunctiva, these primed
lymphocytes (CD4þ) monitor for antigens. When exposed again, clonal
expansion of the cells occurs with differentiation into B-lymphocytes and
plasma cells or T-lymphocytes.

Fig. 1. Eye associated lymphatic tissue. EALT is composed of diffuse lymphoid tissue of T lym-
phocutes and IgA-secreting plasma cells throughout all the organs of EALT. There are also lym-
phoid follicles in the conjunctiva and LDALT, where antigen is taken up from the environment by
overlying follicle-associated epithelium and presented to naive lymphocytes by antigen-present-
ing cells. This leads to lymphocyte activation, proliferation, and eventual differentiation into
effector cells (ie, B or T cells). These activated lymphocytes leave the follicle (green arrows)
and migrate to afferent lymphatic vessels, the bloodstream, and eventually to effector organs:
the lacrimal gland and conjunctiva (red arrows).

225

IMMUNOLOGY OF THE OCULAR SURFACE

Therefore, through the EALT functional mucosal immune unit, lymphoid

follicles detect antigen and produce effector cells in the lacrimal gland and con-
junctiva. Local detection of antigens then provides the ocular tissues with effec-
tor cells and the lacrimal gland with specific IgA producing plasma cells

Antigen Presenting Cells

Antigen presenting cells (APCs) are classified based on their level of constitui-
tive expression of major histocompatibility complex (MHC) class II antigen.
APCs are considered ‘‘professional’’ if they have high expression of MHC II
antigen (and costimulatory molecules). Examples of these include dendritic
cells, macrophages, B lymphocytes, and epithelial Langerhans cells

APCs are considered ‘‘nonprofessional’’ if they have low expression of
MHC class II antigen and include vascular endothelial cells and mesenchymal
cells. Nonprofessional APCs may be induced to express MHC II antigen with
inflammatory stimuli

. APCs, including dendritic cells and macrophages,

capture antigens and then migrate via draining lymph nodes to prime naive
T lymphocytes

The central cornea is mostly devoid of professional APCs (specifically den-

dritic cells and Langerhans cells), which may contribute to the immune privi-
leged (immunologic nonresponsive) nature of the cornea (see below).
However, with inflammatory processes (eg, infectious keratitis), APCs are
recruited from the limbus to the cornea. This recruitment of APCs is preceded
by infiltration of neutrophils and macrophages, over-expression of intercellular
adhesion molecule-1, and subsequent release of cytokines, such as IL-1 and
TNF-a

. Furthermore, with inflammation, there is an up-regulation in the

expression of MHC II and costimulatory molecules (ie, CD80 and CD86)
by resident corneal mesenchymal and endothelial cells. Rejection of corneal
transplants is mediated mainly by corneal dendritic cells that travel to the
regional lymph nodes to initiate the immune reaction

Mucosal Tolerance, Ignorance, and Privilege

The cornea is considered an immune privileged site

[6,12]

. There are multiple

active and passive mechanisms that contribute to this phenomenon, including
the lack of corneal vasculature and lymphatics, reduced MHC class II positive
APCs, and reduced corneal expression of MHC I

. Immune privilege is

provided by the processes of immune ignorance, immune tolerance, and the
development of an immunosuppressive microenvironment

. Ignorance is

when the presentation of an antigen to the immune system is impeded. This
occurs in the cornea, for example, by the strategic intracellular or basal epithe-
lial positioning of Toll-like receptors (TLRs), which respond to microorgan-
isms that have penetrated the surface epithelium and play a major role in the
initiation of the immune response

. Lack of TLRs on the epithelial surface,

however, minimizes the presentation of antigens to APCs and immune
recognition of surface antigens, and thus allows ignorance

Mucosal tolerance, although not well described in the ocular mucosa, is the

active induction of immune unresponsiveness of the systemic immunity to

226

GILGER

antigens (eg, normal ocular flora) present on mucosal surfaces. A similar phe-
nomenon, called the anterior chamber associated immune deviation, is known
to protect the anterior chamber from antigenic stimulation and uncontrolled
inflammatory response

. Ocular surface immunosuppressive microenviron-

ment likely occurs because of the presence of substances, such as transforming
growth factor beta, surface-bound Fas ligand, vascoactive intestinal peptide,
alpha melanocyte-stimulating hormone, lipocalin, and angiogenin

[6,12]

Ocular surface secretory IgA, which is constitutively deposited on the ocular
surface in the tear film, also contributes to the immune privilege of the ocular
surface because it prevents invasion of microorganisms

If tolerance or ignorance is impaired because of damage to the surface epithe-

lium or decrease in production of IgA, inflammatory cytokines are produced
and uncontrolled access to antigens by the immune system can result.

Toll-like Receptors and Ocular Surface Immunity

TLRs are part of the body’s innate immune system and constitute the first-line
of defense against many pathogens, including bacteria, fungi, viruses, protozoa,
and helminthes

. TLRs are type I transmembrane proteins with an extracel-

lular domain for ligand recognition and a cytoplasmic domain for intracellular
signaling. Eleven TLRs have been identified in mammalian cells and each TLR
type recognizes a specific pathogen-associated molecular pattern (PAMP)
(

). Different PAMPs stimulate different TLRs and induce distinct

patterns of cytokines appropriate for that specific pathogen (

). Thus,

TLRs function to both trigger and modulate the activation of the adaptive
immune response

. Bacteria or its components (ligands) that can stimulate

TLRs include lipoproteins (TLR-1 and -2), gram-positive peptidoglycan (TLR-2),
gram-negative lipopolysaccharide (TLR-4), bacterial flagellin (TLR-5), and
microbial unmethylated cytosine-phosphate-guanosine (CpG) DNA (TLR-9).
Viruses may stimulate TLR-3 (double stranded RNA) as well as TLR-4, -7, -8,
and -9. Fungi may stimulate TLR-2 and possibly others. Recently, Onchocerca spp
has been shown to activate TLR-4 (see

)

. TLRs have been identified

in the human eye, and the uvea, retina, choroid, and conjunctiva have been
shown to constituitively express TLR-4

. Equine specific TLR-2 and -4

mRNA sequences have been reported

[22,23]

and TLRs have been shown to

be highly conserved across species

. Preliminary studies in the author’s labo-

ratory have demonstrated expression of TLR-2 and -4 in mRNA in normal
equine uveal tissue

Response of the TLRs to PAMP is dependent on the specific receptor and

cell type but can include phagocytosis of microbes, production of reactive nitro-
gen and oxygen species, production of inflammatory cytokines, and expression
of costimulatory molecules

. TLRs are expressed on cells that are most

likely to initially encounter microbes, such as dendritic cells, neutrophils, and
macrophages. TLR-mediated activation of dendritic cells enhances their anti-
gen presenting capacity by the production of proinflammatory cytokines and
up-regulation of costimulatory molecules

. Ligand recognition by TLRs

227

IMMUNOLOGY OF THE OCULAR SURFACE

triggers activation of cytoplasmic signaling pathways, the activation of the tran-
scription factor NF-jB, and transcription proinflammatory molecules, such as
TNFa, IL-1, IL-2, chemokines, and adhesion molecules (see

TLRs likely to play a pivotal role in the ocular surface immune response,

especially the interplay between the innate and adaptive immune response
and in the immunologic tolerance of ocular surface and environmental
antigens. In fact, TLR regulation may be a main player in ocular tolerance
of surface antigens. In a recent study it was found that human corneal epithelial
cells expressed TLR-2 and -4 intracellularly, but not at the cell surface,
suggesting that this placement of TLR in the normal cells helps to create an
‘‘immuno-silent condition’’ to prevent unnecessary inflammatory response to
normal bacterial flora

. Another study found that the expression of TLR-5

is located at the basal and wing cell layers of human corneal epithelium,
but not at the apical layers, again suggesting a potential mechanism by which

Table 1
Mammalian TLRs and their known ligands

TLR

Exogenous ligands

Endogenous ligands

Located on plasma membrane
TLR-1 with TLR-2

(heterodimer)

Tri-acyl lipopeptides

TLR-2

Lipoproteins/lipopeptides

(various pathogens)

Peptidoglycan and

lipoteichoid acid (gram-
positive bacteria)

Zymosan (fungi)

TLR-2 with TLR-6

(heterodimer)

Di-acyl lipopeptides

TLR-4

LPS (gram-negative

bacteria)

Respiratory syncytial virus

coat protein

Heat-shock proteins,

inflammatory debris?

TLR-5

Flagellin (flagellated bacteria)

Located in cytoplasmic compartments of endoplasmic reticulum or endosome
TLR-3

Double stranded RNA

(viruses)

TLR-7

Imidazoquinolone antiviral

drug

RNA autoantigens

TLR-8

Single stranded RNA

(viruses)

TLR-9

Unmethylated CpG motifs

of bacterial DNA

DNA autoantigens
Neoepitopes

TLR-10

Unknown

Data from Chang JH, McCluskey PJ, Wakefield D. Toll-like receptors in ocular immunity and the immunopa-
thogenesis of inflammatory eye disease. Br J Ophthalmol 2006;90:103–8; and Marshak-Rothstein A, Rifkin
IR. Immunologically active autoantigens: the role of toll-like receptors in the development of chronic inflam-
matory disease. Annu Rev Immunol 2007;25:419–41.

228

GILGER

the corneal epithelium may remain inactive (ie, ignorant) in response to non-
pathogenic bacterial at the apical surface

. However, if the outer epithelial

layers are breached and the wing and basal layers are exposed, an organism
would then trigger a TLR mediated innate immune response

. In fact,

activation of TLR-4 was shown to be a critical step in the pathogenesis of
endotoxin (lipopolysaccharide) induced keratitis in mice

Chronic Ocular Disease, Autoimmunity, and Toll-like Receptors

TLRs help to enable the adaptive immune system to specifically recognize and
attack an invading organism. However, once the inflammatory process has
been initiated, there may be a general loss of tolerance, and in this environment
activated APCs can effectively present autoantigens to the adaptive immune
system

. When the adaptive immune system is up-regulated, cross reaction

Fig. 2. Toll-like receptors are type I transmembrane proteins with an extracellular domain
(yellow) for ligand recognition. Each TLR type recognizes a specific pathogen-associated
molecular pattern and stimulated TLRs activate distinct patterns of cytokines appropriate for
that specific pathogen. Ligand recognition by TLRs triggers activation of cytoplasmic signaling
pathways. Two prominent pathways are known to exist and include the MyD88 dependent (in
green) and the MyD88 independent pathways (in blue). The MyD88 pathways, IRAK and
TRAF6 activate the transcription factors NF-jB, JNK, and p38, resulting in the transcription
of proinflammatory molecules such as TNFa, IL-1, IL-2, chemokines, and adhesion molecules.
The MyD88 independent pathways activate the transcription factor TBK1, through TRIF/
TRAM, resulting in the transcription of interferons (INF). Up-regulation of proinflammatory
cytokines leads to tissue destruction, production of neoepitopes, heat-shock proteins, and other
inflammatory debris that can result in autoantigen recognition and development of autoimmune
disease (red pathway). (Data from Refs.

[16–19]

229

IMMUNOLOGY OF THE OCULAR SURFACE

between infectious agents and self-antigens can proceed to autoimmune disease.
Autoimmune disease develops with contributions of environmental and genetic
factors. Known examples of this antigen cross reactivity include leptospiral
antigens in equine cornea

and retina

[30,31]

When cross-reactivity occurs and self antigens are recognized and processed,

the specific subcellular site, or epitope, is bound and internalized, and proteins
are processed within cytoplasmic endosomes to ultimately present to T-cells.
Autoantigen cleavage can create neoepitopes, either from novel conformations
or realignment of protein sequences

. This process can lead to epitope

spreading, which is the development of autoantibodies against additional
components of the autoantigen

. Similarly, autoantibodies specific for one

antigen may bind to apoptotic bodies or other similar autoantigens produced
or recognized during inflammatory processes, leading to activated APC uptake
of this material and leading to reamplification of inflammatory response. This
intramolecular and intermolecular epitope spreading, respectively, may explain
the relapsing nature of inflammation in many autoimmune diseases, including
arthritis, recurrent uveitis, and immune-mediated keratitis

TLR-7 and -9 are located in cytoplasmic compartments of the endoplasmic

reticulum, endosome, and lysosomes

. TLR-7 and -9 are exposed to

internalized antigens, whether autoantigens or infectious organisms. TLR-7
and -9 likely contribute to the activation of autoreactive B cells and to the pro-
duction of autoantibodies, especially to DNA and RNA associated autoantigens

. Therefore, TLR-7 and -9 may be therapeutic targets for autoimmune

disease

. Other inflammatory by-products, such as heat-shock proteins,

may also activate T-cells through TLR-4

http://www.ncbi.nlm.nih.gov/Genbank/index.html

Acknowledgments

The author would like to thank Jacklyn Salmon for review of this article and
for all of her behind-the-scenes hard work.

References

[1] Berry M, Ellingham R. Mucins on the ocular surface. New York: Taylor & Francis; 2005.
[2] Basbaum C, Lemjabbar H, Longphre M, et al. Control of mucin transcription by diverse

injury-induced signaling pathways. Am J Respir Crit Care Med 1999;160:S44–8.

[3] Hicks SJ, Carrington SD, Kaswan RL, et al. Demonstration of discrete secreted and

membrane-bound ocular mucins in the dog. Exp Eye Res 1997;64:597–607.

[4] Paulsen FP, Varoga D, Steven P, et al. Antimicrobial peptides at the ocular surface. In:

Zierhut M, Stern M, Sullivan D, editors. Immunology of the lacrimal gland, tear film, and
ocular surface. New York: Taylor & Francis; 2005. p. 97–104.

[5] Davidson HJ, Blanchard GL, Montgomery PC. Comparisons of tear proteins in the cow,

horse, dog and rabbit. Adv Exp Med Biol 1994;350:331–4.

[6] Knop E, Knop N. Anatomy and immunology of the ocular surface. Chem Immunol Allergy

2007;92:36–49.

[7] Haynes RJ, Tighe PJ, Dua HS. Antimicrobial defensin peptides of the human ocular surface.

Br J Ophthalmol 1999;83:737–41.

[8] Knop E, Knop N. Lymphocyte homing in the mucosal immune system to the eye-associated

lymphoid tissue (EALT). In: Zierhut M, Stern M, Sullivan D, editors. Immunology of the lacri-
mal gland, tear film, and ocular surface. New York: Taylor & Francis; 2005. p. 13–42.

230

GILGER

[9] Knop E, Knop N. The role of eye-associated lymphoid tissue in corneal immune protection.

J Anat 2005;206:271–85.

[10] Dana MR. Corneal antigen-presenting cells: diversity, plasticity, and disguise: the Cogan

lecture. Invest Ophthalmol Vis Sci 2004;45:722–7, 721.

[11] Streilein J, Arancibia-Carcamo C, Osawa H. Antigen presentation and corneal transplanta-

tion. In: Zierhut M, Stern M, Sullivan D, editors. Immunology of the lacrimal gland, tear film,
and ocular surface. New York: Taylor & Francis; 2005. p. 127–40.

[12] Cursiefen C. Immune privilege and angiogenic privilege of the cornea. Chem Immunol

Allergy 2007;92:50–7.

[13] Ueta M, Nochi T, Jang MH, et al. Intracellularly expressed TLR2s and TLR4s contribution to

an immunosilent environment at the ocular mucosal epithelium. J Immunol 2004;173:
3337–47.

[14] Streilein JW. Anterior chamber associated immune deviation: the privilege of immunity in

the eye. Surv Ophthalmol 1990;35:67–73.

[15] Chaudhuri N, Dower SK, Whyte MK, et al. Toll-like receptors and chronic lung disease. Clin

Sci (Lond) 2005;109:125–33.

[16] Chang JH, McCluskey PJ, Wakefield D. Toll-like receptors in ocular immunity and the

immunopathogenesis of inflammatory eye disease. Br J Ophthalmol 2006;90:103–8.

[17] Marshak-Rothstein A, Rifkin IR. Immunologically active autoantigens: the role of toll-like

receptors in the development of chronic inflammatory disease. Annu Rev Immunol
2007;25:419–41.

[18] Yu FS, Hazlett LD. Toll-like receptors and the eye. Invest Ophthalmol Vis Sci 2006;47:

1255–63.

[19] Kumar A, Yu FS. Toll-like receptors and corneal innate immunity. Curr Mol Med 2006;6:

327–37.

[20] Hise AG, Gillette-Ferguson I, Pearlman E. Immunopathogenesis of Onchocerca volvulus

keratitis (river blindness): a novel role for TLR4 and endosymbiotic Wolbachia bacteria.
J Endotoxin Res 2003;9:390–4.

[21] Chang JH, McCluskey P, Wakefield D. Expression of toll-like receptor 4 and its associated

lipopolysaccharide receptor complex by resident antigen-presenting cells in the human
uvea. Invest Ophthalmol Vis Sci 2004;45:1871–8.

[22] Vandenplas ML, McNeill BW, Barton MH, et al. Cloning and sequencing of equine Toll-like

receptor 4 (TLR4). Available at:

[23] Vandenplas ML, McNeill BW, Cordonnier-Pratt M-M, et al. Cloning and sequencing of equine

toll-like receptor 2 (TLR2). Available at:

http://www.ncbi.nlm.nih.gov/Genbank/index.html

[24] Ignacio G, Nordone S, Howard KE, et al. Toll-like receptor expression in feline lymphoid

tissues. Vet Immunol Immunopathol 2005;106:229–37.

[25] Yi N, Salmon J, Gilger B. Role of toll-like receptors in equine uveitis, in press.
[26] Brentano F, Kyburz D, Schorr O, et al. The role of Toll-like receptor signalling in the patho-

genesis of arthritis. Cell Immunol 2005;233:90–6.

[27] Zhang J, Xu K, Ambati B, et al. Toll-like receptor 5-mediated corneal epithelial inflammatory

responses to Pseudomonas aeruginosa flagellin. Invest Ophthalmol Vis Sci 2003;44:
4247–54.

[28] Khatri S, Lass JH, Heinzel FP, et al. Regulation of endotoxin-induced keratitis by PECAM-1,

MIP-2, and toll-like receptor 4. Invest Ophthalmol Vis Sci 2002;43:2278–84.

[29] Parma AE, Sanz ME, Lucchesi PM, et al. Detection of an antigenic protein of Leptospira inter-

rogans which shares epitopes with the equine cornea and lens. Vet J 1997;153:75–9.

[30] Deeg CA, Hauck SM, Amann B, et al. Major retinal autoantigens remain stably expressed

during all stages of spontaneous uveitis. Mol Immunol 2007;44:3291–6.

[31] Deeg CA, Pompetzki D, Raith AJ, et al. Identification and functional validation of novel

autoantigens in equine uveitis. Mol Cell Proteomics 2006;5:1462–70.

[32] Deeg CA, Amann B, Raith AJ, et al. Inter- and intramolecular epitope spreading in equine

recurrent uveitis. Invest Ophthalmol Vis Sci 2006;47:652–6.

231

IMMUNOLOGY OF THE OCULAR SURFACE

Canine Conjunctivitis and Blepharitis

Teresa Pen˜a, DVM, PhD*, Marta Leiva, DVM

Departament de Medicina i Cirurgia Animals, Facultat de Veterina`ria, Universitat Auto`noma
de Barcelona, 08193 Bellaterra, Barcelona, Spain

he eyelids and conjunctiva are immunologically active structures with an
extensive presence of blood vessels, lymphatics, and immune cells. Sev-
eral immune-mediated phenomena can involve these structures either in

isolation or in association with systemic clinical features, but, fortunately, they
are rare diseases

[1–5]

. Immune-mediated blepharoconjunctival diseases are

divided into two main categories: primary autoimmune disease in which the
disease results from an attack against self-antigens and secondary immune-
mediated disease in which the disorder results from exogenous material induc-
ing the autoimmune disease. Such causes of secondary immune-mediated
disease include infectious agents and drugs

Although the pathogenesis of many ocular autoimmune diseases is known,

most of the eye allergic disorders have not been well characterized in veterinary
ophthalmology. Although conjunctival disorders used to be diagnosed and
treated by ophthalmologists, when the eyelids are affected, the patient can be
seen by either an ophthalmologist or a dermatologist. It is always necessary
to rule out systemic involvement to coordinate therapy so as to treat the under-
lying disease rather than only ocular signs. Although clinical signs are different
for every disorder, immune-mediated eyelid and conjunctival diseases share
some characteristics, including itching, redness, and ocular discharge.

In human medicine, chronic presentation of these conditions can induce ab-

normal cicatrization leading to significant mechanical alterations as a result of
fibrosis. In some chronic cases, the eyelid may need surgery to restore its phys-
iologic function. In veterinary ophthalmology, there are no reports on chronic
immune-mediated conjunctival and eyelid lesions without systemic signs. This
article reviews the most important autoimmune and immune-mediated eyelid
and conjunctival disorders in dogs.

AUTOIMMUNE EYELID AND CONJUNCTIVAL DISORDERS
Medial Canthal Ulcerative Blepharitis

This disease represents a juxtapalpebral disorder, usually affecting the medial
canthus. Breeds most often affected include the German shepherd, long-haired

*Corresponding author. E-mail address: teresa.pena@uab.cat (M.T. Pen˜a).

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.12.001

Vet Clin Small Anim 38 (2008) 233–249

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

dachshund, toy and miniature poodle, and others (

[7,8]

. In the

German shepherd, the medial canthal blepharitis can be associated with pannus
and immune-mediated plasma cell infiltration of the nictitating membrane. In
the long-haired dachshund, medial canthal blepharitis may appear concurrently
with superficial punctate keratitis. Even if it has not been described as a sepa-
rated disease, the lateral marginal canthus can also be affected (

B).

The condition is usually bilateral. Biopsy reveals lymphocytic and plasma

cell infiltrates; sebaceous glandular hyperplasia may be also present. Antibodies
against epithelial cells have been demonstrated in selected cases and may sug-
gest a relationship to pemphigus. The condition usually responds to topical
ophthalmic antibiotics and corticosteroids

Vogt-Koyanagi-Harada-Like Syndrome

The Vogt-Koyanagi-Harada (VKH) syndrome in humans is an immune-
mediated disease in which melanocytes are targeted

. The factor or factors

responsible for the development of cellular hypersensitivity to melanin have
not been elucidated, although specific circulating anti-melanin autoantibodies
and melanin-sensitized lymphocytes have been reported in affected patients

. The possibility that VKH syndrome has an autoimmune pathogenesis

is supported by the statistically significant presence of human leukocyte antigen
DR4 (HLA-DR4 or human major histocompatibility complex [MHC] DR4) in
affected individuals. This antigen has been commonly associated with other
autoimmune diseases

. Darkly pigmented human races are predisposed,

and there may be a genetic component of the disease as well given the high
incidence of the condition among the Japanese. The clinical presentation can
include anterior uveitis, chorioretinitis, exudative retinal detachment, poliosis,
vitiligo, dysacusis, and meningitis. A similar disease syndrome has been
reported in the dog, although meningitis is rarely reported

. In dogs the dis-

ease has been termed VKH-like syndrome or uveodermatologic syndrome. It has been
described in the Akita

, Siberian husky

, golden retriever

beagle

, chow chow

, Old English sheepdog

, Saint Bernard

Fig. 1. (A) Medial canthal ulcerative blepharitis in a 3-year-old Yorkshire terrier. (B) Lateral
canthal affection in a German shepherd with medial canthal ulcerative blepharitis.

234

PEN

˜ A & LEIVA

Australian shepherd

, dachshund

, Brazilian fila dog

, Shetland

sheepdog

, Irish setter

, and Samoyed

. The breed predisposition

in dogs could be also related to the presence of MHC anomalies as in humans
(HLA-DR4). To the authors’; knowledge, there are no studies showing a rela-
tion between MHC anomalies and VKH-like syndrome in dogs. Dogs are typ-
ically affected in adulthood, and ocular lesions usually precede the dermatologic
lesions, which are located in the mucocutaneous junctions. The ocular clinical
signs in dogs are similar to those in humans. Apart from the intraocular signs,
the eyelids are also affected, showing ulceration, hypopigmentation, and crust-
ing (

). Often, loss of pigmentation of the eyelids and nose is the primary

clinical sign recognized by the owner and is the basis for the initial presentation.
A recent study examining ocular and dermatologic tissue from two affected dogs
suggested that skin lesions are the result of a Th1-mediated inflammatory
response, whereas ocular lesions are the result of a Th2-mediated one

Currently, there is no specific diagnostic test for uveodermatologic syndrome.

The diagnosis is made by means of clinical signs and histopathologic examina-
tion of skin biopsies

. Histology of the eyelid skin reveals lichenoid interface

dermatitis with infiltration by histiocytes, lymphocytes, plasma cells, and multi-
nucleated giant cells

. Initial therapy involves immunosuppressive doses of

oral prednisone (1 to 2 mg/kg/d), possibly in combination with azathioprine
(beginning with 2 mg/kg/d and tapering gradually to 0.5 mg/kg/d) or cyclophos-
phamide (1–2 mg/kg/d), and topical eye treatment

consisting of corticoste-

roids and immunosuppressive drugs. The long-term prognosis is poor.

Pemphigus Complex

The pemphigus complex is a group of uncommon autoimmune diseases de-
scribed in dogs that is comparable to human disease. In humans, there are at
least eight varieties of pemphigus

; in dogs, there are five described varieties

(vulgaris, foliaceus, erythematosus, vegetans, and bullous)

Fig. 2. Vogt-Koyanagi-Harada syndrome in a Saint Bernard 4-year-old male with periocular
alopecia, ulcers, conjunctival hyperemia, diffuse corneal edema, and mucopurulent discharge.

235

CANINE CONJUNCTIVITIS AND BLEPHARITIS

In humans, the pemphigus complex is characterized histologically by intrae-

pithelial acantholysis leading to vesicle formation and immunologically by the
presence of autoantibodies to different components of the keratinocyte desmo-
some found in the skin and circulating in the serum

[26,27]

. In dogs, only pem-

phigus vulgaris causes an intraepidermal vesicle or bulla. The other forms of
pemphigus are typically associated with intraepidermal pustules, a major dis-
tinction between human and canine disease

In dogs, facial lesions usually involve mucocutaneous junctions and are char-

acterized by pustules or vesicles that eventually rupture, leaving erosions and
ulcers, crusting, scaling, and hypopigmentation. In pemphigus foliaceus, vulga-
ris, and erythematosus, the facial lesions usually involve the eyelids (

The dermatologic clinical signs are due to a type II hypersensitivity reaction

The pemphigus complex is uncommon in dogs, accounting for about 0.6%

to 1% of all canine skin disorders diagnosed at referral small animal clinics

[29,30]

. The most important diagnostic aspects are the history, physical exam-

ination, and histopathologic findings (

)

. Detection of pemphigus

antibody by direct immunofluorescence or immunohistochemical testing may
also be helpful. Owing to costs, technical problems, and relatively poor diag-
nostic sensitivity and specificity, these tests are not routinely recommended.

The prognosis for canine pemphigus varies with the form and severity of the

disease

[29,32,33]

. On the basis of the small number of cases documented in the

veterinary literature, pemphigus vulgaris appears to be a severe disease that is
often fatal, and, even with treatment, many dogs fail to respond and are eutha-
nized. Pemphigus foliaceus is less severe but, without treatment, may be fatal.
In contrast, pemphigus erythematosus is usually a benign disorder that rarely
produces systemic signs and readily responds to treatment.

The treatment of these diseases requires long-term topical and systemic cor-

ticosteroids (prednisone, 1 to 2 mg/kg/d), with additional immune suppression
thorough the use of cyclophosphamide (1–2 mg/kg/d) or azathioprine (1–2 mg/
kg/d) for refractory cases

[7,34]

. Side effects of these drugs are common and

vary from mild to severe, and close physical and hematologic monitoring of

Fig. 3. Pemphigus foliaceus. Note the severe mucopurulent discharge and eyelid ulceration.
(Courtesy of Mar Bardagi, Barcelona, Spain.)

236

PEN

˜ A & LEIVA

the patient is critical. Some animals may require medication for life; therefore,
the therapeutic regimen must be individualized for each patient, and owner
education is essential

[3,25]

. Occasionally, the cicatricial entropion from these

diseases may require corrective blepharoplasty

[7,35]

Lupus Erythematosus

Lupus erythematosus is a term that encompasses a group of diseases that have differ-
ent clinical syndromes but share a similar underlying autoimmune process

. The

terminology and classification system used in humans, described by Sontheimer, is

Table 1
Histopathologic findings in autoimmune disease affected eyelids

Disease

Histopathologic findings

Pemphigus foliaceus

Intragranular and subcorneal acantholysis is

present with resultant cleft and vesicle or pustule
formation. Neutrophils or eosinophils may
predominate within the vesicle or pustule.
Eosinophilic exocytosis and microabscess
formation occur within the epidermis or
follicular outer root sheath. Acantholytic,
dyskeratotic granular epidermal cells are
found at the surface of erosions.

Pemphigus vulgaris

Suprabasilar acantholysis is present with resultant

cleft and vesicle formation. Basal epidermal
cells remain attached to the basement
membrane zone like a row of tombstones.
The inflammatory reaction may be scant
and perivascular or prominent and interstitial
to lichenoid.

Pemphigus erythematosus

Condition is identical to pemphigus foliaceus

except for the fact that it often has a lichenoid
cellular infiltrate of mononuclear cells, plasma
cells, and neutrophils or eosinophils or both.

Canine discoid lupus erythematosus

Interface dermatitis is present. Findings include

focal hydropic degeneration of basal
epidermal cells, pigmentary incontinence,
focal thickening of the basement membrane
zone, apoptotic keratinocytes, and marked
accumulations of mononuclear cells and
plasma cells around dermal vessels. Dermal
mucinosis occurs of variable degrees.

Canine systemic lupus erythematosus

The dermatohistopathologic changes vary with

the type of gross morphologic lesions and
may be nondiagnostic. Interface dermatitis occurs
involving hair follicle outer root sheaths. Apoptosis
of basal and suprabasal cells may occur, and,
occasionally, these apoptotic cells are associated
with lymphocytic satellitosis. Findings
include subepidermal vacuolar alteration, focal

thickening of the basement membrane zone,

and dermal mucinosis.

237

CANINE CONJUNCTIVITIS AND BLEPHARITIS

currently being used in veterinary medicine. The basis of Sontheimer’s system is that
lupus erythematosus may be systemic or cutaneous (discoid)

Lupus erythematosus is an uncommon autoimmune disorder of dogs, cats,

and humans that has polyclonal lymphocytic involvement

. The exact

etiology is unknown, but, in humans, all forms are characterized by a variety
of autoantibodies to nuclear antigens with or without immune complex depo-
sition. It is considered a type III–mediated (antibody-antigen complex-related)
hypersensitivity reaction.

Canine discoid lupus erythematosus is a relatively benign cutaneous disease

with no systemic involvement

[38,39]

. A relationship or progression to canine

systemic lupus erythematosus has not been reported. Although there is a clear
breed predisposition in German shepherd dogs, it has been demonstrated that
sun exposure aggravates the disease in about 50% of cases, suggesting that pho-
tosensitivity has a role in pathogenesis. The disease is associated with facial
dermatitis consisting of crusts, depigmentation, erosions, and ulcers, which pre-
dominantly affect the nasal planum and muzzle; eyelid and oral lesions are also
described

. The diagnosis is based on history, physical examination, and

skin biopsy (

). Anti-nuclear antibodies (ANA) test results are not reli-

able

. The prognosis for canine discoid lupus erythematosus is good

[29,39]

. The disease can be managed by avoiding exposure to intense sunlight

and by using topical immunosuppressive drugs (glucocorticoids or 0.2% to 1%
cyclosporine A)

. In refractory cases, systemic glucocorticoids (2.2 to 4.4

mg/kg of prednisone or prednisolone given orally every 24 hours) may be
needed. Therapy will probably need to be continued for life.

The most common presentation of canine systemic lupus erythematosus is

fever (constant or irregularly cyclic) with polyarthritis, proteinuria, and skin dis-
ease, present in greater than 50% of cases

[36,37]

. Other relatively common man-

ifestations include anemia, leucopenia, thrombocytopenia, proteinuria,
peripheral lymphadenopathy, splenomegaly, and oral ulcers. Cutaneous mani-
festations of canine systemic lupus erythematosus are extremely diverse and sim-
ilar to those of canine discoid lupus erythematosus. The disease is so variable in
its clinicopathologic presentation that any dogmatic diagnostic categorization is
impossible. The diagnosis is based mainly on ANA tests and skin biopsy (

Table

). The ANA test is currently considered the most sensitive serologic test, but its

specificity is low

[35,41]

. The prognosis in systemic lupus erythematosus is gen-

erally unpredictable and depends on the organs involved. In general, the earlier
the diagnosis is made, the better the prognosis

[38,42,43]

. Therapy for canine sys-

temic lupus erythematosus must be individualized. The initial agent of choice is
probably large doses of systemic glucocorticoids. When systemic glucocorticoids
are unsatisfactory, other immunomodulating drugs may be useful

[38,42,43]

IMMUNE-MEDIATED EYELID AND CONJUNCTIVAL DISORDERS
Canine Juvenile Cellulitis

Canine juvenile cellulitis is a well-recognized lymphocutaneous disease that is
commonly seen in puppies less than 8 months of age

[6,44–48]

; however, adult

238

PEN

˜ A & LEIVA

dogs may become affected by this condition

[49,50]

. Predisposed breeds

include the dachshund, golden retriever, Labrador retriever, Gordon setter,
and Lhasa apso

[6,44]

. Canine juvenile cellulitis is an uncommon granuloma-

tous and pustular disorder of the face, pinnae, and submandibular lymph
nodes. Normally, an acute swollen face with particular involvement of the eye-
lids, lips, and muzzle is accompanied by submandibular lymphadenopathy
(

). Within 24 to 48 hours, papules and pustules develop around the

lips, muzzle, chin, bridge of the nose, and periocular area. Occasionally, lesions
may also appear on the feet, abdomen, thorax, vulva, prepuce, or anus. Lesions
typically fistulate, drain, and crust. Affected eyelids are often painful but not
pruritic. Approximately 50% of affected puppies are lethargic and depressed.
Fever, anorexia, and sterile suppurative arthritis manifesting as joint pain are
inconsistent findings. Leukocytosis with neutrophilia and normocytic, normo-
chromic anemia may also be seen

Canine juvenile cellulitis may be diagnosed primarily on a clinical basis,

although a definitive diagnosis requires cytologic and histopathologic evalua-
tions

. Even if the patient presents with only blepharitis, the diagnosis

should be suspected because of the age of the dog and the bilateral eyelid
involvement

. Cytologic examination of eyelid papulopustular lesions re-

veals pyogranulomatous inflammation with no microorganisms, and carefully
performed cultures are negative. Biopsies of early eyelid lesions reveal multiple
discrete or confluent granulomas and pyogranulomas consisting of clusters of
large epitheloid macrophages with variably sized cores of neutrophils

The cause of canine juvenile cellulitis is unknown, but a bacterial hypersen-

sitivity has been postulated to explain the response to corticosteroids and the

Fig. 4. Dermatologic signs in a 4-month-old English cocker spaniel affected by juvenile cellu-
litis. The animal has been treated with topical diluted clorhexidine. (Courtesy of Mar Bardagi,
Barcelona, Spain.)

239

CANINE CONJUNCTIVITIS AND BLEPHARITIS

explosive course of the disease

[6,46]

. Early and systemic aggressive therapy is

indicated; otherwise, eyelid scarring may be severe. Immunosuppressive doses
of systemic corticosteroids, tapered following 3 to 4 weeks after resolution of
clinical signs, are recommended

[6,45]

. If cytologic or clinical evidence of sec-

ondary bacterial infection exists, systemic bactericidal antibiotics such as ceph-
alexin, cefadroxil, and amoxicillin clavulanate should be prescribed

Nursing care consisting of gentle cleansing or soaking of the skin lesions
may also be attempted. With appropriate treatment, the prognosis is excellent.

Acute Allergic Blepharitis and Conjunctivitis

Acute allergic blepharitis and conjunctivitis can occur at any age and in atopic
or non-atopic dogs. It is considered a hypersensitivity reaction in which aller-
gens (eg, plant pollen, topical drops, insect bites) are inoculated into the eyelid
or the conjunctival surface. It causes intense itching, eye redness, and dramatic
and immediate swelling of the eyelids and conjunctiva which may be so severe
that the eye closes. Several breeds are more affected; the West Highland white
terrier demonstrates an especially high incidence

. It is of relevant impor-

tance to compile a completed history taking in consideration environmental
aspects such as recent exposition to cut grass, plant pollen allergens in the sur-
roundings, cleaning products, and so on. This condition is self-limiting and nor-
mally requires no treatment, although an intensive wash out of the conjunctival
fornix is recommended

[51,52]

. If the allergen is identified, it may be needed to

be avoided. If the inflammation is severe, it is important to protect the eye from
self-trauma and to give some topical or systemic corticosteroids depending on
the severity of the clinical signs. In humans, if the condition becomes recurrent,
it may be helpful to protect patients from allergic challenges with mast cell
inhibitors such as sodium cromoglycate or nedocromil sodium

. The effec-

tiveness of these medications has not been proven in dogs, although anecdotal
evidence suggests that antihistamines such as levocabastine and sodium chro-
moglycate can be valuable when given topically.

Contact hypersensitivity has also been described in the eyelids and conjunc-

tiva

[7,52]

. This condition is related to the use of topical ophthalmic medication.

The reaction can be induced by the active ingredient or by the excipients. Some
reported drugs are benzalkonium chloride, neomycin, pilocarpine

, thimer-

osal

, 2% dorzolamide

, and prostaglandin analogues

. The allergic

reaction can be acute, with immediate chemosis and discomfort, or more
chronic, with conjunctival redness, serous discharge, and swollen medial can-
thus blepharitis

. In chronic cases, affected animals have a history of nonre-

sponsive conjunctivitis. Diagnosis and treatment include cessation of all topical
medication for a week

. It is easier to diagnose and identify the allergen in

acute cases. In the authors’ opinion, some cases could be misinterpreted as
a treatment failure, more so when the allergen is the active ingredient. For
example, an antibiotic could be replaced by another on the basis of a lack of sen-
sibility rather than contact hypersensitivity. Reactions to excipients can be diffi-
cult to evidence. In human ophthalmology, allergic reactions to excipients are

240

PEN

˜ A & LEIVA

a frequent disorder; therefore, excipient-free ophthalmic drugs are increasing in
the market

Another type of acute blepharoconjunctival allergy involves the urticarial

lesions of acute angioneurotic edema. Dogs may develop urticarial lesions of
the head in the skin involving the ears, muzzle, and periorbital areas

[6,51]

Lesions are characterized by the acute onset of skin edema, chemosis, and
edema of the subcutaneous connective tissue of these areas. Swelling around
the eyes may be severe enough to close the palpebral fissures and prevent
the animal from seeing. The cause of urticarial eye disease is usually associated
with the stings of insects, with ingestion of spoiled protein material in foods,
and with the administration of systemic drugs. Treatment of angioneurotic
edema depends on the severity of the clinical signs and includes the following:

Identify and remove the irritating substance if possible. Wash the eye to
eliminate any chemical residues. Stop any ongoing systemic medication.

Administer high doses of a rapidly acting corticosteroid such as hydrocorti-
sone hemisuccinate intravenously (10 to 20 mg/kg). Administer adrenalin
only if angioneurotic edema is severe and swelling of the face and neck
may interfere with normal breathing (1:10,000 epinephrine, 0.5 to 1.0 mL
intravenously) [51].

Necrotizing Marginal Blepharitis

Marginal blepharitis or meibomitis is the term used to describe inflammation of
the lids that involves the meibomian glands. Necrotizing marginal blepharitis is
a meibomitis secondary to the necrotizing direct effect of Staphylococcus toxin

, although an immune-mediated response to Staphylococcus toxin should

not be excluded.

Staphylococcus spp are distributed everywhere in the nature; therefore, expo-

sure of the eye is unavoidable. With such widespread occurrence of Staphylococ-
cus spp, animals would normally carry these organisms in their eyes as potential
commensals or pathogens. In affected animals, the lid margins become swollen,
red, inflamed, and pruritic

[7,59,60]

. In severe cases, crusts of fibrin may

develop on the lid margins, and tear film abnormalities can be present.

The pathogenic mechanism is related to the bacterial presence and the im-

mune-mediated reaction induced by the toxin. For that reason, a combined
treatment based on topical and systemic antibiotics and corticosteroids is indi-
cated in the majority of cases. Autogenous vaccine can be effective in chronic
and seemingly resistant staphylococcal infections

. The prognosis is good if

the disease is diagnosed and treated early.

CHRONIC ALLERGIC BLEPHARITIS AND CONJUNCTIVITIS
Canine Atopic Disease

The eyelids and conjunctiva are exposed structures that come into contact with
a huge number and variety of airborne particles. Chronic ocular allergic dis-
eases also occur in humans and can be concomitant with systemic atopic clin-
ical signs

. The term atopy was introduced to describe the ability to produce

241

CANINE CONJUNCTIVITIS AND BLEPHARITIS

a hypersensitivity reaction against common environmental allergens, a response
that has been identified to be mediated by IgE. An exaggerated IgE response
produces tissue damage as a type I hypersensitivity reaction

. The patho-

physiology of the disease is still under study in humans. A cascade starts by
the activation of mast cells, which release histamine, tryptase, or leukotriene
C4 in tears, mediators that promote eosinophil adhesion and degranulation.
Mast cell proteases also activate the matrix metalloproteinases MMP-2 and
MMP-9

. Several cytokines are involved in the recruitment and activation

of inflammatory cells, many of them produced by conjunctival fibroblasts.
Although cytokine levels in tears can be used to diagnose ocular allergic dis-
eases in humans, they have not been characterized in dogs

Chronic atopic blepharitis and conjunctivitis is characterized by redness,

blepharospasm, erythema, and crusting extending from the eyelid margin
upward for 8 to 10 mm accompanied by excoriation and ulceration. In chronic
situations with persistent eye rubbing it may lead to secondary bacterial mar-
ginal blepharitis, corneal involvement, and secondary visual impairment.
Chronic meibomian gland inflammation can induce production of a more polar
lipid secretion and can induce surface corneal disease due to early preocular
tear film evaporation (

)

. Atopic keratoconjunctivitis is the human

counterpart of the disease. The presence of conjunctival giant papilla is one
of the most common clinical signs. Matrix metalloproteases (MMP-2 and
MMP-9) have been proposed to be the vehicle for the corneal involvement

The diagnosis of atopic eye disorders is based on history, physical examina-

tion, and the use of intradermal and ocular allergy testing. Physical examina-
tion is important to evaluate other dermatologic conditions and to rule out
other causes of pruritus and periocular excoriation. A commercially available
ocular allergy test in humans is the rapid assay for total IgE determination in
tears (Lacrytest, ADIATEC S.A. Diagnostic and Biotechnologies, Nantes,
France)

. There are no reports of the use of this test in dogs.

Fig. 5. Atopic blepharoconjunctivitis in a poodle. Note the alopecic, erythematosus, and red
eyelids.

242

PEN

˜ A & LEIVA

Treatment of chronic atopic blepharitis and conjunctivitis follows a regimen

similar to that described for the skin and involves avoidance of the offending
allergen, pharmacologic modification of the clinical signs, and hyposensitization
of the offending allergens. The most important diagnostic problem is that the
offending allergen can not be identified as easily as in skin diseases. In cases
in which important dermatologic problems are associated, allergy skin tests
can be performed to identify the antigen. Once identified, if the allergen cannot
be removed, the animal may need a change in environment. For the control of
clinical signs, a variety of pharmacologic and supportive measures are available.
Cold compresses can bring relief to the ocular pruritus. In general, all ocular
medications when refrigerated provide additional subjective relief when applied
immediately in a cold state. A deficient tear film may be rectified by giving tear
supplement drops. In more severe cases, topical or systemic glucocorticoids and
antibiotics may be needed to treat lid margin blepharitis, as well as an Elizabe-
than collar to avoid self-trauma. In the authors’ opinion, dexamethasone and
prednisone ointments are the most helpful topical glucocorticoids in these cases.
Systemic and topical antihistamine drugs have been used with benefit in mild
human ocular allergies

. Although topical application has been recommen-

ded in veterinary ophthalmology, there are no reports on its effectiveness. New
therapeutic modalities in humans such as chemokine antagonists have been
proposed to treat chronic allergic disease. One relevant and attractive approach
is to employ CCR3 (chemokine receptor type 3) antagonism. Conjunctival
mast cells express CCR3, which is essential for their maintenance and differen-
tiation. The inhibition of CCR3 has been proven to diminish the immune-
mediated ocular response

. Several pharmaceutical approaches have been

described, including amino terminus modification of natural chemokines, a de-
velopment of peptide-based and nonpeptide-based antagonists, and monoclonal
antibody generation. Nevertheless, these drugs have been described to be spe-
cies specific and to have tissue-specific effects. It would be necessary to improve
research before using these drugs in veterinary ophthalmology

Canine Food Hypersensitivity

Canine food hypersensitivity is a nonseasonal pruritic skin disorder of dogs
associated with the ingestion of allergens found in the diet. Presumably, it is
a hypersensitivity reaction to an antigenic ingredient. Although the pathome-
chanism of food hypersensitivity is unclear, type I hypersensitivity reactions
are well documented as the most common type of hypersensitivity reactions
in humans. Why the skin is a frequent target of food-induced hypersensitivity
is not well known, although it has been recognized in humans that cutaneous
lymphocyte antigen is induced on T cells when cutaneous disease is present

Pruritus with or without a primary eruption is the only consistent finding.

No classic set of cutaneous signs is pathognomonic for food hypersensitivity
in the dog. A variety of primary and secondary skin lesions are noted and
can affect the eyelids. These lesions include papules, plaques, pustules, wheals,

243

CANINE CONJUNCTIVITIS AND BLEPHARITIS

angioedema, erythema, ulcers, excoriation, lichenification, pigment changes,
alopecia, scales, crusts, and moist erosions (

Currently, the definitive diagnosis of food hypersensitivity in dogs is attainable

only on the basis of elimination diets and provocative exposure testing. Routine
laboratory tests are not useful in diagnosing canine food hypersensitivity

Treatment consists of allergen detection and elimination. In the meantime,

eye topical treatment is needed to reduce ocular discomfort and pruritus that
can induce secondary corneal problems. The topical treatment is individual,
depending on ocular discharge, secondary bacterial infection, and conjunctival
involvement. The most used therapy is a topical combination of antibiotics and
glucocorticoids. The ocular signs will disappear as soon as the allergen is de-
tected and eliminated.

Allergic Conjunctivitis without Systemic Clinical Signs

Seasonal and perennial allergic conjunctivitis have been well reported in human
ophthalmology as non–sight-threatening ocular allergies

. Several cytokines

have been found in the tears of patients sustaining allergic conjunctivitis. In
patients who have seasonal and vernal keratoconjunctivitis, the most frequently
found are interleukin-4 (IL-4), IL-10, and interferon-c (IFN-c)

. In most

reports, allergic conjunctivitis in dogs is associated with allergic systemic signs.
Nevertheless, conjunctival signs can also be present alone. The affected animals
present with chronic epiphora and ocular redness without any other ophthal-
mic clinical signs. Diagnosis can be challenging. Conjunctival biopsy shows
a mild lymphoplasmacytic infiltrate with variable amounts of eosinophils, vas-
cular congestion, and dilation. IgE determination in tears is actually used in hu-
man medicine to diagnose ocular allergies

. There are no reports of the use

of this test in dogs.

The response to topical corticosteroids or non-steroidal anti-inflammatory

drugs is poor. Other therapeutic options include vasoconstrictors such as

Fig. 6. Food hypersensitivity in a golden retriever. Note the periocular alopecia and hyper-
pigmentation. (Courtesy of Mar Bardagi, Barcelona, Spain.)

244

PEN

˜ A & LEIVA

phenylephrine or tetrahydrozoline. These drugs are sympathomimetic agents
that decrease vascular congestion and eyelid edema via a-adrenoceptor stimu-
lation but have no effect in diminishing the allergic response. Topical antihis-
tamines are useful in human ophthalmology, but there are no studies
showing their efficacy in veterinary ophthalmology

Follicular Conjunctivitis

Follicular conjunctivitis consists of a macroscopic proliferation of the conjunc-
tival-associated lymphoid tissue of the palpebral or bulbar conjunctiva. Follicles
appear primarily on the bulbar surface of the nictitating membrane, being
outnumbered and larger than those normally seen. In more severe cases, the
follicles can involve palpebral and bulbar conjunctiva. Concurrent mucous
or serous ocular discharge and redness are present (

. This condition

occurs most frequently in dogs younger than 18 months of age, although older
animals can also be affected. It develops secondary to chronic antigenic stimu-
lation. Vernal keratoconjunctivitis is a clinically similar entity present in
humans

. Although there is a histologic and pathogenetic difference

between follicular conjunctivitis and vernal keratoconjunctivitis, the human
disease is also more frequently diagnosed in children and usually resolves with-
out treatment. Clinical signs include the presence of conjunctival giant papillae;
in severe cases, corneal inflammation can also be present

. IFN-c levels in

tears have been correlated with disease severity and have been suggested to
have a role in the inflammatory phase of chronic eye allergy

The diagnosis is made by clinical signs and conjunctival cytology. Cytologic

results of conjunctival scraping demonstrate the presence of lymphocytes.
Most cases respond to treatment with saline irrigation and topical corticosteroids.
Some authorities describe that nonresponsive cases can be treated by mechani-
cally debriding the follicles. The debridement should be performed after

Fig. 7. Follicular conjunctivitis involving the palpebral conjunctiva in an 8-month-old dog.

245

CANINE CONJUNCTIVITIS AND BLEPHARITIS

instillation of ophthalmic topical anesthetic with a dry cotton-tipped applicator.
In the authors’; experience, follicle debridement can sometimes worsen the situ-
ation, increasing local inflammation and predisposing to chronic conjunctivitis.

The conjunctiva is the most immunologically active tissue of the external eye

and undergoes lymphoid hyperplasia in response to stimuli. The conjunctival-
associated lymphoid tissue is mainly located in the conjunctival superficial
layers and in normal situations is CD8 dominated. The purpose of conjuncti-
val-associated lymphoid tissue is to receive antigen and to present it to the
circulating mononuclear cells, acting as the first line of the ocular defense sys-
tem

. In the authors’ opinion, the lymphoid tissue is critical to conjunctival

immunity. It is important to maintain its integrity as much as possible, trying
not to eliminate the conjunctival follicles. Dogs with chronic nonresponsive
conjunctivitis should be reevaluated for previous follicle debridement.

Acknowledgments

The authors thank the Dermatology Service of the Veterinary Teaching Hos-
pital of the University Autonomous of Barcelona, especially Drs. Mar Bardagı´
and Giordana Zanna for their photographic support. They also thank Drs.
Carolina Naranjo and Xavier Roura for reviewing this article.

References

[1] Halliwell RE. Skin diseases associated with autoimmunity. Part I. The bullous autoimmune

skin diseases. Compendium on Continuing Education for the Practicing Veterinarian
1980;2:911–5.

[2] Marsella R. Canine pemphigus complex. Part I. Compendium on Continuing Education for

the Practicing Veterinarian 2000;22:568–74.

[3] Olivry T, Chan LS. Autoimmune blistering dermatoses in domestic animals. Clin Dermatol

2001;19:750–60.

[4] Ackerman LJ. Canine and feline pemphigus and pemphigoid. Part I. Pemphigus. Compen-

dium on Continuing Education for the Practicing Veterinarian 1985;7:89–99.

[5] White SD, Rosychuk RA, Reinke SI, et al. Use of tetracycline and niacinamide for treatment

of autoimmune skin disease in 31 dogs. J Am Vet Med Assoc 1992;200:1497–502.

[6] Cullen CL, Webb AA. Ocular manifestation of systemic diseases. In: Gelatt KN, editor.

Veterinary ophthalmology. 4th edition. Oxford (UK): Blackwell Publishing; 2007.
p. 1470–537.

[7] Stades FC, Gelatt KN. Diseases and surgery of the canine eyelid. In: Gelatt KN, editor.

Veterinary ophthalmology. 4th edition. Oxford (UK): Blackwell Publishing; 2007.
p. 563–617.

[8] Abegneli PL. Pathologie des paupie`res chez les carnivores domestiques. Rev Med Vet

(Toulouse) 1989;165:217–25 [in French].

[9] Sigle KJ, McLellan GJ, Haynes JS, et al. Unilateral uveitis in a dog with uveodermatologic

syndrome. J Am Vet Med Assoc 2006;228(4):543–8.

[10] Yamaki K, Cocho K, Hayakawa K, et al. Tyrosinase family proteins are antigens specific to

Vogt-Koyanagi-Harada disease. J Immunol 2000;165:7323–9.

[11] Denerolle P, Tessier M, Molon-Noblot S. [Nerve lesions in a Siberian husky suffering from

a uveodermatological syndrome]. Pratique Medicale and Chirurgicale de l’Animal de
Compagnie 2000;35:273–8 [in French].

[12] Carter WJ, Crispin SM, Gould DJ, et al. An immunohistochemical study of uveodermato-

logic syndrome in two Japanese Akita dogs. Vet Ophthalmol 2005;8(1):17–24.

246

PEN

˜ A & LEIVA

[13] Bedford PG. Uveodermatological syndrome in the Japanese Akita. Vet Rec 1986;1:134.
[14] Morgan RV. Vogt-Koyanagi-Harada syndrome in humans and dogs. Compendium on

Continuing Education for the Practicing Veterinarian 1989;11:1211–4.

[15] Hasegawa A. Mucocutaneous ocular syndrome in Akita dogs. Animal Eye Research

1985;4:17–21.

[16] Massa KL, Gilger BC, Miller TL, et al. Causes of uveitis in dogs: 102 cases (1989–2000).

Vet Ophthalmol 2002;5:93–8.

[17] Cottrell BD, Barnett KC. Harada’s disease in the Japanese Akita. J Small Anim Pract

1987;28:517–21.

[18] Kern TJ, Walton DK, Riis RC, et al. Uveitis associated with poliosis and vitiligo in six dogs.

J Am Vet Med Assoc 1985;187:408–14.

[19] Lindley DM, Boosinger TR, Cox NR. Ocular histopathology of Vogt-Koyanagi-Harada-like

syndrome in an Akita dog. Vet Pathol 1990;27:294–6.

[20] Laus JL, Sousa MG, Cabral VP, et al. Uveodermatologic syndrome in a Brazilian Fila dog.

Vet Ophthalmol 2004;7(3):193–6.

[21] Campbell KL, McLauchlin SA, Reynolds HA. Generalized leukoderma and poliosis follow-

ing uveitis in the dog. J Am Anim Hosp Assoc 1986;22:121–4.

[22] Bussanich MN, Rootman J, Dolman CL. Granulomatous panuveitis and dermal depigmen-

tation in dogs. J Am Anim Hosp Assoc 1982;18:131–8.

[23] Herrera HD, Duchene AG. Uveodermatological syndrome (Vogt-Koyanagi-Harada-like

syndrome) with generalized depigmentation in a Dachshund. Vet Ophthalmol 1998;1:
47–51.

[24] Robinson ND, Hashimoto T, Amagai M, et al. The new pemphigus variants. J Am Acad

Dermatol 1999;40:649–71.

[25] Scott DW, Miller WH, Griffin CE. Immune-mediated disorders. In: Muller, Kirks, editors.

Small animal dermatology. 6th edition. Philadelphia: WB Saunders Company; 2001.
p. 667–779.

[26] Amagai M, Tsunoda K, Zillikens D, et al. The clinical phenotype of pemphigus is defined by

the antidesmoglein autoantibody profile. J Am Acad Dermatol 1999;40:167–70.

[27] Anhalt GJ. Making sense of antigens and antibodies in pemphigus. J Am Acad Dermatol

1999;40:763–6.

[28] Griffin CE. Controversies in immunologic diseases from a clinician’s standpoint. In: Contro-

versies in veterinary dermatology. Bad Kreuznach (Germany); 1992.

[29] Scott DW. Immune-mediated dermatoses in domestic animals: ten years after. Part II.

Compendium on Continuing Education for the Practicing Veterinarian 1987;9:539–47.

[30] Ihrke PJ, Stannard AA, Ardans AA, et al. Pemphigus foliaceus in dogs: a review of 37 cases.

J Am Vet Med Assoc 1985;186:59–66.

[31] Marsella R. Canine pemphigus complex: pathogenesis and clinical presentation. Compen-

dium on Continuing Education for the Practicing Veterinarian 2000;22(6):568–72.

[32] Scott DW, Lewis RM. Pemphigus and pemphigoid in dogs and man: comparative aspects.

J Am Acad Dermatol 1981;5:148–67.

[33] Scott DW. Pemphigus in domestic animals. Clin Dermatol 1983;1:141–52.
[34] Beale KM. Azathioprine for the treatment of immune-mediated diseases of dogs and cats.

J Am Vet Med Assoc 1988;192:1316–8.

[35] Rashid S, Dana MR. Cicatrizing and autoimmune diseases. Chem Immunol Allergy

2007;92:195–202.

[36] Chabanne L. Canine systemic lupus erythematosus. Part I. Clinical and biologic aspects.

Compendium on Continuing Education for the Practicing Veterinarian 1999;21:135–47.

[37] Sontheimer RD. The lexicon of cutaneous lupus erythematosus: a review and personal

perspective on the nomenclature and classification of the cutaneous manifestations of lupus
erythematosus. Lupus 1997;6:84–95.

[38] Scott DW. Immune-mediated dermatoses in domestic animals: ten years after. Part I.

Compendium on Continuing Education for the Practicing Veterinarian 1987;9:423–32.

247

CANINE CONJUNCTIVITIS AND BLEPHARITIS

[39] Scott DW. Canine lupus erythematosus II: discoid lupus erythematosus. J Am Anim Hosp

Assoc 1983;19:481–90.

[40] White SD. Investigation of antibodies to extractable nuclear antigen in dogs. Am J Vet Res

1992;53:1019–21.

[41] McVey DS, Shuman W. Use of multiple antigen substrates to detect antinuclear antibody in

canine sera. Vet Immunol Immunopathol 1991;28:37–43.

[42] Scott DW. Canine lupus erythematosus I: systemic lupus erythematosus. J Am Anim Hosp

Assoc 1983;19:461–80.

[43] Chabanne L. Canine systemic lupus erythematosus. Part II. Diagnostic and treatment.

Compendium on Continuing Education for the Practicing Veterinarian 1999;21:402–19.

[44] Reimann KA, Evans MG, Chalifoux LV, et al. Clinicopathologic characterization of canine

juvenile cellulitis. Vet Pathol 1989;26(6):499–504.

[45] White SD, Rosychuk RA, Stewart LJ, et al. Juvenile cellulitis in dogs: 15 cases (1979–1988).

J Am Vet Med Assoc 1989;195(11):1609–11.

[46] Hutchings SM. Juvenile cellulitis in a puppy. Can Vet J 2003;44(5):418–9.
[47] Bassett RJ, Burton GG, Robson DC. Juvenile cellulitis in an 8-month-old dog. Aust Vet J

2005;83(5):280–2.

[48] Mason IS, Jones J. Juvenile cellulitis in Gordon setters. Vet Rec 1989;124:642.
[49] Neuber AE, van den Broek AH, Brownstein D, et al. Dermatitis and lymphadenitis

resembling juvenile cellulitis in a four-year-old dog. J Small Anim Pract 2004;45(5):254–8.

[50] Jeffers JG, Duclos DD, Goldschmidt MH. A dermatoses resembling juvenile cellulitis in an

adult dog. J Am Anim Hosp Assoc 1995;31(3):204–8.

[51] Bistner S. Allergic and immunologic-mediated diseases of the eye and adnexae. Vet Clin

North Am Small Anim Pract 1994;24(4):711–34.

[52] Hendrix DVH. Canine conjunctiva and nictitating membrane. In: Gelatt KN, editor.

Veterinary ophthalmology. 4th edition. Oxford (UK): Blackwell Publishing; 2007.
p. 662–89.

[53] Whitley RD, Gelatt KN, Cum GG. Dose response of topical pilocarpine in the normotensive

and glaucomatous beagle. Am J Vet Res 1980;41:417–24.

[54] Willis AM, Diehl KA, Robbin TE. Advances in topical glaucoma therapy. Vet Ophthalmol

2002;5:9–17.

[55] Schumer RA, Camras CB, Mandahl AK. Putative side effects of prostaglandin analogs. Surv

Ophthalmol 2002;47(Suppl 1):S219–30.

[56] Glaze MB. Ocular allergy. Semin Vet Med Surg (Small Anim) 1991;6:296–302.
[57] Bielory L. Ocular allergy guidelines: a practical treatment algorithm. Drugs 2002;62(11):

1611–34.

[58] Chamber ED, Severin GA. Staphylococcal bacterin for treatment of chronic staphylococcal

blepharitis in the dog. J Am Vet Med Assoc 1984;185:422–5.

[59] Barrie KP, Parshall CJ. Eyelid pyogranulomas in four dogs. J Am Anim Hosp Assoc 1979;14:

433–7.

[60] Samson J, Heinrich C, Featherstone H. Pyogranulomatous blepharitis in two dogs. J Small

Anim Pract 2000;41:80–3.

[61] Buckley RJ. Allergic eye disease: a clinical challenge. Clin Exp Allergy 1998;28(Suppl 6):

39–43.

[62] Scott DW, Miller WH, Griffin CE. Skin immune system and allergic skin diseases. In: Muller,

Kirks, editors. Small animal dermatology. 6th edition. Philadelphia: WB Saunders
Company; 2001. p. 543–666.

[63] Miyazaki D, Nakamura T, Komatsu N, et al. Roles of chemokines in ocular allergy and pos-

sible therapeutic strategies. Cornea 2004;23(1):48–54.

[64] Leonardi A, Curnow SJ, Zhan H, et al. Multiple cytokines in human tear specimens in

seasonal and chronic allergic eye disease and in conjunctival fibroblast cultures. Clin Exp
Allergy 2006;36:777–84.

248

PEN

˜ A & LEIVA

[65] Moore CP. Qualitative tear film disease. Vet Clin North Am Small Anim Pract 1990;20(3):

565–81.

[66] Doan S, Belayachi N, Ren L, et al. Tear IgE level measurement using semi-quantitative

methods in children with vernal keratoconjunctivitis (VKC). In: Programs and abstracts of
the ARVO meeting. 2002. p. 187–8.

[67] Hingorani M, Calder VL, Buckley RJ, et al. The role of conjunctival epithelial cells in chronic

ocular allergic disease. Exp Eye Res 1998;67(5):491–500.

[68] Leonardi A, Fregona IA, Plebani M, et al. Th1- and Th2-type cytokines in chronic ocular

allergy. Graefes Arch Clin Exp Ophthalmol 2006;244:1240–5.

249

CANINE CONJUNCTIVITIS AND BLEPHARITIS

Immunopathogenesis
of Keratoconjunctivitis Sicca
in the Dog

David L. Williams, MA, VetMB, PhD, CertVOphthal, FRCVS

a,b,

Department of Veterinary Medicine, University of Cambridge, Madingley Road, Cambridge,

CB3 0ES, England, UK

Veterinary Medicine and Pathology, St. John’s College, Cambridge CB2 1TP, England, UK

eratoconjunctivitis sicca (KCS), more commonly known as dry eye, is an
inflammatory condition of the ocular surface caused by pathologic re-
duction in the aqueous component of the tear film. It is seen commonly

in the dog and defined as a Schirmer tear test (STT) reading of less than
10 mm/min with concomitant ocular surface pathologic findings

. KCS

can be divided into two types: one in which tear production is deficient and
other cases in which tear evaporation accounts for the ocular surface tear de-
ficiency. In the dog, the latter is seen in brachycephalic dogs in which lagoph-
thalmos, that is to say failure of complete eyelid closure, leads to a central area
of tear film deficiency or in dogs in which a deficiency in tear film lipid leads to
increased tear loss through evaporation. In this review, the author concentrates
on the former condition in which pathologically reduced aqueous tear produc-
tion leads to the ocular pathologic findings of corneal vascularization, pigmen-
tation, and, in several cases, frank ulceration (

Figs. 1–3

The first attempt to document the prevalence of dry eye in canine keratocon-

junctivitis sicca (cKCS) was undertaken in a university clinic population more
than 30 years ago

and suggested a figure of 0.4% of animals affected,

whereas a more recent report documented a prevalence as high as 35% in
460 dogs

. The author and his colleagues

have performed STTs on

1000 randomly selected animals from the general canine population and deter-
mined that 4% had STT values less than 10 mm/min. Clearly, cKCS is a com-
mon and probably underrecognized condition in the general canine population.
Its treatment has been revolutionized over the past 15 years by the introduction
of topical cyclosporine as an effective lacrimogenic agent

, yet the etiopatho-

genesis of the condition is still poorly understood. The same holds true for dry
eye in human beings, which is a widely recognized problem in postmenopausal

*Department of Veterinary Medicine University of Cambridge Madingley Road Cambridge
CB3 OES, England, UK. E-mail address: doctordlwilliams@aol.com

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.12.002

Vet Clin Small Anim 38 (2008) 251–268

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

women. In some of these cases, it is associated with other exocrinopathies, such
as xerostomia from salivary gland involvement, and more generalized connec-
tive tissue diseases, such as rheumatoid arthritis, in which case it is known as
Sjo¨gren’s syndrome (SS); however, in many cases, it exists on its own as an un-
comfortable ocular surface condition for which optimal long-term treatment is
taxing. Rodent models abound, even though measuring tear production in rats
and mice is difficult. The purebred mouse strains mirroring SS have allowed
considerable dissection of the immunopathogenesis of the disease, yet how

Fig. 1. This 6-year-old West Highland white terrier neutered bitch shows the characteristic
appearance of canine keratoconjunctivitis sicca (cKCS), with sterile mucoid discharge as the
predominant feature.

Fig. 2. In this 12-year-old pug with tear deficiency and evaporative dry eye, the desiccated
and hyperpigmented ocular surface shows the classic signs of chronic canine keratoconjunc-
tivitis sicca (cKCS).

252

WILLIAMS

much this similarity holds for human beings and dogs is somewhat unclear.
The nonobese diabetic (NOD) mouse and the MRL and MRL/lpr congenic
strains are important rodent models for human SS, and also potentially for
cKCS, whereas other strains, such as the transforming growth factor (TGF)-
b

1 knockout mouse and New Zealand Black hybrid mouse, may also provide

valuable models for immune-mediated tear deficiency. In this review, the au-
thor provides some novel findings regarding the immunopathogenesis of
cKCS, discusses immunologic abnormalities in human KCS and SS, and as-
sesses the contributions made by research on the rodent models outlined pre-
viously, explaining how understanding of these noncanine examples can
inform our treatment of the disease in the dog.

ANATOMY AND PHYSIOLOGY OF TEAR PRODUCTION

For many years, the tear film has been characterized as having three layers:
a mucus layer close to the corneal epithelial surface, an aqueous layer, and,

Fig. 3. In this 9-year-old golden retriever, canine keratoconjunctivitis sicca (cKCS), again
characterized by profound mucoid discharge (A) has led to a central deepening corneal ulcer
(B), which is only seen once the discharge has been cleared from the eye. (C) Tear deficiency is
neurologic in origin, as seen by the dry external nares. The dog has otitis media, which
accounts for the neurologic cKCS.

253

KERATOCONJUNCTIVITIS SICCA IN THE DOG

finally, a most superficial layer of meibomian lipid that acts to reduce evapora-
tion

. More recently, studies have shown that this cut-and-dried distinction

between layers is not strictly correct—the corneal epithelium has a surface gly-
cocalyx composed of membrane-spanning mucins (including MUC1, 3, 4, 12,
13, and 16), with these have signaling capabilities through their cytoplasmic tail
and extracellular epithelial growth factor-like domains

. The aqueous layer,

far from being solely aqueous in nature, is filled with cleaved membrane-bound
mucins, small soluble mucins (MUC 7 and 9), and much larger gel-forming
mucins (including MUC2, 5, 6, and 19)

. Mucins are formed in conjunctival

goblet cells, whereas tear film lipid arises from the lid meibomian glands. The
aqueous component of the tear film is produced in the dog from the lacrimal
gland dorsolateral to the globe and is closely associated with it; it is also pro-
duced by the gland of the nictitating membrane. Recently, it has become clear
that to understand the physiology of tear production and the pathophysiology
of dry eye, the entire unit of the lacrimal glands, ocular surface, and innerva-
tion connecting them needs to be considered fully to describe the pathologic
events occurring in KCS. Although this review’s discussion of immunopatho-
genesis primarily concerns the glandular tissue producing the aqueous compo-
nent of the tear film, we must not forget that this is merely a part of this
‘‘lacrimal functional unit’’

Lacrimal gland secretion is under neural and hormonal control; the lacrimal

gland has parasympathetic and sympathetic innervation with a reflex arc iden-
tified from sensory nerves in the cornea activating efferent parasympathetic
and sympathetic nerves originating in the parasympathetic motor nucleus of
the facial nerve but traveling with the trigeminal nerve to the lacrimal gland.
The neurotransmitters acetylcholine, vasoactive intestinal peptide, substance
P, noradrenaline, and calcitonin gene-related peptide are all important in lacri-
mal secretion, but acetylcholine and noradrenaline are the most potent stimuli
to water and electrolyte secretion in tears

. Hormonal control by means of

the hypothalamopituitary-gonadal axis has a profound effect on tear secretion,
with adrenocorticotrophic hormone, a-melanocyte–stimulating hormone, pro-
lactin, androgens, estrogens, and progestagens all having an influence on lacri-
mal gland function

. In particular, androgens are potent stimulators of tear

production with important gender- and age-related effects on tear production.

OPHTHALMIC EXAMINATION OF THE TEAR FILM

Standard direct ophthalmoscopy and indirect ophthalmoscopy are important
in the evaluation of the ocular surface and tear film. In particular, evaluation
of the reflection of a focal light source shone onto the ocular surface demon-
strates the integrity of the epithelial surface and the tear film (see

). The

normal measure for tear production in the dog is the STT, alone (STT I) or after
topical anesthetic (STT II), with these being well recognized and reported in
the veterinary literature

[1,10,11]

. Commonly used to measure tear production,

the STT, in reality, assesses a combination of the rate of tear production and the
volume of the tear lake

. The phenol red thread test provides less of

254

WILLIAMS

a stimulus to lacrimation and is more appropriate for small mammals than is
the STT but has yet to find a place in routine assessment of the canine tear
film

. In human ophthalmology, the STT is used in combination with sev-

eral other methods of assessing the tear film, which should probably play
a wider part in investigation of canine lacrimation. Tear film breakup time is
becoming more commonly used to assess the combination of reduced tear pro-
duction and increased evaporation in ocular surface tear film deficiency

Determination of the tear volume by meniscometry has been championed by
some ophthalmic researchers

but has yet to be reported in the veterinary

sphere. Rose bengal staining assesses pathologic changes to the epithelial sur-
face

but has yet to be used sufficiently in canine ophthalmology to allow

correlation of the test with other measures of canine tear deficiency. The mei-
bomian gland plays an important part in providing the lipid that reduces tear
film evaporation, and assessment of meibomian gland function has recently
been reported in the normal dog

, but changes in animals with cKCS

have yet to be defined.

Overall, the STT is still an important criterion by which to judge ocular sur-

face health, with readings less than 10 mm/min indicative of dry eye. All too
often, the problem is not that other tests for tear production should be used
in ocular examination but that the STT is not used sufficiently in general vet-
erinary practice. The author and his colleagues

have shown in their study of

1000 dogs that the average STT reading is 18.6 mm/min and that for the
breeds classically associated with cKCS, the tear production even in normal
dogs is significantly lower. Increasing age is associated with decreasing produc-
tion of tears

, and in a study of dogs with endocrinopathies, the author and

his colleagues

have demonstrated that animals with diabetes mellitus,

hyperadrenocorticism, and hypothyroidism all have reduced tear production.
Thus, older dogs of such breeds as those in

; dogs with the previous

endocrinopathies; and any dog with corneal ulceration, corneal inflammation,
or a red eye should be subject to careful ocular surface evaluation, including the
STT.

IMMUNOLOGIC ASPECTS OF CANINE
KERATOCONJUNCTIVITIS SICCA

Early research on the etiopathogenesis of cKCS centers around the work of
Dr. Rene Kaswan in the 1980s. The first report of Kaswan and colleagues

documented histologic evaluation of 40 nictitating membrane glands

(NMGs) and 9 main lacrimal gland (MLGs) from 28 dogs affected by cKCS
with varying diagnoses of concurrent systemic disease from distemper, diabetes
mellitus, systemic lupus erythematosus (SLE), and hypothyroidism. This pop-
ulation differs substantially from the case load seen by the current author, in
which most dogs have KCS as their only immune-mediated disorder. In
the study by Kaswan and colleagues

, 20% of the NMGs and 22% of the

MLGs had insufficient glandular tissue on biopsy for evaluation, whereas the
remaining histologic findings in 10% were considered within normal limits.

255

KERATOCONJUNCTIVITIS SICCA IN THE DOG

Sixteen percent of glands were classified as having stage I inflammatory disease
with minimal numbers of periductal and periacinal lymphocytes and some fi-
brous connective tissue replacing tubuloacinar elements. Thirty-three percent
of glands exhibited stage II inflammatory change (

) with lymphatic nod-

ules, squamous metaplasia of ductal epithelial lining, and, in dogs with distem-
per only, the presence of numerous neutrophils. Thirteen percent of glands
were classified as having stage III inflammatory change replacement of glandu-
lar elements with fibrous connective tissue and moderate numbers of

Table 1
Canine breeds predisposed to keratoconjunctivitis sicca

Breed

Relative risk

Cavalier King Charles

spaniel

11.5

English bulldog

10.8

Lhasa apso

9.8

Shih tzu

6.2

West Highland white terrier

5.5

Pug

5.2

Bloodhound

4.5

American cocker spaniel

4.1

Pekingese

4.0

Boston terrier

2.0

Miniature schnauzer

1.8

Samoyed

1.7

Other breeds

1.0

Data from Kaswan RL, Salisbury MA. A new perspective on canine keratoconjunctivitis sicca. Vet Clin North
Am Small Anim Pract 1990;20:595.

Fig. 4. Mild inflammatory cell infiltrate and fibrosis in the lacrimal gland of a West Highland
white terrier with cKCS.

256

WILLIAMS

mononuclear inflammatory cells. Duct dilation and epithelial squamous meta-
plasia were seen in several dogs.

More recently, Izci and colleagues

showed a significant decrease in

CD8þ lymphocytes and reversal of the CD4þ/CD8þ ratio in NMGs of
dogs with KCS after 30 days of treatment with topical 2% cyclosporine giving
concurrent regression of clinical signs. The study, however, failed to include
information on the initial lymphocyte populations in the affected NMGs.
The author and his colleagues

have recently performed immunohisto-

chemistry on NMG biopsies from nine dogs with normal STT values, six
with idiopathic cKCS, and two with cKCS associated neurologic etiopathologic
findings characterized by dry nostrils and dry eyes. The number of CD3þ T
lymphocytes as a proportion of cells in the lacrimal tissue in normal dogs was
0.058 compared with 0.143 in dogs that had idiopathic KCS and 0.079 in sam-
ples from dogs that had neurogenic KCS (

). The proportion of CD79a-

expressing B cells in the NMG of dogs with normal tear production was
0.087, whereas in dogs that had idiopathic KCS, it was 0.181, and in neurologic
cases of KCS, the proportion was 0.202. Numbers of T and B cells were signif-
icantly increased in idiopathic cKCS (P ¼ .002 and P ¼ .044, respectively) but
not in neurologic cKCS (P ¼ .07 and P ¼ .18, respectively). These results show
that the increase in T-lymphocyte numbers is likely to be the cause of the dis-
ease and not the result of ocular surface drying, in which case, numbers would
be significantly increased in idiopathic (ie, presumed to be immune-mediated)
and neurologic cases. Unfortunately, we do not currently have access to anti-
bodies against CD4 and CD8 epitopes in canine fixed embedded tissue, and
thus cannot evaluate T-lymphocyte subpopulations at present. The inflamma-
tory infiltrate in these cases is periacinal, presumably directed at antigens on
acinal cells of the lacrimal glandular tissue. This is not, however, the only
mechanism of reduction in tear production.

Fig. 5. Immunohistochemistry shows CD3-expressing T cells in the periacinal tissue of the
lacrimal gland in a cocker spaniel with cKCS.

257

KERATOCONJUNCTIVITIS SICCA IN THE DOG

cKCS in leishmaniosis is the result not of periacinal inflammation but of peri-

ductal infiltration

, with constriction of the outflow of tears and subsequent

dilation of lacrimal ductules (

). Quite what the inciting antigen resulting in

this periductal lymphocytic infiltrate is remains unclear, but it is unlikely to be
the same as that enticing cellular infiltration in idiopathic immune-mediated
cKCS.

Kaswan and colleagues

continued work on the immunology of cKCS

with reports documenting systemic immunologic changes in the disease—
autoantibodies against undefined lacrimal antigens in one article and rheuma-
toid factor in another

. Autoantibodies are important in human SS as

detailed elsewhere in this article and in rodent models of immune-mediated
KCS. As such, the nature of autoantibodies in cKCS should be more carefully
elucidated. In the report of Kaswan and colleagues

, 40% of the cases had

concurrent immunologic disease, such as SLE, pemphigus foliaceous, or rheu-
matoid arthritis, or diseases with a possible immunologic component in their
etiopathology, such as hypothyroidism or diabetes mellitus, generalized demo-
dectic mange, ulcerative colitis, or glomerulonephritis; thus, again, these cases
may not reflect the population of dogs we see with cKCS, in which, generally,
tear abnormality is the only complaint. Five of 9 dogs tested had autoantibodies
against lacrimal antigens on direct immunofluorescence, whereas 9 of 31 dogs
demonstrated autoantibodies to ductal tissue in the NMG. Sixty-seven percent
of dogs had hypergammaglobulinemia, and 16% had elevated serum IgA.
Forty-two percent of dogs tested had circulating antinuclear antibodies. The rel-
evance of these findings in understanding the pathogenesis of cKCS is unclear,
because it is impossible to say whether such immunologic changes are a cause
or effect of lacrimal inflammation. Clearly, more work is needed to repeat and
extend this work.

Fig. 6. Periductal inflammatory cell infiltrate in a dog with cKCS associated with leishmanio-
sis. (Courtesy of Teresa Pen˜a, DVM, PhD, Barcelona, Spain.)

258

WILLIAMS

It has to be added, however, that apart from the ductal occlusion in leishma-

niosis, the data presented here do not explain why tear production should be
lower. In many cases, neurologic cKCS is also characterized by a dry nose
(

C) because the innervation to the medial nasal gland giving wetting of

the external nares is also affected. It is relatively straightforward to explain
how efferent denervation impairs lacrimal secretion, but the exact mechanism
by which inflammation reduces tearing is less easily understood. For research
attempting to explain this, we need to examine the work on rodent models pre-
sented elsewhere in this article, with discussion of the debate that still rages re-
garding the mechanism of glandular hypofunction in SS. Before that, however,
the treatment of cKCS requires discussion.

TREATMENT OF CANINE KERATOCONJUNCTIVITIS SICCA

A key feature suggesting that immunologic changes are central to the etiopatho-
genesis of cKCS is the ameliorative effect of cyclosporine, a drug first used to
treat transplant rejection in human patients

and now employed as a sys-

temic immunomodulator in canine patients that have immune-mediated condi-
tions as diverse as atopy

, anal furunculosis

, and myasthenia gravis

. Topical use of this drug started at 2%

[5,30]

and was reduced in later stud-

ies to 1%

, and, finally, in the commercially marketed medication Optim-

mune (Union, New Jersey), to 0.2%

. Cyclosporine is a specific

immunomodulator that prevents lymphokine production through its action
as a calcineurin inhibitor

, whereas other more recently developed agents

within this family of drugs, such as tacrolimus

and pimecrolimus

have been suggested to have a more potent effect, although our findings are
that in cases resistant to the lacrimogenic effects of 2% cyclosporine, topical
tacrolimus, at least, shows little added clinical benefit

, whereas other inves-

tigators have somewhat contrary findings

, with tacrolimus yielding a better

lacrimomimetic effect than cyclosporine. It has to be said that these two results
are not mutually exclusive—one might say that although the lacrimomimetic
effect of tacrolimus is clearer than that of cyclosporine when some remaining
glandular tissue exists, when there is no exocrine glandular tissue left and
lacrimal gland pathologic change has reached the fibrotic stage III of the study
by Kaswan and colleagues

, no immunomodulatory lacrimogen acts to in-

crease tear production. It is at that stage that parotid duct transposition is the
only effective therapeutic way forward

. The original hypothesis that the

lacrimomimetic effect of cyclosporine is through a local immunosuppressant
is called into question by its lacrimogenic effects in dogs without cKCS, in
which tear production is increased

, and by experimental studies on neuro-

modulatory effects

, which are discussed further elsewhere in this article.

Other work on dogs rendered lacrimally deficient by removal of the NMG
and the MLG demonstrated that the drug has effects on mucus production
and the ocular surface separate from those on aqueous secretion

Another immunomodulatory treatment is that using a-interferon (IFN)

orally

. Just more than half of the 20 treated animals showed a favorable

259

KERATOCONJUNCTIVITIS SICCA IN THE DOG

response to an escalating dose regimen of cytokine administration from 20 to
80 IU/d. Similar beneficial effects were noted with regard to lacrimal and sali-
vary secretory function in human patients who have SS

. Given that the

cytokine is now recognized to act as a balancing agent in the immune system

[42]

, such treatment has a firm rationale, although little further development

of a-IFN seems to have been undertaken after the preliminary report by Gilger
and colleagues

Finally, the use of parasympathomimetics should not be forgotten. Topical

use of pilocarpine at concentrations between 0.25% and 2% caused blepharo-
spasm, conjunctival hyperemia, and miosis of the treated eye without signifi-
cant increase in tear production

[43]

, whereas systemic administration by use

of drops applied on the tongue stimulates tear production in normal dogs
and in a proportion of KCS-affected animals

MECHANISMS OF TEAR HYPOSECRETION:
AN INTRODUCTION

Quite apart from the diagnosis and treatment of cKCS, there remains the rather
more academic question concerning the mechanism by which the inflammatory
changes in KCS more generally lead to a profound reduction in tear produc-
tion. One suggestion is that lymphocyte-associated cytotoxicity of lacrimal
tissue is central to the pathologic effects on lacrimation. A second is that apo-
ptosis of glandular epithelial cells is critical in tear hyposecretion. A third is
that cytokine release from inflammatory cells alters tear production. Finally, in-
flammatory cells or their associated cytokines or autoantibodies may influence
neurotransmitter function in the lacrimal gland, inhibiting neurologic stimula-
tion of tear secretion. These possible mechanisms are not, of course, self-
excluding, and more than one may be in action in one or several of the tear
deficiency syndromes seen in dogs, in human beings, and in experimental an-
imal models. It is with these possible routes to tear deficiency in mind that the
reader is invited to consider the immunopathologic research in human beings
and rodents detailed here.

IMMUNOPATHOLOGY OF KERATOCONJUNCTIVITIS
SICCA IN HUMAN PATIENTS

Similar pathologic changes to those described previously for cKCS are seen in
the lacrimal and salivary glands of human patients who have SS: CD4þ T cells
predominate in the lesions, with increased major histocompatibility complex
(MHC) class II expression on glandular epithelial cells

[45,46]

. These T cells

express a limited number of T-cell receptor phenotypes shared between lacri-
mal and salivary glands, suggesting that they are clonally expanding in re-
sponse to a shared antigen between these two glands

. Several cell lines

cloned from infiltrated salivary gland tissue showed autoreactivity to the
Ro/SSA 52-kDa antigen against which autoantibodies are present in affected pa-
tients

. Ro/SSA is a ribonuclear polypeptide complex, whereas La/SSB is

260

WILLIAMS

a 48-kDa protein against which autoantibodies are also directed in SS, SLE, and
several other connective tissue autoimmune diseases. These proteins are nor-
mally nonimmunogenic but can associate with stress proteins, such as calregu-
lin, which, although normally regulating calcium homeostasis in the
endoplasmic reticulum (ER), when overexpressed, increases calcium levels in
the ER and aberrantly protects the cell from apoptosis

. Interaction between

these ribonuclear polypeptides and calregulin may result in failed clearance of
apoptotic debris by macrophages and provoke a cascade of autoimmune reac-
tions that lead to connective tissues diseases, such as SS and SLE.

Ironically, although the CD4þ hyper-T cell is reported as central to the path-

ogenesis of SS, there is also evidence that the disease is primarily a B-cell–dom-
inated condition

. B cells can differentiate to have the characteristics of

T helper (Th) cells, secrete Th1 or Th2 cytokines, and thus be termed B-effec-
tor (Be) lymphocytes. It may be that these cells are instrumental in SS patho-
genesis. Rituximab, a humanized anti-CD20 (and thus anti-B cell)
monoclonal antibody, has a profound ameliorative effect on several patients
who have SS, showing how important B cells are in disease pathogenesis. In-
deed, in some patients, there is a fine line to be drawn between reactive B-lym-
phocyte expansion in SS and mucosal-associated lymphoid tissue (MALT)
lymphoma

. Generally, autoimmune diseases are characterized by lesions

in which Th1-lymphocyte populations predominate, with cytokines, such as in-
terleukin (IL)-1, IL-2, and c-IFN. Some researchers find these cytokines in
SS lesions

, whereas others, working with lesions at a different stage of dis-

ease progression, find Th2-cell populations

. Indeed, the most recent work

published at the time of writing this review shows different lymphocyte popu-
lations at different time points in the development of disease

. The Th2

cytokines IL-4 and IL-5 are seen in early lesions, although as the inflammation
progresses and worsens in severity, the balance swings in favor of Th1 cyto-
kines and a more aggressive cytotoxic immune response. Quite how this influ-
ences lacrimal function, and what it means for treatment options, remains
unclear at present.

LABORATORY MODELS OF KERATOCONJUNCTIVITIS SICCA

Several rodent models of SS and its associated KCS are available, and each
adds to the overall understanding regarding pathogenic mechanisms in human
SS-associated KCS and also, potentially, in cKCS. A good review is that pro-
vided by Barabino and Dana

. In this article, the author concentrates on

the immunologic aspects of the MRLþ, MRL/lpr, and NOD mice.

For many years, the MRLþ mouse strain has been recognized as a model

for human SS. As in human patients who have SS, female mice are more sus-
ceptible to infiltration of lacrimal tissue by T lymphocytes

. A more aggres-

sive lacrimal gland inflammation develops in the MRL/Mp-lpr/lpr strain in
which the lpr gene, encoding the proapoptotic fas protein, is defective

Lymphocytes in this mouse strain fail to apoptose, that is, to commit pro-
grammed cell death; thus, massive accretions of activated lymphocytes develop

261

KERATOCONJUNCTIVITIS SICCA IN THE DOG

in several tissues, particularly in the lacrimal glands. Apoptotic cell death of
lacrimal gland epithelium may be important in the initiation of disease in other
mouse models, as discussed further elsewhere in this article. Note the potential
similarity to alterations in apoptosis in the human patients as discussed
previously.

T-cell populations in the MRL/lpr mice are CD4þ in character

, and an-

tibodies against CD4 ameliorate disease

. Cytokines in these lacrimal gland

lesions are predominantly Th2 in character with IL-4 RNA transcripts being
present in between 100 and 1000 times great number by reverse transcriptase
polymerase chain reaction (RT-PCR) than c-IFN transcripts. Glandular epithe-
lial cells, which can act as nonprofessional antigen-presenting cells

[60]

, express

the costimulatory molecule B7-2 (CD86), which is associated with generation
of a Th2 response, rather than B7-1 (CD80), which produces a Th1 inflamma-
tory phenotype

. Were it to be thought that this indicates a benign anti-

body-mediated immune response rather than an aggressive cytotoxic
reaction, it must be remembered that the proinflammatory mediators nitric ox-
ide and tumor necrosis factor (TNF)-a have also been detected in these lesions

[62,63]

; thus, not everything on the lacrimal front is quiet in the disease.

Lacrimal inflammatory disease in the NOD mouse offers another autoim-

mune murine model of SS with a Th1 CD4 lymphocyte inflammatory cell pop-
ulation

. Several proinflammatory cytokines, including IL-1b, IL-6, IL-7,

IL-10, c-IFN, and TNFa, were detected, together with inducible nitric oxide
synthase (iNOS), but IL-4 synthesis was absent in lacrimal and salivary glands.
Cytokines detected in lacrimal tissue appeared earlier and more intensely in
the submandibular glands

[65,66]

. Interestingly, other research groups find

that defective neurotransmitter signaling in the lacrimal and salivary glands
of these NOD mice precede the inflammatory infiltrate in these mice

Other changes within the gland may be nonimmunologic in origin. In
NOD–severe combined immunodeficiency (SCID) transgenic mice, in which
the NOD genotype occurs in the absence of T or B cells, there are changes
in the lacrimal gland with age

. NOD-IgM null mice, which lack B cells

but still have T cells, do not lose secretory function even though they have
a T-cell infiltrate in their lacrimal glands

. Autoantibodies to M3 acetylcho-

line receptors have been shown to play a pivotal role in the reduction of glan-
dular secretion in NOD mice, as is discussed further elsewhere in this article

[70,71]

, and complement also plays an important part in disease pathogenesis

[72]

Two questions remain after this large volume of work on the immunologic

phenotype of these dry eye mouse models. One must continue to enquire as to
the nature of the initial autoantigen that sets the inflammatory process off. Sec-
ond, we have much still to learn regarding the link between the inflammatory
process occurring in all these individuals, be they rodents, dogs, or people, and
the reduction in secretory output that is the key feature leading to clinical dis-
ease of the ocular surface. Perhaps the M3 muscarinic receptor lies at the heart
of the answers to both of these questions.

262

WILLIAMS

MECHANISMS OF TEAR HYPOSECRETION: REVIEWING
THE EVIDENCE

The classic model of glandular dysfunction or hypofunction regards the loss of
gland secretion as being caused by immune destruction of gland tissue and sub-
sequent apoptosis of glandular epithelium

[73]

. Yet, although apoptosis has

been reported in dogs, people, and rodents with KCS

, it is not an invari-

able finding. Perhaps more relevantly, in many cases of human SS and in a size-
able proportion of dogs with cKCS, a significant amount of glandular tissue
appears apparently normal. Laboratory studies show this tissue to be functional
in vitro

, and cases unresponsive to immunomodulators, such as cyclospor-

ine, may revert to normal function with application of systemic parasympatho-
mimetic lacrimologues

. Defective neurotransmitter-mediated signaling has

been described in the NOD mouse

, and other researchers have suggested

that such defects point to a nonimmunologic factor in the etiopathogenesis of
dry eye

. Elevations in levels of matrix metalloproteinases in the disease

again point to different disease processes occurring concurrent with the inflam-
matory infiltrates, but determining whether these are a causative agent or an
effect of inflammatory change is problematic

What predisposing abnormalities actually initiate the disease? Clearly, the

genetic background is important, not to say central. Although we do not yet
know the dog leukocyte antigen haplotypes of predisposed breeds, as listed
in

, it is likely that these breed predispositions mirror the population

genetics of human SS patient groups in which the human leukocyte antigen
(HLA) haplotypes HLA-Dw3 and HLA-B8 are associated with SS disease

. Hormonal changes reflected in the gender assignment of most patients

who have SS and in MRLþ and MRL/lps mouse models (but interestingly
not in the NOD mouse)

have a critical part to play in the generation of

dry eye in most rodent models, in human patients who have SS and non-Sjo¨g-
ren’s dry eye, and in dogs

. Androgen deficiency is a key feature of human

and mouse models of the disease, in which testosterone treatment re-

duces the disease severity, increasing tear secretion and IgA content

[81]

. An-

drogen receptors have been reported in the acini of the glandular epithelium
and not within the inflammatory cell infiltrate in lacrimal lesions in mouse
models, such as the MRL/lpr mouse

, suggesting that the response to andro-

gen therapy is primarily caused by glandular physiologic changes and not ini-
tially by an amelioration of the inflammatory pathologic changes noted.
Having said that, the same research group has previously reported that andro-
gen treatment does markedly reduce the inflammatory cell population in the
lacrimal gland

. Proapoptotic genes, such as pcl-2, are expressed in the lac-

rimal gland of these mice in a strikingly gender-specific manner

[84]

, with this

influenced by testosterone administration. In addition, normal organogenesis of
the salivary gland has been reported in NOD mice, which progresses to inflam-
matory disease of this structure, suggesting that there may be initial nonimmu-
nologic factors involved in disease pathogenesis

[85]

, but similar results have

not been reported for the lacrimal gland.

263

KERATOCONJUNCTIVITIS SICCA IN THE DOG

A hypothesis that draws these divergent strands of disease pathogenesis to-

gether might be one that sees the effects of antimuscarinic receptor autoanti-
bodies on neurotransmitter release, giving the secretory failure and also
provoking a further inflammatory response that eventually leads to glandular
destruction. Defects in apoptosis may rest at the heart of the original autoanti-
genic challenge to the immune system, with apoptosis of secretory epithelial
cells exposing previously cryptic autoantigens, such as a-fodrin, a calmodu-
lin-binding protein, to the immune system

[86]

. The lymphocytes apparently

at the heart of SS congregate around acini with apoptotic cells

[87]

, and anti-

gens unmasked in apoptotic fragments seem to be important in lymphocyte-
associated cell death

. The occurrence, or lack of it, of autoimmune responses

in mouse models seems to be linked to display and cleavage of autoantigens in
apoptotic cells in work from some researchers

, whereas other investigators

propose a model for lacrimal hypofunction that does not involve apoptosis

[90]

but in which local neurologic disturbance, caused primarily by proinflamma-
tory cytokines, explains most of the clinical and immunohistologic signs seen
in the disease.

The truth of the matter probably lies in a combination of these mechanisms,

with various factors (genetic, environmental, apoptotic, and, possibly, infec-
tious) leading to the exposure of self-antigens, production of danger signals
as first suggested by Matzinger

[91]

, and development of an autoantibody-

mediated inhibition of neurotransmitter function giving tear deficiency

SUMMARY

It may be possible to confirm such a train of pathologic events as outlined here
in rodent models and even in human patients who have SS; however, as we
have seen here, cKCS languishes a long way behind in research terms even
if, as veterinarians, we lead the field, and have done so for nearly 2 decades,
in effective lacrimogenic treatment for cKCS in the form of topical cyclospor-
ine. We have not yet even fully characterized the inflammatory cell population
in the lacrimal glands of dogs affected by cKCS or evaluated their genetic basis.
It is hoped that this review spurs further research into the etiopathogenic fac-
tors in cKCS.

References

[1] Gelatt KN, Peiffer RL Jr, Erickson JL, et al. Evaluation of tear formation in the dog, using

a modification of the Schirmer tear test. J Am Vet Med Assoc 1975;166:368–70.

[2] Helper LC. Keratoconjunctivitis sicca in dogs. Trans Sect Ophthalmol Am Acad Ophthalmol

Otolaryngol 1976;81:624–8.

[3] Kaswan R, Pappas C Jr, Wall K, et al. Survey of canine tear deficiency in veterinary practice.

Adv Exp Med Biol 1998;438:931–9.

[4] Peirce V, Harmer E, Williams DL. Prevalence of keratoconjunctivitis in the canine population:

Shirmer tear tests in one thousand normal dogs. Veterinary Journal submitted.

[5] Kaswan RL, Salisbury MA. Spontaneous canine keratoconjunctivitis sicca. A useful model

for human keratoconjunctivitis sicca: treatment with cyclosporine eye drops. Arch Ophthal
1989;107:1210–6.

264

WILLIAMS

[6] Lambert DW. Physiology of the tear film. In: Smolin G, Thoft RA, editors. The cornea. New

York: Little Brown & Co; 1994. p. 439–55.

[7] Gipson IK, Yankauckas M, Spurr-Michaud SJ, et al. Characteristics of a glycoprotein in the

ocular surface glycocalyx. Invest Ophthalmol Vis Sci 1992;33:218–27.

[8] Gipson IK, Hori Y, Argueso P. Character of ocular surface mucins and their alteration in dry

eye disease. Ocul Surf 2004;2:131–48.

[9] Stern ME, Gao J, Siemasko KF, et al. The role of the lacrimal functional unit in the pathophys-

iology of dry eye. Exp Eye Res 2004;78:409–16.

[10] van der Woerdt A, Adamcak A. Comparison of absorptive capacities of original and mod-

ified Schirmer tear test strips in dogs. J Am Vet Med Assoc 2000;216:1576–7.

[11] Afonso AA, Monroy D, Stern ME, et al. Evaluation of results for Schirmer tear tests conducted

with and without application of a topical anesthetic in clinically normal dogs of 5 breeds.
Am J Vet Res 2000;61:1422–5.

[12] Williams DL. Analysis of tear uptake by the Schirmer tear test strip in the canine eye. Vet

Ophthalmol 2005;8:325–30.

[13] Brown MH, Galland JC, Davidson HJ, et al. The phenol red thread test in dogs. Progress in

Veterinary and Comparative Ophthalmology 1996;6:274–7.

[14] Saito A, Kotani T. Estimation of lacrimal level and testing methods on normal beagles. Vet

Ophthalmol 2001;4:7–11.

[15] Yokoi N, Bron A, Tiffany J, et al. Reflective meniscometry: a non-invasive method to measure

tear meniscus curvature. Br J Ophthalmol 1999;83:92–7.

[16] Saito A, Kotani T. Tear production in dogs with epiphora and corneal epitheliopathy. Vet

Ophthalmol 1999;2:173–8.

[17] Ofri R, Orgad K, Kass PH, et al. Canine meibometry: establishing baseline values for mei-

bomian gland secretions in dogs. Vet J 2007;174:536–40.

[18] Hartley C, Williams DL, Adams VJ. Effect of age, gender, weight, and time of day on tear

production in normal dogs. Vet Ophthalmol 2006;9:53–7.

[19] Williams DL, Pierce V, Mellor P, et al. Reduced tear production in three canine endocrinopa-

thies. J Small Anim Pract 2007;48:252–6.

[20] Kaswan RL, Martin CL, Chapman WL Jr. Keratoconjunctivitis sicca: histopathologic study

of nictitating membrane and lacrimal glands from 28 dogs. Am J Vet Res 1984;45:
112–8.

[21] Izci C, Celik I, Alkan F, et al. Histologic characteristics and local cellular immunity of the

gland of the third eyelid after topical ophthalmic administration of 2% cyclosporine for treat-
ment of dogs with keratoconjunctivitis sicca. Am J Vet Res 2002;63:688–94.

[22] Tighe A, Williams DL. Analysis of lymphocyte populations in nictitans glands of dogs with

keratoconjunctivitis sicca. Veterinary Journal submitted.

[23] Naranjo C, Fondevila D, Leiva M, et al. Characterization of lacrimal gland lesions and pos-

sible pathogenic mechanisms of keratoconjunctivitis sicca in dogs with leishmaniosis. Vet
Parasitol 2005;133:37–47.

[24] Kaswan RL, Martin CL, Dawe DL. Keratoconjunctivitis sicca: immunological evaluation of

62 canine cases. Am J Vet Res 1985;46:376–83.

[25] Kaswan RL, Martin CL, Dawe DL. Rheumatoid factor determination in 50 dogs with kerato-

conjunctivitis sicca. J Am Vet Med Assoc 1983;183:1073–5.

[26] Kahan BD, van Buren CT, Flechner SM, et al. Clinical and experimental studies with cyclo-

sporine in renal transplantation. Surgery 1985;97:125–40.

[27] Olivry T, Rivierre C, Jackson HA, et al. Cyclosporine decreases skin lesions and pruritus in

dogs with atopic dermatitis: a blinded randomized prednisolone-controlled trial. Vet Derma-
tol 2002;13:77–87.

[28] Hardie RJ, Gregory SP, Tomlin J, et al. Cyclosporine treatment of anal furunculosis in

26 dogs. J Small Anim Pract 2005;46:3–9.

[29] Bexfield N, Watson PJ, Herrtage M. Management of myasthenia gravis using cyclosporine

in 2 dogs. J Vet Intern Med 2006;20:1487–90.

265

KERATOCONJUNCTIVITIS SICCA IN THE DOG

[30] Morgan RV, Abrahams KL. Topical administration of cyclosporine for treatment of keratocon-

junctivitis sicca in dogs. J Am Vet Med Assoc 1991;99:1043–6.

[31] Olivero DK, Davidson MG, English RV, et al. Clinical evaluation of 1% cyclosporine for top-

ical treatment of keratoconjunctivitis sicca in dogs. J Am Vet Med Assoc 1991;199:
1039–42.

[32] Sansom J, Barnett KC, Neumann W, et al. Treatment of keratoconjunctivitis sicca in dogs

with cyclosporine ophthalmic ointment: a European clinical field trial. Vet Rec 1995;137:
504–7.

[33] Williams DL. A comparative approach to topical cyclosporine therapy. Eye 1997;11:

453–64.

[34] Berdoulay A, English RV, Nadelstein B. Effect of topical 0.02% tacrolimus aqueous suspen-

sion on tear production in dogs with keratoconjunctivitis sicca. Vet Ophthalmol 2005;8:
225–32.

[35] Nell B, Walde I, Billich A, et al. The effect of topical pimecrolimus on keratoconjunctivitis

sicca and chronic superficial keratitis in dogs: results from an exploratory study. Vet Ophthal-
mol 2005;8:39–46.

[36] Williams DL. Efficacy of topical tacrolimus in dogs with cyclosporine-non-responsive kerato-

conjunctivitis sicca. Birmingham, UK: British Small Animal Veterinary Association; 2006.

[37] Glenn JB, Lawson DD. A modified technique of parotid duct transposition for the treatment of

keratoconjunctivitis sicca in the dog. Vet Rec 1971;88:210–3.

[38] Yoshida A, Fujihara T, Nakata K. Cyclosporin A increases tear fluid secretion via release of

sensory neurotransmitters and muscarinic pathway in mice. Exp Eye Res 1999;68:541–6.

[39] Moore CP, McHugh JB, Thorne JG, et al. Effect of cyclosporine on conjunctival mucin in a ca-

nine keratoconjunctivitis sicca model. Invest Ophthalmol Vis Sci 2001;42:653–9.

[40] Gilger BC, Rose PD, Davidson MG, et al. Low-dose oral administration of interferon-alpha

for the treatment of immune-mediated keratoconjunctivitis sicca in dogs. J Interferon Cyto-
kine Res 1999;19:901–5.

[41] Ferraccioli GF, Salaffi F, De Vita S, et al. Interferon alpha-2 (IFN alpha 2) increases lacrimal

and salivary function in Sjogren’s syndrome patients. Preliminary results of an open pilot trial
versus OH-chloroquine. Clin Exp Rheumatol 1996;14:367–71.

[42] Amadori M. The role of IFN-alpha as homeostatic agent in the inflammatory response: a bal-

ance between danger and response? J Interferon Cytokine Res 2007;27:181–9.

[43] Smith EM, Buyukmihci NC, Farver TB. Effect of topical pilocarpine treatment on tear produc-

tion in dogs. J Am Vet Med Assoc 1994;205:1286–9.

[44] Rubin LF, Aguirre G. Clinical use of pilocarpine for keratoconjuctivitis in dogs and cats. J Am

Vet Med Assoc 1967;151:313–20.

[45] Adamson TC, Fox RI, Frisman DM, et al. Immunohistologic analysis of lymphoid infiltrates in

primary Sjogren’s syndrome using monoclonal antibodies. J Immunol 1983;130:203–8.

[46] Lindahl G, Hedfors E, Klareskog L, et al. Epithelial HLA-DR expression and T lymphocyte sub-

sets in salivary glands in Sjogren’s syndrome. Clin Exp Immunol 1985;61:475–82.

[47] Matsumoto I, Tsubota K, Satake Y, et al. Common Tcell receptor clonotype in lacrimal glands

and labial salivary glands from patients with Sjogren’s syndrome. J Clin Invest 1996;97:
1969–77.

[48] Namekawa T, Kuroda K, Kato T, et al. Identification of Ro(SSA) 52 kDa reactive T cells in

labial salivary glands from patients with Sjogren’s syndrome. J Rheumatol 1995;22:
2092–9.

[49] Eggleton P. Stress protein-polypeptide complexes acting as autoimmune triggers. Clin Exp

Immunol 2003;134:6–8.

[50] Daridon C, Guerrier T, Devauchelle V, et al. Polarization of B effector cells in Sjogren’s syn-

drome. Autoimmun Rev 2007;6:427–31.

[51] Carbone A, Gloghini A, Ferlito A. Pathological features of lymphoid proliferations of the sal-

ivary glands: lymphoepithelial sialadenitis versus low-grade B-cell lymphoma of the malt
type. Ann Otol Rhinol Laryngol 2000;109:1170–5.

266

WILLIAMS

[52] Konttinen YT, Kemppinen P, Koski H, et al. Th1 cytokines are produced in labial salivary

glands in Sjogren’s syndrome, but also in healthy individuals. Scand J Rheumatol
1999;28:106–12.

[53] Pertovaara M, Antonen J, Hurme M. Th2 cytokine genotypes are associated with a milder

form of primary Sjogren’s syndrome. Ann Rheum Dis 2006;65:666–70.

[54] Mitsias DI, Tzioufas AG, Veiopoulou C, et al. The Th1/Th2 cytokine balance changes with

the progress of the immunopathological lesion of Sjogren’s syndrome. Clin Exp Immunol
2002;128:562–8.

[55] Barabino S, Dana MR. Animal models of dry eye: a critical assessment of opportunities and

limitations. Invest Ophthalmol Vis Sci 2004;45:1641–6.

[56] Jabs DA, Enger C, Predergast RA. Murine models of Sjogren’s syndrome. Evolution of the

lacrimal gland inflammatory lesions. Invest Ophthalmol Vis Sci 1991;32:371–80.

[57] Jabs DA, Lee B, Whittum-Hudson J, et al. The role of Fas-Fas ligand-mediated apoptosis in

autoimmune lacrimal gland disease in MRL/MpJ mice. Invest Ophthalmol Vis Sci
2001;42:399–401.

[58] Jabs DA, Lee B, Prendergast RA. Role of T cells in the pathogenesis of autoimmune lacrimal

gland disease in MRL/Mp-lpr/lpr mice. Curr Eye Res 1997;16:909–16.

[59] Jabs DA, Burek CL, Hu Q, et al. Anti-CD4 monoclonal antibody therapy suppresses autoim-

mune disease in MRL/Mp-lpr/lpr mice. Cell Immunol 1992;141:496–507.

[60] Jabs DA, Prendergast RA, Rorer EM, et al. Cytokines in autoimmune lacrimal gland disease

in MRL/MpJ mice. Invest Ophthalmol Vis Sci 2001;42:2567–71.

[61] Jabs DA, Lee B, Whittum-Hudson JA, et al. Th1 versus Th2 immune responses in autoimmune

lacrimal gland disease in MRL/Mp mice. Invest Ophthalmol Vis Sci 2000;41:826–31.

[62] Akpek EK, Jabs DA, Gerrard HC, et al. Chemokines in autoimmune lacrimal gland disease

in MRL/MpJ mice. Invest Ophthalmol Vis Sci 2004;45:185–90.

[63] Jabs DA, Gerard HC, Wei Y, et al. Inflammatory mediators in autoimmune lacrimal gland

disease in MRL/MpJ mice. Invest Ophthalmol Vis Sci 2004;45:2293–8.

[64] Robinson CP, Yamachika S, Bounous DI, et al. A novel NOD-derived murine model of pri-

mary Sjogren’s syndrome. Arthritis Rheum 1998;41:150–6.

[65] Robinson CP, Cornelius J, Bounous DE, et al. Characterization of the changing lymphocyte

populations and cytokine expression in the exocrine tissues of autoimmune NOD mice.
Autoimmunity 1998;27:29–44.

[66] Gao J, Killedar S, Cornelius JG, et al. Sjogren’s syndrome in the NOD mouse model is an

interleukin-4 time-dependent, antibody isotype-specific autoimmune disease. J Autoimmun
2006;26:90–103.

[67] Rosignoli F, Roca V, Meiss R, et al. Defective signalling in salivary glands precedes the au-

toimmune response in the non-obese diabetic mouse model of sialadenitis. Clin Exp Immunol
2005;142:411–8.

[68] Humphreys-Beher MG, Yamachika S, Yamamoto H, et al. Salivary gland changes in the

NOD mouse model for Sjogren’s syndrome: is there a non-immune genetic trigger? Eur
J Morphol 1998;36:247–51.

[69] Humphreys-Beher MG, Peck AB. New concepts for the development of autoimmune exocrin-

opathy derived from studies with the NOD mouse model. Arch Oral Biol 1999;44:S21–5.

[70] Robinson CP, Brayer J, Yamachika S, et al. Transfer of human serum IgG to nonobese dia-

betic Igmu null mice reveals a role for autoantibodies in the loss of secretory function of exo-
crine tissues in Sjogren’s syndrome. Proc Natl Acad Sci U S A 1998;95:7538–43.

[71] Nguyen KH, Brayer J, Cha S, et al. Evidence for antimuscarinic acetylcholine receptor anti-

body-mediated secretory dysfunction in NOD mice. Arthritis Rheum 2000;43:2297–306.

[72] Nguyen C, Cornelius J, Singson E, et al. Role of complement and B lymphocytes in Sjogren’s

syndrome-like autoimmune exocrinopathy of NOD.B10-H2b mice. Mol Immunol 2006;43:
1332–9.

[73] Zouhkri D. Effect of inflammation on lacrimal gland function. Exp Eye Res 2006;82:

885–98.

267

KERATOCONJUNCTIVITIS SICCA IN THE DOG

[74] Gao J, Schwalb TA, Addeo JV, et al. The role of apoptosis in the pathogenesis of canine ker-

atoconjunctivitis sicca: the effect of topical Cyclosporin A therapy. Cornea 1998;17:
654–63.

[75] Zoukhri D, Kublin DL. Impaired neurotransmitter release from lacrimal and salivary gland

nerves of a murine model of Sjogren’s syndrome. Invest Ophthalmol Vis Sci 2001;2:
925–32.

[76] Papas AS, Fernandez MM, Castano RA, et al. Oral pilocarpine for symptomatic relief of dry

mouth and dry eyes in patients with Sjogren’s syndrome. Adv Exp Med Biol 1998;438:
973–8.

[77] Yamachika S, Nanni JM, Nguyen KH, et al. Excessive synthesis of matrix metalloproteinases

in exocrine tissues of NOD mouse models for Sjogren’s syndrome. J Rheumatol 1998;25:
2371–80.

[78] Fye KH, Terasaki PI, Michalski JP, et al. Relationship of HLA-Dw3 and HLA-B8 to Sjogren’s

syndrome. Arthritis Rheum 1978;21:337–42.

[79] Toda I, Sullivan BD, Rocha EM, et al. Impact of gender on exocrine gland inflammation in

mouse models of Sjogren’s syndrome. Exp Eye Res 1999;69:355–66.

[80] Porola P, Laine M, Virkki L, et al. The influence of sex steroids on Sjogren’s syndrome. Ann

N Y Acad Sci 2007;1108:426–32.

[81] Sullivan DA, Edwards JA. Androgen stimulation of lacrimal gland function in mouse models

of Sjogren’s syndrome. J Steroid Biochem Mol Biol 1997;60:237–45.

[82] Ono M, Rocha FJ, Sullivan DA. Immunocytochemical location and hormonal control of an-

drogen receptors in lacrimal tissues of the female MRL/Mp-lpr/lpr mouse model of Sjogren’s
syndrome. Exp Eye Res 1995;61:659–66.

[83] Sato H, Ariga EH, Sullivan DA. Impact of androgen therapy in Sjogren’s syndrome:

hormonal influence on lymphocyte populations and Ia expression in lacrimal glands of
MRL/Mp-lpr/lpr mice. Invest Ophthalmol Vis Sci 1992;33:2537–45.

[84] Toda I, Wickham LA, Sullivan DA. Gender and androgen treatment influence the expression

of proto-oncogenes and apoptotic factors in lacrimal and salivary tissues of MRL/lpr mice.
Clin Immunol Immunopathol 1998;86:59–71.

[85] Cha S, van Blockland SC, Versnel MA, et al. Abnormal organogenesis in salivary gland de-

velopment may initiate adult onset of autoimmune exocrinopathy. Exp Clin Immunogenet
2001;18:143–60.

[86] Wang Y, Virji AS, Howard P, et al. Detection of cleaved alpha-fodrin autoantigen in Sjog-

ren’s syndrome: apoptosis and co-localisation of cleaved alpha-fodrin with activated cas-
pase-3 and cleaved poly(ADP-ribose) polymerase (PARP) in labial salivary glands. Arch
Oral Biol 2006;51:558–66.

[87] Fujihara T, Fujita H, Tsubota K, et al. Preferential localization of CD8þ alpha E beta

7þ T cells around acinar epithelial cells with apoptosis in patients with Sjogren’s syn-
drome. J Immunol 1999;163:2226–35.

[88] Nagaraju K, Cox A, Casciola-Rosen L, et al. Novel fragments of the Sjogren’s syndrome

autoantigens alpha-fodrin and type 3 muscarinic acetylcholine receptor generated during
cytotoxic lymphocyte granule-induced cell death. Arthritis Rheum 2001;44:2376–86.

[89] Saegusa K, Ishimaru N, Yanagi K, et al. Prevention and induction of autoimmune exocrin-

opathy is dependent on pathogenic autoantigen cleavage in murine Sjogren’s syndrome.
J Immunol 2002;169:1050–7.

[90] Dawson LJ, Fox PC, Smith PM. Sjogren’s syndrome—the non-apoptotic model of glandular

hypofunction. Rheumatology (Oxford) 2006;45:792–8.

[91] Matzinger P. Tolerance, danger and the extended family. Annu Rev Immunol 1994;12:

991–1045.

[92] Bredberg A, Henriksson G, Larsson A, et al. Sjogren’s syndrome and the danger model.

Rheumatology (Oxford) 2005;44:965–70.

268

WILLIAMS

Immune-Mediated Canine and Feline
Keratitis

Stacy E. Andrew, DVM

Georgia Veterinary Specialists, 455 Abernathy Road NE, Atlanta, GA 30328, USA

he transparent cornea of dogs and cats is composed of five layers: an
outer stratified squamous epithelium, epithelial basement membrane,
stroma, Descemet’s membrane, and the inner endothelium

. The main

tissue components are collagen (types I, III, IV, V, VI, VII, VIII, and XII) and
glycans (laminin, fibronectin, hyaluronans, heparin sulfate, chondroitan 6-sul-
fate, chondroitan 4-sulfate, dermatan sulfate, tenascin, and P component)

The cornea is anatomically designed to be a clear structure without blood ves-
sels, pigment, or lymphatics. Normal cell components include a sparse
population of lymphocytes in the epithelium and occasional leukocytes in the
stroma

. The cornea is nerve rich with numerous sensory nerves found in

the superficial layers.

CORNEAL IMMUNITY

The cornea is usually described as having a limited immune response because
of the lack of lymphatics and blood vessels. In fact, it has been stated that the
cornea may even be considered an immune-privileged tissue

, particularly

the central cornea

. Corneal immune responses are inhibited by the presence

of transforming growth factor-beta, alpha-melanocyte–stimulating hormone,
and Fas ligand expression

. Because of this, corneal tissues are quite vulner-

able to extension of inflammation from surrounding tissues such as the
conjunctiva

When corneal disease occurs, the primary cause of tissue damage is more

often an immune reaction than an immune deficiency

. Immune responses

in the cornea are adaptive and there are corneal antigen-presenting cells
(APC) that infiltrate the cornea following an inflammatory stimulus as well
as a lesser number of APCs that reside in the normal cornea

. Corneal

APC regulation is mediated by vascular endothelial growth factor receptor-3
(VEGFR-3) signaling

, which is expressed on corneal dendritic cells

. Until

recently, it was believed that the corneal APCs were found only in peripheral
cornea and that the central cornea was an immune-privileged site. However, it
has been demonstrated that the central cornea also has its own population of
epithelial Langerhans cells and anterior stromal precursor dendritic cells

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.11.007

Vet Clin Small Anim 38 (2008) 269–290

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

[5–7]

. When corneal inflammation occurs, the resident dendritic cells mature

by expressing major histocompatibility complex class II antigens, as well as
B7 (CD80/CD86) and CD40 costimulatory molecules

[5,6]

. These new data

may cause a rethinking of the corneal immune privilege, as it appears that
the cornea is capable of mounting immune responses

There are also other molecules that are expressed in normal corneas of dogs

and possibly cats that may be related to the inflammatory response. Nerve
growth factor and its receptor (trkA) have been detected in normal canine
corneal epithelium

. Cyclooxygenase-2 (COX-2), which has been shown

to be overexpressed at inflammatory sites and causes formation of prostanoids,
has been detected in normal as well as inflamed canine corneas using immuno-
histochemistry

CHRONIC SUPERFICIAL KERATITIS (PANNUS)
Clinical Presentation

Chronic superficial keratitis (CSK) typically presents as a pink, vascularized
lesion in the anterior stroma, near or at the limbus in the lateral quadrant of
the cornea (

). In the largest study to date, the lesions were found in the

temporal quadrant (96.80%), nasal quadrant (55.40%), inferior quadrant
(44.25%), and superior quadrant (20.40%) at initial presentation

. Initially

lesions may be unilateral, but in most cases (93% in one study

) it is a bilat-

eral although often asymmetric lesion. The tissuelike growth and pigmentation
progress toward the center from the primary quadrant(s) (

). Sometimes

a white area infiltrating into the clear cornea at the leading edge of the lesion
can be appreciated, which has been shown to be composed of CD4-expressing
lymphocytes

. Chronically, there is pigment infiltration of the cornea and of

the adjacent bulbar conjunctiva

. The lesions are usually progressive and

can cover the entire cornea (

), resulting in blindness. Depigmentation

Fig. 1. Left eye of a 6-year-old German shepherd dog with chronic superficial keratitis
advancing from the dorsolateral aspect of the cornea.

270

ANDREW

and thickening of the leading edge of the nictitating membrane can also be seen
in some dogs

[10,13]

CSK is not ulcerative so affected dogs are not in pain. Owners are usually

not aware of the condition until there is development of the pink fleshy (gran-
ulation) tissue or pigmentation that follows vascular ingrowth.

Diagnosis

The diagnosis of CSK is made based on clinical findings and breed predispo-
sition as well as response to treatment. Keratoconjunctivitis sicca (KCS),

Fig. 2. Right eye of a 3-year-old greyhound with early, lateral chronic superficial keratitis.
Note the increased perfusion of conjunctival vessels.

Fig. 3. (A) Profound right eye chronic superficial keratitis in a German shepherd dog. Both
eyes were affected similarly and the dog was visually impaired. (B) The same eye as
Fig. 3A following treatment for 4 weeks with topical prednisolone. The lesion is much improved
although there is quite a bit of haziness to the cornea as well as mild lateral pigment.

271

IMMUNE-MEDIATED CANINE AND FELINE KERATITIS

irritation caused by eyelash or eyelid disease, as well as granulation tissue due
to chronic corneal ulceration must be ruled out.

There is a distinct breed predisposition to CSK, with most cases being

reported in the German shepherd dog. In one study, 82% of cases were
German shepherd dogs

. However, it can occur in any breed and has

been reported in the following breeds: akita, Australian shepherd, Belgian ter-
vuren, border collie, bull mastiff, collie, dachshund, dalmatian, golden
retriever, greyhound, Labrador retriever, mixed breed, poodle, Shetland sheep-
dog, Siberian husky, viszla, and weimaraner

[10,13–17]

. Most animals are 3 to

6 years old

; however, the mean age of greyhounds is significantly lower

than other breeds

. Most believe there is no sex predilection, although

one study

found the prevalence to be increased in females. There is also

an increased risk for development of CSK in dogs at high altitude. Dogs that
reside at more than 7000 feet above sea level were shown to be 7.75 times
more likely to be diagnosed with CSK in one study

. In the largest study

of CSK, 95.3% of affected animals lived more than 4500 feet above sea level

. Therefore, ultraviolet exposure is thought to play a role in the

pathogenesis.

Laboratory Testing

Schirmer tear testing is recommended in cases of CSK to ensure that there is
not concomitant KCS, which is similarly characterized by corneal vasculariza-
tion, pigmentation, and granulation tissue. Fluorescein staining of the cornea is
also recommended before beginning therapy, as the syndrome is not usually
ulcerative and corneal ulceration could require a change in therapy. Cytology
will reveal increased numbers of plasma cells, lymphocytes, and mast cells

[13,18]

Histologically, there is mild to severe hyperplasia of the corneal epithelium

and a thickening of the epithelial cell basement membrane

. In early lesions,

there are increased numbers of lymphocytes and plasma cells and in more
chronic lesions there are also melanocytes and histiocytes

. In another

study, CD4-expresing lymphocytes predominated early with increased num-
bers of the fibrocytes and fibroblasts normally seen in the cornea, while
more chronic lesions demonstrated plasma cells, macrophages, and neutrophils

. There is also an increase in the diameter of collagen fibrils in affected

dogs, associated with a reduction in sulfated glycosaminoglycans, which may
be a result of lymphocytic activity in the extracellular matrix

Treatment

Owners must be made aware that CSK is controllable but not curable. Many
potential therapies have been reported, and the most commonly employed
treatment is with a topical immune-modulatory medication. Most authors rec-
ommend either 1% prednisolone acetate four times a day, 0.1% dexamethasone
three times a day

, or cyclosporine twice a day

[17,20]

. Frequent treatment

is often needed for 3 to 4 weeks, at which time the animal should be reex-
amined to determine if tapering of therapy is indicated. Some animals will

272

ANDREW

have fairly long periods between episodes of active disease and others must be
maintained on continuous immune-modulatory therapy. One study compared
topical dexamethasone three times daily and cyclosporine twice daily treatment
and found that the drugs at these dosage frequencies were equally effective

A recent experimental study evaluated the use of 1% pimecrolimus twice daily
and found a positive response in 4 (66.7%) of 6 dogs with CSK

. Subcon-

junctival corticosteroids in additional to topical therapy have also been used in
certain cases

[10,18,22]

More invasive treatments such as superficial keratectomy

[10,11,18]

, stron-

tium-90 irradiation

[10,23]

, cryosurgery

, or lamellar corneoscleral transpo-

sition

have been used successfully in cases that are unresponsive to topical

corticosteroid therapy. It is also recommended to use UV protection such as
tinted goggles for affected dogs.

Pathogenesis and Immunology

The definitive cause of CSK has yet to be determined; however, it appears to
be an immune-mediated disease condition based on the clinical signs, histopa-
thology, immunohistology, and response to topical steroid treatment. Conflict-
ing results as to the possibility of cellular response to corneal antigens have
been reported

[15,26]

. Immunohistochemical studies have shown that there

is immunoglobulin present in the adjacent conjunctiva (12/14, 86%) but far
less frequently in the cornea (2/14, 14%)

. CSK inflammatory lesions

involve stromal infiltration by CD4þ T lymphocytes with the majority contain-
ing the cytokine gamma interferon, as well as a smaller number of CD8þ lym-
phocytes

. One of the hallmarks of ocular autoimmunity is the presence of

CD4þ lymphocytes

. The majority of cells in the lesions have been deter-

mined to be CD4þ lymphocytes based on studies of CD4 and CD5 antigen
expression

. There is also an increased expression of major histocompat-

ability complex (MHC) class II in the corneas of CSK dogs, which may be par-
tially responsible for the continuing inflammation with this disease or may be
a result of gamma interferon production by the helper T cells populating the
lesions

It is difficult to explain the relationship of CSK with high altitude and con-

sequent increased ultraviolet light exposure. It is known that ultraviolet light
emitting sources with different levels of UVA and UVB have differential effects
on the modulation of immunoregulatory molecules such as interleukin-6,
tumor necrosis factor alpha, and interferon gamma

. There may also be

interactions between various light wavelengths

and possibly with tissues

exposed to the wavelengths.

SUPERFICIAL PUNCTATE KERATITIS
Clinical Presentation

Superficial punctate keratitis is a presumed immune-mediated syndrome that
presents as diffuse, multiple, punctate, corneal epithelial defects or superficial
corneal opacities. The lesions may or may not be ulcerated (

), and

273

IMMUNE-MEDIATED CANINE AND FELINE KERATITIS

patients are usually affected bilaterally although not necessarily symmetrically.
Lesions progress in depth over time and there may or may not be vessel in-
growth (

. The problem is seen almost exclusively in dachshunds,

and the longhaired dachshund has been reported to be affected most com-
monly

. Affected dogs are usually presented for ocular discomfort that is

manifested as squinting and tearing. Chronically, diffuse pigmentation and opa-
cification result (

Fig. 4. Superficial punctate keratitis in the right eye of a young miniature dachshund. The mul-
tifocal erosions were fluorescein negative but Rose Bengal positive. Superficial neovasculariza-
tion to each punctate lesion from the limbus eventually developed. (Courtesy of David T.
Ramsey, DVM, Williamstown, MI.)

Fig. 5. Superficial vascularization from the dorsal limbus to multifocal superficial ulcers in the
right eye of a miniature dachshund. (Courtesy of Dennis E. Brooks, DVM, PhD, Gainseville, FL.)

274

ANDREW

Diagnosis

The diagnosis is based on breed as well as clinical signs of diffuse, bilateral, su-
perficial corneal opacities that may be ulcerated. KCS and CSK are the main
differential diagnoses

Laboratory Testing

Schirmer tear testing is indicated to rule out KCS as the cause of ulceration and
pigmentation. Fluorescein staining should also be performed to determine the
depth and extent of corneal ulcerations. Rose Bengal staining may also be of
use if there is no staining with fluorescein in cases where the epithelial erosion
does not reach as deep as the corneal stroma.

Treatment

Although generally not used in corneal ulceration, this is a situation in which
topical steroid administration can be valuable. Neomycin-polymyxin B-dexa-
methasone or 1% prednisolone acetate in conjunction with a broad-spectrum
topical antibiotic is used four times daily in active ulcer cases. We assume
that the deleterious side effects of corticosteroids on epithelialization of the
ulcers is less important than the immunosuppression of disease

. Some

clinicians also favor the use of cyclosporine

in either the 0.2% ointment

or 1% to 2% in oil.

Pathogenesis and Immunology

The etiology of superficial punctate keratitis is unknown. This appears to be an
immune-mediated phenomenon based on the rapid response to topical immu-
nosuppressive therapy. There have been no known etiologic, histopathologic,
or immunohistochemical studies of this condition. One group has suggested
that herpesvirus may play a role in the pathogenesis

based on some ulcers

Fig. 6. Chronic inactive superficial punctate keratitis. Note the vascularization, lateral pig-
ment deposition, and overall cloudy appearance to the corneal epithelium and stroma.
(Courtesy of Dennis E. Brooks, DVM, PhD, Gainseville, FL.)

275

IMMUNE-MEDIATED CANINE AND FELINE KERATITIS

being linear. Canine herpesvirus has not been investigated but animals improve
without antiviral treatment.

CANINE ADENOVIRUS ENDOTHELIITIS
Clinical Presentation

Canine adenovirus (CAV) type-1 is a DNA virus in the family Adenoviridae
and is the cause of infectious canine hepatitis (ICH). CAV-1 has been deter-
mined to have systemic, renal, liver, and ocular symptoms, while CAV-2
causes mostly respiratory signs

. Natural systemic infection most com-

monly occurs in young, unvaccinated dogs and may be fatal. Clinical signs in-
clude depression, anorexia, recumbency, jaundice, petechiae, and corneal
opacification

Ocular clinical signs include conjunctival hyperemia that is related to mild

anterior uveitis, with corneal edema that develops 1 to 3 weeks following
vaccination

. Mild anterior uveitis can be detected within 1 week of

infection, and usually goes undetected by pet owners

. Very astute owners

and clinicians may notice mild conjunctival hyperemia, slight aqueous flare,
and possibly pupil constriction. The hallmark ocular lesion in both postvacci-
nation and naturally infected cases is corneal edema, and affected animals pres-
ent with a blue-white to completely opaque-appearing cornea (

Depending on the amount of corneal edema, visualization of moderate to
severe uveitis may be detected

Natural infection results in ocular disease approximately 20% of the time, as

well as in 0.4% of animals following modified live vaccine administration

The incidence is much lower now since CAV-1 modified live vaccine is no

Fig. 7. Diffuse edema in the right eye of an Afghan hound with canine adenovirus vaccine–
induced corneal changes. Intraocular structures cannot be visualized.

276

ANDREW

longer used. The corneal stromal edema is bilateral in 10%

to 30%

cases and the detection of clinical signs in the contralateral eye may vary by up
to 4 days

. Edema begins at the limbus

and may remain focal or prog-

ress to complete edema. At this point, there is conjunctival hyperemia as well as
anterior uveitis that may not be visualized depending on the extent of the
edema. Corneal thickness increases up to three times normal, either focally
or diffusely, and keratic precipitates may also be present

. Resolution of

the edema may occur spontaneously starting peripherally

and usually be-

gins to disappear in 2 to 3 days

. In some cases, edema may take months to

resolve or may be permanent

Diagnosis

Clinical signs with history of vaccination or in unvaccinated and exposed ani-
mals should arouse suspicion for CAV-1 endotheliitis. Afghan hounds may
have an increased susceptibility to infection and therefore more lesions

Afghans have been suggested as being more prone to developing glaucoma
than other breeds

, but secondary glaucoma has also been reported more

frequently in greyhounds, Norwegian elkhounds, samoyeds, and Siberian
huskies

. Definitive diagnosis requires laboratory testing (see the following

paragraph). Tonometry is definitely recommended to help differentiate
between uveitis and glaucoma.

Laboratory Testing

Dogs naturally infected with ICH will have leukopenia (lymphopenia, neutro-
penia) in early stages of disease, and later demonstrate neutrophilia and lym-
phocytosis during recovery

. Other suggestive laboratory test findings

include transient increase in alpha-2 globulin and delayed increase in
gamma-globulin on serum protein electrophoresis, increases in liver enzyme
activity related to hepatic necrosis, bilirubinuria, proteinuria, coagulation
abnormalities suggestive of disseminated intravascular coagulation, and some-
times increased protein in cerebrospinal fluid samples

. Definitive diagnosis

of infection with CAV-1 is made via polymerase chain reaction (PCR) that
identifies viral DNA in infected tissue

, and because CAV-1 readily repli-

cates in cell cultures, it is fairly easy to isolate

. PCR and immunohisto-

chemistry have been used to detect CAV-1 in fixed liver sections of dogs
with infectious canine hepatitis

Treatment

CAV-1–associated corneal edema and anterior uveitis are generally self-limit-
ing

. If therapy is withheld, the clinical course is usually 2 to 4 weeks

and the edema often clears without treatment as a result of endothelial regen-
eration

. However, there can be persistent corneal edema because of endo-

thelial damage. Symptomatic treatment, including topical corticosteroids such
as 1% prednisolone acetate or 0.1% dexamethasone, is suggested for the uveitis
to prevent persistent edema, secondary glaucoma, and possible phthisis bulbi.
Up to 10% of affected dogs have been shown to have these complications

277

IMMUNE-MEDIATED CANINE AND FELINE KERATITIS

Topical hyperosmotics may be useful if corneal bullae are present but are gen-
erally not useful for treatment of the edema

Pathogenesis and Immunology

Natural infection with CAV-1 occurs following oronasal exposure and tonsil
localization

. Ocular lesions develop during viremia when the virus enters

the anterior chamber

. Uveitis begins 7 days following infection

and

signs are attributable to immune complex (Arthus type III) reaction that is de-
tected 10 days to 21 days post-vaccination or in the recovery stages of a natural
infection. The initial theory was that viral replication and growth occurred in
the corneal endothelium, causing edema, and viral particles were released into
the aqueous humor, causing uveitis

. It is now known that immune com-

plex deposition in uveal tissues and in the corneal endothelium is the underly-
ing mechanism for the ocular damage that occurs

[33,35]

. Viral antigen is

released from the endothelial cells resulting in immune complex formation in
the aqueous humor

. These are phagocytosed by neutrophils and macro-

phages

. Inflammatory and chemotactic factors are stimulated with resul-

tant lysosomal enzyme release

, which is directed at the corneal

endothelial viral antigens and results in corneal edema and endothelial damage

Virus has been isolated from the aqueous fluid during the initial mild ante-

rior uveitis and viral replication has been identified in corneal endothelial cells
using electron microscopy

. With increasing severity of uveitis and corneal

edema, anterior chamber inflammatory cells and membrane-bound viral-anti-
body complexes were found

. The immune complex deposition created

by experimental intravenous CAV-1 injection was similar to natural or sponta-
neous infection

. Viral-antibody immune complexes release neutrophil

chemotactic factors in the presence of complement, complement activation
occurs, immune complexes are phagocytosed, and leukocytes then release lyso-
somal enzymes

. IgG and CAV-1 antigen complexes have been identified in

the glomeruli of young dogs with naturally occurring CAV

and are prob-

ably a type IV hypersensitivity

. Both type III (glomerular) and type IV

(interstitial) hypersensitivity lesions occur with the renal form of CAV-1

Vaccination using CAV-2 in place of CAV-1 decreases the incidence of corneal
edema to less than 1%

[42]

. CAV-1 modified live vaccines are no longer avail-

able in the United States

or the United Kingdom

EOSINOPHILIC KERATITIS
Clinical Presentation

Eosinophilic keratitis (EK) is also termed proliferative keratoconjunctivitis and
it occurs almost exclusively in cats and rarely in horses

[43,44]

. The most com-

mon presentation is typically a pink-white vascularized mass starting laterally
or medially in the peripheral cornea (

Fig. 8

). In a series of 15 cats, the predom-

inant lesion was lateral (n ¼ 5), medial (n ¼ 4), dorsal (n ¼ 4), ventral (n ¼ 1),
or central (n ¼ 1)

. The third eyelid and/or bulbar conjunctiva may also be

278

ANDREW

involved. Lesions are more often unilateral but may be bilateral. This is a pro-
gressive keratopathy and clinical signs of pain or discharge are variable

Up to 24% of cats may have accompanying corneal ulceration

. There is

a wide age range and no definite breed predisposition

, although young

adult mixed breed cats tend to be overrepresented

Diagnosis

The clinical lesion of a white-pink, proliferative, gritty or granular, irregular,
tissue ingrowth or vascularized infiltrate is highly suggestive of EK. Differential
diagnoses include fungal keratitis, neoplasia, and foreign bodies

. Lesions

may be quite variable in size and chronically the lesion can cover the entire
cornea (

Fig. 9

). Definitive diagnosis is based on cytologic sampling of the lesion.

Laboratory Testing

Cytology samples are obtained following application of topical anesthetic agent
(0.5% proparacaine or 1% tetracaine). A Kimura platinum spatula or the handle
end of a #15 Bard-Parker blade is used to scrape the lesion and samples are
placed on glass slides and air dried. Eosinophils, lymphocytes, neutrophils,
and mast cells are typically identified on cytologic samples (

Fig. 10

)

. A

study compared cytology to histopathology and found the cell types were sim-
ilar but exocytosis of mast cells and eosinophils, basement membrane thicken-
ing, sub-basal cleft formation, and corneal excrescences were recognized on
histopathologic examination

. It was theorized that sloughing of the excres-

cences may be responsible for the white nodules seen in EK

Fig. 8. (A) Typical eosinophilic keratitis lesion in the right eye of a 4-year-old domestic short-
hair cat. Note the pink infiltrate invading the cornea from the dorsolateral aspect as well as the
raised white accretions near the leading edge. (B) The same eye as Fig. 8A after 2 weeks of
three times daily topical corticosteroid treatment. While still present, the lesion is resolving with
less hyperemic conjunctiva, absence of white concretions and subjective thinning of corneal
vascularization.

279

IMMUNE-MEDIATED CANINE AND FELINE KERATITIS

Schirmer tear testing was also performed on most cats (46/54, 85%) in the

larger study of EK and was found to not be statistically different between af-
fected and normal eyes

. Various authors have speculated on the relation-

ship between EK and feline herpesvirus-1 (FHV-1). A PCR study found 45
(76.3%) of 59 corneal scrapings in EK cases to be positive for FHV-1, which
strongly suggests that FHV-1 may be related to the pathogenesis of EK

Fig. 9. Proliferative, irregular tissue ingrowth in the left eye of a 9-year-old domestic shorthair.
The pupil is visible ventrally but the lesion is affecting most of the cornea.

Fig. 10. Cytology sample from a corneal scraping of a cat with eosinophilic keratitis. Eosin-
ophils, mast cells, and neutrophils are seen.

280

ANDREW

Treatment

EK generally responds well to topical corticosteroid administration. Predniso-
lone acetate 1.0% or 0.1% dexamethasone starting four times a day and grad-
ually weaning over the course of several weeks to one to two times daily will
usually result in resolution of lesions. In one study, 31 (57.4%) of 54 cats were
treated with topical steroids alone, and only 3 failed to improve long term

Additional therapies include subconjunctival triamcinolone (4 mg)

[46,49]

, top-

ical cyclosporine (0.2% to 1.0%)

or, last, oral megesterol acetate (5 mg/day

initially then 2.5 to 5.0 mg once weekly for control)

[46,49–51]

, which should

be used with great caution because of potential adverse side effects. Therapy
can be discontinued following regression of lesions, but recurrence is common
(65.5%) with long-term follow-up

Pathogenesis and Immunology

The etiology and pathogenesis of EK have yet to be determined. It is a chronic,
progressive lesion that tends to recur despite treatment. Based on the cell types
present in the lesion, it is most likely a type I hypersensitivity (IgE, degranulat-
ing mast cells, tissue injury) or a type IV hypersensitivity (T cells, interleuki-
n[IL]-5 production, eosinophil stimulation, and related injury) reaction

If FHV-1 plays a role, then direct cytolysis and/or an immune response medi-
ated by T cells may also be occurring

Eosinophils in ocular tissues usually indicate a host response to environmen-

tal or parasite allergens

. Thus far, parasites have not been detected in cy-

tologic or histopathologic studies of EK. IL-4 and IL-13 have a prominent role
in regulating the expression of intercellular cell adhesion molecule-1 and the de-
pendent recruitment of eosinophils to the cornea in Onchocerca volvulus keratitis
in humans

. The role of these immune factors has not been investigated in

cats.

FELINE HERPESVIRUS–RELATED DISEASE
Clinical Presentation

Feline herpesvirus type 1 (FHV-1) is a DNA alpha-herpesvirus that damages
mucosal epithelial cells during replication and is the causative agent of viral rhi-
notracheitis

. The natural infection routes for FHV-1 are oral, nasal, and

conjunctival. Infection is typically through direct contact with an infected cat.
Indirect infections can also occur via contaminated cages, utensils, and person-
nel. Infection and carrier states are fairly widespread in the general cat popula-
tion, despite vaccination, with a higher prevalence in multicat households and
catteries

. Following infection, approximately 80% of susceptible animals

become FHV-1 carriers, and 45% of those spontaneously reactivate and either
asymptomatically shed or develop clinical disease manifestations

. Many

ocular diseases are directly caused by or are thought to be related to FHV-1
infection including conjunctivitis, corneal ulceration, stromal keratitis, ophthal-
mia neonatorium, corneal sequestrum, symblepharon, EK, KCS, and anterior
uveitis

281

IMMUNE-MEDIATED CANINE AND FELINE KERATITIS

Most cats that present with FHV-1–related ocular disease have ulcerative

keratitis and/or conjunctivitis. Infection and replication of FHV-1 with second-
ary bacterial infection result in conjunctivitis. After a short (2- to 6-day) incuba-
tion, clinical signs of serous ocular and nasal discharge, sneezing, decreased
appetite, and fever are noted

. FHV-1 conjunctivitis is usually bilateral,

and presents as hyperemia with serous progressing to mucopurulent discharge.
In adult cats that have had reactivation of latent virus, corneal ulceration is
more frequently noted. Branching or dendritic ulcers are the earliest type of
corneal ulcer noted, as the virus damages epithelium and the basement mem-
brane along its path from the trigeminal ganglion

. Clinical signs with

corneal ulcers depend on the depth and as well as chronicity of the lesion.
Conjunctivitis, blepharospasm, and discharge are often seen acutely, whereas
stromal edema and vascularization are noted more chronically. Stromal kerati-
tis refers to infection and inflammation of the deeper corneal tissue, and is a less
common manifestation of FHV-1 but significantly vision-threatening (

Fig. 11

Stromal FHV-1 keratitis may result from chronic recurrent episodes of keratitis
causing stromal collagen damage and opacification

. It does not appear to

be a manifestation of primary FHV-1 infection as the epithelium must be absent
for a prolonged period of time before the occurrence of stromal keratitis

KCS can also occur in cats that have chronic or recurrent conjunctivitis.

Clinical signs can include hyperemia of the conjunctiva, a dry appearance to
the cornea, as well as corneal ulceration and epithelial hyperplasia. In kittens,
mucopurulent conjunctivitis can cause distension of the eyelids (ophthalmia
neonatorium) if FHV-1 infection occurs before the eyelids open at 10 to 14

Fig. 11. Stromal keratitis in the left eye of a middle-aged domestic shorthair with chronic, re-
current feline herpesvirus-1 infection. Vascularization and opacification of the corneal stroma
have resulted.

282

ANDREW

days. Symblepharon (adhesion of the conjunctiva to the cornea or to itself) is
not uncommon in young animals with historical FHV-1 infection, and is likely
caused by conjunctival epithelial cell necrosis

Corneal sequestra are also theoretically related to FHV-1, as they are most

likely to form following corneal ulceration and stromal damage. Sequestra
appear as a tan to dark brown pigment deposition or plaque that usually occurs
centrally and at varying depths in the cornea. Cats with sequestra may or may
not be painful (blepharospastic, epiphora) and the main presenting clinical sign
is the colored lesion. EK has been described.

Diagnosis

A thorough physical and ophthalmic examination on any cat suspected of
having FHV-1 infection should be performed. An acute FHV-1 infection is
suspected based on ocular and respiratory clinical signs, but laboratory testing
can be used to confirm the diagnosis. Staining with Rose Bengal may help to
identify early dendritic lesions before the epithelium is disrupted and fluorescein
retention will occur. If immunofluorescent antibody testing is going to be
performed, samples must be obtained before instilling topical fluorescein dye.
Clinical suspicion and response to therapy are typically the mainstays in the
‘‘diagnosis’’ of FHV-1 keratitis cases. Definitive diagnosis requires PCR testing
with a reliable lab, and results must be interpreted in light of clinical signs.

The diagnosis of FHV-1 KCS is based on clinical signs and Schirmer tear test

measurement (normal is 17.0 5.7 mm wetting/60 seconds

), and KCS is

defined as tear production of less than 5 mm wetting/60 seconds. Laboratory
testing can be used to help determine if FHV-1 is associated with corneal
sequestra or EK but is not often used with ophthalmia neonatorium or
symblepharon.

Laboratory Testing

There are numerous laboratory tests that can be used to support the suspicion
of FHV-1 being related to corneal diseases in cats. A study by Maggs and col-
leagues

[60]

compared virus isolation (VI), immunofluorescent antibody (IFA),

serum neutralization, and ELISA in normal cats, those with upper respiratory
tract infections, and those with chronic ocular disease and found that none of
the testing methods were of more than limited value diagnostically. PCR and
nested PCR have met with more success in diagnosing FHV-1 in affected cats.
Various PCR methods have identified FHV-1 DNA in conjunctivitis (27/50,
54.0%

and 3/7, 43.0%

), corneal sequestra (5/28, 18.0%

; 3/12,

25.0%

; and 86/156, 55.1%

), EK (45/59, 76.3%

), epithelial keratitis

(5/6, 83.0%

), stromal keratitis (3/11, 27.0%

), and normal corneas (6/13,

46.0%

; 1/17, 5.9%

[48,62]

A recent paper evaluated PCR testing with six assays and found that they

were all equally likely to detect vaccine virus as a wild-type virus and there
was a high variability in detection rates (29% to 86%)

. Normal animals

can shed FHV-1 and thus clinicians should remember that a positive result
may not correlate with a high sensitivity in testing

. Many commercial

283

IMMUNE-MEDIATED CANINE AND FELINE KERATITIS

and university laboratories offer PCR testing for FHV-1 and clinicians should
ask about detection rates and sensitivity before using testing. The number of
cycles used for PCR most likely has the highest impact on reliability consider-
ing the large number of variables in testing

. It is also important to remem-

ber that a PCR-positive test result does not necessarily mean that an active
infection is occurring at the time of sample collection, only that the animal
was infected at some point in time

Treatment

Diagnosis and treatment of FHV-1–related corneal disease can be difficult, frus-
trating, and expensive. Owners must be made aware of this from the onset, and
they should be educated that recurrences are common in infected animals.
There are multiple treatment modalities for FHV-1–related corneal diseases.
Viral conjunctivitis cases are typically self-limiting with resolution in 7 to 14
days; however, topical antibiotic treatment with either oxytetracycline or eryth-
romycin TID for 7 days is recommended to treat secondary infections caused
by Mycoplasma or Chlamydophila. Experimentally, oral L-lysine has been shown
to decrease the severity of conjunctivitis

. It is recommended that adult

cats receive 500 mg oral L-lysine in food twice daily and kittens should receive
250 mg in food twice daily

. Chronic conjunctivitis may require treatment

with a topical or oral antiviral medication as described in the following
paragraphs.

FHV-1 corneal ulceration and stromal keratitis generally require more dili-

gent treatment including antiviral medication(s), and response to therapy can
be variable as well as unpredictable. Epithelial keratitis, such as dendritic or
geographic ulcers, usually has a better prognosis than stromal keratitis. Antivi-
ral medications are static not cidal, and consequently frequent topical adminis-
tration is recommended. The in vitro efficacy of topical antivirals against
FHV-1 is trifluridine > idoxuridine > vidarabine > bromovinyldeoxyuridine >
acyclovir

. Trifluridine or trifluorothymidine should be administered

four to six times daily for 2 days, and then the frequency is gradually reduced
over the next 14 to 21 days. Cidofivir, a nucleoside analog of deoxycytidine
monophosphate, at a recommended concentration of 0.02 mg/mL, has been
shown to be highly efficacious against FHV-1 in cell culture

. Clinically,

cidofivir 0.5% is recommended for topical use twice daily for 14 days.

Systemic antiviral drugs have also been experimentally investigated and are

used clinically in cats with keratitis and, sometimes, chronic conjunctivitis. Acy-
clovir, while effective against human herpesvirus, does not reach effective
plasma concentration in cats

. Valacyclovir has a higher bioavailability

but is extremely toxic to cats and its use is not recommended

. Although

there are no known experimental studies evaluating the safety or efficacy of
famciclovir, which is the oral prodrug of penciclovir, it has been used empiri-
cally at a dose of 125 to 250 mg orally per cat once daily for 2 to 4 weeks with
no known adverse side effects and impressive therapeutic benefits. Famciclovir
has been shown to reduce severity of corneal lesions, decrease the number of

284

ANDREW

trigeminal ganglion HSV-1 genomes, and improve survival in a rabbit model

. It is the author’s opinion that oral famciclovir is particularly useful in cases

of recurrent ulceration, chronic conjunctivitis, and stromal keratitis.

Interferons (IFNs) are regulatory proteins that have many activities including

defense against viral infection

[70]

. Topical and oral IFN therapies have been

used with success in cats with FHV-1 keratitis. IFN-alpha did not have toxic
effects and was efficacious against FHV-1 in cell culture, and was recommen-
ded at a concentration of 105 IU/mL

. A synergistic effect has also been

noted when cats were pretreated with recombinant human IFN-alpha before
acyclovir administration

[72]

. A recent study compared human IFN-alpha-2b

with feline recombinant IFN-omega and found that the antiviral effect of
IFN-omega was better than IFN-alpha-2b on cell culture FHV-1

[73]

. Topical

and oral administration of recombinant feline IFN-omega take advantage of
mucosal tolerance to prevent neutralizing antibody formation to IFN and it
was determined that the activity in conjunctival cells and white blood cells
was dependent on the dose administered, route of administration (topical, sys-
temic), and the cell type evaluated (conjunctival, corneal)

Therapies for the other corneal manifestations of FHV-1–mediated disease

include topical cyclosporine A (Optimmune 0.02%) for the treatment of
KCS, although the safety and efficacy have not been evaluated in cats. There
is also a slight trepidation about topical cyclosporine use in a cat with FHV-1,
because of local immunosuppressive effects and possible enhancement of latent
infection.

Corticosteroids have known local immunosuppressive effects as well as caus-

ing retardation of corneal epithelialization. Consequently, topical corticoste-
roids are rarely used with FHV-1–related ocular disease. The general
exceptions to this rule are EK and possibly chronic stromal keratitis. Stromal
keratitis is most likely a sequelae to viral immune reaction (see the next section)
and there is a possible indication for topical corticosteroid administration to de-
crease the FHV-1 antigenic immune response and conceivably decrease corneal
scar size

. However, clinicians must monitor such cases extremely fre-

quently as FHV-1 infection may be intensified. Topical nonsteroidal anti-
inflammatory therapy (flurbiprofen twice a day to three times a day) or
cyclosporine twice daily may be a better option in most cases of stromal
keratitis if anti-inflammatory therapy is deemed necessary.

Pathogenesis and Immunology

Feline herpesvirus-1 is a well-studied disease complex that has many similarities
with herpes simplex virus (HSV)-1 in humans. Because HSV-1 stromal keratitis
is one of the leading causes of blindness in humans, there has been a great deal
of research trying to elucidate the mechanisms of disease. HSV-1 is considered
a disease continuum

and it is wise to approach FHV-1 in a similar manner.

After cats are infected with FHV-1, approximately 80% become carriers, and
up to 45% of those will spontaneously reactivate and either shed virus or de-
velop clinical disease

. FHV-1 is an alpha-herpesvirus, which means that

285

IMMUNE-MEDIATED CANINE AND FELINE KERATITIS

recovered cats become latently infected carriers, and the primary site of pre-
sumed lifelong latency is the trigeminal ganglion

. Virus reactivation, par-

ticularly after a stress, may also occur

Conjunctivitis during primary infection is typically the initial lesion in most

affected cats due to preferential viral replication in conjunctival epithelium

Once latent virus becomes reactivated, it travels from the trigeminal ganglion
to corneal nerves. Dendritic corneal ulcers resulting from direct cytopathic ef-
fect of the virus in the basal cell layer of the corneal epithelium

and sup-

pression of local immune responses allows FHV-1 to reach the corneal
stroma. The resulting keratitis is likely mediated by an immune response to vi-
ral antigen, which causes stromal damage independent of virus replication

Stromal keratitis is an infection or inflammation of the corneal stroma that is

at least partly related to an immunopathologic response to herpesvirus

The majority of stromal damage that occurs after HSV-1 infection is not due
to viral replication but an immune and inflammatory reaction to viral particles

. While the exact mechanism of disease is not known, chronic or recurrent

episodes of FHV-1 keratitis do result in collagen damage and opacification

In FHV-1 stromal keratitis samples evaluated histopathologically, neutrophils
are followed by B and T lymphocytes, which correlates with an antiviral im-
mune response

. Neutrophils are the predominant infiltrating cells that

are likely the source of cytokines (IL-6, IL-10, IL-12, IFN-alpha) in a mouse
model of HSV-1 stromal keratitis

. The responses are not definitively

T-helper (Th)1 or Th2, although a Th1 response is more likely based on the
finding of IFN-gamma–positive and IL-12–positive cells with almost no IL-4

. This results in a proinflammatory cytokine production response

Currently, there are three proposed mechanisms for the role of CD4þ T

cells in HSV-1 keratitis

. Bystander activation of the CD4þ T cells by cyto-

kines produced as a response to infection

[80,81]

, molecular mimicry by a viral

protein causing an autoimmune response to corneal tissue

, and virus-spe-

cific activation whereby the virus-specific CD4þ T cells are involved in induc-
tion and progression of viral disease

are all potential mechanisms for CD4þ

T cells in the role of FHV-1. Regardless of the cause of the chronic inflamma-
tion, it is at best a complex process

with many potential target systems for

future treatment.

References

[1] Samuelson DA. Ophthalmic anatomy. In: Gelatt KN, editor. Veterinary ophthalmology. 4th

edition. (IA): Blackwell Publishing; 2007. p. 37–148.

[2] Lepisto AJ, Frank GM, Hendricks RL. How herpes simplex virus type 1 rescinds corneal priv-

ilege. Chem Immunol Allergy 2007;92:203–12.

[3] Biros DJ. Ocular immunity. In: Gelatt KN, editor. Veterinary ophthalmology. 4th edition. (IA):

Blackwell Publishing; 2007. p. 223–35.

[4] Dana R. Corneal antigen presentation: molecular regulation and functional implications.

Ocul Surf 2005;3(Suppl 4):S169–72.

[5] Hamrath P, Dana MR. Corneal antigen-presenting cells. Chem Immunol Allergy 2007;92:

58–70.

286

ANDREW

[6] Hamrath P, Hug SO, Liu Y, et al. Corneal immunity is mediated by heterogeneous popula-

tions of antigen-presenting cells. J Leukoc Biol 2003;74(2):172–8.

[7] Williams DL. Major histocompatibility class II expression in the normal canine cornea and in

canine chronic superficial keratitis. Vet Ophthalmol 2005;8(6):395–400.

[8] Woo HM, Bentley E, Campbell SF, et al. Nerve growth factor and corneal wound healing in

dogs. Exp Eye Res 2005;80(5):633–42.

[9] Sellers RS, Silverman L, Khan KN. Cyclooxygenase-2 expression in the cornea of dogs with

keratitis. Vet Pathol 2004;41(2):116–21.

[10] Slatter DH, Lavach JD, Severin GA, et al. Uberreiter’s syndrome (chronic superficial keratitis)

in dogs in the Rocky Mountain area–a study of 463 cases. J Small Anim Pract 1977;18(12):
757–72.

[11] Bedford PGC, Lonstaffe JA. Corneal pannus (chronic superficial keratitis) in the German

Shepherd dog. J Small Anim Pract 1979;20:41–56.

[12] Williams DL. Histological and immunohistochemical evaluation of canine chronic superfi-

cial keratitis. Res Vet Sci 1999;67(2):191–5.

[13] Eichenbaum JD, Lavach JD, Gould DH, et al. Immunohistochemical staining patterns of ca-

nine eyes affected with chronic superficial keratitis. Am J Vet Res 1986;47(9):1952–5.

[14] Chavkin MJ, Roberts SM, Salman MD, et al. Risk factors for development of chronic super-

ficial keratitis in dogs. J Am Vet Med Assoc 1994;204(10):1630–4.

[15] Peiffer RL, Gelatt KN, Gwin RM. Chronic superficial keratitis (pannus) in related Grey-

hounds. Vet Med Small Anim Clin 1977;72(1):35–7.

[16] Rapp E, Kolbl S. Ultrastructural study of unidentified inclusions in the cornea and iridocor-

neal angle of dogs with pannus. Am J Vet Res 1995;56(6):779–85.

[17] Williams DL, Hoey AJ, Smitherman P. Comparison of topical cyclosporin and dexameth-

asone for the treatment of chronic superficial keratitis in dogs. Vet Rec 1995;137(25):
635–9.

[18] Stanley RG. Superficial stromal keratitis in the dog. Aust Vet J 1988;65(10):321–3.
[19] Williams DL, Cribb A, Scott JE. Proteoglycan-collagen interactions in chronic superficial ker-

atitis in the dog. Biochem Soc Trans 1991;19(4):353S.

[20] Jackson PA, Kaswan RL, Merideth RE, et al. Chronic superficial keratitis in dogs: a placebo con-

trolled trial of topical cyclosporine treatment. Veterinary & Comparative Ophthalmology
1991;1(4):269–75.

[21] Nell B, Walde I, Billich A, et al. The effect of topical pimecrolimus on keratoconjunctivitis

sicca and chronic superficial keratitis in dogs: results from an exploratory study. Vet Ophthal-
mol 2005;8(1):39–46.

[22] Austad R, Oen EO. Chronic superficial keratitis in the dog. II. Results of a combined non-

surgical treatment. J Small Anim Pract 1978;19:203–6.

[23] Harwell GM, Angell JA, Merideth RE, et al. Chronic superficial keratitis in a Mexican

wolf. J Am Vet Med Assoc 1985;187(11):1268.

[24] Holmberg DL, Scheifer HB, Parent J. The cryosurgical treatment of pigmentary keratitis in

dogs: an experimental and clinical study. Vet Surg 1986;15(1):1–4.

[25] Kuhns EL, Keller W, Blanchard GL. The treatment of pannus in dogs by use of a corneal-

scleral graft. J Am Vet Med Assoc 1973;162(11):950–2.

[26] Campbell LH, Okuda HI, Lipton DE, et al. Chronic superficial keratitis in dogs: detection of

cellular hypersensitivity. Am J Vet Res 1975;36(5):669–71.

[27] De Smet MD, Chan CC. Regulation of ocular inflammation—what experimental and human

studies have taught us. Prog Retin Eye Res 2001;20(6):761–97.

[28] Clingen PH, Berneburg M, Petit-Frere C, et al. Contrasting effects of an ultraviolet B and an

ultraviolet A tanning lamp on interleukin-6, tumour necrosis factor-alpha and intercellular
adhesion molecule-1 expression. Br J Dermatol 2001;145(1):54–62.

[29] Stades FC, Wyman M, Boeve MH, et al. Cornea and sclera. In: Ophthalmology for the vet-

erinary practitioner. Hannover (Germany): Schlutersche GmbH & Co.,; 1998. p. 107–27.

[30] Clerc B, Jegou J. Superficial punctate keratitis. Canine Pract 1996;21(3):6–11.

287

IMMUNE-MEDIATED CANINE AND FELINE KERATITIS

[31] Wright NG. Canine adenovirus: its role in renal and ocular disease: a review. J Small Anim

Pract 1976;17(1):85–95.

[32] Morrison WI, Nash AS, Wright NG. Glomerular deposition of immune complexes in dogs

following natural infection with canine adenovirus. Vet Rec 1975;96(24):522–4.

[33] Carmichael LE. The pathogenesis of ocular lesions of infectious canine hepatitis. I. Pathol-

ogy and virologic observations. Pathol Vet 1964;1:73–95.

[34] Curtis R, Barnett KC. The ‘blue eye’ phenomenon. Vet Rec 1983;112(15):347–53.
[35] Carmichael LE, Medic BL, Bistner SI, et al. Viral-antibody complexes in canine adenovirus

type 1 (CAV-1) ocular lesions: leukocyte chemotaxis and enzyme release. Cornell Vet
1975;65(3):331–51.

[36] Cullen CL, Webb AA. Ocular manifestations of systemic disease. Part 1: the dog. In:

Gelatt KN, editor. Veterinary ophthalmology. 4th edition. (IA): Blackwell Publishing;
2007. p. 1470–537.

[37] Curtis R, Barnett KC. Canine adenovirus-induced ocular lesions in the Afghan hound. Cor-

nell Vet 1981;71:85–95.

[38] Greene CE. Infectious canine hepatitis and canine acidophil cell hepatitis. In: Greene CE,

editor. Infectious diseases of the dog and cat. 3rd edition. St. Louis (MO): Saunders Elsevier;
2006. p. 41–7.

[39] Chouinard L, Martineau D, Forget C, et al. Use of polymerase chain reaction and immuno-

histochemistry for detection of canine adenovirus type 1 in formalin-fixed, paraffin-embed-
ded liver of dogs with chronic hepatitis or cirrhosis. J Vet Diagn Invest 1998;10(4):320–5.

[40] Curtis R, Barnett KC. The ocular lesions of infectious canine hepatitis. 1. Clinical features.

J Small Anim Pract 1973;14:375–89.

[41] Aguirre G, Carmichael L, Bistner S. Corneal endothelium in viral induced anterior uveitis.

Ultrastructural changes following canine adenovirus type 1 infection. Arch Ophthalmol
1975;93(3):219–24.

[42] Gilger BC. Diseases and surgery of the canine cornea and sclera. In: Gelatt KN, editor. Vet-

erinary ophthalmology. 4th edition. (IA): Blackwell Publishing; 2007. p. 690–752.

[43] Ramsey DT, Whiteley HE, Gerding PA, et al. Eosinophilic keratoconjunctivitis in a horse.

J Am Vet Med Assoc 1994;205(9):1308–11.

[44] Yamagata M, Wilkie DA, Gilger BC. Eosinophilic keratoconjunctivitis in seven horses. J Am

Vet Med Assoc 1996;209(7):1283–6.

[45] Paulsen ME, Lavach JD, Severin GA, et al. Feline eosinophilic keratitis: a review of 15 cases.

J Am Anim Hosp Assoc 1987;23(1):63–9.

[46] Morgan RV, Abrams KA, Kern TJ. Feline eosinophilic keratitis: a retrospective study of 54

cases: (1989–1994). Vet Comp Ophthalmol 1996;6(2):131–4.

[47] Prasse KW, Winston SM. Cytology and histopathology of feline eosinophilic keratitis.

Veterinary & Comparative Ophthalmology 1996;6(2):74–81.

[48] Nasisse MP, Glover TL, Moore CP, et al. Detection of feline herpesvirus 1 DNA in corneas of

cats with eosinophilic keratitis or corneal sequestration. Am J Vet Res 1998;59(7):856–8.

[49] Brightman AH, Vestre WA, Helper LC, et al. Chronic eosinophilic keratitis in the cat. Feline

Pract 1979;9(3):21–3.

[50] Martin CL. Feline ophthalmologic diseases. Mod Vet Pract 1982;63(2):115–22.
[51] Bedford PGC, Cotchin E. An unusual chronic keratoconjunctivitis in the cat. J Small Anim

Pract 1983;24(2):85–102.

[52] Berger RB, Blackwell NM, Lass JH, et al. IL-4 and IL-13 regulation of ICAM-1 expression and

eosinophil recruitment in Onchocerca volvulus keratitis. Invest Ophthalmol Vis Sci
2002;43(9):2992–7.

[53] Gaskell RM, Dawson S, Radford A. Feline respiratory disease. In: Greene CE, editor. Infec-

tious diseases of the dog and cat. 3rd edition. St. Louis (MO): Saunders Elsevier; 2006.
p. 145–58.

[54] Gaskell RM, Povey RC. Experimental induction of feline viral rhinotracheitis virus re-excre-

tion in FVR-recovered cats. Vet Rec 1977;100(7):128–33.

288

ANDREW

[55] Andrew SE. Ocular manifestations of feline herpesvirus. J Feline Med Surg 2001;3(1):

9–16.

[56] Liesegang TJ. Classification of herpes simplex virus keratitis and anterior uveitis. Cornea

1999;18(2):127–43.

[57] Nasisse MP. Feline herpesvirus ocular disease. Vet Clin North Am Small Anim Pract

1990;20(3):667–80.

[58] Nasisse MP, Guy JS, Davidson MG, et al. Experimental and ocular herpesvirus infection in

the cat: sites of virus replication, clinical features, and effects of corticosteroid administra-
tion. Invest Ophthalmol Vis Sci 1989;30(8):1758–68.

[59] Veith LA, Cure TH, Gelatt KN. The Schirmer tear test in cats. Mod Vet Pract 1970;51:48–9.
[60] Maggs DJ, Lappin MR, Reif JS, et al. Evaluation of serologic and viral detection methods for

diagnosing feline herpesvirus-1 infection in cats with acute respiratory tract or chronic ocu-
lar disease. J Am Vet Med Assoc 1999;214(4):502–7.

[61] Stiles J, McDermott M, Bigsby D, et al. Comparison of nested polymerase chain reaction to

identify feline herpesvirus in ocular tissue from clinically normal cats and cats with corneal
sequestra or conjunctivitis. Am J Vet Res 1997;58(4):338–42.

[62] Volopich S, Venetka V, Schwendenwein I, et al. Cytologic findings, and feline herpesvirus

DNA and Chlamydophila felis antigen detection rates in normal cats and cats with conjunc-
tival and corneal lesions. Vet Ophthalmol 2005;8(1):25–32.

[63] Maggs DJ, Clarke HE. Relative sensitivity of polymerase chain reaction assays used for de-

tection of feline herpesvirus type 1 DNA in clinical samples and commercial vaccines. Am
J Vet Res 2005;66(9):1550–5.

[64] Lutz H, Leutenegger C, Hofman-Lehmann R. The role of polymerase chain reaction and its

newer developments in feline medicine. J Feline Med Surg 1999;1(2):89–100.

[65] Stiles J, Townsend WM, Rogers QR, et al. Effect of oral administration of L-lysine on conjunc-

tivitis caused by feline herpesvirus in cats. Am J Vet Res 2002;63(1):99–103.

[66] Nasisse MP, Guy JS, Davidson MG, et al. In vitro susceptibility of feline herpesvirus-1 to vi-

darabine, idoxuridine, trifluridine, acyclovir, or bromovinyldeoxyuridine. Am J Vet Res
1989;50(1):158–60.

[67] Sandmeyer LS, Keller CB, Bienzle D. Effects of cidofivir on cell death and replication of

feline herpesvirus-1 in cultured feline corneal epithelial cells. Am J Vet Res 2005;66(2):
217–22.

[68] Nasisse MP, Dorman DC, Jamison KC, et al. Effects of valacyclovir in cats infected with feline

herpesvirus 1. Am J Vet Res 1997;58(10):1141–4.

[69] Loutsch JM, Sainz B, Marquart ME, et al. Effect of famciclovir on herpes simplex virus type 1

corneal disease and establishment of latency in rabbits. Anitmicrob Agents Chemother
2001;45(7):2044–53.

[70] Samuel CE. Antiviral actions of interferons. Clin Microbiol Rev 2001;14(4):778–809.
[71] Sandmeyer LS, Keller CB, Bienzle D. Effects of interferon-alpha on cytopathic changes and

titers for feline herpesvirus-1 in primary culture of feline corneal epithelial cells. Am J Vet Res
2005;66(2):210–6.

[72] Weiss RC. Synergistic antiviral activities of acyclovir and recombinant human leukocyte

(alpha) interferon on feline herpesvirus replication. Am J Vet Res 1989;50(10):
1673–7.

[73] Siebeck N, Hurley DJ, Garcia M, et al. Effects of human recombinant alpha-2b interferon

and feline recombinant omega interferon on in vitro replication of feline herpesvirus-1.
Am J Vet Res 2006;67(8):1406–11.

[74] Bracklein T, Theise S, Metzler A, et al. Activity of feline interferon-omega after ocular or oral

administration in cats as indicated by Mx protein expression in conjunctival and white blood
cells. Am J Vet Res 2006;67(6):1025–32.

[75] Gaskell R, Dawson S, Radford A, et al. Feline herpesvirus. Vet Res 2007;38(2):337–54.
[76] Roberts SR, Dawson CR, Coleman V, et al. Dendritic keratitis in a cat. J Am Vet Med Assoc

1972;161(3):285–9.

289

IMMUNE-MEDIATED CANINE AND FELINE KERATITIS

[77] Nasisse MP, English RV, Tompkins MB, et al. Immunologic, histologic, and virologic features

of herpesvirus-induced stromal keratitis in cats. Am J Vet Res 1995;56(1):51–5.

[78] Stumpf TH, Case R, Shimeld C, et al. Primary herpes simplex virus type 1 infection of the eye

triggers similar immune responses in the cornea and skin of the eyelids. J Gen Virol
2002;83(Pt 7):1579–90.

[79] Stumpf TH, Shimeld C, Easty DL, et al. Cytokine production in a murine model of recurrent

herpetic stromal keratitis. Invest Ophthalmol Vis Sci 2001;42(2):372–8.

[80] Gangappa S, Deshpande SP, Rouse BT. Bystander activation of CD4þ T cells accounts for

herpetic ocular lesions. Invest Ophthalmol Vis Sci 2000;41(2):453–9.

[81] Newell CK, Martin S, Sendele D, et al. Herpes simples virus-induced stromal keratitis: role of

T-lymphocyte subsets in immunopathology. J Virol 1989;63(2):769–75.

[82] Zhao ZA, Granucci F, Yeh L, et al. Molecular mimicry by herpes simplex virus-type 1: auto-

immune disease after viral infection. Science 1998;279(5355):1344–7.

290

ANDREW

Canine Episcleritis, Nodular
Episclerokeratitis, Scleritis,
and Necrotic Scleritis

Bruce H. Grahn, DVM*,
Lynne S. Sandmeyer, DVM, DVSc

Department of Small Animal Clinical Sciences, Western College of Veterinary Medicine,
University of Saskatchewan, 52 Campus Drive, Saskatoon, Saskatchewan, Canada S7N 5B4

everal inflammatory conditions develop within the episclera and sclera of
dogs, and most of these are assumed to be primary and immune medi-
ated

. This is based on the clinical response to immunosuppression;

the lymphocytic, plasmacytic, and macrophage cell infiltrates; and a lack of
detection of specific infectious etiologies with light microscopic and other labo-
ratory techniques. Some immune-mediated bilateral episcleritis disorders may
be inherited in specific breeds of dogs (ie, collies, American cocker spaniels,
golden retrievers)

[1–3]

. Episcleritis and scleritis in dogs also develops occasion-

ally secondary to some infections (Ehrlichia canis); parasitic infestations (Onchocerca
spp)

; and commonly related to ocular trauma (ocular surgery, injuries) and

penetrating foreign bodies

. The disorders with known etiologies are best

categorized as secondary episcleritis and scleritis.

To understand adequately the pathogenesis and clinical manifestations of

episcleritis and scleritis the anatomy and physiology of this outer ocular tunica
should be reviewed. The sclera in domestic animals covers approximately 80%
of the eye, and it is continuous with the cornea anteriorly. The posterior sclera
is penetrated by several posterior ciliary arteries, veins, and nerves, and the
optic nerve. The optic meninges are fused to the episclera at the posterior
pole of the globe around the optic nerve. The sclera is also penetrated around
the equator by several anterior ciliary arteries, veins, and nerves that enter and
exit the uvea. The junction of the cornea and sclera is termed the ‘‘limbus.’’
This region is commonly affected by the inflammatory response of the episclera
and scleral inflammation in dogs. Corneal edema and degeneration are fre-
quently noted in focal and diffuse episcleral and conjunctival hyperemia and
edema

. These manifestations are useful in establishing a clinical diagnosis

(

Figs. 1 and 2

). The sclera merges with the choroid at suprachoroid. With

*Corresponding author. E-mail address: bruce.grahn@usask.ca (B.H. Grahn).

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.11.003

Vet Clin Small Anim 38 (2008) 291–308

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

the exception of the penetrating arteries and veins, the sclera is relatively avas-
cular and receives most of its nutrients by diffusion from the vasculature of the
suprachoroid, and a loose fibrovascular net on its outer surface called ‘‘Tenon’s
capsule.’’ The sclera is rigid and helps maintain the globe as a sphere. Its major
functions include protection of the ocular contents, and provision of anchorage
for the extraocular muscles, bulbar conjunctiva, and the optic nerve and its
meninges. Because the sclera and episclera are intimately associated with the
choroid, bulbar conjunctiva, and orbital tissues, uveitis, conjunctivitis, and
orbital cellulitis result in some secondary inflammation. The focus of this arti-
cle, however is on primary and secondary inflammatory conditions that origi-
nate in the episclera and sclera because of idiopathic or known etiologies.

The classification of immune-mediated episcleritis and scleritis is somewhat

arbitrary, and is based on clinical and pathologic changes. Fischer

classified

episcleritis in dogs as simple, or nodular, and scleritis as superficial, or deep
nonnecrotizing, and necrotizing respectively. This is similar to classification
of episcleritis and scleritis in human patients. Yanoff and Fine

and Okhravi

and coworkers

categorize episcleritis as simple or nodular, and scleritis as

anterior diffuse, and anterior nodular, anterior necrotizing with inflammation,
anterior necrotizing without inflammation (scleromalacia perforans), and poste-
rior scleritis. Recent clinical and immunohistochemical investigations of
episcleritis in dogs have identified similarities and occasional significant differ-
ences in unilateral, bilateral, nodular, and diffuse forms of episcleritis in dogs

[8–10]

. Based on these studies and clinical and therapeutic variations, the

Fig. 1. The right eye of a mature neutered male American cocker spaniel with histologically
confirmed idiopathic unilateral episcleritis. Note the perilimbal bulbar conjunctival hyperemia,
and the rim of episcleral masses that extend around the limbus from 2 o’clock to 10 o’clock.
The adjacent cornea is vascularized and edematous.

292

GRAHN & SANDMEYER

authors categorize episcleritis in dogs as primary (idiopathic) immune-mediated
unilateral or bilateral focal or diffuse episcleritis, and bilateral nodular episcler-
okeratitis of collie-type dogs, and secondary episcleritis where the etiology is
readily discernable. Scleritis is categorized as idiopathic nonnecrotizing scleritis;
necrotic scleritis; and scleritis that develops secondary to trauma, known infec-
tious organisms, and surgery.

This article provides a brief review of human episcleritis and scleritis because

the pathogenesis, diagnosis, and treatment of these disorders are perhaps the
best understood of all species of animals. A review of the clinical and light
microscopic manifestations, therapeutic options, and prognosis for primary
and secondary episcleritis and scleritis in dogs is also provided.

EPISCLERITIS AND SCLERITIS OF HUMANS

Episcleritis is classified as diffuse or nodular. Scleritis may be divided into an-
terior scleritis and posterior scleritis based on location, and these can be further
classified as diffuse, nodular, necrotizing with inflammation, or necrotizing
without inflammation (scleromalacia perforans)

[11–13]

. The episclera and

sclera are also frequently affected secondary to inflammation originating in ad-
jacent structures, such as the conjunctiva, cornea, uvea, and orbit. Episcleritis
tends to be benign and self-limiting, whereas scleritis is a painful and a more
vision-threatening condition, which is often associated with systemic autoim-
mune disease.

Fig. 2. Nodular episclerokeratitis in a collie dog; note the inflamed masses on the eyelid mar-
gin, temporal limbus and cornea, and third eyelid. These lesions were bilateral and symmetric,
and they are often considered pathognomonic for nodular episclerokeratitis of collie-type dogs.

293

EPISCLERITIS AND SCLERITIS

Episcleritis

Episcleritis is inflammation confined to the superficial episcleral tissue. Episcler-
itis is a mild non–vision-threatening form of ocular inflammation that is idio-
pathic in nature and is not usually associated with involvement of other
ocular structures, although adjacent limbal corneal involvement may be seen

. Diffuse episcleritis is the most frequently seen clinical form

. It presents

as ill-defined, intense redness and edema with engorged superficial episcleral
vessels occurring most commonly in the interpalpebral region. The condition
is usually acute in onset with only mild ocular discomfort. Nodular episcleritis
presents as localized redness and edema with a 2- to 3-mm intraepiscleral
nodular elevation that is mobile over the underlying sclera. Occasionally, mul-
tiple nodules may form. The condition is usually gradual in onset with localized
tenderness. Both forms of episcleritis occur unilaterally most often, but can
present bilaterally. The histopathologic lesion of simple and nodular episcleritis
is usually a nongranulomatous inflammation, where lymphocytes, plasma cells,
and edema are found in the episcleral tissue, although rarely a chronic granu-
lomatous inflammatory infiltrate may be seen. Episcleritis is often self-limiting;
simple episcleritis may subside in 5 to 10 days and nodular episcleritis may dis-
sipate within 4 to 5 weeks without therapy

. Most patients with episcleritis

respond to topical nonsteroidal or steroidal anti-inflammatory medications that
are administered for symptomatic relief, although occasionally systemic nonste-
roidal anti-inflammatory drugs (NSAIDs) may be required. Both forms of epis-
cleritis may be recurrent with episodes over several years; however, only rarely
do they progress to scleritis

. Ocular complications of episcleritis are rare,

but can include anterior uveitis, glaucoma, and cataract formation

Scleritis

Scleritis is defined as inflammation of the sclera. It is typically a severe painful
inflammatory process centered in the sclera that may involve the cornea, adja-
cent episclera, and underlying uvea. It is a more severe disease than episcleritis,
and is more likely to lead to visual loss

. The characteristic feature of scleritis

is the severe pain that may involve the eye and orbit, and radiates to involve
the ear, scalp, face, and jaw

. Scleritis is more prevalent in middle-aged to

older women

. The diagnostic feature differentiating scleritis from epis-

cleritis is maximal involvement of the deep episcleral vascular plexus, which
is displaced outward by edematous swollen sclera. This displacement of the
deep episcleral vessels is only seen in scleritis

. The vascular engorgement

of the deep episcleral plexus in scleritis has a characteristic bluish violet hue.
A small area or the whole of the anterior segment of the eye may be involved

Anterior scleritis accounts for 90% of cases of scleritis and approximately

50% of anterior scleritis cases are bilateral. Diffuse anterior scleritis is the
most common form. Clinical signs are diffuse involvement of the anterior
sclera with edema and dilation of the deep episcleral vascular plexus. It may
be localized to a patch of the sclera, or may involve the entire sclera. The

294

GRAHN & SANDMEYER

underlying trabecular meshwork can be involved with trabeculitis, which in
addition to elevated episcleral venous pressure, may result in elevation of the
intraocular pressure

[7,16]

. Diffuse anterior scleritis may progress to other

types of scleritis. Nodular scleritis is characterized by more localized scleral
inflammation with nodule formation. Nodules may be single or multiple. In-
flammation occurs most often near the limbus and nodules are usually deep,
red, totally immobile, and separate from the overlying congested episcleral
tissues

Necrotizing scleritis is the most painful and destructive form of scleritis and

accounts for about 23% of cases

. Scleral involvement is characterized by

severe vasculitis and closure of the episcleral vascular bed such that there are
visible areas of capillary nonperfusion, infarction, and necrosis of the involved
sclera. Necrosis may be subtle or profound, localized or generalized, and may
progress rapidly to expose the uvea. Inflammation commonly spreads to
involve the cornea, ciliary body, and trabecular meshwork resulting in keratitis,
anterior uveitis, and elevated or depressed intraocular pressures, and may even
lead to staphyloma formation

[6,7,11]

Rarely, necrotizing scleritis may occur without signs of pain or inflammation

in patients with longstanding rheumatoid arthritis and is termed ‘‘scleromalacia
perforans’’

. Scleromalacia perforans accounts for only 3% of scleritis cases

. The onset is often insidious, and pain may be minimal or nonexistent.

It results from an obliterative arteritis involving the deep episcleral vascular
plexus

[15,17,18]

. Clinical manifestations include a white, avascular, thin patch

on the sclera without inflammation and may progress to complete dissolution
of sclera and episclera with exposure of uvea and staphyloma formation

[6,7,11]

Posterior scleritis may be an extension of anterior scleritis, or may be isolated

inflammation of the sclera posterior to the ora serrata

. Most cases are

unilateral. It may be diffuse, nodular, or necrotizing. Clinical manifestations
of posterior scleritis include marked ocular pain; decreased vision; ocular prop-
tosis; and fundus abnormalities, which may consist of choroidal folds, cystoid
macular edema, choroidal detachment, retinal detachment, and optic neuritis

SCLERITIS AND SYSTEMIC DISEASE ASSOCIATIONS

Approximately 50% of humans with scleritis have a known associated systemic
disease, most often an autoimmune disease

[7,14,15,19]

. In a high percentage of

patients, scleritis is actually the first manifestation of systemic disease

[20,21]

Rheumatoid arthritis, a connective tissue disorder, is most frequently associ-
ated with scleritis, followed by Wegener’s granulomatous, a systemic vasculitis

[7,15,22,23]

. These are followed in incidence by relapsing polychondritis,

inflammatory bowel disease, and systemic lupus erythematosus

[6,7,11,14]

Other less frequently associated diseases include Reiter’s syndrome, psoriatic
arthritis, polyarteritis nodosa, ankylosing spondylitis, Behc¸et disease, giant
cell arteritis, Cogan’s syndrome, and Goodpasture’s syndrome

[15,22]

295

EPISCLERITIS AND SCLERITIS

INFECTIOUS SCLERITIS

Scleritis may develop secondary to infection with bacterial, viral, parasitic, fun-
gal, or amoebal organisms

. Infection is a cause of scleritis in less than 10%

of cases; however, clinical signs may be identical to those caused by systemic
autoimmune disease

[19,24]

. Infectious scleritis may develop after surgical or

nonsurgical trauma or as an extension from infectious keratitis or severe
endophthalmitis

. Pseudomonas aeruginosa is the most common cause of infec-

tious scleritis

. Other causative organisms include Streptococcus pneumoniae,

Staphylococcus aureus, coagulase-negative Staphylococcus, Proteus sp, Corynebacterium,
Serratia, and Nocardia

[7,11]

. Systemic infections, such as toxoplasmosis, syphi-

lis, tuberculosis, Lyme disease, leprosy, and toxocariasis, have also been
reported to cause scleritis

[7,11]

. Fungal scleritis is rare, however, and may

develop secondary to hematogenous spread from a systemic infection, or
may spread from a fungal-infected corneal ulcer, or result from traumatic im-
plantation. Acanthamoeba keratitis may develop secondary to contact lens
wear, or exposure to contaminated water

. Herpes zoster is the most com-

mon viral cause of scleritis. Scleritis usually appears months to years after an
episode of herpes zoster ophthalmicus and is most likely an immune-mediated
reaction to the virus

[12,19,22,24,26,27]

. Other viral etiologies include herpes

simplex virus, Coxsackie B5, and Epstein-Barr virus

SURGICALLY INDUCED SCLERITIS

Surgically induced necrotizing scleritis (SINS) may occur postoperatively as
a focal area of intense scleral inflammation that develops adjacent to the site
of previous scleral, or limbal incision

. The most common cause is catarac-

tous lens extraction; however, SINS may occur following glaucoma, strabis-
mus, and retinal surgery. The mean time to presentation after surgery is
reported as 9 months; however, some may remain latent for several years

[28,29]

. The etiology is thought to be autoimmunity or hypersensitivity,

because immune complexes have been found in and around episcleral vessel
walls by immunofluorescence and systemic immunosuppression has been
successful in treatment of SINS

[28–30]

. In one study up to 90% of patients

who developed SINS were later diagnosed with autoimmune vasculitic disease

HISTOPATHOLOGY OF SCLERITIS

All forms of scleritis are histologically similar but vary in distribution, intensity,
and extent. The basic histopathologic lesion is a granulomatous inflammation
surrounding necrotic scleral collagen

[22,31]

. Often, inflammatory cell infiltra-

tion may extend into the underlying uvea and episclera causing the tissues to
thicken

. In histopathologic studies scleritis associated with systemic auto-

immune disease is more likely to have zonal granulomatous scleral inflamma-
tion, in which there was a central necrotic sclera surrounded by an inner zone
of polymorphonuclear leukocytes, histiocytes, epithelioid cells, and giant cells,

296

GRAHN & SANDMEYER

and an outer zone of lymphocytes and plasma cells; idiopathic cases are more
likely to have nonzonal diffuse mainly nongranulomatous inflammation in
which there was diffuse involvement of the scleral wall with chronic inflamma-
tory cells with or without occasional giant cells, and few small foci of scleral
necrosis

[22,31]

. Infectious scleritis was characterized histologically by necrotic

sclera surrounded by microabscesses

. In mild cases, in which only small

foci of sclera are involved, the process may heal with minimal damage. In
more severe cases, after inflammation has subsided, the affected portions of
sclera may become thin and staphylomas may form

TREATMENT AND PROGNOSIS

The therapeutic goals of scleritis treatment are control of inflammation, elimi-
nation of pain, reduction of complications, and treatment of any underlying
systemic disease. Nonnecrotizing scleritis often responds to systemic NSAIDs.
Topical NSAIDS and corticosteroids are used to improve comfort; however,
these do not suppress scleral inflammation. Systemic corticosteroids are used
if inflammation is nonresponsive to systemic NSAIDs or if posterior or necro-
tizing scleritis is present. Adjunctive immunosuppressive therapies (azathio-
prine, cyclophosphamide, cyclosporine, mycophenolate, methotrexate, and
anti–tumor necrosis factor blockers) are initiated if the inflammation is unre-
sponsive to systemic corticosteroids or if high corticosteroid doses are needed
for extended periods of time

. Immunosuppressive therapy is more likely to

be required in necrotizing scleritis

. Surgical therapy is rarely necessary

except in cases of necrotizing scleritis because these may require surgical inter-
vention to reinforce large areas of scleral thinning or uveal prolapse

[6,11,32]

Ocular complications of scleritis include decreased vision, anterior uveitis, cat-
aract formation, peripheral ulcerative keratitis, glaucoma, and fundus abnor-
malities. These findings are most commonly associated with necrotizing
scleritis

. The ocular prognosis of scleritis may vary depending on underly-

ing systemic disease. Scleritis associated with systemic lupus erythematosus
tends to be a benign and self-limiting condition; scleritis associated with rheu-
matoid arthritis or relapsing polychondritis is a disease of intermediate severity;
whereas, scleritis associated with Wegener’s granulomatosis is severe and can
lead to permanent blindness

. Systemic disease has been associated with an

increased mortality in patients with scleritis. Patients with systemic disease,
such as rheumatoid arthritis, often have cardiovascular or respiratory disease,
and connective tissue disorders that may limit their lifespan

CANINE UNILATERAL AND BILATERAL FOCAL AND DIFFUSE
EPISCLERITIS

These are the most common clinical disorders of the episclera and sclera of
dogs. Unilateral or bilateral primary focal or diffuse episcleritis is often misdiag-
nosed as nonresponsive conjunctivitis, uveitis, orbital cellulitis, and keratitis.
The lack of response to therapies aimed at these diagnoses may increase the
referrals of episcleritis cases to veterinary ophthalmologists. There disorders

297

EPISCLERITIS AND SCLERITIS

have not been reproduced by trial breeding and confirmed breed, age, or sex
predelictions for unilateral focal or diffuse episcleritis have not been reported.
The American cocker spaniel and the golden retriever breeds, however, seem
to be predisposed to these disorders

[2,10]

. The clinical manifestations are often

considered diagnostic by veterinary ophthalmologists (see

). The con-

gested vascular response is located near, or centered on, the limbus and it
may manifest as a focal or diffuse lesion either unilaterally or bilaterally.
The adjacent cornea is usually edematous and is gray to blue in color. Super-
ficial to mid-stromal corneal vascular ingrowths and associated lipid and min-
eral deposits are often present. After the episclerokeratitis has responded to
therapy these degenerative corneal lesions remain. These clinical manifesta-
tions, however, are nonspecific and similar for immune-mediated, infectious,
parasitic, and foreign body–associated secondary episcleritis, and some neo-
plasms (lymphosarcoma), and scleritis may manifest with similar clinical signs.
In addition, some of the focal episcleritis is quite nodular in appearance and
very similar to bilateral nodular episclerokeratitis of collie-type dogs. Biopsies
of the inflamed masses are required to confirm an etiologic diagnosis. The light
microscopic manifestations of episcleritis and episclerokeratitis are similar and
are reviewed together later in this article. Similarly, most types of episcleritis
and nodular episclerokeratitis respond to similar immunosuppressive therapies,
and the prognoses are very similar and are also discussed later.

BILATERAL NODULAR EPISCLEROKERATITIS OF COLLIE DOGS
AND SHETLAND SHEEPDOGS

Numerous synonyms for bilateral nodular episclerokeratitis of collie-type dogs
exist including pseudotumors, fibrous histiocytoma, nodular fasciitis, prolifera-
tive keratoconjunctivitis, limbal granulomas, and collie granulomas, and are
commonly used by ophthalmologists for this syndrome

[1–5,34–39]

. The au-

thors recommend the term ‘‘nodular episclerokeratitis’’ to avoid confusion,
and these nodular masses are almost without exception bilaterally symmetric
pink nodular tumors and involve primarily the cornea and episclera. In con-
trast to previous reports

, the authors do not include nodular as a descriptor

for episcleritis to avoid confusion because nodular episclerokeratitis is a devel-
opmental condition that may be inherited. The collie and collie-type dogs are
uniquely predisposed, and Shetland sheepdogs may also be affected

Despite this breed predisposition, however, breeding trials, mode of inheri-
tance, and mutations responsible are unknown

. The clinical lesions are

virtually pathognomonic and include bilateral fleshy proliferative subepithelial
temporal corneal limbal masses, and hyperemic third eyelid thickening (see

)

[35,38,39]

. Variant forms of nodular granulomatous episclerokeratitis

have also been reported in dogs

. Three of four dogs reported as variant

nodular episclerokeratitis had anterior uveal granulomas and one of four had
granulomatous blepharitis in addition to the third eyelid and temporal limbal
corneal tumors

. Biopsies that are sectioned and examined with light

microscopy reveal chronicity with many similarities to focal and diffuse

298

GRAHN & SANDMEYER

episcleritis of other breeds of dogs described later. Immunohistochemical eval-
uations have identified some unique features of bilateral episclerokeratitis
and episclerokeratitis that often require lifelong topical or systemic
immunosuppression.

LABORATORY EVALUATIONS OF IMMUNE-MEDIATED FOCAL
AND DIFFUSE EPISCLERITIS AND NODULAR
EPISCLEROKERATITIS OF DOGS

Complete blood counts, serum biochemistry profiles, and urinalyses from dogs
with unilateral and bilateral focal or diffuse episcleritis and nodular episclero-
keratitis were not significantly different

Biopsy and light microscopic evaluation categorize the more common im-

mune-mediated episcleritis and less common secondary episcleritis (parasitic
or bacterial or foreign body granulomas), and neoplasia that has masqueraded
as episcleritis

[1–5]

. The cellular infiltrates of all types of episcleritis include

lymphocytes, plasma cells, macrophages, fibroblasts, and occasional neutro-
phils and giant cells (

). This inflammatory infiltrate varies by cell type

and nodular episclerokeratitis biopsies have significantly more fibroblasts and
less histiocytes, lymphocytes, and plasma cells compared with focal and diffuse
episcleritis

[1–3,8–10]

. Focal vascular hyalinization has been reported as a non-

specific finding in nodular episclerokeratitis

. The vascular hyalinization does

not seem to be associated with a vasculitis or collagen vascular disorder and is
likely the effect of chronic inflammation, not the etiology

. The epithelial and

subepithelial regions of biopsies of episcleritis and nodular episclerokeratitis are

Fig. 3. Light microscopic section of focal immune-mediated episcleritis in a dog. Note the dif-
fuse mixed inflammatory infiltrate that includes lymphocytes, plasma cells, macrophages, fibro-
blasts, and occasional neutrophils. This mixed inflammatory infiltrate and a lack of detectable
infectious organisms supports the diagnosis immune-mediated episcleritis (hematoxylin-eosin,
original magnification 40).

299

EPISCLERITIS AND SCLERITIS

usually nonulcerated and relatively free of inflammatory cells, respectively. Un-
derlying the relatively normal epithelium and subepithelial layer, the inflamma-
tion is centered on the episcleral tissues and corneal stroma

[1–3,35–39]

Characteristically, there are minimal mitotic figures, a very mixed inflamma-
tory cellular infiltrate, and distinct lack of a monotonous proliferation of invad-
ing cells that signify neoplasia.

Focal and diffuse episcleritis, superficial keratitis of large breed dogs (pan-

nus), and pigmentary keratitis of brachycephalic dogs have similar inflamma-
tory cell infiltrates. Pannus and pigmentary keratitis, however, has less
extensive cellular infiltrates of lymphocytes, plasma cells, histiocytes, and fibro-
blasts that predominately invade the cornea from the limbus. Usually there are
significantly less inflammatory nodules associated with these conditions, and
these conditions have pathognomonic clinical manifestations, and only occa-
sionally are biopsies of pannus and pigmentary keratitis submitted. The differ-
entiating histologic features of pannus and pigmentary keratitis are much more
subtle than the clinical manifestations. These include increased epithelial and
subepithelial pigmentation and fibrovascular tissue that are associated predom-
inately with the superficial cornea. Typically, there is much less tissue mass and
a more diffuse conjunctival lymphocytic plasma cell subepithelial infiltration
compared with focal episcleritis and nodular episclerokeratitis, where the
inflammation is centered in the episcleral and corneal stromal tissues.

Secondary episcleritis caused by foreign bodies and infectious organisms

usually has cellular infiltrates that are granulomatous and contain more macro-
phages, giant cells, and leucocytes, and special stains may reveal the organisms
or foreign bodies (

)

[1,4,40,41]

Fig. 4. Granulomatous episcleritis in this case is characterized by foamy macrophages,
which contained acid-fast bacteria, confirming episcleritis and orbital cellulitis secondary to
Mycobacterium spp. Note also the diffuse lymphocytic and plasmacytic infiltrate and perivas-
cular cuffing (hematoxylin-eosin, original magnification 40).

300

GRAHN & SANDMEYER

Immunohistochemical staining of focal and diffuse episcleritis and nodular

episclerokeratitis have been reported

[8–10]

. The numbers of T and B lympho-

cytes and histiocytes were not significantly different between dogs with unilat-
eral or bilateral focal, diffuse, or nodular episcleritis

[8,10]

. The numbers of B

cells were significantly elevated in biopsies of lesions from dogs that required
long-term therapy, however, because the lesions did not resolve and would
recrudesce when topical corticosteroid therapy was discontinued

. Inflam-

matory and proliferative episcleritis have also been categorized immunohisto-
chemically

, and most of the proliferative lesions were characterized by

cells that are nonreactive to histiocytic markers. These cells seem to be of
different origin and may result in differences noted in response to therapy or
the need for ongoing therapy

. Increased amounts of collagen, thickened hy-

alinized collagen fibers, and reticulin fibers have also been reported with rou-
tine histochemical stains of focal and diffuse episcleritis biopsies

[2,38]

Additional prospective investigations of unilateral and bilateral episcleritis
and nodular episclerokeratitis are warranted to reveal the pathogenesis of these
lesions.

TREATMENT AND PROGNOSIS FOR UNILATERAL
AND BILATERAL EPISCLERITIS AND BILATERAL
NODULAR EPISCLEROKERATITIS OF DOGS

Topical immunosuppression with 1% prednisolone acetate or 0.1% dexameth-
asone solutions are the preferred initial therapies for unilateral and bilateral
focal or diffuse episcleritis

[3,4,34–38,42]

. Either of these corticosteroids should

be administered topically every 6 hours until the episcleritis and episcleroker-
atitis subsides. The topical therapy is then tapered by reducing the frequency
on a 2- to 4-week basis from four times a day, to three times a day, to twice
a day, and finally to once a day. If signs of inflammation do not recrudesce,
all medications may be discontinued, and complete remission is expected.
Recrudescence is uncommon in unilateral episcleritis, but frequent in bilateral
episcleritis and nodular episclerokeratitis. Bilateral focal and diffuse episcleritis
that do recrudesce have significantly different cellular infiltrates

[8,10]

. Lifelong

therapy, or at least long-term therapies with topical corticosteroids, with or
without intralesional steroids, oral steroids, or other oral immunosuppressants
(azathioprine) is often administered to resolve promptly the clinical manifesta-
tions. The topical corticosteroids then can be reduced to maintain inflamma-
tory control

[3,4,34]

. The prognosis for maintaining a noninflamed and

visual eye is excellent.

Nodular episclerokeratitis tends to be more proliferative and is often more

resistant to topical immunosuppression. These dogs may be treated effectively
with topical corticosteroids and systemically with azathioprine (2 mg/kg orally)

[37,43]

. Once remission is established, the azathioprine is gradually reduced

[37,43]

. Intralesional long-lasting corticosteroid therapy is a useful adjuvant

therapy and surgical excision or debulking, strontium therapy, cryotherapy,
and electrocautery at the time of biopsy may also assist in establishing prompt

301

EPISCLERITIS AND SCLERITIS

remission

[3,34,38]

. Lifelong therapy is required more often in lesions with

a predominating B-lymphocyte population

[8,10]

SCLERITIS OF DOGS

Nonnecrotizing immune-mediated scleritis of dogs is an uncommon unilateral or
bilateral disorder and manifests with a diffusely thickened sclera, and conjuncti-
val and episcleral hyperemia. If it is extensive or chronic an accompanying cho-
roiditis with focal flat serous retinal detachments and retinal degeneration are
often detectable (

[1,2]

. The clinical manifestations are very similar to dif-

fuse episcleritis with the exception of the posterior uveitis and increased severity
of inflammation. The clinical presentation is also very similar to necrotic scleritis,
except for the absence of collagen necrosis. The pathogenesis of nonnecrotizing
scleritis is poorly understood and most often the etiology is assumed to be
immune-mediated. The diagnosis is confirmed by a scleral biopsy and light
microscopic examination that reveals a granulomatous or inflammatory scleritis,
an associated choroiditis, and a distinct lack of collagen necrosis (see

;

. Given that scleral areas affected with collagen necrosis may be small, the

authors often use resection scleral biopsies to rule out collagen necrosis. Whether
or not idiopathic (primary) scleritis and necrotic scleritis are related or completely
separate entities is unknown; however, nonnecrotizing scleritis tends to be a mild-
er disease clinically. The treatment of choice is immunosuppression with oral
prednisolone, and azathioprine

[4,34]

. When the lesions are focal, intralesional

injections of corticosteroids may have beneficial effects. The prognosis for the
globe is guarded and scleral grafts may be required if staphylomas develop.

Fig. 5. Pale pink peripheral choroidal granulomas are present in this dog secondary to a focal
scleritis.

302

GRAHN & SANDMEYER

Necrotic Scleritis of Dogs

Necrotic scleritis is a rare unilateral or bilateral aggressive inflammatory condi-
tion of the sclera that affects the choroid and episclera. The clinical manifesta-
tions are diverse and often include a markedly thickened and inflamed sclera,
although scleral lysis and staphyloma with disensertion of varied recti muscles
also develop (

). Conjunctivitis, episcleritis, choroiditis, and choroidal

granulomas are usually present (

Fig. 8

). The diagnosis is confirmed with light

microscopic examination of biopsies of affected sclera. Scleral graft material
(fresh or frozen sclera or cornea) should be available to repair full-thickness
lesions that may be encountered or created during the biopsy. Light micro-
scopic examination of biopsies confirms the diagnosis by revealing collagen
necrosis, collagen lysis, and scleral and episcleral inflammation (

Fig. 9

). The

inflammatory cell infiltrates vary from region to region and mononuclear cells
are common including lymphocytes and plasma cells. Occasionally, giant cells
and epithelioid macrophages are present. The pathogenesis is unknown and
the etiology is assumed to be immune-mediated based on similar diseases in
humans

. Necrotic scleritis of dogs is challenging to manage, and immuno-

suppression with oral azathioprine and steroids is required. Lifelong therapy
is required and remission is uncommon, although the systemic medications
may be reduced. Large staphyloma are best repaired surgically with scleral
grafts. The prognosis for retaining sight and the eye is poor. Given the rarity,
lack of an experimental model, and lack of retrospective and prospective stud-
ies to date, the pathogenesis, etiology, effective therapies, and prognosis are
largely unknown for necrotic scleritis in dogs.

Fig. 6. An idiopathic nonnecrotizing scleritis with orbital cellulitis is present in this adult dog.
Note the diffuse infiltrate of lymphocytes and plasma cells, which extend from the choroid
through the sclera (hematoxylin-eosin, original magnification 20).

303

EPISCLERITIS AND SCLERITIS

Unilateral and Bilateral Episcleritis and Scleritis Secondary to Bacterial
Infections, Parasitic Infestations, Subconjunctival Foreign Bodies,
and Traumatic Injuries

Episcleritis has been reported to develop secondary to E canis

[4,40]

. Scleritis

that develops secondary to infectious organisms (E canis) and parasitic infesta-
tions (Onchocerca spp) are uncommon, except in geographic regions where these

Fig. 8. The fundus photograph of the dog in

. Note the multiple focal choroidal gran-

ulomas that accompanied the necrotic scleritis in this dog.

Fig. 7. The left eye of a dog with histologically confirmed necrotic scleritis. The dorsal rectus
muscle is avulsed and this is inducing a ventral and slightly medial strabismus. Multiple bilat-
eral areas of collagen necrosis were present in this dog.

304

GRAHN & SANDMEYER

organisms are endemic (southern United States, Greece). The clinical signs are
similar to idiopathic immune-mediated episcleritis and scleritis, and the diagno-
ses is confirmed with biopsy, culture, polymerase chain reaction, and immuno-
fluorescence laboratory procedures. E canis scleritis and episcleritis are treated
with doxycycline, 5 mg/kg, orally for a minimum of 3 weeks, and topical
corticosteroids, NSAIDs, and mydriatics cycloplegics as required

. Oncho-

cerciasis is treated by debulking the granulomas surgically and administering
melarsomine, 2.5 mg/kg, every 24 hours intramuscularly, and then ivermectin,
50 lg/kg, every 24 hours for 30 days

Scleritis and episcleritis are common sequelae to intraocular surgery, ocular

trauma, penetrating foreign bodies, and endophthalmitis. The clinical manifes-
tations are similar to primary immune-mediated episcleritis and scleritis,
although the history usually identifies a predisposing trauma. Traumatic scler-
itis and episcleritis is usually mild and responds promptly to symptomatic anti-
inflammatory therapy, and is seldom examined with light microscopy unless
the globe is enucleated because of nonresponsive endophthalmitis. The inflam-
matory infiltrate varies with the etiology and generally has neutrophils and
macrophages acutely, and chronic lesions have lymphocytes and plasma cells
and have fibrous tissue repair.

SUMMARY

Idiopathic episcleritis is a common condition in dogs that may be categorized
into unilateral and bilateral focal or diffuse forms, and bilateral nodular epis-
clerokeratitis, a unique likely developmental condition of collie and Shetland
sheepdog breeds. The clinical manifestations of all forms of episcleritis are

Fig. 9. Light microscopic features of necrotic scleritis include diffuse inflammatory cells and
collagen necrosis in dogs. Note the diffuse mononuclear cellular infiltrate, and eosin-stained
areas of collagen (collagen necrosis) in the sclera biopsy of this dog (hematoxylin-eosin, orig-
inal magnification 40).

305

EPISCLERITIS AND SCLERITIS

similar and include focal to diffuse hyperemia, congestion, and edema of the
episclera often near the limbus; neighboring cornea is affected with focal
edema, vascularization, and degeneration. Light microscopic examination of
episcleritis is surprisingly uniform with a mixed inflammatory infiltrate with
lymphocytes, plasma cells, fibroblasts, and macrophages predominating. Im-
munohistochemical stains have identified a predominance of B-cell lympho-
cytes in bilateral recurrent episcleritis, and increased fibroblasts in
proliferative variant. Most episcleritis cases respond to topical corticosteroid
immunosuppression.

Idiopathic scleritis, unlike episcleritis, is an uncommon condition and mani-

fests with two clinical and pathologic conditions: nonnecrotizing scleritis and
necrotizing scleritis. Scleritis may be unilateral or bilateral and unlike episcler-
itis these conditions are malignant inflammatory conditions that induce uveitis
and often are vision and globe threatening. The light microscopic examination
of scleritis and necrotic scleritis is also similar with a variable infiltration of the
lymphocytes, plasma cells, and macrophages, and necrotic scleritis is unique with
focal areas of collagen necrosis and collagenolysis. Scleritis and necrotic scleritis
require long-term systemic immunosuppression and the prognosis is guarded.

Secondary episcleritis and scleritis are common and develop secondary to

foreign body penetration, surgery, choroiditis, and conjunctivitis. They are
usually ignored diagnostically as the primary disorder (uveitis) and are usually
treated successfully.

References

[1] Grahn BH, Peiffer RL. Fundamentals of veterinary ophthalmic pathology. In: Gelatt KN,

editor. Veterinary ophthalmology, vol. 1. 4th edition. Ames (IA): Blackwell Publishing;
2007. p. 355–437.

[2] Deykin AR, Guandalini A, Ratto A. A retrospective histopathologic study of primary

episcleral and scleral inflammatory disease in dogs. Progress in Veterinary and Com-
parative Ophthalmology 1997;7:245–8.

[3] Paulsen ME, Lavach JD, Snyder SP, et al. Nodular granulomatous episcleritis in dogs:

19 cases (1973–1985). J Am Vet Med Assoc 1987;190:1581–7.

[4] Martin CL. Cornea and sclera. In: Martin CL, editor. Ophthalmic disease in veterinary med-

icine. London: Manson Publishing; 2005. p. 241–97.

[5] Fischer C. A clinicopathologic classification of episcleritis and scleritis in the dog. Programs

and abstracts of the 13th American College of Veterinary Ophthalmologists, Lake Tahoe,
NA; 1982. p. 1–18.

[6] Yanoff M, Fine BS. Ocular pathology. 4th edition. London: Mosby-Wolfe; 1996. p. 285–8.
[7] Okhravi N, Odufuwa B, McCluskey P, et al. Scleritis. Surv Ophthalmol 2005;50(4):

351–63.

[8] Breaux CB, Grahn BH, Sandmeyer LS. Retrospective immunohistochemical and clinical in-

vestigation of episcleritis in the dog. Programs and abstracts of the 36th American College
of Veterinary Ophthalmologists, Nashville, TN; 2005. p. 1.

[9] Williams CO, Dietrich UM, Moore PA, et al. The immunopathogenesis of canine nodular

ocular episcleritis. Programs and abstracts of the 36th American College of Veterinary Oph-
thalmologists, Nashville, TN; 2005. p. 37.

[10] Breaux CB, Sandmeyer LS, Grahn BH. Immunohistochemical investigation of canine epis-

cleritis. Vet Ophthalmol 2007;10:168–72.

306

GRAHN & SANDMEYER

[11] Gold DH, Lewis RA. Clinical eye atlas. USA: American Medical Association; 2002.

p. 457–72.

[12] Watson PG, Heyreh SS. Scleritis and episcleritis. Br J Ophthalmol 1976;60(3):163–91.
[13] Spencer WH. Sclera. In: Spencer WH, editor. Ophthalmic pathology: an atlas and text-

book. 5th edition. Philadelphia: WB Saunders Co.; 1988. p. 389–422.

[14] Jabs DA, Mudun A, Dunn JP, et al. Episcleritis and scleritis: clinical features and treatment

results. Am J Ophthalmol 2000;130(4):469–76.

[15] Sainz de la Maza M, Jabbur NS, Foster CS. Severity of scleritis and episcleritis. Ophthalmol-

ogy 1994;101(2):389–96.

[16] Wilhelmus KR, Grierson I, Watson PG. Histopathologic and clinical associations of scleritis

and glaucoma. Am J Ophthalmol 1981;91(6):697–705.

[17] Mader TH, Stulting RD, Crosswell HH. Bilateral paralimbal scleromalacia perforans.

Am J Ophthalmol 1990;109(2):233–4.

[18] Perlstein SH, Yablonski ME. Spontaneous remission of glaucoma in scleromalacia perfo-

rans: a case report. Ann Ophthalmol 1984;16(3):229–30.

[19] Akpek EK, Thorne JE, Qazi FA, et al. Evaluation of patients with scleritis for systemic disease.

Ophthalmology 2004;111(3):501–6.

[20] Carrasco MA, Cohen EJ, Rapuano CJ, et al. Therapeutic decision in anterior scleritis: our

experience at a tertiary care eye centre. J Fr Ophtalmol 2005;28(10):1065–9.

[21] Harper SL, Letko E, Samson CM, et al. Wegener’s granulomatosis: the relationship between

ocular and systemic disease. J Rheumatol 2001;23(5):1025–32.

[22] Riono WP, Hidayat AA, Rao NA. Scleritis: a clinicopathologic study of 55 cases. Ophthal-

mology 1999;106(7):1328–33.

[23] Pavesio CE, Meier FM. Systemic disorders associated with episcleritis and scleritis. Curr

Opin Ophthalmol 2001;12(6):471–8.

[24] McCluskey PJ, Wakefield D. Current concepts in the management of scleritis. Aust N Z J

Ophthalmol 1988;16(3):169–76.

[25] Reynolds MG, Alfonso E. Treatment of infectious scleritis and keratoscleritis. Am J Ophthal-

mol 1991;112(5):543–7.

[26] Liesegang TJ. Corneal complications from herpes zoster ophthalmicus. Ophthalmology

1985;92(3):316–24.

[27] Gungor IU, Ariturk N, Beden U, et al. Necrotizing scleritis due to varicella zoster infection:

a case report. Ocul Immunol Inflamm 2006;14(5):317–9.

[28] O’Donoughue E, Lightman S, Tuft S, et al. Surgically induced necrotizing sclerokeratitis

(SINS) precipitating factors and response to treatment. Br J Ophthalmol 1992;76(1):
17–21.

[29] Alsagoff C, Tan DT, Chee SP. Necrotizing scleritis after bare sclera excision of pterygium.

Br J Ophthalmol 2000;84(9):1050–2.

[30] Fong LP, Sainz de la Maza M, Rice BA, et al. Immunopathology of scleritis. Ophthalmology

1991;98(4):472–9.

[31] Rao NA, Marak GE, Hidayat AA. Necrotizing scleritis: a clinico-pathologic study of 41

cases. Ophthalmology 1985;92(11):33–41.

[32] Davidson RS, Erlanger M, Taravella M, et al. Tarsoconjunctival pedicle flap for the manage-

ment of a severe scleral melt. Cornea 2007;26(2):235–7.

[33] Sainz de la Maza M, Foster CS, Jabbur NS. Scleritis associated with systemic vascular dis-

eases. Ophthalmology 1995;102(4):687–92.

[34] Gilger BC, Ollivier FJ, Gelatt KN. Bentley E Diseases and surgery of the canine cornea and

sclera. In: Veterinary ophthalmology. 4th edition. Ames (IA): Blackwell Publishing; 2007.
p. 690–752.

[35] Lavignette AM, Carlton WW. A case of ocular nodular fasciitis in a dog. J Am Anim Hosp

Assoc 1974;10:503–6.

[36] Smith JS, Bistner S, Riis R. Infiltrative corneal lesions resembling fibrous histiocytoma: clinical

and pathologic findings in six dogs and one cat. J Am Vet Med Assoc 1976;169:722–6.

307

EPISCLERITIS AND SCLERITIS

[37] Dugan SJ, Ketring KL, Severin GA, et al. Variant nodular granulomatous episclerokeratitis in

four dogs. J Am Anim Hosp Assoc 1993;29:403–9.

[38] Gwin RM, Gelatt KN, Peiffer RL. Ophthalmic nodular fasciitis in the dog. J Am Vet Med

Assoc 1977;170:611–4.

[39] Bellhorn RW, Henkind P. Ocular nodular fasciitis in a dog. J Am Vet Med Assoc 1967;150:

212–3.

[40] Willis M, Martin C, Stiles J. Infectious disease as a possible cause for scleritis and episcleritis

in dogs. Programs and abstracts of the 27th American College of Veterinary Ophthalmolo-
gists, Wailea HI; 1996. p. 45.

[41] Komnenou AA, Mylonakis ME, Kouti V, et al. Ocular manifestations of natural canine mono-

cytic ehrlichiosis (Ehrlichia canis): a retrospective study of 90 cases. Vet Ophthalmol
2007;10:137–42.

[42] Peiffer RL, Gelatt KN, Gwin RM. Proliferative episcleritis in the dog. Vet Med Small Anim Clin

1976;1273–8.

[43] Latimer CA, Wyman M, Szymanski C, et al. Azathioprine in the management of fibrous his-

tiocytoma in two dogs. J Am Anim Hosp Assoc 1983;19:155–8.

[44] Zarfoss MK, Dubielzig RR, Eberhard ML, et al. Canine ocular onchocerciasis in the United

States: two new cases and a review of the literature. Vet Ophthalmol 2005;8:51–7.

308

GRAHN & SANDMEYER

Anterior Chamber-Associated Immune
Deviation

Daniel Biros, DVM

Angell Animal Medical Center, 350 South Huntington Avenue, Boston, MA 02130, USA

he healthy eye possesses an exceptional form of regional immunity
referred to as anterior chamber–associated immune deviation (ACAID).
ACAID has been described in rodents, rabbits, and nonhuman primates

and is likely present in human beings

[1,2]

. This local immune response is an

extreme type of immune privilege initially explored almost 130 years ago in Eu-
rope. In the first known published study documenting the special properties of
the anterior chamber of the eye, Van Dooremaal

, in the nineteenth century,

demonstrated that murine skin grafts had a prolonged survival when placed in
the anterior chamber of the dog eye. Seventy-five years later, Medawar

pro-

posed that the graft survival in the anterior chamber was attributable to immu-
nologic ignorance because he observed no specific lymphatic drainage within
the eye, presuming that the sequestered antigens in the anterior chamber es-
caped detection from the immune system in general. It was for this work in
immune tolerance that Medawar received the 1960 Nobel Prize in Medicine.

It is only in the past few decades, however, that we have discovered the

unique properties of the anterior chamber are, in reality, a result of an active
immune system, beginning with the capture of antigen in the eye and culminat-
ing in a systemic immune response that we now know as ACAID. In brief,
Kaplan and colleagues

were the first to show that antigen-specific systemic

cell-mediated (T helper [Th] 1) immunity was suppressed when alloantigens
were naively presented to the anterior chamber. ACAID has been widely stud-
ied using soluble proteins, particulate antigens, histocompatibility antigens, and
tumor antigens

[6–12]

. In ACAID, intraocular antigen-presenting cells (APCs)

capture the alloantigen and travel directly to the spleen, wherein they begin an
intricate set of interactions with other immune cells resulting in the activation of
T-regulatory lymphocytes (T

reg

cells). These T

reg

cells significantly suppress

inflammation that would otherwise actively destroy the delicate and mostly
irreparable tissue of the eye, such as the neuroretina, cornea, and lens.

The primary cellular output of ACAID and the reason why ACAID is

unique among regional immune responses is the generation of two types of

E-mail address: dbiros@angell.org

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.12.006

Vet Clin Small Anim 38 (2008) 309–321

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

antigen-specific T

reg

cells: afferent T

reg

cells and efferent T

reg

cells. The afferent

reg

cell is CD4þ and suppresses the initial activation and differentiation of

T cells into Th1 effector cells. Unlike conventional CD4þCD25þ T

reg

cells,

however, the afferent CD4þ cells are unique in that they do not require cell-
to-cell contact to carry out their regulatory functions. The efferent T

reg

cell is

a CD8þ T cell that inhibits the expression of Th1 type immunity, such as
delayed-type hypersensitivity (DTH). Afferent regulators generated in ACAID
work at the level of the secondary lymphoid organs, whereas efferent regula-
tors generated in ACAID work in the periphery (ie, the eye)

[13–15]

. The

single most important function of ACAID is to spare vision by preserving
ocular tissue that is extremely vulnerable to innate and adaptive immunity.
For this article, ACAID is discussed in terms of the ocular phase (induction),
thymic and splenic phases (T

reg

cell activation), and, in brief, the recently pro-

posed integral role of the sympathetic nervous system (neuroregulation). The
influence of ocular inflammation in ACAID and the therapeutic potential of
ACAID are also addressed.

ACAID is an antigen-specific suppression of Th1 immunity (eg, DTH,

complement-fixing antibodies) by means of antigen presentation directly
into the anterior chamber. ACAID does preserve the generation of non–
complement-fixing antibodies, such as the IgG1 isotype in the mouse

[2,16]

ACAID also has a specific cytokine profile

[17–19]

: interferon-c (IFNc) produc-

tion is suppressed and transforming growth factor-b (TGFb) production is
enhanced in ACAID. The Th2 cytokine interleukin (IL)-10 is also produced
in ACAID, which led many to believe that ACAID is a Th2 type response
countering the effects of Th1 immunity. IL-4, another Th2 cytokine, was not
required for ACAID, however, and ACAID actually can modulate Th2 type
allergic inflammatory lung diseases

. Further, no Th2 cytokine-producing

cells were found in spleen and lymph node tissue after anterior chamber anti-
gen presentation

. Genetically altered mice unable to mount a Th2 response

because of deficiencies in IL-4, IL-13 or to stimulate signal transducer and
activator (STAT) of signal 6 genes (all critical for Th2 development) still could
generate ACAID

. As discussed elsewhere in this article, these observations

suggest that ACAID is not a specialized immune response favoring Th2 reacti-
vity over Th1. ACAID is a systemic type of immunity that involves specific
cytokine production and the interaction of many types of immune cells able
to modulate Th1 and Th2 immunity.

OCULAR PHASE, ANTERIOR CHAMBER–ASSOCIATED IMMUNE
DEVIATION INDUCTION

When antigen is presented into the anterior chamber, it begins a completely
different type of immune response by contrast to antigen presentation in other
places in the body. At first, it was thought that ACAID was the same deviant
immune response achieved by intravenous presentation of antigen, but there
are distinct differences

. Most notable was the difference in the requirement

of certain cytokines necessary to elicit the respective intravenous immune

310

BIROS

deviation or ACAID responses: IL-4 was required for the intravenously
induced immune deviation, and IL-10, B cells, efferent suppressor cells,
blood-borne APCs, invariant natural killer T (iNKT) cells, and b

microglobu-

lin were all essential for ACAID

[18,24–31]

. Furthermore, the eye had to

remain intact for ACAID to happen, because enucleation within 3 days of
anterior chamber antigen injection abrogated ACAID

Antigen that enters the anterior chamber leaves the eye and travels not only

to the spleen and thymus but, as recent evidence suggests, to the cervical lymph
node in approximately 3 days by means of class II-negative F4/80þ macro-
phages that reside in the iris and ciliary body

. F4/80 is a surface molecule

with adhesion and signaling function and is found on macrophages in primarily
T-cell–independent areas

[34–37]

. These F4/80þ macrophages must have the

complement 3b receptor ligated for ACAID to occur

. This ligation was

associated with the upregulation of IL-10 and TGFb and the suppression of
IL-12 production, which are hallmarks of ACAID. In primates, APCs laden
with processed antigen are thought to exit through the uveoscleral pathway,
because there are no major resident lymphatic vessels in the eye

. Most

(98%) of the antigen placed in the anterior chamber leaks into the blood, but
only a small amount of antigen is carried to the spleen by F4/80þ ocular
APCs. In the circulation, these cells travel to the spleen and thymus and are
potent at inducing ACAID; in some experiments, only 20 activated F4/80þ
APCs were required by means of adoptive transfer to induce ACAID in naive
mice

. CD1d, a major histocompatibility complex (MHC) class I molecule,

must be expressed on the surface of the ocular APCs in addition to F4/80 for
ACAID to happen

[9,40]

. The role of CD1d is primarily in the splenic phase of

ACAID, as discussed elsewhere in this article.

The anterior chamber has a distinct composition of immunomodulatory

cytokines and cell surface molecules that have an early impact on antigen pre-
sentation in ACAID. The ubiquitous presence of TGFb in the aqueous humor
induces a higher concentration of IL-10 and suppression of IL-12

. TGF-b2

is the TGFb isotype found within the aqueous humor

. Macrophage inflam-

matory protein-2 (MIP-2) is present in the anterior chamber and is a potent
recruiter of CXCR2þ NKT cells to the spleen

. Fas ligand (FasL) is expres-

sed on most ocular surfaces and must bind the Fas receptor on the surface of
antigenic cells in the anterior chamber to induce apoptosis, which is a prerequi-
site for ACAID

[42,43]

. In the herpes simplex virus 1 (HSV-1) ACAID model,

inflammatory cells recruited to the eye after anterior chamber virus inoculation
underwent Fas-FasL–mediated apoptosis within 48 hours

[42]

. In mice deficient

in Fas or FasL, the apoptosis did not occur and DTH ensued. Tumor necrosis
factor-a (TNFa) in the anterior chamber may also aid in the induction of
ACAID by upregulating the Fas receptor, and thereby promoting Fas-FasL–
mediated apoptosis

[44,45]

. TNFa interaction in ACAID is also thought to

be associated with enhanced corneal allograft survival in mice

. As briefly

mentioned previously, complement is also important in the induction of
ACAID

. Mice deficient in C3, the third component of complement, could

311

ANTERIOR CHAMBER-ASSOCIATED IMMUNE DEVIATION

not produce ACAID. Neutralizing the iC3b receptor, OX42, with neutralizing
antibodies before the induction of ACAID experimentally did not suppress
DTH. Adding excessive soluble OX42 to in vitro cultures of ACAID APCs
also abrogated the ACAID response. In other words, iC3b must bind its recep-
tor CR3 (OX42) on the ocular APCs to produce TGFb and IL-10 and to
downregulate IL-12, leading to an ACAID response.

THYMIC PHASE OF ANTERIOR CHAMBER–ASSOCIATED
IMMUNE DEVIATION

The thymus is integral to ACAID

. Thymectomized mice did not elicit an

ACAID response experimentally with intravenous adoptive transfer of stimu-
lated F4/80þ cells or direct anterior chamber injection of antigen

. Approx-

imately 3 days after presentation of antigen to the anterior chamber, the
thymus generated CD4CD8 NK1.1 thymocytes that trafficked preferen-
tially to the spleen, wherein they participated in the production of splenic
CD8þ suppressor cells able to suppress Th1 immunity

SPLENIC PHASE OF ANTERIOR CHAMBER–ASSOCIATED
IMMUNE DEVIATION

The F4/80þ ocular APCs migrate to the marginal zone of the spleen, wherein
they interact with NKT cells and B cells, leading to the generation of CD4þ
and CD8þ efferent suppressor cells

[28,30,49]

. In this process, there is an intri-

cate mix of cytokines and cell surface molecules that interact in a specific fashion
to create the CD8þ T

reg

cells for which ACAID is known. Without the spleen,

these cell clusters could not form; ACAID did not occur if the spleen was
removed within 7 days of antigen presentation to the eye

. The F4/80þ

ocular cell must express CD1d and produce IL-10, IL-13, MIP-2, and STAT6

[17,22,28,30,40,51]

. One of the first cells the ocular APCs come into contact

with are CD4þ iNKT cells by means of CD1d. Ocular APCs transporting anti-
gen to the spleen attracted iNKT cells by releasing MIP-2

. iNKT cells are

rare lymphocyte bearing markers of NK cells and T cells. iNKT cells express
the invariant Va14a18 T cell receptor (TCR), which preferentially binds
some Vb chains. iNKT cells produced IL-10 but not IL-4

. The iNKT cells

also produced regulated upon activation, normal T-cell expressed, and secreted
(RANTES), which recruited more F4/80þ ocular APCs and T cells to the
marginal zone of the spleen. The primary cells in these clusters were F4/80þ oc-
ular APCs, CD4þ iNKT cells, and T cells

B cells are also integral to the induction of ACAID

. The marginal zone of

the spleen is rich in CD1þ B cells. It is thought that B cells act as APCs by
capturing and processing antigen released by F4/80þ ocular APCs. These B cells
had the ability to transfer antigen-specific ACAID to naive mice experimentally

[24,25]

. The B cells presented the antigen to T cells in the marginal zone of the

spleen. b

-microglobulin expression was required on the B cells and the F4/80þ

ocular APCs

. The nonclassic class Ib molecule Qa-1 was found on these

CD1þ B cells that were necessary for the generation of CD8þ T-suppressor cells

312

BIROS

[16,52,53]

. One hypothesis of the role of B cells in ACAID stated that F4/80þ

CD1dþ ocular APCs interact with CD1d-dependent invariant CD4þ
NK1.1T cells in the marginal zone of the spleen. Antigen fragments released
by the F4/80þ ocular APCs are picked up and processed by the CD1dþ B cells
and, in turn, present the antigen to the CD8þ T cells, leading to their generation
as CD8þ ACAID suppressor T cells. It was marginal zone B cells rather than
follicular B cells that were required to induce ACAID

The Ly49 molecule may also have an integral role in ACAID. Ly49 molecules

are generally known as inhibitory molecules on NK cells, NKT cells, and some T
cells. When ligated to their appropriate MHC class I molecule, they downregu-
late IFNc

[54,55]

and suppress NKT cell proliferation

[56,57]

and cytotoxic

activity

. When Ly49C/I was blocked experimentally, NKT cells not only

produced IFNc but decreased IL-10 production. It may be hypothesized that
Ly49C/I function may be to induce IL-10 and thereby decrease IFNc and other
lytic functions so as to allow the generation of CD8þ T

reg

cells in ACAID.

ACAID induction also requires cdT cells, but the precise mechanisms remain

unknown

[38,59,60]

. cdT cells make up only 2% to 10% of the total T-cell pop-

ulation and are known to play a role in other forms of immune tolerance

[61,62]

These cells are prolific cytokine producers, and it is thought that they may con-
tribute a significant amount of TGFb and IL-10 in ACAID. These cells also have
the ability to present antigen; thus, they may have a role not unlike B cells in the
sense that they act as ancillary APCs during ACAID

. Furthermore, cdT cells

have been shown to inhibit IFNc production and interfere with Th1 immunity.

NEUROREGULATION AND ANTERIOR CHAMBER–ASSOCIATED
IMMUNE DEVIATION

All three organs involved in ACAID—the eye, spleen, and thymus—have signif-
icant sympathetic innervation. Experimentally chemical sympathectomy did
not alter the function of F4/80þ ocular APCs, but ACAID was impaired

. It was thought that the denervation affected the generation of CD4þ

NKT cells that were integral to the generation of the ACAID CD8þ
T-suppressor cells. Superior cervical ganglionectomy also significantly decreased
the levels of active TGFb in the aqueous humor

. DTH and antibody

responses can be significantly altered with sympathetic denervation

[66–69]

;

thus, it is plausible that other immune responses, such as ACAID, may be
affected in the absence of a functional sympathetic nervous system.

It has also been suggested that corneal innervation contributes to ACAID.

When circumferential corneal denervation was experimentally performed,
ACAID could not be induced

[70]

. This suggests that afferent neural stimuli

may be important in the generation of ACAID.

ANTERIOR CHAMBER–ASSOCIATED IMMUNE DEVIATION
AND CORNEAL ALLOGRAFTS

The cornea is in direct contact with the anterior chamber, and the axial cornea
is considered an immune privileged site. Full-thickness central corneal allografts

313

ANTERIOR CHAMBER-ASSOCIATED IMMUNE DEVIATION

had almost 100% survival as a result of immune privilege and ACAID

Mice with accepted allografts showed key components of ACAID

. Further,

if ACAID was blocked in mice experimentally, corneal allograft survival was
greatly reduced

. Anterior chamber immunization with corneal alloantigens

before corneal allografting enhanced transplant survival, suggesting a direct
role of ACAID in a graft setting

[72,73]

ANTERIOR CHAMBER–ASSOCIATED IMMUNE DEVIATION
AND INFLAMMATION

Recently, it has been questioned whether ACAID would remain intact or be able
to recover despite the presence of intense uveitis. Transgenic mice that produce
intraocular IFNc similar to the Th1 immunity found in experimental autoim-
mune uveitis (EAU) could not generate ACAID

. When ovalbumin (OVA)

was injected into the anterior chamber of uveitic mice, which were then immu-
nized subcutaneously 1 week later with OVA to test for ACAID by means of
DTH suppression, the results of ACAID induction depended on the uveitis
model used. In an experimental study of EAU and ACAID, mice that were
presented with antigen in the anterior chamber just after the peak of EAU at
the time of the anterior chamber antigen injection could not generate ACAID
and DTH was not suppressed

. By contrast, mice with active endotoxin-

induced uveitis (EIU) at the time of the antigen injection into the anterior cham-
ber did demonstrate ACAID when challenged subcutaneously with the same
antigen 1 week later (DTH was suppressed)

. The intensity of the inflamma-

tion was based on the total protein levels and total leukocyte counts in the anterior
chamber. It may be proposed that the type of inflammation present rather than
the magnitude of the inflammation may influence the success of ACAID.

In Mycobacterium tuberculosis adjuvant–induced uveitis (MTU), for example,

intravitreal Complete Freund’s adjuvant induced an upregulation of intraocular
IL-12 induction that was able to suppress ACAID at 3 hours after MTU

. On

day 8 of MTU when IL-12 levels dropped significantly, ACAID was induced.
IL-12 may be the determining factor or merely an indicator of a microenviron-
ment that supports or abolishes ACAID. This remains to be seen.

The DBA/2J mouse model also has been used to study ACAID. DBA/2J mice

spontaneously develop uveitis and glaucoma by means of progressive uveal
hyperpigmentation, pigment dispersion, and Th1 type inflammation

. Upre-

gulation of gene expression of the IL-18 gene, a known inducing factor for IFNc,
was detected in uveal tissue and aqueous humor of DBA/2J mice. This evidence
suggests that Th1 immunity may be inherent to the progression of ocular disease
in these mice. DBA-2J mice aged 2, 4, and 6 months could not generate a signif-
icant ACAID response to OVA

. DBA/2J mice aged 7 months did generate

an ACAID response, however, despite ongoing and progressive uveitis. This
may be linked to an observed increase in the density of nerve terminals in the
uvea containing calcitonin gene–related peptide (CGRP), a potent immunomod-
ulatory factor found in the aqueous humor

. These nerve terminals were

found in close proximity to the F4/80þ ocular APCs and may be a reason

314

BIROS

why ACAID was generated in the ageing but not younger DBA/2J mice. In other
experimental models, CGRP was able to suppress Th1 immunity in the skin,
possibly through the influence of anti-Th1 cytokines, such as IL-10 and IL-4

[81–83]

; thus, it is plausible that CGRP may also influence the generation of

ACAID in DBA/2J mice through similar mechanisms.

OTHER MODELS OF IMMUNE TOLERANCE SIMILAR TO
ANTERIOR CHAMBER–ASSOCIATED IMMUNE DEVIATION

Peripheral tolerance is not exclusive to the eye. The testis, brain, and maternal/
fetal interfaces, for example, are also immune privileged sites where foreign
grafts enjoy enhanced survival as the result of a variant form of regional immu-
nity promoting tolerance

[84–88]

. There are other organs, such as the gut and

vasculature, that may also produce a type of non–immune-privileged immune
tolerance

. Other types of peripheral tolerance are not identical to ACAID.

Intravenous tolerance, for example, did not produce CD8þ T

reg

cells and did

not require CD1d or iNKT cells

[28,49]

. Certain tumors showed a type of

peripheral tolerance that requires CD1d similar to ACAID

. In the tumor

model, however, iNKT cells produced IL-13 rather than IL-10.

IS THERE A DOWNSIDE TO ANTERIOR CHAMBER–ASSOCIATED
IMMUNE DEVIATION?

If ACAID so effectively suppresses DTH, one may consider that with ACAID,
DTH suppression leads to intraocular pathogen survival causing more damage
over time to the eye, and possibly a threat to the host. One observation that sup-
ports this theory was noted in human patients who have acute retinal necrosis
(ARN)

[1,90]

. Patients who had ARN attributable to acute varicella-zoster virus

(VZV) showed a lack of DTH to viral antigens similar to an ACAID response

[1,91]

. The antigen-specific DTH was restored as the patients recovered from

the viral disease. Therefore, if ACAID was present in response to VZV, it
may have participated in the pathogenesis of ARN by promoting viral survival.
In another example, certain parasites had the ability to release immunomodula-
tory cytokine homologues, such as TGFb, TGFb receptor, and migration inhibi-
tory factor, which may themselves lead to a deviant immune response or
contribute to and propagate an ongoing ACAID response

[92,93]

. If ACAID

does ‘‘allow’’ pathogens to survive DTH immunity, it may do so by drawing
a balance between pathogen destruction and ocular tissue survival. Cases of
infectious uveitis from Toxoplasma are often treated only with anti-inflamma-
tories, because drugs that kill toxoplasmic organisms may also lead to worsening
uveitis and vision loss. Clinicians may be placed in a position to decide between
treatment that preserves vision and treatment that kills the pathogen. At times,
both goals cannot be achieved simultaneously.

Nonetheless, the principles of ACAID have been used successfully in animal

models to suppress not only uveitis

[94,95]

but nonocular diseases, including

experimental autoimmune encephalomyelitis

[96]

, hapten immune pulmonary

interstitial fibrosis

[97]

, and OVA-induced allergic pulmonary inflammation

315

ANTERIOR CHAMBER-ASSOCIATED IMMUNE DEVIATION

(a Th2-mediated disease)

[20,22]

. ACAID APCs can be induced in vitro and used

to promote peripheral tolerance as a cell-based therapy in immune-mediated
disease without anterior chamber antigen inoculation. In vitro ACAID, or
‘‘ACAID in a dish,’’ was generated when TGF-b2–treated antigen-pulsed perito-
neal exudate cells (PECs)

[1,8,29,31,94,98–101]

, macrophage hybridoma cells

[100,102,103]

, or bone marrow–derived APCs

[97]

were cultured with naive

spleen cells. In general, these APCs acted like semimature dendritic cells similar
to other types of tolerogenic APCs, with low expression of CD40 and IL-12 and
high expression of IL-10 and TGFb

[104]

. The resulting mix of cells produced

reg

cells that were able to suppress DTH in vivo. In addition, intravenous injec-

tion of the resulting TGF-b2–treated PECs induced the same tolerogenic cellular
interactions as antigen presentation in the anterior chamber. One could imagine
peripheral blood APCs treated with TGF-b2 and antigen ex vivo and then given
back to the patient to ameliorate signs of immune-mediated disease.

SUMMARY

The eye exhibits a highly specialized regional immune response, ACAID, which
involves two other organs: the spleen and the thymus. The autonomic nervous
system, vasculature, and specialized immune cells link these three organs in
ACAID, creating a confluent response that provides the eye with protection
from Th1 immunity and allows the tissues of the eye to work in the precise
and complex manner for which they are made. Perhaps in many veterinary oph-
thalmic disease conditions, including cataract-induced uveitis, Encephalitozoon cuni-
culi infection in rabbits, and uveal melanoma in cats and dogs, ACAID may have
a central role in directing disease pathogenesis. Without ACAID, the eye would
be much more vulnerable to scarring, adhesions, bleeding, and cellular infiltrates,
which would ultimately lead to more severe permanent vision loss and pain.

After decades of research, we now know many of the processes in ACAID

and have made much progress since the early studies of Van Dooremaal

. In

reality, however, we only have a partial understanding of ACAID’s complex
mechanisms. The future of ACAID research is primarily to continue character-
ization of the components of ACAID and to move steadily toward use of
ACAID principles to benefit the eye, and possibly other organs, in clinical
situations, such as refractory Th1 inflammation in autoimmunity, infection,
and cancer.

References

[1] Kezuka T, Sakai J, Usui N, et al. Evidence for antigen-specific immune deviation in patients

with acute retinal necrosis. Arch Ophthalmol 2001;119:1044–9.

[2] Streilein JW. Ocular immune privilege: therapeutic opportunities from an experiment of

nature. Nat Rev Immunol 2003;3:879–89.

[3] van Dooremaal J. Die Entwicklung der in fremden Grund versetzten lebenden Gewebe.

Albrecht von Graefes Arch Ophthalmol 1873;19:358–73.

[4] Medawar P. Immunity to homologous grafted skin. III. The fate of skin homografts trans-

planted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye. Br
J Exp Pathol 1948;29:58–69.

316

BIROS

[5] Kaplan HJ, Streilein JW, Stevens TR. Transplantation immunology of the anterior chamber

of the eye. II. Immune response to allogeneic cells. J Immunol 1975;115:805–10.

[6] Benson JL, Niederkorn JY. Immune privilege in the anterior chamber of the eye: alloantigens

and tumour-specific antigens presented into the anterior chamber simultaneously induce
suppression and activation of delayed hypersensitivity to the respective antigens. Immuno-
logy 1992;77:189–95.

[7] Streilein JW, Okamoto S, Hara Y, et al. Blood-borne signals that induce anterior chamber-

associated immune deviation after intracameral injection of antigen. Invest Ophthalmol Vis
Sci 1997;38:2245–54.

[8] Wilbanks GA, Mammolenti M, Streilein JW. Studies on the induction of anterior chamber-

associated immune deviation (ACAID). II. Eye-derived cells participate in generating
blood-borne signals that induce ACAID. J Immunol 1991;146:3018–24.

[9] Lin HH, Faunce DE, Stacey M, et al. The macrophage F4/80 receptor is required for the

induction of antigen-specific efferent regulatory T cells in peripheral tolerance. J Exp
Med 2005;201:1615–25.

[10] Ferguson TA, Herndon JM. The immune response and the eye: the ACAID inducing signal is

dependent on the nature of the antigen. Invest Ophthalmol Vis Sci 1994;35:3085–93.

[11] Ferguson TA, Kaplan HJ. The immune response and the eye. I. The effects of monoclonal

antibodies to T suppressor factors in anterior chamber-associated immune deviation
(ACAID). J Immunol 1987;139:346–51.

[12] Waldrep JC, Kaplan HJ. Anterior chamber-associated immune deviation induced by TNP-

splenocytes (TNP-ACAID). II. Suppressor T-cell networks. Invest Ophthalmol Vis Sci
1983;24:1339–45.

[13] Skelsey ME, Mayhew E, Niederkorn JY. CD25þ, interleukin-10-producing CD4þ T cells

are required for suppressor cell production and immune privilege in the anterior chamber
of the eye. Immunology 2003;110:18–29.

[14] Stein-Streilein J, Streilein JW. Anterior chamber associated immune deviation (ACAID):

regulation, biological relevance, and implications for therapy. Int Rev Immunol
2002;21:123–52.

[15] Streilein JW, Masli S, Takeuchi M, et al. The eye’s view of antigen presentation. Hum

Immunol 2002;63:435–43.

[16] Niederkorn JY. Immune privilege in the anterior chamber of the eye. Crit Rev Immunol

2002;22:13–46.

[17] D’Orazio TJ, Niederkorn JY. A novel role for TGF-beta and IL-10 in the induction of immune

privilege. J Immunol 1998;160:2089–98.

[18] Kosiewicz MM, Alard P, Streilein JW. Alterations in cytokine production following intraoc-

ular injection of soluble protein antigen: impairment in IFN-gamma and induction of TGF-
beta and IL-4 production. J Immunol 1998;161:5382–90.

[19] Li XY, D’Orazio LT, Niederkorn JY. Role of Th1 and Th2 cells in anterior chamber-associated

immune deviation. Immunology 1996;89:34–40.

[20] Katagiri K, Zhang-Hoover J, Mo JS, et al. Using tolerance induced via the anterior chamber

of the eye to inhibit Th2-dependent pulmonary pathology. J Immunol 2002;169:84–9.

[21] Kosiewicz MM, Streilein JW. Is deviant immunity induced by intraocular injection of anti-

gen dependent on Th2 cells? Invest Ophthalmol Vis Sci 1996;37:S1136.

[22] Nakamura T, Terajewicz A, Stein-Streilein J. Mechanisms of peripheral tolerance following

intracameral inoculation are independent of IL-13 or STAT6. J Immunol 2005;175:
2643–6.

[23] Streilein JW, Wilbanks GA. Evidence that anterior chamber and intravenous antigen injec-

tions have distinctly different systemic immune consequences. Ocular Immunol Today
1990;255–8.

[24] D’Orazio TJ, Niederkorn JY. Splenic B cells are required for tolerogenic antigen presenta-

tion in the induction of anterior chamber-associated immune deviation (ACAID). Immuno-
logy 1998;95:47–55.

317

ANTERIOR CHAMBER-ASSOCIATED IMMUNE DEVIATION

[25] Skelsey ME, Mayhew E, Niederkorn JY. Splenic B cells act as antigen presenting cells for

the induction of anterior chamber-associated immune deviation. Invest Ophthalmol Vis Sci
2003;44:5242–51.

[26] Vella AT, Scherer MT, Schultz L, et al. B cells are not essential for peripheral T-cell tolerance.

Proc Natl Acad Sci U S A 1996;93:951–5.

[27] Wilbanks GA, Streilein JW. Characterization of suppressor cells in anterior chamber-

associated immune deviation (ACAID) induced by soluble antigen. Evidence of two func-
tionally and phenotypically distinct T-suppressor cell populations. Immunology 1990;71:
383–9.

[28] Faunce DE, Sonoda KH, Stein-Streilein J. MIP-2 recruits NKT cells to the spleen during

tolerance induction. J Immunol 2001;166:313–21.

[29] Hara Y, Okamoto S, Rouse B, et al. Evidence that peritoneal exudate cells cultured with eye-

derived fluids are the proximate antigen-presenting cells in immune deviation of the ocular
type. J Immunol 1993;151:5162–71.

[30] Sonoda KH, Faunce DE, Taniguchi M, et al. NK T cell-derived IL-10 is essential for the

differentiation of antigen-specific T regulatory cells in systemic tolerance. J Immunol
2001;166:42–50.

[31] Wilbanks GA, Streilein JW. Studies on the induction of anterior chamber-associated

immune deviation (ACAID). 1. Evidence that an antigen-specific, ACAID-inducing, cell-
associated signal exists in the peripheral blood. J Immunol 1991;146:2610–7.

[32] Whittum JA, Niederkorn JY, McCulley JP, et al. Intracameral inoculation of herpes simplex

virus type I induces anterior chamber associated immune deviation. Curr Eye Res 1982;2:
691–7.

[33] Egan RM, Yorkey C, Black R, et al. Peptide-specific Tcell clonal expansion in vivo following

immunization in the eye, an immune-privileged site. J Immunol 1996;157:2262–71.

[34] Gordon S, Lawson L, Rabinowitz S, et al. Antigen markers of macrophage differentiation in

murine tissues. Curr Top Microbiol Immunol 1992;181:1–37.

[35] Austyn JM, Gordon S. F4/80, a monoclonal antibody directed specifically against the

mouse macrophage. Eur J Immunol 1981;11:805–15.

[36] McKnight AJ, Gordon S. The EGF-TM7 family: unusual structures at the leukocyte surface.

J Leukoc Biol 1998;63:271–80.

[37] McKnight AJ, Gordon S. EGF-TM7: a novel subfamily of seven-transmembrane-region

leukocyte cell-surface molecules. Immunol Today 1996;17:283–7.

[38] Sohn JH, Bora PS, Suk HJ, et al. Tolerance is dependent on complement C3 fragment iC3b

binding to antigen-presenting cells. Nat Med 2003;9:206–12.

[39] Sherman SH, Green K, Laties AM. The fate of anterior chamber fluorescein in the monkey

eye. 1. The anterior chamber outflow pathways. Exp Eye Res 1978;27:159–73.

[40] Sonoda KH, Stein-Streilein J. CD1d on antigen-transporting APC and splenic marginal

zone B cells promotes NKT cell-dependent tolerance. Eur J Immunol 2002;32:
848–57.

[41] Taylor AW. Immunoregulation of the ocular effector responses by soluble factors in aque-

ous humor. Reg Immunol 1994;6:52–7.

[42] Griffith TS, Yu X, Herndon JM, et al. CD95-induced apoptosis of lymphocytes in an immune

privileged site induces immunological tolerance. Immunity 1996;5:7–16.

[43] Kawashima H, Yamagami S, Tsuru T, et al. Anterior chamber inoculation of splenocytes

without Fas/Fas-ligand interaction primes for a delayed-type hypersensitivity response
rather than inducing anterior chamber-associated immune deviation. Eur J Immunol
1997;27:2490–4.

[44] Elzey BD, Griffith TS, Herndon JM, et al. Regulation of Fas ligand-induced apoptosis by

TNF. J Immunol 2001;167:3049–56.

[45] Niederkorn JY, Mayhew E, Mellon J, et al. Role of tumor necrosis factor receptor expression

in anterior chamber-associated immune deviation (ACAID) and corneal allograft survival.
Invest Ophthalmol Vis Sci 2004;45:2674–81.

318

BIROS

[46] Wang Y, Goldschneider I, Foss D, et al. Direct thymic involvement in anterior chamber-

associated immune deviation: evidence for a nondeletional mechanism of centrally
induced tolerance to extrathymic antigens in adult mice. J Immunol 1997;158:2150–5.

[47] Wang Y, Goldschneider I, O’Rourke J, et al. Blood mononuclear cells induce regulatory

NK T thymocytes in anterior chamber-associated immune deviation. J Leukoc Biol
2001;69:741–6.

[48] Goldschneider I, Cone RE. A central role for peripheral dendritic cells in the induction of

acquired thymic tolerance. Trends Immunol 2003;24:77–81.

[49] Sonoda KH, Exley M, Snapper S, et al. CD1-reactive natural killer T cells are required for

development of systemic tolerance through an immune-privileged site. J Exp Med
1999;190:1215–26.

[50] Streilein JW, Niederkorn JY. Induction of anterior chamber-associated immune deviation

requires an intact, functional spleen. J Exp Med 1981;153:1058–67.

[51] Faunce DE, Stein-Streilein J. NKTcell-derived RANTES recruits APCs and CD8þ Tcells to the

spleen during the generation of regulatory Tcells in tolerance. J Immunol 2002;169:31–8.

[52] D’Orazio TJ, Mayhew E, Niederkorn JY. Ocular immune privilege promoted by the presen-

tation of peptide on tolerogenic B cells in the spleen. II. Evidence for presentation by Qa-1.
J Immunol 2001;166:26–32.

[53] Niederkorn JY, Chiang EY, Ungchusri T, et al. Expression of a nonclassical MHC class Ib

molecule in the eye. Transplantation 1999;68:1790–9.

[54] Ikarashi Y, Mikami R, Bendelac A, et al. Dendritic cell maturation overrules H-2D-mediated

natural killer T (NKT) cell inhibition: critical role for B7 in CD1d-dependent NKT cell inter-
feron gamma production. J Exp Med 2001;194:1179–86.

[55] Maeda M, Lohwasser S, Yamamura T, et al. Regulation of NKT cells by Ly49: analysis of

primary NKT cells and generation of NKT cell line. J Immunol 2001;167:4180–6.

[56] Held W, Cado D, Raulet DH. Transgenic expression of the Ly49A natural killer cell receptor

confers class I major histocompatibility complex (MHC)-specific inhibition and prevents
bone marrow allograft rejection. J Exp Med 1996;184:2037–41.

[57] Skold M, Cardell S. Differential regulation of Ly49 expression on CD4þ and CD4CD8

(double negative) NK1.1þ T cells. Eur J Immunol 2000;30:2488–96.

[58] Ortaldo JR, Winkler-Pickett R, Mason AT, et al. The Ly-49 family: regulation of cytotoxicity

and cytokine production in murine CD3þ cells. J Immunol 1998;160:1158–65.

[59] Skelsey ME, Mellon J, Niederkorn JY. Gamma delta T cells are needed for ocular immune

privilege and corneal graft survival. J Immunol 2001;166:4327–33.

[60] Xu Y, Kapp JA. Gammadelta T cells are critical for the induction of anterior chamber-asso-

ciated immune deviation. Immunology 2001;104:142–8.

[61] Cai JL, Tucker PW. Gamma-delta T cells: immunoregulatory functions and immunoprotec-

tion. Chem Immunol 2001;79:99–138.

[62] Holtmeier W, Kabelitz D. Gamma-delta Tcells link innate and adaptive immune responses.

Chem Immunol Allergy 2005;86:151–83.

[63] Brandes M, Willimann K, Moser B. Professional antigen-presentation function by human

gamma-delta T cells. Science 2005;309:264–8.

[64] Li X, Taylor S, Zegarelli B, et al. The induction of splenic suppressor T cells through an

immune-privileged site requires an intact sympathetic nervous system. J Neuroimmunol
2004;153:40–9.

[65] Vega J, Keino H, Streilein JW. Immune privilege in the anterior chamber of the eye is

disrupted by removal of the superior cervical ganglion. Fort Lauderdale: American Uveitis
Society Meeting; April 24, 2004.

[66] Bellinger DL, Lorton D, Romano TD, et al. Neuropeptide innervation of lymphoid organs.

Ann N Y Acad Sci 1990;594:17–33.

[67] Madden KS, Felten SY, Felten DL, et al. Sympathetic neural modulation of the immune sys-

tem. I. Depression of T cell immunity in vivo and vitro following chemical sympathectomy.
Brain Behav Immun 1989;3:72–89.

319

ANTERIOR CHAMBER-ASSOCIATED IMMUNE DEVIATION

[68] Rice PA, Boehm GW, Moynihan JA, et al. Chemical sympathectomy increases the innate

immune response and decreases the specific immune response in the spleen to infection
with Listeria monocytogenes. J Neuroimmunol 2001;114:19–27.

[69] Rook GA, Lightman SL, Heijnen CJ. Can nerve damage disrupt neuroendocrine immune

homeostasis? Leprosy as a case in point. Trends Immunol 2002;23:18–22.

[70] Streilein JW, Bradley D, Sano Y, et al. Immunosuppressive properties of tissues obtained

from eyes with experimentally manipulated corneas. Invest Ophthalmol Vis Sci
1996;37:413–24.

[71] Sonoda Y, Streilein JW. Impaired cell-mediated immunity in mice bearing healthy ortho-

topic corneal allografts. J Immunol 1993;150:1727–34.

[72] She SC, Steahly LP, Moticka EJ. Intracameral injection of allogeneic lymphocytes enhances

corneal graft survival. Invest Ophthalmol Vis Sci 1990;31:1950–6.

[73] Niederkorn JY, Mellon J. Anterior chamber-associated immune deviation promotes corneal

allograft survival. Invest Ophthalmol Vis Sci 1996;37:2700–7.

[74] Geiger K, Sarvetnick N. Local production of IFN-gamma abrogates the intraocular immune

privilege in transgenic mice and prevents the induction of ACAID. J Immunol 1994;153:
5239–46.

[75] Ohta K, Wiggert B, Taylor AW, et al. Effects of experimental ocular inflammation on ocular

immune privilege. Invest Ophthalmol Vis Sci 1999;40:2010–8.

[76] Mo JS, Streilein JW. Immune privilege persists in eyes with extreme inflammation induced

by intravitreal LPS. Eur J Immunol 2001;31:3806–15.

[77] Mo JS, Streilein JW. Analysis of immune privilege in eyes with Mycobacteria tuberculosa

adjuvant-induced uveitis. Ocul Immunol Inflamm 2005;13:139–47.

[78] John SW, Smith RS, Savinova OV, et al. Essential iris atrophy, pigment dispersion, and

glaucoma in DBA/2J mice. Invest Ophthalmol Vis Sci 1998;39:951–62.

[79] Mo JS, Wang W, Kaplan HJ. Impact of inflammation on ocular immune privilege. Chem

Immunol Allergy 2007;92:155–65.

[80] Adamek-Kotowicz E, Lutjen-Drecoll E. Development of F4/80 positive cells and CGRP

nerve terminals in the mouse uvea (abstract). Invest Ophthalmol Vis Sci 2003;44:
744.

[81] Kitazawa T, Streilein JW. Hapten-specific tolerance promoted by calcitonin gene-related

peptide. J Invest Dermatol 2000;115:942–8.

[82] Levite M. Nerve-driven immunity. The direct effects of neurotransmitters on T-cell function.

Ann N Y Acad Sci 2000;917:307–21.

[83] Streilein JW, Alard P, Niizeki H. Neural influences on induction of contact hypersensitivity.

Ann N Y Acad Sci 1999;885:196–208.

[84] Wenkel H, Streilein JW, Young MJ. Systemic immune deviation in the brain that does not

depend on the integrity of the blood-brain barrier. J Immunol 2000;164:5125–31.

[85] Aluvihare VR, Kallikourdis M, Betz AG. Regulatory T cells mediate maternal tolerance to

the fetus. Nat Immunol 2004;5:266–71.

[86] Moffett-King A. Natural killer cells and pregnancy. Nat Rev Immunol 2002;2:656–63.
[87] Robertson SA, Ingman WV, O’Leary S, et al. Transforming growth factor beta—a mediator

of immune deviation in seminal plasma. J Reprod Immunol 2002;57:109–28.

[88] Fijak M, Meinhardt A. The testis in immune privilege. Immunol Rev 2006;213:66–81.
[89] Terabe M, Matsui S, Noben-Trauth N, et al. NKT cell-mediated repression of tumor

immunosurveillance by IL-13 and the IL-4R-STAT6 pathway. Nat Immunol 2000;1:
515–20.

[90] Kezuka T. Immune deviation and ocular infections with varicella zoster virus. Ocul Immunol

Inflamm 2004;12:17–24.

[91] Rochat C, Polla BS, Herbort CP. Immunological profiles in patients with acute retinal necro-

sis. Graefes Arch Clin Exp Ophthalmol 1996;234:547–52.

[92] Maizels RM, Balic A, Gomez-Escobar N, et al. Helminth parasites—masters of regulation.

Immunol Rev 2004;201:89–116.

320

BIROS

[93] Goodridge HS, Wilson EH, Harnett W, et al. Modulation of macrophage cytokine

production by ES-62, a secreted product of the filarial nematode Acanthocheilonema
viteae. J Immunol 2001;167:940–5.

[94] Hara Y, Caspi RR, Wiggert B, et al. Analysis of an in vitro-generated signal that induces

systemic immune deviation similar to that elicited by antigen injected into the anterior
chamber of the eye. J Immunol 1992;149:1531–8.

[95] Okamoto S, Kosiewicz MM, Caspi RR, et al. ACAID as a potential therapy for established

experimental autoimmune uveitis. In: Nussenblatt RB, Whitcup SM, Caspi RR, et al, editors.
Advances in ocular immunology. Amsterdam: Elsevier; 1994. p. 195–8.

[96] Faunce DE, Terajewicz A, Stein-Streilein J. Cutting edge: in vitro-generated tolerogenic

APC induce CD8þ T regulatory cells that can suppress ongoing experimental autoimmune
encephalomyelitis. J Immunol 2004;172:1991–5.

[97] Zhang-Hoover J, Stein-Streilein J. Tolerogenic APC generate CD8þ T regulatory cells that

modulate pulmonary interstitial fibrosis. J Immunol 2004;172:178–85.

[98] Cousins SW, McCabe MM, Danielpour D, et al. Identification of transforming growth fac-

tor-beta as an immunosuppressive factor in aqueous humor. Invest Ophthalmol Vis Sci
1991;32:2201–11.

[99] Kezuka T, Streilein JW. In vitro generation of regulatory CD8þ Tcells similar to those found

in mice with anterior chamber-associated immune deviation. Invest Ophthalmol Vis Sci
2000;41:1803–11.

[100] Wilbanks GA, Mammolenti M, Streilein JW. Studies on the induction of anterior chamber-

associated immune deviation (ACAID). III. Induction of ACAID depends upon intraocular
transforming growth factor-beta. Eur J Immunol 1992;22:165–73.

[101] Wilbanks GA, Streilein JW. Fluids from immune privileged sites endow macrophages with

the capacity to induce antigen-specific immune deviation via a mechanism involving trans-
forming growth factor-beta. Eur J Immunol 1992;22:1031–6.

[102] Kuchroo VK, Minami M, Diamond B, et al. Functional analysis of cloned macrophage

hybridomas. VI. Differential ability to induce immunity or suppression. J Immunol
1988;141:10–6.

[103] Masli S, Turpie B, Hecker KH, et al. Expression of thrombospondin in TGFbeta-treated

APCs and its relevance to their immune deviation-promoting properties. J Immunol
2002;168:2264–73.

[104] Rutella S, Lemoli RM. Regulatory T cells and tolerogenic dendritic cells: from basic biology

to clinical applications. Immunol Lett 2004;94:11–26.

321

ANTERIOR CHAMBER-ASSOCIATED IMMUNE DEVIATION

Canine and Feline Uveitis

Wendy M. Townsend, DVM, MS

Department of Small Animal Clinical Sciences, D208 Veterinary Medical Center,
Michigan State University, East Lansing, MI 48824–1314, USA

MORPHOLOGY

The uvea is composed of the highly vascular and often pigmented iris, ciliary
body, and choroid. The iris and ciliary body comprise the anterior uvea. The
choroid comprises the posterior uvea. The uveal tract regulates the quantity of
light allowed into the eye by varying the pupillary aperture; produces aqueous
humor, which provides nourishment and removes waste from the cornea and
lens; and provides nutrients to the outer layers of the retina

TERMINOLOGY

Uveitis is simply inflammation within the uveal tract. More precise terminology
to describe the portion of the uveal tract involved includes iritis (inflammation
of the iris), iridocyclitis or anterior uveitis (inflammation of the iris and ciliary
body), choroiditis or posterior uveitis (inflammation of the choroid), and pan-
uveitis (inflammation of the entire uveal tract)

. Differentiating whether

inflammation arises from the iris or the ciliary body can be difficult because
of their close anatomic proximity and the similar clinical signs

. Posterior

uveitis can occur independent of anterior uveitis

. Endophthalmitis occurs

when ocular inflammation is confined to three or more tissues inside the eye

. Panophthalmitis indicates that ocular inflammation involves all layers of

the eye, including the sclera

. Determining the extent of involvement is clin-

ically important, because involvement of the posterior segment may decrease
the likelihood of maintaining a visual globe. Therapeutic agents must also be
administered systemically to achieve effects within the posterior segment.

PATHOPHYSIOLOGY

Uveitis is a significant cause of ocular disease in dogs and cats

[4,5]

. Uveitis

occurs after damage to uveal tissue or vasculature disrupts the ocular blood-
aqueous barrier (BAB) or the blood-retinal barrier

. The BAB is composed

of an epithelial barrier at the level of the nonpigmented ciliary epithelium
and an endothelial barrier at the level of the iridal blood vessels

[7,8]

. The

blood-retinal barrier is created by tight junctions within retinal capillaries and

E-mail address: townsend@cvm.msu.edu

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.12.004

Vet Clin Small Anim 38 (2008) 323–346

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

the cells of the retinal pigment epithelium

[9,10]

. The BAB prevents the move-

ment of molecules across the vascular endothelial surface

, which results in

aqueous humor protein concentrations 200 times less than that of plasma

Disruption of the BAB causes the aqueous humor protein concentration to
increase to greater than normal levels, and the resultant light scattering
(Tyndall effect) makes the beam from the slit-lamp visible as it traverses the
anterior chamber

. This phenomenon is known as flare and is a hallmark

of uveitis. Disruption of the blood-retinal barrier results in retinal edema, reti-
nal hemorrhage, and detachment of the neurosensory retina

The acute inflammatory phase of uveitis begins with brief arteriolar vasocon-

striction, followed by prolonged vascular dilation

. Prostaglandins and leuko-

trienes mediate the vasodilation and cause increased vascular permeability,
which results in the breakdown of the BAB. During episodes of anterior uveitis,
the intraocular prostaglandin concentrations may increase 200-fold

. Prosta-

glandins also induce hyperemia and reduction in intraocular pressure (IOP)

PGF

constricts the iris sphincter muscle, causing miosis and pain. In a recently

reported study measuring inflammatory mediators in aqueous humor after ante-
rior chamber paracentesis, an increase in PGE

levels was noted along with ele-

vations in inducible cyclooxygenase (COX)-2 and nitrites and nitrates

Breakdown of the BAB allows proteins, cells, and additional inflammatory

mediators entry into the iridal stroma and aqueous humor. Cytokines and che-
mokines are important chemotactic factors that recruit inflammatory cells. One
particularly important cytokine seems to be leukotriene B4, a classic chemoat-
tractant that triggers adherence of leukocytes to the endothelium and recruits
granulocytes and macrophages to the site of inflammation

. In an experi-

mental model of uveitis in mice, blockade of the leukotriene B4 receptor greatly
reduced the intensity of the ongoing disease

CLINICAL SIGNS
Nonspecific Signs

Blepharospasm, photophobia, excessive lacrimation, and enophthalmos are
nonspecific signs of ocular discomfort noted with uveitis but also with ulcera-
tive keratitis, scleritis, and glaucoma

. The globe often appears red because

of hyperemia of the deep perilimbal anterior ciliary vessels

[3,17]

. The en-

gorgement of these radially oriented vessels is called ciliary flush. The vascular
dilation occurs secondary to the elevation in prostaglandin levels. Deep vessels
may be distinguished from superficial conjunctival vessels by manipulation of
the conjunctiva and application of 1:1000 epinephrine or 2.5% phenylephrine
solution

. Deep vessels do not move with the conjunctiva and do not read-

ily blanch after the application of topical epinephrine. Uveitis may result in cor-
neal edema. Corneal edema may occur in association with uveitis as a result of
reduction of the endothelial sodium potassium (NaK)-ATPase or epithelial Na-
chlorine (Cl) pump activities or may be related to a rupture of the endothelial
cell-cell junction barrier

. Either mechanism allows hydration of the corneal

stroma noted clinically as corneal edema.

324

TOWNSEND

Anterior Segment Clinical Signs

The presence of aqueous flare confirms a diagnosis of anterior uveitis. Aqueous
flare denotes breakdown of the BAB and increased permeability of the ocular
vasculature. Because of this increase in vascular permeability, inflammatory
cells may be visualized within the aqueous humor or vitreous body. An accu-
mulation of purulent material within the anterior chamber is termed hypopyon
(

). Blood within the anterior chamber is termed hyphema (

). Both

may be noted during episodes of anterior uveitis. Aggregates of inflammatory
cells may adhere to the corneal endothelium and are then called keratic precip-
itates (

). Typically, the keratic precipitates are visualized on the ventral

one half to one third of the cornea, where they are deposited by the aqueous
humor thermal convention currents. Blepharospasm or an elevated third eyelid
may prevent visualization of the keratic precipitates by obscuring the ventral
portion of the cornea. Larger fatty-looking clusters of keratic precipitates
have been termed mutton fat keratic precipitates (

) and often indicate gran-

ulomatous inflammation

Miosis, sometimes quite marked, occurs in response to prostaglandins, par-

ticularly PGF

[20,21]

, and other inflammatory mediators that act directly on

the iris sphincter muscle

. The miosis and associated ciliary muscle spasm

contribute greatly to the pain associated with anterior uveitis. Failure to dilate
completely in response to the topical application of tropicamide 1% ophthalmic
solution can be a subtle sign of anterior uveitis

. Iridal swelling may cause the

iris to appear engorged or darker in color, possibly even yellow in animals with
normally blue irides.

The IOP typically decreases during uveitis, because the inflammatory pro-

cess leads to a reduction in active secretion of aqueous humor, possibly by

Fig. 1. Hypopyon fills the ventral quarter of this canine globe. Mild diffuse corneal edema is
present.

325

CANINE AND FELINE UVEITIS

means of interference with active transport mechanisms

. Prostaglandin

release may also contribute to the ocular hypotony by increasing aqueous hu-
mor outflow through the uveoscleral route

. Subtle ocular hypotony, for ex-

ample, an IOP within the normal range but 5 mm Hg less than the fellow eye is
a significant finding that may be an early indication of inflammation

[3,24,25]

Fig. 2. Moderate corneal edema obscures visualization of the hyphema present in the eye of
a dog with uveitis secondary to immune-mediated thrombocytopenia.

Fig. 3. Fine keratitic precipitates in a cat with idiopathic uveitis as seen with a slit-beam.
(Courtesy of David L. Williams, MA, VetMB, PhD, CertVOphthal, FRCVS, Cambridge,
England.)

326

TOWNSEND

Posterior Segment Clinical Signs

Examination of the posterior segment may reveal a cellular infiltrate within the
vitreous body as inflammatory cells diffuse into the vitreous from the pars pla-
na and pars plicata of the ciliary body. A complete fundic examination is nec-
essary to evaluate alteration within the retina and choroid. Because of the close
anatomic proximity of the retina and choroid, the choroid is infrequently in-
flamed as a sole process. The retina is typically involved primarily or second-
arily. Areas of grayish discoloration over the tapetal fundus and grayish to
white areas within the nontapetal fundus may occur as a result of retinal edema
or cellular infiltration

. More extensive inflammation may result in areas of

retinal detachment (

). The detachments may be bullous with a fluid exu-

date in the subretinal space, which allows visualization of the underlying tape-
tum (

). Retinal detachments may also be characterized by a cellular

infiltrate in the subretinal space, which appears as grayish to pink-white accu-
mulations of material beneath the retinal detachment. Hemorrhage may be
present within the vitreous, retina, or subretinal space. Close inspection of
the retinal vasculature may reveal changes in vascular caliber and tortuosity.
Sheathing of the retinal vessels by inflammatory cells, called perivascular cuff-
ing, occurs with some forms of uveitis.

Sequelae to Uveitis

Chronic inflammation of previous bouts of inflammation may incite multiple
changes within the globe. Chronic inflammation stimulates ingrowth of periph-
eral corneal vascularization. The vessels bud from the limbal vasculature

Matrix metalloproteinase 2 within the anterior chamber may be one of the
stimuli that incite corneal angiogenesis

The combination of inflammatory cells, fibrin, fibroblasts, iridal swelling,

and miosis may create adhesions of the iris to the lens capsule or cornea

Posterior synechiae occur if the iris is adherent to the anterior lens capsule.

Fig. 4. Larger mutton-fat keratic precipitates in a cat with feline infectious peritonitis. (Courtesy
of David L. Williams, MA, VetMB, PhD, CertVOphthal, FRCVS, Cambridge, England.)

327

CANINE AND FELINE UVEITIS

Anterior synechiae occur if the iris is adherent to the corneal endothelium. If
extensive, posterior synechiae may occlude the flow of aqueous from the pos-
terior chamber through the pupil into the anterior chamber, causing iris bombe´,
which is an anterior ballooning of the iris, and secondary glaucoma

. Pre-

iridal fibrovascular membranes may form on the anterior surface of the iris.
Angiogenic factors released by ischemic retina, neoplasms, or leukocytes
involved in ocular inflammation can incite endothelial budding from vessels

Fig. 5. Retinal detachment with perivascular infiltrates in a dog with idiopathic posterior
uveitis. (Courtesy of David L. Williams, MA, VetMB, PhD, CertVOphthal, FRCVS, Cambridge,
England.)

Fig. 6. Complete bullous retinal detachment is present in the right eye of a dog diagnosed
with canine ehrlichiosis.

328

TOWNSEND

in the anterior iridal stroma

. The membranes may extend onto the anterior

lens capsule or into the iridocorneal angle. Clinically, the term rubeosis iridis is
applied, because the neovascular membrane on the anterior iridal surface
causes a reddish cast to the iris (

)

Chronic inflammation often induces the formation of cataracts, presumably

by the diffusion of inflammatory mediators across the lens capsule causing lens
epithelial metaplasia, necrosis, or posterior migration and lens fiber degenera-
tion, liquefaction, and necrosis

[29,30]

. Chronic uveitis may also result in

lens luxation. The inflammatory products within the aqueous humor cause
degradation of the zonular fibers, which then allows the lens to move from
its normal position within the patellar fossa

. Although this process seems

to be relatively rare in dogs

, chronic uveitis seems to be a frequent cause

of lens luxation in the cat

. Secondary glaucoma may occur from pupillary

block as a result of iris bombe´ or lens luxation or may be attributable to occlu-
sion of the iridocorneal angle by peripheral anterior synechiae

Resolution of chorioretinitis may leave areas of retinal degeneration demar-

cated as areas of tapetal hyperreflectivity in the tapetal fundus as the overlying
retina has thinned and mottled pigmentation in the nontapetal fundus. Hyper-
trophy of the retinal pigment epithelium may be noted as areas of dense pig-
mentation in areas of previous retinal detachment. If marked choroidal
inflammation was present, there may be changes in tapetal coloration, pigment
clumping, or loss of choroidal pigment, which exposes the choroidal vessels or
sclera. Retinal vascular attenuation may be generalized or occur overlying the
areas of retinal degeneration. Finally, phthisis bulbi may occur as chronic cycli-
tis, and the resultant tissue atrophy and fibrosis destroy the ability of the ciliary
body to produce aqueous humor. Because the normal IOP can no longer be
maintained, the globe begins to shrink

. The histologic hallmarks are an atro-

phic and disorganized globe typically characterized by a cyclitic membrane and
variable degrees of chronic inflammation

. Fibrous or osseous metaplasia

may occur as well.

Fig. 7. The fine branching vasculature noted on the iridal surface is termed rubeosis iridis and
has developed secondary to chronic idiopathic uveitis in this cat.

329

CANINE AND FELINE UVEITIS

Diagnostic Tests

Fluorescein staining should be performed routinely to rule out the presence of
ulcerative keratitis and the possibility of a reflex uveitis

. The presence of

ulcerative keratitis precludes the use of topical corticosteroids.

Because many significant systemic diseases can induce uveitis, a thorough

physical examination, complete blood cell count, serum biochemistry profile,
and urinalysis should be performed. During the physical examination, particu-
lar attention should be paid to the integument, lymphatic system, thoracic aus-
cultation, and abdominal palpation. Thoracic radiographs, abdominal
ultrasonographic examination, and select serologic titers are often valuable
tools in the diagnostic investigation.

If a marked cellular infiltrate is present, cytologic evaluation of aqueous hu-

mor may be beneficial in identifying etiologic agents or neoplastic cells, partic-
ularly in cases of lymphoma

. The risks of aqueous centesis are low in the

hands of an experienced ophthalmologist but do include cataract formation if
one contacts the anterior lens capsule, hyphema

, and exacerbation of the

uveitis. If marked vitreal infiltrates or cellular retinal detachments are present,
cytologic examination of those infiltrates is often more rewarding

. In a re-

port by Brightman and colleagues

, an etiologic agent was identified in 13

(65%) of 20 of cases using vitreous centesis and cytologic examination.

Aqueocentesis can also be used to determine the level of intraocular antibody

production. The ratio of aqueous antibody titer to serum titer is known as the
Goldman-Witmer coefficient, or C value. If greater than 1, this shows that an
intraocular infectious agent is causing iridal plasma cells to produce antibody,
thus demonstrating that an infectious agent is causing the uveitis rather than
merely being a bystander

[38,39]

Therapy

Primary treatment goals for uveitis are halting inflammation, stabilizing the
BAB, minimizing sequelae, decreasing pain, and preserving vision. The agents
used to attain those goals include topical mydriatics, topical (and, in select
cases, systemic) corticosteroids, and nonsteroidal anti-inflammatory drugs
(NSAIDs). If an underlying etiology can be detected, therapy should be
directed toward removal of the inciting agent or alleviating the associated sys-
temic disease.

Mydriatics

The mydriatic agent most often selected is 1% atropine ointment or solution.
Atropine is a selective, reversible, direct-acting anticholinergic agent

. Top-

ical administration results in pharmacologic blockage of the sphincter muscles
of the iris and ciliary body, leading to pupillary dilation and cycloplegia

The resultant pupillary dilation decreases the possibility of posterior synechia
formation. The pupillary dilation may exacerbate congestion of the iridocor-
neal angle, however, and thereby decrease aqueous outflow. Therefore, atro-
pine must be used with caution in patients that have or are at risk for
secondary glaucoma. The cycloplegia greatly lessens the pain associated with

330

TOWNSEND

ciliary spasm. Administration of atropine also decreases the permeability of oc-
ular blood vessels to proteins and intravenously administered fluorescein,
thereby stabilizing the BAB

[41,42]

In a normal canine or feline globe, the onset of action is within 30 to 60 min-

utes

[43,44]

and mydriasis may persist for up to 10 days after administration of

the last dose

. In uveitic globes, atropine is initially applied up to four times

daily to achieve mydriasis and is then administered once to twice daily to main-
tain mydriasis. Because the agent is bitter tasting, hypersalivation or, infre-
quently, vomiting may be noted after the instilled topical solution travels
through the lacrimal puncta and nasolacrimal ducts and is tasted

. Cats

may have particularly marked reactions

. Therefore, use of the ointment

preparation is indicated in feline or potentially sensitive patients.

Corticosteroids

Anti-inflammatory therapy is a key element of therapy for uveitis. Corticoste-
roids are frequently used to suppress inflammation because they reduce pro-
duction of metabolites of the arachidonic acid (inflammatory) cascade

Glucocorticoids upregulate lipocortin expression

. Lipocortin is a key inhib-

itor of proinflammatory substances, such as phospholipase A

, the enzyme re-

sponsible for initiating the arachidonic acid cascade

. Glucocorticoids can

also directly reduce PGE synthesis and increase vascular stability

To control anterior uveitis, topical application of the corticosteroid is often the

preferred route because it allows for high local drug concentrations and minimal
systemic side effects

. Prednisolone acetate 1% ophthalmic suspension and

dexamethasone 0.1% are able to penetrate an intact corneal epithelium

Therefore, they are the only topical corticosteroids that can achieve therapeutic
concentrations in the aqueous humor

. Each of these agents may inhibit 40%

to 50% of protein exudation from the iris and ciliary body

. The initial fre-

quency of application may be every 2 to 4 hours depending on the severity of
the inflammation. Once clinical improvement is noted, the frequency may be de-
creased but the therapy should be gradually tapered to diminish the likelihood of
recurrence

. Topical corticosteroids are contraindicated in the presence of cor-

neal ulceration because they delay normal wound healing.

Use of glucocorticoids, such as prednisone, prednisolone, or dexamethasone,

to suppress inflammation within the posterior segment, for example, chorioreti-
nitis, requires systemic administration of anti-inflammatory or, occasionally, im-
munosuppressive doses

. The dose is then incrementally decreased based on

the response to therapy

. When using corticosteroids, one risks exacerbation

of clinical signs if an infectious cause is present but has not yet been identified

Systemic side effects, including endocrinopathies, may result with long-term use.

Nonsteroidal Anti-Inflammatory Drugs

The NSAIDs can be particularly useful in cases of mild anterior uveitis, as
adjunctive therapy when combined with topical corticosteroids, and to control
posterior segment inflammation when an infectious etiology has not been com-
pletely ruled out. The NSAIDs block the conversion of arachidonic acid to

331

CANINE AND FELINE UVEITIS

prostaglandins by COX

. Prostaglandins are key mediators of ocular

inflammation, causing breakdown of the BAB, exacerbating photophobia,
and lowering the ocular pain threshold

. The NSAIDs are also beneficial

because they have been shown to suppress polymorphonuclear cell locomotion
and chemotaxis

, decrease expression of inflammatory cytokines

, and

function as free radical scavengers

. Although many systemic NSAIDs are

currently available, only a few have been evaluated to assess their efficacy in
controlling ocular inflammation. Flunixin meglumine and aspirin have been
shown to stabilize the BAB in experimental models of uveitis

[57,58]

. Flunixin

meglumine had good effects, whereas the effects of aspirin were moderate
when compared with the placebo. Both drugs induced some gastrointestinal
bleeding, however. In a pilocarpine-induced model of uveitis in dogs, carprofen
resulted in a 68% inhibition of aqueous flare

. In a clinical study, tolfenamic

acid was shown to control postoperative intraocular inflammation

[60]

. When

selecting a systemic NSAID, one must be cognizant of the potential systemic
side effects, particularly the adverse gastrointestinal effects in all species

and the potential for bone marrow suppression and hemorrhage in cats

Topical NSAIDs may be used to control mild inflammation or may be com-

bined with topical corticosteroids to improve control of more severe ocular
inflammation

. The application frequency typically varies from two to

four times per day

[62,63]

. The relative efficacies have been studied in an

anterior chamber paracentesis model and the order of BAB stabilizing efficacy
was as follows: diclofenac greater than flurbiprofen, flurbiprofen greater than
suprofen, and suprofen greater than tolmetin, which was equal to the control
solution

. One must use caution in canine eyes with the potential for sec-

ondary glaucoma because the topical NSAIDs have been found to elevate
the IOP

. One must also exercise caution when using topical NSAIDs in

the presence of corneal ulceration. In people, topical NSAIDs have been asso-
ciated with marked corneal collagenolysis

Cases of immune-mediated uveitis that require long-term maintenance with

systemic glucocorticoids or that fail to respond to conventional therapy may
require use of immunosuppressive drugs, such as azathioprine. Azathioprine
is a purine analogue with relatively select cytotoxicity for T helper lymphocytes

. Conditions like uveodermatologic syndrome and pigmentary uveitis are

the conditions in which azathioprine is most frequently used

[67,68]

. The initial

dose in dogs is 2 mg/kg every 24 hours

. The dose for long-term therapy is

typically decreased to 0.5 to 1 mg/kg every other day. The lag period before
successful treatment is recognized ranges from 3 to 5 weeks

. Complete

blood cell counts should be monitored, because bone marrow suppression is
a concern. Gastrointestinal side effects and hepatotoxicity may be noted as
well

Causes of Anterior Uveitis

A plethora of etiologies may incite uveitis. Infectious diseases, neoplasia, and
immune-mediated conditions may all present with clinical signs of uveitis.

332

TOWNSEND

For the purposes of this discussion, the etiologies are grouped into noninfec-
tious and infectious causes of uveitis.

NONINFECTIOUS CAUSES OF UVEITIS
Idiopathic Uveitis

Unfortunately, most cases of uveitis remain idiopathic despite intensive sys-
temic evaluations. In a study by Massa and colleagues

, 60% of cases of

dogs that had uveitis were classified as idiopathic or immune mediated, because
an underlying systemic cause could not be identified. The dogs with idiopathic
uveitis were typically middle aged, did not exhibit any signs of systemic illness,
and more often were presented with unilateral uveitis. In the study by Massa
and colleagues

, the degree of inflammation and ocular lesions did not vary

between those dogs with infectious, neoplastic, or idiopathic uveitis. In studies
of feline uveitis, approximately 30% to 62% of affected cats had no identifiable
concurrent systemic disease

[70,71]

. Although the underlying etiology may

often remain obscure, a complete diagnostic investigation remains essential
because of the severity of the systemic diseases associated with uveitis.

Lens-Induced Uveitis

An excellent review article on this topic has been written by van der Woerdt

[72]

, and the reader is referred to that text for additional information. Two dif-

ferent forms of uveitis may be initiated by lenticular pathologic findings. A lym-
phoplasmacytic inflammatory process, termed phacolytic uveitis, occurs in
association with hypermature or rapidly forming cataracts

. Phacolytic uve-

itis is typically mild. The prevalence has been reported to be as high as 71% in
dogs screened for cataract surgery

[73]

. The high prevalence is not surprising,

because fluorophotometric studies have demonstrated breakdown of the BAB
in association with all stages of cataracts

. The lens-associated inflammation

is proposed to occur after deviation of the normal low level of T-cell–mediated
tolerance to lens proteins

[72,75]

A more dramatic form of uveitis, termed phacoclastic uveitis, occurs in associ-

ation with rupture of the lens capsule and release of lens proteins and mem-
brane-associated antigens

. Histologic examination of affected globes

revealed intralenticular neutrophils and a surrounding inflammatory response
that ranges from suppurative to lymphocytic in nature

. In a study by Da-

vidson and colleagues

, prompt surgical removal of the lens material re-

sulted in a visual eye in most cases. In contrast, attempts at medical
management resulted in the loss of vision. In cats, because of the risk for trau-
matic ocular sarcoma, surgical removal of the lens material or globe is recom-
mended if the eye cannot be salvaged. Rupture of the lens capsule is believed to
induce traumatic ocular sarcoma, although the time from trauma to detection
of the tumor averages 5 years

[5,78]

Trauma

Blunt trauma and penetrating trauma can incite uveitis and are common causes
of uveitis in domestic animals

. Hyphema and varying amounts of fibrin are

333

CANINE AND FELINE UVEITIS

often present in cases of traumatic uveitis

. In cases of penetrating trauma,

one must assess the extent of damage within the globe, including whether
the lens capsule has been ruptured. Ultrasonographic examination may be
required to evaluate the extent of the damage fully. The administration of
broad-spectrum systemic antibiotics is strongly suggested, because bacterial
or fungal contamination may occur at the time of globe penetration

. En-

dophthalmitis can progress rapidly and cause loss of the globe. If marked hy-
phema is present, administration of tropicamide can mobilize the pupil and
assist in preventing the development of synechia

. Administration of tropica-

mide is preferable to atropine, because tropicamide produces greater iridoplegia
than cycloplegia, and thus less often produces increases in IOP

Golden Retriever Pigmentary Uveitis

Primarily reported in the golden retriever, pigmentary uveitis is characterized
by anterior segment inflammation and the deposition of pigment on the ante-
rior lens capsule, often in a radial fashion

. No systemic signs are associated

with this condition. The mean age at presentation is 8.6 years

. In a report

by Sapienza and colleagues

, common sequelae were cataract formation

(37%) and secondary glaucoma (46%). In a report by Deehr and Dubielzig

, uveal cysts were noted in 15 of 18 eyes and were thought to be an impor-

tant factor in the development of glaucoma. In the report by Sapienza and col-
leagues

, uveal cysts were a common finding on histologic examination of

the enucleated glaucomatous blind eyes, whereas they were only noted clini-
cally in 13% of cases. Interestingly in those globes, microscopically, little inflam-
mation was noted. Therapy often consists of combinations of topical and
systemic corticosteroids, topical NSAIDs, medications to control secondary
glaucoma, and azathioprine

. Administration of topical NSAIDs seems to

exacerbate ocular hypertension frequently

[65,67]

Uveodermatologic Syndrome

Uveodermatologic syndrome, or Vogt-Koyanagi-Harada-like syndrome, is an
autoimmune condition of dogs in which melanocytes become the target of
the cellular response

. An immunohistochemical examination of affected tis-

sues from two dogs revealed that the skin lesions were mediated by T cells and
macrophages (T helper [Th] 1 immunity), whereas the ocular lesions were
more consistent with a B-cell and macrophage response (Th2 immunity).
The breeds primarily affected are the akita, samoyed, Siberian husky, and Shet-
land sheepdog

. The condition does occur sporadically in other breeds, how-

ever. Affected patients are usually presented with anterior uveitis or panuveitis
characterized by iridal or choroidal depigmentation, bullous retinal detach-
ment, or blindness

. The ocular lesions may precede the cutaneous lesions,

which include poliosis and vitiligo of the facial mucocutaneous junctions, nasal
planum, scrotum, and footpads

[81,82]

. Generalized vitiligo may also occur

[81]

. Because of the chronic nature of the disease, affected patients typically

develop extensive posterior synechia, iris bombe´, cataract, and secondary glau-
coma. Immunosuppressive drugs are the mainstay of therapy

. Azathioprine

334

TOWNSEND

is often combined with or substituted for corticosteroids to avoid the side ef-
fects associated with chronic systemic corticosteroid administration.

Neoplasia-Associated Uveitis

The presence of any neoplastic process, whether primary or metastatic, within
the globe may induce clinical signs of uveitis, such as iris hyperpigmentation,
intraocular fibrin exudation, and hemorrhage

. Therefore, the possibility

of associated neoplasia must always be considered in patients that have uveitis,
particularly if the inflammatory response or secondary glaucoma precludes
complete visualization of the intraocular compartments

. The more com-

mon primary intraocular tumors include melanomas (dogs and cats)

[5,84]

and iridociliary epithelial tumors (dogs)

[85]

. Lymphosarcoma is the most

frequent metastatic intraocular tumor in cats and dogs

[5,86,87]

. Ocular

involvement in canine lymphoma may include anterior uveitis, posterior uve-
itis, panuveitis, retinal hemorrhage, and superficial disease

[87]

. Metastasis of

angioinvasive pulmonary carcinoma has been described in four cats

. Oph-

thalmic examination revealed wedge-shaped tan discoloration of the tapetal
fundus, variable but mild serous exudation under the retina, and profoundly
attenuated retinal vasculature

Infectious Causes of Uveitis in Cats

The more common infectious causes of uveitis in cats are feline infectious peri-
tonitis (FIP), feline leukemia virus (FeLV), feline immunodeficiency virus
(FIV), toxoplasmosis, and the systemic mycoses. Bartonella henselae has also
been proposed as a frequent cause of feline uveitis

. According to previ-

ously published reports, between 38% and 70% of cats with uveitis have an as-
sociated systemic disease

[70,71]

Feline Infectious Peritonitis

FIP is caused when the immune response to feline coronavirus induces granu-
lomatous necrotizing phlebitis and periphlebitis, protein-rich effusions into
body cavities, and granulomatous inflammatory lesions in multiple organs

[90,91]

. The associated clinical signs include febrile episodes, weight loss, an-

orexia, depression, debility, and variable thoracic and abdominal involvement

. The uveitis associated with FIP is typically a panuveitis or panophthalmitis

with diffuse and severe corneal edema, marked anterior uveitis, marked cellu-
lar infiltration of the vitreous (ie, vitritis), chorioretinitis, inflammatory retinal
detachments, or optic neuritis

. Mutton fat keratic precipitates with occa-

sional admixed hemorrhage occur most often with FIP because of the granulo-
matous nature of the infection. The granulomatous periphlebitis may be noted
as perivascular cuffing surrounding the retinal vasculature.

The diagnosis of FIP can be challenging, because enteric coronaviruses cross-

react and cause positive results on serologic tests and reverse transcriptase
(RT)–polymerase chain reaction (PCR) assays. According to recommendations
from the FIP workshop symposium

, the first step in establishing a diagnosis

of FIP is to compare the signalment, history, and clinical findings with those of

335

CANINE AND FELINE UVEITIS

the typical individual infected with FIP. Most cats with FIP are from 6 months
to 3 years of age, come from shelters or catteries, and show signs of cyclic an-
tibiotic-resistant fevers and specific physical manifestations depending on the
form of the disease and location of lesions

. Diagnostic test findings support-

ive of a diagnosis of FIP include characteristic analysis of peritoneal or pleural
effusions, leukocytosis with neutrophilia and lymphopenia, hyperglobulinemia,
hypoalbuminemia, increased fibrinogen, and nonregenerative anemia of
chronic disease

. Finally, in a cat with suspect FIP, one may also submit ef-

fusions or surgical biopsies for immunohistochemistry or RT-PCR

Serology or RT-PCR performed on serum samples can confirm exposure to
feline coronavirus but must be paired with appropriate clinical signs to ensure
an accurate diagnosis

. No therapy has been proved effective in the manage-

ment of FIP

. Therapy remains symptomatic.

Bartonella spp

B henselae was first suggested as a cause of feline anterior uveitis by Lappin and
Black

in 1999 after a feline patient had a C value for IgG antibodies to

Bartonella spp of 4.42, which indicated antibody production by ocular tissues

[38,39]

, and no clinical response to therapy was noted until doxycycline was

administered

. Since the initial report, Ketring and colleagues

[93]

have

demonstrated elevated serum antibody production against Bartonella in cats
with uveitis. In a more recent study by Fontanelle and colleagues

[94]

, how-

ever, healthy cats were more likely to have elevated Bartonella titers than cats
with uveitis. Therefore, serum antibody tests alone do not seem to be sufficient
to confirm a diagnosis of Bartonella-induced uveitis. A definitive therapeutic pro-
tocol to resolve Bartonella infection does not currently exist

[95,96]

. The authors

of one article

[93]

suggest azithromycin, doxycycline, or rifampin.

Feline Leukemia Virus

Illness in FeLV-infected cats results from the direct effects of the virus on bone
marrow or lymphoid tissue

[97]

. The uveitis in association with FeLV infection

is primarily a manifestation of lymphosarcoma. In the study by Peiffer and
Wilcock

[98]

, FeLV-associated lymphosarcoma was the third most frequent

cause of uveitis in cats after idiopathic lymphoplasmacytic uveitis and FIP-asso-
ciated uveitis. FeLV-induced lesions range from inflammatory cells and fibrin
within the anterior chamber to small iris nodules or extensive neoplastic infil-
tration

[99]

. Funduscopically, one may note the characteristic lesions of retinop-

athy of anemia, which may occur secondary to FeLV-related anemia

[100]

Cytologic examination of aqueous humor usually reveals variable amounts

of lymphocytes and occasional plasmacytes and neutrophils

. The presence

of lymphosarcoma may be confirmed based on bone marrow aspiration, lymph
node biopsy, and direct biopsy of intraocular masses

[101]

. An ELISA antigen

test of peripheral blood can demonstrate the presence of the FeLV antigen

[99]

The diagnosis of a persistent infection should be confirmed by performing an
immunofluorescent antibody (IFA) test or repeating the ELISA antigen test in 3
to 4 months

[99]

336

TOWNSEND

Feline Immunodeficiency Virus

Direct viral damage of ocular tissues, initiation of secondary immune phenom-
ena, or opportunistic infection secondary to immunosuppression may all con-
tribute to the uveitis associated with FIV infection

[102]

. Clinically, pars

planitis is often a significant feature of the disease, creating a ‘‘snow banking’’
phenomenon as cells are deposited throughout the anterior vitreous with
a greater density toward the periphery

[102]

. Keratic precipitates are uncommon

findings

[102]

. Coinfection with Toxoplasma gondii seems to increase the severity

of clinical signs in cats that are infected with FIV

[71,103]

, perhaps because

the defense mechanisms for T gondii depend on CD4þ cells

[103]

ELISA tests may be used to detect antibodies to FIV

[104]

. According to a recent

study

[105]

, the Snap Combo Plus (IDEXX Laboratories, Atlanta, GA) is the

best performing in-hospital test kit, and the MAPIC FIV test (Sinovus Biotech,
Inc., Lund, Sweden) should not be used because of the large number of invalid
results or results that are difficult to interpret. Unfortunately, vaccination of cats
for FIV produces antibodies that are indistinguishable from those used to diag-
nose FIV infection

[104]

. Currently, there is no method by which to differentiate

vaccinal antibodies from those produced by natural infection. Therefore, attain-
ing a definitive diagnosis in a cat whose vaccination status is unknown becomes
nearly impossible. Symptomatic therapy is used to control the uveitis.

Toxoplasmosis

T gondii is a well-recognized cause of retinitis, choroiditis, and anterior uveitis

. Tachyzoites target the eye and multiply intracellularly within ocular tis-

sues

[107]

. The classic funduscopic appearance is multifocal dark gray lesions

in the tapetal fundus and fluffy white infiltrates in the nontapetal fundus

T gondii has frequently been implicated as a main cause of acute and chronic
idiopathic feline anterior uveitis

. Supportive evidence for this hypothesis

includes the higher seroprevalence rate to T gondii in cats with uveitis

The only definitive diagnosis is histologic identification of the organism

[108]

. Because a histologic diagnosis is frequently not available, however, sero-

logic testing is the primary diagnostic modality. Exposure to the organism is
widespread within the feline population. Therefore, paired serologic titers are
recommended, and most cases remain suspect rather than confirmed

An ELISA test for IgM antibodies with a single titer greater than 1:256 or rising
IgG titers is considered consistent with an active infection

[108,109]

. More re-

cently, PCR (B1 gene) tests for T gondii–specific IgM and IgG have been eval-
uated in serum and aqueous humor samples

[110,111]

. T gondii–specific IgM or

IgG was detected in the serum but not in the aqueous humor in 34.8% of
healthy cats. In cats that had uveitis, T gondii–specific IgM or IgG was detected
in the serum of 72% and in the aqueous humor of 39.5%, suggesting that the
combination of serum and aqueous humor T gondii titers may be the most in-
formative and useful method of testing. Anti-inflammatory therapy should be
used to control the uveitis in conjunction with clindamycin hydrochloride at
a dose of 12.5 mg/kg administered orally twice daily for 14 to 21 days

[112]

337

CANINE AND FELINE UVEITIS

Systemic Mycoses

A granulomatous anterior uveitis, often with concurrent chorioretinitis, may be
associated with cryptococcosis

[113]

, histoplasmosis

[114]

, blastomycosis

[115]

or coccidioidomycosis

[116]

. From the literature, the incidence of blastomycosis

in cats seems to be rare, except for sporadic clusters

[117,118]

. In most cases,

hematogenous spread is the likely route of ocular involvement. Ocular crypto-
coccosis seems to be an exception, however, because extension from the central
nervous system occurs more frequently. The diagnostic protocol for mycotic
uveitides includes a complete physical examination, hematology, and clinical
chemistries

. Diagnosis is often achieved by identification of the organism

during cytologic examination of aspirates or impression smears from lymph
nodes, bone marrow aspirates, cerebrospinal fluid, or cutaneous lesions

Aqueous or vitreous paracentesis may be useful depending on the degree of
ocular involvement (ie, anterior or posterior segment)

. Histoplasmosis

may be effectively treated using itraconazole

[119]

. Blastomycosis may be

treated with itraconazole or fluconazole, although the response is often poor

[117,118]

INFECTIOUS CAUSES OF UVEITIS IN DOGS
Brucella canis

An excellent review article on this disease has been published by Wanke

Endophthalmitis, chorioretinitis, and hyphema have all been reported in asso-
ciation with Brucella canis infection

. Vinayak and colleagues

reported

that 14.2% of patients that had B canis infection demonstrated ocular signs.
Other clinical signs include reproductive tract lesions (eg, abortions, epididymi-
tis, failure to conceive)

, diskospondylitis, osteomyelitis, splenomegaly,

glomerulopathy, and meningoencephalitis

[122,123]

. Isolation of the organism

is considered the ‘‘gold standard’’ diagnostic test

. This can be difficult,

however. The rapid slide agglutination test is sensitive and can be performed
early in the stage of infection

. Positive results are confirmed with other

tests, including the tube agglutination test, agar-gel immunodiffusion, indirect
fluorescent antibody test, and ELISA

[122]

. Achieving complete eradication

of the organism is difficult

. Various suggested therapeutic regimens in-

clude minocycline and streptomycin, tetracycline and streptomycin, long-acting
oxytetracycline

, enrofloxacin

[124]

, and gentamicin

. Complete

resolution of ocular clinical signs and clearance of the organism have only
been reported in one case in the literature

Tick-Borne Diseases

Borrelia burgdorferi (Lyme disease), Ehrlichia spp, including canis, platys, and risticii,
and Rickettsia rickettsii (Rocky Mountain spotted fever [RMSF]) have all been im-
plicated as causative agents in cases of uveitis

[122,125]

. The ocular lesions are

similar and include anterior uveitis, hyphema, retinal hemorrhage, and retinal
detachment (see

). The diagnosis is typically based on the combination

338

TOWNSEND

of clinical signs and serologic testing. Serum ELISA and IFA serologic tests can
be used to document exposure to Lyme disease

[126]

. In patients previously

vaccinated for Lyme disease, Western blot immunoassays may be used to dif-
ferentiate natural exposure from vaccinal response

[126]

. The serum fluores-

cent antibody test is most reliable for the diagnosis of ehrlichial agents

[125]

IFA and ELISA serum antibody titers are available for the diagnosis of
RMSF

[125]

. The current therapeutic recommendations are also similar. Doxy-

cycline is administered at a dosage of 10 mg/kg every 24 hours for 1 month as
the primary therapy for Lyme disease in patients with positive serology and
clinical signs of disease

[126]

. Two case reports document favorable responses

to administration of doxycycline, sometimes paired with anti-inflammatory
doses of corticosteroids in cases of canine ehrlichiosis

[127,128]

. Doxycycline

is also used in the therapy of RMSF, and combination with systemic prednis-
olone has not been shown to decrease efficacy

[129]

Leptospirosis

Uveitis is a relatively infrequent presentation of this re-emerging disease

[122,130–132]

. Reported cases have had anterior uveitis

[122]

and, in one

case, bilateral serous retinal detachment

[132]

. The diagnosis is most com-

monly achieved using a serum microscopic agglutination test. A single high ti-
ter or rising titers are considered indicative of infection

[133–137]

Therapy is

directed toward elimination of the organism. High doses of penicillin, ampicil-
lin, and amoxicillin can clear the leptospiremia phase, usually within hours of
administration

[138]

. These drugs do not eliminate the carrier state, however

[133,134]

. Current recommendations are to use a 2-week course of doxycycline

to clear the carrier state in dogs

[138]

Systemic Mycoses

Blastomycosis, cryptococcosis, coccidiomycosis, and histoplasmosis are the sys-
temic mycoses most commonly involved with uveitis

[139,140]

. Patients may

be presented with anterior uveitis, chorioretinitis, panuveitis, endophthalmitis,
or optic neuritis. A complete physical examination with particular attention to
the cutaneous examination, thoracic auscultation, and abdominal palpation can
aid in establishing a diagnosis. As discussed previously, cytologic identification
of the organism within aspirates or impression smears is the gold standard for
establishing a diagnosis. Serologic tests are available. The latex cryptococcal ag-
glutination test detects cryptococcal antigen and can be a useful test for estab-
lishing a diagnosis and monitoring the response to therapy. The serologic tests
for histoplasmosis, blastomycosis, and cocciciomycosis detect antibody produc-
tion. False-negative responses, at least for blastomycosis, are not uncommon

[141]

. The preferred systemic therapy for each type of mycosis varies

The current preferred therapy for blastomycosis is the administration of itraco-
nazole

[142]

. Although many clinicians do not advocate systemic corticosteroid

treatment in dogs with systemic mycoses

[143]

, a recent retrospective study of

dogs infected with blastomycosis by Finn and colleagues

[144]

did not note any

339

CANINE AND FELINE UVEITIS

change in survival times and suggested that combination therapy of systemic
prednisone and itraconazole may have assisted in the maintenance of vision.

SUMMARY

The clinical signs of uveitis occur as a result of inflammation within the vascu-
lar coat of the eye, which causes breakdown of the BAB and blood-retinal
barrier. Clinical signs include blepharospasm, photophobia, conjunctival
hyperemia, circumlimbal corneal vascularization, corneal edema, aqueous flare,
inflammatory cells within the anterior chamber, keratic precipitates, iridal con-
gestion, ocular hypotony, retinal hemorrhage, and retinal detachment. Se-
quelae to uveitis include cataracts, posterior synechiae, secondary glaucoma,
and retinal degeneration. Many infectious and noninfectious causes can incite
episodes of uveitis. Therefore, complete ocular and physical examinations are
recommended for all patients that have uveitis. A complete blood cell count,
serum biochemistry profile, urinalysis, thoracic radiographs, and select sero-
logic tests may be performed in an effort to identify any underlying etiologic
agents. Despite exhaustive workups, however, the underlying cause is not de-
termined in many cases. The goals of therapy are preserving vision if possible,
minimizing pain, and halting inflammation. Additional therapeutic agents may
be used if the underlying etiologic agent can be identified.

References

[1] Samuelson D. Ophthalmic anatomy. In: Gelatt KN, editor. Veterinary ophthalmology,

vol. 1. 4th edition. Ames (IA): Blackwell Publishing Professional; 2007. p. 62–98.

[2] Chan CC, Li Q. Immunopathology of uveitis. Br J Ophthalmol 1998;82:91–6.
[3] Hendrix D. Diseases and surgery of the canine anterior uvea. In: Gelatt KN, editor. Veteri-

nary ophthalmology, vol. 2. 4th edition. Ames (IA): Blackwell Publishing; 2007. p. 812–58.

[4] Massa K, Gilger B, Miller T, et al. Causes of uveitis in dogs: 102 cases (1989–2000). Vet

Ophthalmol 2002;5(2):93–8.

[5] Stiles J, Townsend WM. Feline ophthalmology. In: Gelatt KN, editor. Veterinary ophthal-

mology, vol. 2. 4th edition. Ames (IA): Blackwell Publishing; 2007. p. 1117–24.

[6] Rankin AJ, Krohne SG, Glickman NW, et al. Laser flaremetric evaluation of experimentally

induced blood-aqueous barrier disruption in cats. Am J Vet Res 2002;63(5):750–6.

[7] Raviola G. The structural basis of the blood-ocular barriers. Exp Eye Res 1977;25(Suppl):

27–64.

[8] Cuhna-Vaz J. The blood-ocular barriers. Surv Ophthalmol 1979;23:279–96.
[9] Cunha-Vaz JG, Shakib M, Ashton N. Studies on the permeability of the blood-retinal bar-

rier. I. On the existence, development, and site of a blood-retinal barrier. Br J Ophthalmol
1966;50(8):441–53.

[10] Hudspeth A, Yee A. The intercellular junctional complexes of retinal pigment epithelia. In-

vest Ophthalmol Vis Sci 1973;12:354.

[11] Gabelt B, Kaufman PL. Aqueous humor hydrodynamics. In: Kaufman PL, Alm A, editors.

Adler’s physiology of the eye. 10th edition. St. Louis (MO): Mosby; 2003. p. 237–89.

[12] Caprioli J. The ciliary epithelia and aqueous humor. In: Hart W, editor. Adler’s physiology

of the eye: clinical application. 9th edition. St. Louis (MO): Mosby; 1992. p. 228–47.

[13] Narfstrom K, Petersen-Jones S. Diseases of the canine ocular fundus. In: Gelatt K, editor.

Veterinary ophthalmology, vol. 2. 4th edition. Ames (IA): Blackwell Publishing; 2007.
p. 944–1025.

[14] Floman N, Zor U. Mechanism of steroid action in ocular inflammation: inhibition of pros-

taglandin production. Invest Ophthalmol Vis Sci 1977;16(1):69–73.

340

TOWNSEND

[15] Pinard C, Gauvin D, Moreau M, et al. Measurement of inflammatory mediators in aqueous

humor following paracentesis of the anterior chamber in dogs. Presented at 38th Annual
Conference of the American College of Veterinary Ophthalmology. Kona, Hawaii,
October 22–27, 2007.

[16] Liao T, Ke Y, Shao W, et al. Blockade of the interaction of leukotriene B4 with its receptor

prevents development of autoimmune uveitis. Invest Ophthalmol Vis Sci 2006;47(4):
1543–9.

[17] Hakanson N, Forrester SD. Uveitis in the dog and cat. Vet Clin North Am Small Anim Pract

1990;20(3):715–35.

[18] Beekhuis W, McCarey B. Corneal epithelial Cl-dependent pump quantified. Exp Eye Res

1986;43:707–11.

[19] Cassin B, Rubin M. Dictionary of eye terminology. 4th edition. Gainesville (FL): Triad Pub-

lishing Company; 2001.

[20] Yoshitomi T, Ito Y. Effects of indomethacin and prostaglandins on the dog iris sphincter and

dilator muscles. Invest Ophthalmol Vis Sci 1988;29:127–32.

[21] Dziezyc J, Millichamp N, Keller C, et al. Effects of prostaglandin F2 alpha and leukotriene

D4 on pupil size, intraocular pressure, and blood-aqueous barrier in dogs. Am J Vet Res
1992;53:1302–4.

[22] Toris C, Pederson J. Aqueous humor dynamics in experimental iridocyclitis. Invest Ophthal-

mol Vis Sci 1987;28(3):477–81.

[23] Stern F, Bito L. Comparison of the hypotensive and other ocular effects of prostaglandins E2

and F2a on cat and rhesus monkey eyes. Invest Ophthalmol Vis Sci 1982;22(5):588–98.

[24] Keller W. The canine anterior uvea. In: Gelatt K, editor. Textbook of veterinary ophthalmol-

ogy. Philadelphia: Lea & Febiger; 1981. p. 383–5.

[25] Slatter D. Fundamentals of veterinary ophthalmology. 2nd edition. Philadelphia: WB Sa-

unders Company; 1990.

[26] Inomata H, Smelser G, Polack F. Corneal vascularization in uveitis and graft rejection. In-

vest Ophthalmol Vis Sci 1971;10(11):840–50.

[27] Wong T, Sethi C, Daniels J, et al. Matrix metalloproteinases in disease and repair pro-

cesses in the anterior segment. Surv Ophthalmol 2002;47(3):239–56.

[28] Peiffer R, Wilcock B, Yin H. The pathogenesis and significance of pre-iridal fibrovascular

membrane in domestic animals. Vet Pathol 1990;27(1):41–5.

[29] Streeten B. Pathology of the lens. In: Albert D, Jakobiec F, editors. Principles and practice of

ophthalmology. Philadelphia: WB Saunders; 2000. p. 3685–749.

[30] Eagle R, Spencer W. Lens. In: Spencer W, editor. Ophthalmic pathology. 4th edition. Phil-

adelphia: WB Saunders; 1996. p. 372–427.

[31] Davidson M, Nelms S. Disease of the canine lens and cataract formation. In: Gelatt K, ed-

itor. Veterinary ophthalmology, vol. 2. 4th edition. Ames (IA): Blackwell Publishing; 2007.
p. 877.

[32] Curtis R, Barnett K, Lewis S. Clinical and pathological observations concerning the aetiol-

ogy of primary lens luxation in the dog. Vet Rec 1983;112:238–46.

[33] Olivero DK, Riis RD, Dutton AG, et al. Feline lens displacement: a retrospective analysis of

345 cases. Progress in Veterinary and Comparative Ophthalmology 1991;1(4):239–44.

[34] Grahn B, Peiffer R. Fundamentals of veterinary ophthalmic pathology. In: Gelatt KN, edi-

tor. Veterinary ophthalmology. 4th edition. Ames (IA): Blackwell Publishing; 2007.
p. 355–437.

[35] Olin D. Examination of the aqueous humor as a diagnostic aid in anterior uveitis. J Am Vet

Med Assoc 1977;171:557–9.

[36] Vande Lelij A, Rothova A. Diagnostic anterior chamber paracentesis in uveitis: a safe pro-

cedure. Br J Ophthalmol 1997;81:976–9.

[37] Brightman A, Brown M, Johnson B, et al. Vitreous centesis as a diagnostic procedure. Pre-

sented at Proceedings of the Annual Meeting of the American College of Veterinary Oph-
thalmologists. November 6–9, 1986.

341

CANINE AND FELINE UVEITIS

[38] Lappin MR, Roberts S, Davidson MG, et al. Enzyme linked immunosorbent assays for the

detection of Toxoplasma gondii specific antibodies and antigens in the aqueous humor of
cats. J Am Vet Med Assoc 1992;201(7):1010–6.

[39] Chavkin MJ, Lappin MR, Powell CC, et al. Toxoplasma gondii-specific antibodies in the

aqueous humor of cats with toxoplasmosis. Am J Vet Res 1994;55(9):1244–9.

[40] Klauss G, Constantinescu G. Nonhypotensive autonomic agents in veterinary ophthalmol-

ogy. Vet Clin North Am Small Anim Pract 2004;34:777–800.

[41] Stocker F. Experimental studies on the blood-aqueous barrier. Arch Ophthalmol 1947;37:

583–90.

[42] Swan K, Hart W. A comparative study of the effects of mecholyl, doryl, eserine, pilocar-

pine, atropine, and epinephrine on the blood aqueous barrier. Am J Ophthalmol
1940;23:1311–9.

[43] Rubin L, Wolfes R. Mydriatics for canine ophthalmoscopy. J Am Vet Med Assoc

1962;140(2):137–41.

[44] Gelatt K, Boggess T, Cure T. Evaluation of mydriatics in the cat. J Am Anim Hosp Assoc

1973;9:283–7.

[45] Rubin L. Applications of adrenergic drugs in ophthalmology. J Am Anim Hosp Assoc

1968;4:250–5.

[46] Lynch R, Rubin L. Salivation induced in dogs by conjunctival instillation of atropine. J Am Vet

Med Assoc 1965;147(5):511–3.

[47] Holmberg B, Maggs D. The use of corticosteroids to treat ocular inflammation. Vet Clin

North Am Small Anim Pract 2004;34:693–705.

[48] Flower R. Lipocortin and the mechanism of action of the glucocorticoids. Br J Pharmacol

1988;94:987–1015.

[49] Goppelt-Struebe M, Wolter D, Resch K. Glucocorticoids inhibit prostaglandin synthesis not

only at the level of phospholipase A2 but also at the level of cyclo-oxygenase/PGE iso-
merase. Br J Pharmacol 1989;98:1287–95.

[50] Wilkie DA. Control of ocular inflammation. Vet Clin North Am Small Anim Pract

1990;20(3):693–713.

[51] Krohne SG, Vestre W. Effects of flunixin meglumine and dexamethasone on aqueous

protein values after intraocular surgery in the dog. Am J Vet Res 1987;48(3):
420–2.

[52] Giuliano E. Nonsteroidal anti-inflammatory drugs in veterinary ophthalmology. Vet Clin

North Am Small Anim Pract 2004;34(3):707–23.

[53] Nichols J, Snyder R. Topical nonsteroidal anti-inflammatory agents in ophthalmology. Curr

Opin Ophthalmol 1998;9(4):40–4.

[54] Perianin A, Roch-Arveiller M, Giround J, et al. In vivo effects of indomethacin and flurbipro-

fen on the locomotion of neutrophils elicited by immune and non-immune inflammation in
the rat. Eur J Pharmacol 1985;106(2):327–33.

[55] Leonardi A, Busato F, Fregona I, et al. Anti-inflammatory and antiallergic effects of ketoro-

lac tromethamine in the conjunctival provocation model. Br J Ophthalmol 2000;84(11):
1228–32.

[56] Flach A. Cyclo-oxygenase inhibitors in ophthalmology. Surv Ophthalmol 1992;36(4):

259–84.

[57] Millichamp N, Dziezyc J, Rhohde B, et al. Acute effects of anti-inflammatory drugs on neo-

dymium:yttrium aluminum garnet laser-induced uveitis in dogs. Am J Vet Res 1991;52(8):
1279–84.

[58] Ward D, Ferguson D, Ward S, et al. Comparison of the blood-aqueous barrier stabilizing

effects of steroidal and nonsteroidal anti-inflammatory agents in the dog. Progress in Vet-
erinary and Comparative Ophthalmology 1992;2(3):117–24.

[59] Krohne S, Gionfriddo JR, Morrison E. Inhibition of pilocarpine-induced aqueous humor

flare, hypotony, and miosis by topical administration of anti-inflammatory and anesthetic
drugs to dogs. Am J Vet Res 1998;52(8):1279–84.

342

TOWNSEND

[60] Roze M, Thomas E, Davot J. Tolfenamic acid in the control of ocular inflammation in the

dog: pharmacokinetics and clinical results obtained in an experimental model. J Small
Anim Pract 1996;37(8):371–5.

[61] Reimer ME, Johnston SA, Leib MS, et al. The gastroduodenal effects of buffered aspirin,

carprofen, and etodolac in healthy dogs. J Vet Intern Med 1999;13(5):472–7.

[62] Powell CC, Lappin MR. Diagnosis and treatment of feline uveitis. Compendium on Continu-

ing Education for the Practicing Veterinarian 2001;23(3):258–66.

[63] Gaynes B, Fiscella R. Topical nonsteroidal anti-inflammatory drugs for ophthalmic use:

a safety review. Drug Saf 2002;25(4):233–50.

[64] Ward D. Comparative efficacy of topically applied flurbiprofen, diclofenac, tolmetin, and

suprofen for the treatment of experimentally induced blood-aqueous barrier disruption in
dogs. Am J Vet Res 1996;57(6):875–8.

[65] Millichamp N, Dziezyc J, Olsen J. Effect of flurbiprofen on facility of aqueous outflow in the

eyes of dogs. Am J Vet Res 1991;52(9):1448–51.

[66] Hemady R, Tauber J, Foster CS. Immunosuppressive drugs in immune and inflammatory oc-

ular disease. Surv Ophthalmol 1991;35(5):369–85.

[67] Sapienza JS, Simo FJ, Prades-Sapienza A. Golden retriever uveitis: 75 cases (1994–

1999). Vet Ophthalmol 2000;3(4):241–6.

[68] Morgan R. Vogt-Koyanagi-Harada syndrome in humans and dogs. Compendium of Con-

tinuing Education for the Practicing Veterinarian 1989;11:1211–8.

[69] Papich M. Drugs used to treat immune-mediated disorders in animals. Presented at WSAVA

Congress. Granada, October 3–6, 2002.

[70] Gemensky A, Lorimer D, Blanchard G. Feline uveitis: a retrospective study of 45 cases.

Presented at Annual Meeting of the American College of Veterinary Ophthalmologists,
October 17–22, 1996.

[71] Chavkin M, Lappin M, Powell C, et al. Seroepidemiologic and clinical observations of 93

cases of uveitis in cats. Progress in Veterinary and Comparative Ophthalmology 1992;2:
29–36.

[72] Van der Woerdt A. Lens-induced uveitis. Vet Ophthalmol 2000;3:227–34.
[73] Paulsen M, Lavach J, Severin GA, et al. The effect of lens-induced uveitis on the success of

extracapsular cataract extraction: a retrospective study of 65 lens removals in the dog.
J Am Anim Hosp Assoc 1986;22(1):49–56.

[74] Dziezyc J, Millichamp N, Smith W. Fluorescein concentrations in the aqueous of dogs with

cataracts. Veterinary and Comparative Ophthalmology 1997;7:267–70.

[75] Simpson W, Wild G, Figg K, et al. Human T-cell mediated response to homologous lens

antigen. Exp Eye Res 1989;48(1):49–54.

[76] Wilcock BP, Peiffer RL Jr. The pathology of lens-induced uveitis in dogs. Vet Pathol

1987;24(6):549–53.

[77] Davidson MG, Nasisse MP, Jamieson VE, et al. Traumatic anterior lens capsule disruption.

J Am Anim Hosp Assoc 1991;27:410–4.

[78] Dubielzig R. Feline ocular sarcomas. In: Peiffer RL, Simons K, editors. Ocular tumors in an-

imals and humans. Ames (IA): Iowa State Press; 2002. p. 283–8.

[79] Deehr A, Dubielzig R. A histopathological study of iridociliary cysts and glaucoma in

golden retrievers. Vet Ophthalmol 1998;1(2–3):153–8.

[80] Yamaki K, Takiyama N, Itho N, et al. Experimentally induced Vogt-Koyanagi-Harada dis-

ease in two Akita dogs. Exp Eye Res 2005;80(2):273–80.

[81] Herrera H, Duchene A. Uveodermatologic syndrome (Vogt-Koyanagi-Harada-like syn-

drome) with generalized depigmentation in a Dachshund. Vet Ophthalmol 1998;1:47–51.

[82] Carter WJ, Crispin SM, Gould DJ, et al. An immunohistochemical study of uveodermato-

logic syndrome in two Japanese Akita dogs. Vet Ophthalmol 2005;8(1):17–24.

[83] Peiffer RL. Ciliary body epithelial tumors in the dog and cat: a report of 13 cases. J Small

Anim Pract 1988;24:347–70.

[84] Peiffer R. Primary intraocular tumors in the dog. Mod Vet Pract 1979;60:383.

343

CANINE AND FELINE UVEITIS

[85] Dubielzig RR, Steinberg H, Garvin H, et al. Iridociliary epithelial tumors in 100 dogs and

17 cats: a morphological study. Vet Ophthalmol 1998;1(4):223–31.

[86] Cello R, Hutcherson B. Ocular changes in malignant lymphoma of dogs. Cornell Vet

1961;51:492–523.

[87] Krohne SG, Henderson N, Richardson RC, et al. Prevalence of ocular involvement in dogs

with multicentric lymphoma: prospective evaluation of 94 cases. Veterinary and Compar-
ative Ophthalmology 1994;4:127–35.

[88] Cassotis NJ, Dubielzig RR, Gilger BC, et al. Angioinvasive pulmonary carcinoma with pos-

terior segment metastasis in four cats. Vet Ophthalmol 1999;2(2):125–31.

[89] Lappin MR, Black J. Bartonella spp infection as a possible cause of uveitis in a cat. J Am Vet

Med Assoc 1999;214(8):1205–7.

[90] Kipar A, Bellmann S, Kremendahl J, et al. Cellular composition, coronavirus antigen ex-

pression and production of specific antibodies in lesions in feline infectious peritonitis.
Vet Immunol Immunopathol 1998;65(2–4):243–57.

[91] Kipar A, May H, Menger S, et al. Morphologic features and development of granuloma-

tous vasculitis in feline infectious peritonitis. Vet Pathol 2005;42(3):321–30.

[92] Addie DD, Paltrinieri S, Pedersen NC. Recommendations from workshops of the Second

International Feline Coronavirus/Feline Infectious Peritonitis Symposium. J Feline Med
Surg 2004;6(2):125–30.

[93] Ketring KL, Zuckerman EE, Hardy WD Jr. Bartonella: a new etiological agent of feline oc-

ular disease. J Am Anim Hosp Assoc 2004;40(1):6–12.

[94] Fontanelle J, Powell CC, Hill A, et al. Prevalence of serum antibodies against Bartonella

species in the serum of cats with or without uveitis. J Feline Med Surg; 2007, in press.

[95] Dowers KL, Olver C, Radecki SV, et al. Use of enrofloxacin for treatment of large-form Hae-

mobartonella felis in experimentally infected cats. J Am Vet Med Assoc 2002;221(2):
250–3.

[96] Kordick DL, Papich MG, Breitschwerdt EB. Efficacy of enrofloxacin or doxycycline for treat-

ment of Bartonella henselae or Bartonella clarridgeiae infection in cats. Antimicrob Agents
Chemother 1997;41(11):2448–55.

[97] Hardy W. Hematopoietic tumors of cats. J Am Anim Hosp Assoc 1981;17:921–40.
[98] Peiffer RL, Wilcock BP. Histopathologic study of uveitis in cats: 139 cases (1978–1988).

J Am Vet Med Assoc 1991;198(1):135–8.

[99] Willis AM. Feline leukemia virus and feline immunodeficiency virus. Vet Clin North Am

Small Anim Pract 2000;30(5):971–86.

[100] Brightman AH, Ogilvie GK, Tompkins M. Ocular disease in FeLV-positive cats: 11 cases

(1981–1986). J Am Vet Med Assoc 1991;198(6):1049–51.

[101] Silverstein D, Greene C. Laboratory testing for infectious diseases of dogs and cats. In:

Greene C, editor. Infectious diseases of the dog and cat. 2nd edition. Philadelphia: WB
Saunders; 1998. p. 769.

[102] English RV, Davidson MG, Nasisse MP, et al. Intraocular disease associated with feline im-

munodeficiency virus infection in cats. J Am Vet Med Assoc 1990;196(7):1116–9.

[103] Davidson MG, Rottman JB, English RV, et al. Feline immunodeficiency virus predisposes

cats to acute generalized toxoplasmosis. Am J Pathol 1993;143(5):1486–97.

[104] Levy JK, Crawford PC, Slater MR. Effect of vaccination against feline immunodeficiency

virus on results of serologic testing in cats. J Am Vet Med Assoc 2004;225(10):1558–61.

[105] Hartmann K, Griessmayr P, Schulz B, et al. Quality of different in-clinic test systems for fe-

line immunodeficiency virus and feline leukaemia virus infection. J Feline Med Surg
2007;9(6):439–45.

[106] Davidson MG. Toxoplasmosis. Vet Clin North Am Small Anim Pract 2000;30(5):

1051–62.

[107] Dubey Je. Toxoplasmosis of animals and man. Boca Raton (FL): CRC Press; 1988.
[108] Dubey JP, Carpenter JL. Histologically confirmed clinical toxoplasmosis in cats: 100 cases

(1952–1990). J Am Vet Med Assoc 1993;203(11):1556–66.

344

TOWNSEND

[109] Lappin MR, Greene CE, Winston S, et al. Clinical feline toxoplasmosis. Serologic diagno-

sis and therapeutic management of 15 cases. J Vet Intern Med 1989;3(3):139–43.

[110] Burney DP, Chavkin MJ, Dow SW, et al. Polymerase chain reaction for the detection of

Toxoplasma gondii within aqueous humor of experimentally-inoculated cats. Vet Parasitol
1998;79(3):181–6.

[111] Burney DP, Lappin MR, Spilker M, et al. Detection of Toxoplasma gondii parasitemia in ex-

perimentally inoculated cats. J Parasitol 1999;85(5):947–51.

[112] Colitz CMH. Feline uveitis: diagnosis and treatment. Clin Tech Small Anim Pract

2005;20(2):117–20.

[113] Wilkinson G. Feline cryptococcosis: a review and seven case reports. J Small Anim Pract

1979;20:749–68.

[114] Gwin RM, Makley TA Jr, Wyman M, et al. Multifocal ocular histoplasmosis in a dog and

cat. J Am Vet Med Assoc 1980;176(7):638–42.

[115] Nasisse MP, van Ee RT, Wright B. Ocular changes in a cat with disseminated blastomyco-

sis. J Am Vet Med Assoc 1985;187(6):629–31.

[116] Greene RT, Troy CG. Coccidioidomycosis in 48 cats: a retrospective study (1984–1993).

J Vet Intern Med 1995;9(2):86–91.

[117] Blondin N, Baumbardner D, Moore G, et al. Blastomycosis in indoor cats: suburban Chi-

cago, Illinois, USA. Mycopathologia 2007;163:59–66.

[118] Gilor C, Graves TK, Barger AM, et al. Clinical aspects of natural infection with Blastomy-

ces dermatitidis in cats: 8 cases (1991–2005). J Am Vet Med Assoc 2006;229(1):96–9.

[119] Pearce J, Giuliano E, Galle L, et al. Management of bilateral uveitis in a Toxoplasma gon-

dii-seropositive cat with histopathologic evidence of fungal panuveitis. Vet Ophthalmol
2007;10(4):216–21.

[120] Wanke MM. Canine brucellosis. Anim Reprod Sci 2004;82–83:195–207.
[121] Vinayak A, Greene CE, Moore PA, et al. Clinical resolution of Brucella canis induced oc-

ular inflammation in a dog. J Am Vet Med Assoc 2004;224(11):1804–7.

[122] Dziezyc J. Canine systemic bacterial infections. Vet Clin North Am Small Anim Pract

2000;30(5):1103–17.

[123] Carmichael L, Shin S. Canine brucellosis: a diagnostician’s dilemma. Semin Vet Med Surg

(Small Anim) 1996;11:161–5.

[124] Wanke MM, Delpino MV, Baldi PC. Use of enrofloxacin in the treatment of canine brucel-

losis in a dog kennel (clinical trial). Theriogenology 2006;66(6–7):1573–8.

[125] Stiles J. Canine rickettsial infections. Vet Clin North Am Small Anim Pract 2000;30(5):

1135–49.

[126] Littman MP, Goldstein RE, Labato MA, et al. ACVIM small animal consensus statement on

Lyme disease in dogs: diagnosis, treatment, and prevention. J Vet Intern Med 2006;20(2):
422–34.

[127] Leiva M, Naranjo C, Pena MT. Ocular signs of canine monocytic ehrlichiosis: a retrospec-

tive study in dogs from Barcelona, Spain. Vet Ophthalmol 2005;8(6):387–93.

[128] Komnenou AA, Mylonakis ME, Kouti V, et al. Ocular manifestations of natural canine

monocytic ehrlichiosis (Ehrlichia canis): a retrospective study of 90 cases. Vet Ophthalmol
2007;10(3):137–42.

[129] Breitschwerdt EB, Davidson MG, Hegarty BC, et al. Prednisolone at anti-inflammatory or

immunosuppressive dosages in conjunction with doxycycline does not potentiate the sever-
ity of Rickettsia rickettsii infection in dogs. Antimicrob agents chemother 1997;41(1):
141–7.

[130] Bolin CA. Diagnosis of leptospirosis: a reemerging disease of companion animals. Semin

Vet Med Surg (Small Anim) 1996;11:166–71.

[131] Lajeunesse J, DiFruscia R. Leptospirosis: a re-emergent zoonosis. Medecin Veterinaire du

Quebec 1999;29(4):209–10.

[132] Townsend W, Stiles J, Krohne SG. Leptospirosis panuveitis in a dog. Vet Ophthalmol

2006;9(3):169–73.

345

CANINE AND FELINE UVEITIS

[133] Harkin K, Gartrell C. Canine leptospirosis in New Jersey and Michigan: 17 cases (1990–

1995). J Am Anim Hosp Assoc 1996;32:495–501.

[134] Greene CE, Miller MA, Brown CA. Leptospirosis. In: Greene CE, editor. Infectious dis-

eases of the dog and cat. 2nd edition. Philadelphia: W.B. Saunders Company; 1998.
p. 273–81.

[135] Adin CA, Cowgill LD. Treatment and outcome of dogs with leptospirosis: 36 cases (1990–

1998). J Am Vet Med Assoc 2000;216(3):371–5.

[136] Birnbaum N, Barr S, Center S, et al. Naturally acquired leptospirosis in 36 dogs: serolog-

ical and clinicopathological features. J Small Anim Pract 1998;39:231–6.

[137] Baldwin C, Atkins C. Leptospirosis in dogs. Compendium of Continuing Education for the

Practicing Veterinarian 1987;9(5):499–507.

[138] Langston CE, Heuter KJ. Leptospirosis a re-emerging zoonotic disease. Vet Clin North Am

Small Anim Pract 2003;33:791–807.

[139] Carlton W. Oculomycosis. In: Peiffer R, editor. Comparative ophthalmic pathology.

Springfield (IL): Charles C. Thomas; 1983. p. 299–317.

[140] Wolf A. Systemic mycoses. J Am Vet Med Assoc 1989;194:1192–6.
[141] Legendre A. Blastomycosis. In: Green C, editor. Infectious disease of the dog and cat. 3rd

edition. St. Louis (MO): Saunders Elsevier; 2006. p. 569–76.

[142] Legendre AM, Rohrbach BW, Toal RL, et al. Treatment of blastomycosis with itraconazole

in 112 dogs. J Vet Intern Med 1996;10(6):365–71.

[143] Collins BK, Moore CP. Diseases and surgery of the canine anterior uvea. In: Gelatt KN,

editor. Veterinary ophthalmology. 3rd edition. Philadelphia: Lippincott, Williams & Wil-
kins; 1999. p. 759–75.

[144] Finn MJ, Stiles J, Krohne SG. Visual outcome in a group of dogs with ocular blastomycosis

treated with systemic antifungals and systemic corticosteroids. Vet Ophthalmol
2007;10(5):299–303.

346

TOWNSEND

Extraocular Myositis in the Dog

David L. Williams, MA, VetMB, PhD, CertVOphthal, FRCVS

a,b,

Department of Veterinary Medicine, University of Cambridge, Madingley Road,

Cambridge CB3 OES, England, UK

Veterinary Medicine and Pathology, St. John’s College, Cambridge CB2 1TP, England, UK

xtraocular myositis (EOM) is a rare condition in the dog and is poorly
reported in the peer-reviewed veterinary literature. The publications
documenting the condition can be numbered on the fingers of one hand;

thus, a literature review on its own would be a short and perhaps not particu-
larly worthwhile exercise. Personally, this author’s experience is limited to five
cases, and other veterinary ophthalmologists around the world do not seem to
have a much higher case load than that, with the exception of Dr. David
Ramsey and his colleagues, whose experience is discussed elsewhere in this
article. Thus, this article includes a retrospective multicenter study from several
ophthalmologists, giving a reasonable case series to illustrate the condition. The
author thus starts with acknowledgments to those colleagues who have
graciously allowed him to use their patients in this series. This really should
be a multiauthored article, but the constraints of publication in this format ren-
der it easier to acknowledge them as a group here and individually by case in

. The animals reported here confirm the view that this is, for the most

part, a surprisingly specific condition: most animals are young entire female
golden retrievers with a striking presentation of exophthalmos but no third
eyelid protrusion, which responds well to systemic steroids at anti-inflamma-
tory doses. This in itself causes problems in further evaluation of the condition.
Systemic evaluation of the cases is certainly possible, but extensive evaluation
by MRI or CT is generally not necessary for a diagnosis to be made and inva-
sive procedures, such as biopsies of relevant muscles, are not only unnecessary
but probably unethical in most cases. In this manner, EOM is quite different
from myositis in other muscle groups, such as the masticatory muscles or other
skeletal muscles, in which relatively easy investigations ranging from electro-
physiology to biopsy allow detailed evaluation of disease pathogenesis. Com-
parison with similar conditions in human beings and experimental rodents is
possible but fraught with difficulty, because the apparent similarities may not
actually reflect parallels in the pathogenesis of the condition.

*Department of Veterinary Medicine, University of Cambridge, Madingley Road,
Cambridge CB3 OES, England, UK. E-mail address: doctordlwilliams@aol.com

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.11.010

Vet Clin Small Anim 38 (2008) 347–359

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

Table 1
Cases of extraocular myositis

Case

Breed

Age
(mos)

Gender

Disease signs

Treatment regimen and response to treatment

Clinician

Golden

retriever

Female

Bilateral exophthalmos,

exotropia, epiphora

Prednisolone, 1 mg/kg po sid, 6/52 tapered,

no relapses

Golden

retriever

Female

Bilateral exophthalmos,

exotropia, mild visual
disturbance

Prednisolone, 1 mg/kg po, changed to azathioprine,

2 mg/kg, mild exotropia on near focus, otherwise
no relapses

Golden

retriever

Female

Bilateral exophthalmos,

exotropia, conjunctivitis,
chemosis

Prednisolone, 1 mg/kg po bid for 3 weeks and sid

for 3 weeks, resolved but with mild exotropia
on stimulation

Golden

retriever

Female

neutered

Bilateral exophthalmos

with ventrolateral strabismus

Prednisolone, 1 mg/kg po sid, resolved completely

within 3 weeks

Golden

retriever

Female

Bilateral exophthalmos

with mild exotropia

Prednisolone, 1 mg/kg po sid, tapered over 14 days,

resolved completely but recurred after 1 month,
resolved on steroid as previously with more
gradually tapering dose

Golden

retriever

Female

Bilateral exophthalmos

Prednisolone, 1 mg/kg po sid, tapered over 4 weeks,

resolved in 7 days with no recurrence

Golden

retriever

Female

Bilateral exophthalmos

with exotropia

Prednisolone, 1 mg/kg po sid, resolved in 5 days

Golden

retriever

Female

Bilateral exophthalmos

with exotropia

Prednisolone, 1 mg/kg po sid, resolved in 3 days

Golden

retriever

Female

neutered

Bilateral exophthalmos

Prednisolone, 2 mg/kg po bid, resolved in 2 days,

tapered dose led to recurrence after 10 weeks when
dog was excited, azathioprine was used at 1 mg/kg
sid tapered to half the dose for 5 weeks, no further
recurrence noted

348

WILLIAMS

Golden

retriever

Female

Bilateral exophthalmos

Azathioprine, 2 mg/kg, with prednisone, 1 mg/kg,

divided po bid, responded well, tapered over weeks

Golden

retriever

Male entire

Bilateral exophthalmos

Azathioprine, 2 mg/kg, with prednisone,

1 mg/kg po divided bid, responded well,
tapered over weeks

Golden

retriever

Female

Bilateral exophthalmos

Azathioprine, 2 mg/kg, with prednisone,

1 mg/kg po divided bid, responded well,
tapered over weeks

Golden

retriever

Female

Bilateral exophthalmos

Azathioprine, 2 mg/kg, with prednisone,

1 mg/kg po divided bid, responded well,
tapered over weeks

Golden

retriever

Female

Bilateral exophthalmos

Azathioprine, 2 mg/kg, with prednisone,

1 mg/kg po divided bid, responded well,
tapered over weeks

Golden

retriever

Female

Bilateral exophthalmos

Prednisolone, 2 mg/kg po divided daily bid,

responded well

Golden

retriever

Female

Bilateral exophthalmos

with poor blink reflex

1 mg/kg prednisolone po, resolved

within 24 hours

Golden

retriever

Female

Bilateral exophthalmos

Prednisolone, 2.5 mg/kg po divided bid,

resolved within 1 week, recurrence at 10 months
shortly after routine vaccination,
controlled with prednisolone po

Golden

retriever

Female

Bilateral exophthalmos

Prednisolone, 2 mg/kg po, responded rapidly

Golden

retriever

Female

Bilateral exophthalmos

Prednisolone, 2 mg/kg po, responded rapidly,

tapered dose but with relapse when
treatment stopped by owner

(continued on next page)

349

EXTRAOCULAR

MYOSITIS

IN
THE

DOG

Table 1

(continued)

Case

Breed

Age
(mos)

Gender

Disease signs

Treatment regimen and response to treatment

Clinician

Golden

retriever

Male entire

Mild bilateral exophthalmos

Dog already under steroid treatment at first

presentation, continued treatment tapering
over 2 months

Golden

retriever

Male neuter

Bilateral exophthalmos MM2

antibody positive

Prednisolone 2 mg/kg po, responded well

Labrador

retriever

Male neuter

Bilateral exophthalmos with

chemosis

Prednisolone 1 mg/kg and azathioprine

2 mg/kg, responded well

Labrador

retriever

Female

Bilateral exophthalmos

Prednisolone, 2 mg/kg po, responded well,

gradually tapered

Labrador

retriever

Female

Bilateral exophthalmos

Prednisolone, 2 mg/kg po, responded well,

gradually tapered

Labrador

retriever

Female

Bilateral exophthalmos

Azathioprine, 2 mg/kg, with prednisone,

1 mg/kg divided bid, responded well

Labradoodle

Female

Bilateral exophthalmos

Azathioprine, 2 mg/kg, with prednisone,

1 mg/kg divided bid, responded well

Welsh corgi

Female

Bilateral exophthalmos

Azathioprine, 2 mg/kg, with prednisone,

1 mg/kg divided bid, responded well

Hovart

Female

Bilateral exophthalmos,

external ophthalmoplegia

Azathioprine, 1 mg/kg, responded in

36 hours, tapered over time

Great Dane

Male

entire

Bilateral chemosis, external

ophthalmoplegia, moderate
exophthalmos

Prednisolone, 2 mg/kg po, tapered but

requires 0.25 mg/kg every other day
to prevent recurrence

Bernese

mountain
dog

Male

entire

Bilateral exophthalmos,

external ophthalmoplegia

Prednisolone, 1 mg/kg po sid, resolved

slowly over 2 weeks

350

WILLIAMS

Japanese

spitz

Male

entire

Exophthalmos, external

ophthalmoplegia, slight
elevation of nictitating
membrane

Prednisolone, 2 mg/kg po, responded rapidly

Rotweiler

Male

neutered

Bilateral chemosis with mild

exophthalmos, external
ophthalmoplegia, mild
elevation of nictitating
membrane

Prednisone, 1 mg/kg by injection, resolved

within 12 hours, chemosis returned within
24 hours but resolved on prednisolone,
1 mg/kg po, and remains in remission on
prednisolone, 0.5 mg/kg every other day

Cross-bred

Female

Exophthalmos with increased

scleral show but no
ophthalmoplegia, concurrent
diabetes and
hyperadrenocorticism

Resolved without treatment but note elevated

circulating steroids as part
of hyperadrenocorticism

Dachshund

Female

neutered

Bilateral exophthalmos

Prednisolone, 2 mg/kg, responded rapidly

German

pointer

Female

Bilateral exophthalmos

Prednisolone, 2 mg/kg po, responded rapidly

Cross-bred

Female

Bilateral exophthalmos

Prednisolone 2 mg/kg, responded rapidly

German

shepherd

132

Female neutered

Bilateral exophthalmos

Prednisolone 2 mg/kg, responded rapidly

Abbreviations: BH, Dr. Bradford Holmberg; bid, twice daily; BS, Dr. Bernard Spiess; DE, Dr. Jason Evans; DW, Dr. David Williams; EB, Dr. Ellen Bjerkas; JW, Dr. Joe Wolfer; KK,
Dr. Kerry Ketring; NM, Dr. Natasha Mitchell; po, per os; sid, once daily.

351

EXTRAOCULAR

MYOSITIS

IN
THE

DOG

EXTRAOCULAR MYOSITIS: A CASE SERIES

The retrospective multicenter study reported here yielded 37 dogs with EOM, as
detailed in

. The retrospective nature of this case series means that not all

the clinical evidence one might want to form a full ophthalmic examination was
available for all dogs. Yet, even so, the study yields some interesting results.

All cases showed marked bilateral exophthalmos (

) characterized by

increased scleral show, sometimes predominantly dorsally (

A, B), some-

times predominantly ventrally (

C), and sometimes with accompanying

lateral strabismus or exotropia (see

D). In some animals, chemo-

sis was an important presenting sign (see

E). Although Schirmer tear test

results and intraocular pressure were not recorded for all cases, in those for
which this information was available, tear intraocular pressure was within nor-
mal limits (15.6 2.6 mm Hg), whereas Schirmer tear test data were slightly
elevated (24.8 3.1 mm/min), although the relevance of this is unclear; it may
be that increased corneal exposure increases the reflex tearing, although had
the results been lower than normal, one could have argued that increased evap-
oration was giving an evaporative dry eye. Clearly, from the Schirmer tear test
data and the ocular surface health of the central cornea in these dogs, this is not
occurring in these animals.

Fig. 1. Cases of canine EOM. (A) Case 7 shows increased superior scleral show, exotropia
and the classic ‘‘startled’’ expression. (B) Case 6 shows exophthalmos with a frontal gaze and
increased superior scleral show without third eyelid protrusion. (C) Case 1 shows ventromedial
scleral show and chemosis with exotropia. (D) After medical treatment, case 2 shows exotropia
on stimulation. (E) Case 24 shows pronounced chemosis and scleral show with some protru-
sion of the nictitating membrane. (F) Case 28 shows increased scleral show but normal ocular
movement as the dog looked to a stimulus to the right.

352

WILLIAMS

Of the 37 dogs, 21 were golden retrievers and 5 were Labrador retrievers or

retriever crosses. The mean age of the dogs was 24 months, with a range from
6 months to 11 years of age and a median of 18 months. Of the golden re-
trievers, the mean age was 17.1 months, with a median of 12 months.
Twenty-four were entire female dogs, with 3 neutered female dogs, 5 entire
male dogs, and 2 castrated male dogs; there is thus a clear gender bias in these
cases, with bitches being four times more likely to be affected than male dogs. It
might seem that entire animals are more likely to be affected than ones who
have been spayed or castrated, although it is possible that the young age of
the animals skewed the ratio of entire-to-neutered animals and that sexual
entirety is not necessarily a necessary feature of the condition.

In most cases, a diagnosis can be made on the clinical appearance of the

animal alone. Further diagnostic steps that can be rewarding include imaging
studies; orbital ultrasonography might be thought to be particularly helpful,
but in many cases, the increased volume of the extraocular muscles is not
specifically obvious because a generalized increase in orbital echogenicity de-
creases resolution of individual orbit contents. MRI (

) is particularly valu-

able in defining increase in intraocular muscle volume and water content,
which increases with inflammatory changes in the muscle bellies.

Histopathologic evaluation was rarely obtained because the diagnosis is

evident enough not to require histologic confirmation. When it was obtained,
however, a lymphocytic inflammatory infiltrate was evident (

Response to oral prednisolone at an anti-inflammatory dose of between

1 and 2 mg/kg was generally rapid and complete, although some veterinary

Fig. 2. MRI scan for case 9. Note the increased volume of all extraocular muscles but partic-
ularly of the medial recti. (Courtesy of B. Holmberg, DVM, Davis, CA.)

353

EXTRAOCULAR MYOSITIS IN THE DOG

ophthalmologists used azathioprine in preference, and azathioprine was used in
three cases in which side effects of prolonged steroid medication were problem-
atic. Five animals had recurrence of signs after tapering of the anti-inflamma-
tory dose of steroid, and three had persistent mild exophthalmos or
strabismus that required low-dose oral prednisolone for stabilization.

LITERATURE REVIEW

The first publication reporting this condition appeared in 1991

and reported

two cases of EOM, one in an 18-month-old spayed female golden retriever and
the other in an 8-month-old male dog of the same breed. Both dogs exhibited
characteristic exophthalmos without third eyelid protrusion. The male dog was
euthanized at the owner’s request, allowing histopathologic evaluation,
whereas the bitch had biopsies taken from the dorsal rectus and temporalis
muscles. Macroscopically, the rectus and oblique muscles were enlarged with
pale-yellow centers to the bellies, in contrast to the tendinous extremities of
the muscles, which were normal. The retractor bulbui muscles were similarly
unremarkable. Microscopically, the muscles had patchy but intense infiltrates

Fig. 3. (A) Severe lymphocytic infiltration involving the entire lateral rectus muscle, now mas-
sively swollen in case 17 (original magnification 40). (Courtesy of D. Shelton, DVM, PhD, La
Jolla, CA.) (B) Severe lymphocytic infiltration with myofibrillar necrosis in case 17 (original
magnification 100). (Courtesy of D. Shelton, DVM, PhD, La Jolla, CA.) (C) A less severe ex-
ample of EOM shows lymphocytic infiltration around myofibrils from case 2 (original magnifi-
cation 100). (Courtesy of D. Shelton, DVM, PhD, La Jolla, CA.)

354

WILLIAMS

of lymphocytes and histiocytes, with only a few plasma cells and neutrophils.
Some areas showed complete destruction of muscle fibers, whereas others had
segmental myonecrosis. No other orbital contents showed pathologic change.
No other tissues showed significant pathologic change. The bitch from which
biopsies were taken showed severe lymphocytic and histiocytic infiltration of
the dorsal rectus muscle with myonecrosis and mild fibrosis, whereas the
temporal muscle was entirely normal. In this animal, clinical signs resolved
on treatment with systemic dexamethasone. In the same article, Carpenter
and colleagues

noted five other dogs seen with similar signs over a 7-year

period, including two golden retrievers, one German shepherd dog, one Dober-
man pinscher, and one cross-bred dog. Two dogs were male, whereas three
were female, one of which was spayed.

The largest series of dogs with EOM is that of Ramsey and colleagues

which was presented in the proceedings of the American College of Veterinary
Ophthalmologists from 1995, and thus, unfortunately, was not widely available
to nonophthalmologists. Of the 35 dogs reported, 21 were golden retrievers and
2 were golden retriever-cross dogs, whereas others included Doberman pinsch-
ers, German shepherds, German short-haired pointers, and a Labrador mixed-
breed dog. Twelve were female entire dogs, 10 were neutered bitches, 8 were
male entire dogs, and 4 were male neutered animals. Bilateral but not necessar-
ily symmetric exophthalmos was present in all dogs, importantly, without pro-
trusion of the nictitating membrane as is seen in exophthalmos associated with
most orbital space–occupying lesions. Histopathologic examination showed an
infiltrate of CD3-expressing mononuclear cells in extraocular muscles but with-
out pathologic findings of orbital connective tissue or the tendinous insertions of
extraocular muscles. Mild diffuse myofibrosis, myodegeneration, and myofibril-
lar regeneration were evident, as was perivascular mucopolysaccharide deposi-
tion. Treatment with systemic prednisolone was successful when used at or
greater than 1.1 mg/kg every 12 hours, but recrudescence of signs occurred
when the initial dose was reduced in fewer than 21 days or when the dosage in-
terval was increased. Eighty percent of the dogs had recurrences, and 10 had
more than one recurrence. A single case report from Germany further supports
these findings

. A report of EOM, histologically diagnosed, has been pub-

lished from a group of German short-haired pointers with repeated Neospora in-
fection

. The prevalence of this parasitic infection as a cause of EOM is

unknown, but the fact that this single case report is the only published instance
of the link with the parasite does not suggest that it is particularly high.

A more recent report on 10 dogs with EOM and a different clinical picture is

provided by Allgoewer and colleagues

. These dogs were, again, young

animals, with a mean age of 21.9 months, but were of significantly different
breeds from those reported herein and by Carpenter and colleagues

and

Ramsey and colleagues

. There was only one golden retriever, but there

were 3 Irish wolfhounds, 3 akitas, 2 shar peis, and 1 dalmatian. In these ani-
mals, the clinical presentation was not one of exophthalmos and mild exotropia
without third eyelid protrusion but one of enophthalmos and extreme ventral,

355

EXTRAOCULAR MYOSITIS IN THE DOG

ventromedial, or medial strabismus. Biopsies showed myonecrosis in 3 dogs,
myofibrillar atrophy in 3, and mononuclear cell infiltration in 4 dogs, with fi-
brosis in all 10 dogs. Although it might be suggested that these are the signs
of a chronic fibrotic sequel to what we might call classic EOM, these animals
did not have a history of an acute exophthalmic phase of disease and all cases
were unilateral in nature rather than the classic bilateral presentation of EOM
as seen in the dogs in the reports of Carpenter and colleagues

or Ramsey

and colleagues

or in the cases documented here. Medical therapy was not

effective, and surgery was required to normalize globe position. Although
this report describes EOM, these clinical cases are clearly not the same as those
covered in this review.

COMPARISON OF MYOSITIS IN EXTRAOCULAR
AND SKELETAL MUSCLES

The work of Shelton

has been instrumental in documenting the myopatho-

logic and immunopathologic findings of myositis in the dog. Inflammatory
myopathies are a heterogeneous group of disorders characterized by nonsup-
purative cellular infiltration of skeletal muscles. Generalized myositis includes
specific conditions, such as polymyositis (PM) and infectious and preneoplastic
syndromes, whereas focal inflammatory myopathies include masticatory mus-
cle myositis (MMM) and EOM

. Cellular infiltrates in the latter condition

specifically involve muscles innervated by the trigeminal nerve, including the
masseter, temporalis, pterygoids, tensor tympani, and tensor palatine. These
muscles contain a unique myofiber, type 2M, against which autoantibodies
are found specifically in MMM but generally not in EOM

. This probably

correlates with the different embryologic origin of extraocular muscles from
all other skeletal muscles; they develop from mesenchymal condensations,
and hence have different morphologic features, immunohistologic characteris-
tics, and innervation patterns from other skeletal muscles. A key autoantigen
recognized by antibodies in MMM has recently been defined as a novel mem-
ber of the myosin-binding protein-C family

, but the relevance of that to the

genesis of the disease is unclear at present. In PM, CD8-expressing lympho-
cytes predominate, with T lymphocytes using the ab T-cell receptor, whereas
in MMM, CD4-expressing T lymphocytes are found in greater numbers than
CD8 lymphocytes and T cells use the ab and the cd T-cell receptors

Other differences involve B cells, which are seen in multifocal follicular clusters
in MMM, whereas they are absent in PM. The immune response in MMM is
a CD8 major histocompatibility complex (MHC)-I–restricted pathologic pro-
cess

. The immunopathologic changes seen in canine myopathies mirror

those in their human counterparts

, in which a similar tissue-directed auto-

immunity occurs.

Having said all that, the difficulty in obtaining biopsies from extraocular

muscles and the comparative rarity of EOM has meant that little research
has been possible on the immunopathogenesis of the disease, which may differ
quite substantially from MMM. The animals with EOM are not in pain, and in

356

WILLIAMS

the acute phase represented by the animals in the case series presented here,
there is no pain or muscle spasm, unlike the cases of MMM.

COMPARATIVE APPROACH TO EXTRAOCULAR MYOSITIS

There would seem to be a human condition superficially similar to the EOM
described in young dogs here. Thyroid-associated ophthalmopathy is a com-
mon manifestation of autoimmune thyroid disease in human patients

occurring in between 25% and 50% of those who have Graves’ thyroiditis
and in a smaller proportion of those who have Hashimoto’s thyroiditis

Two subtypes of the condition occur; the first involves a congestive ophthalm-
opathy, whereas the second is an ophthalmic myopathy. Congestive ophthalm-
opathy involves eyelid and periocular swelling, whereas ocular myopathy is the
result of autoimmune attack against the extraocular muscles. It is clearly this
latter syndrome that might be seen as a human counterpart to the canine
EOM recorded here. Shared antigens between the thyroid and extraocular
muscles are seen as the explanation for the pathogenesis of thyroid-associated
ophthalmopathy

[14,15]

. Some see an initial immune reaction in the orbital

connective tissue that has secondary effects on extraocular muscles. The thyro-
tropic receptor has been localized in orbital fat and extraocular muscle

The other hypothesis is that antibodies directed against antigens intrinsic to
extraocular muscles are central to thyroid-associated ophthalmopathy. The
unique developmental origin of the extraocular muscles as individual conden-
sations of tissue and their specific innervation lead them to have a different
antigen expression than other muscles

. Calsequestrin is a protein expressed

at almost five times the level in human extraocular muscles than in other
skeletal muscles

and was seen in the serum of 75% of patients who had oph-

thalmopathy compared with 5% of those who had Graves’ disease without oph-
thalmopathy

. In that study, there was also a correlation between

ophthalmopathy and serologic evidence for an antibody response to collagen
XIII and a 67-kDa flavoprotein subunit of mitochondrial succinate dehydroge-
nase. The problem here is knowing which of these antibodies is present as
a cause of ophthalmopathy or as a result of extraocular muscle damage

The same problem would occur when looking at serologic changes in canine
EOM. Antigen release after immune damage to extraocular muscles may
explain this seropositivity better than an attempt to argue that the antibodies
are a cause of myopathy.

What then of experimental models of thyroid autoimmunity? Several

models exist in mice and rats

[18,19]

, ranging from spontaneously occurring

disease in the BB rat

to that occurring after immunization of mice with

the thyroid-stimulating hormone (TSH) receptor

; however, perhaps signif-

icantly, none of these animals develop exophthalmopathy. The canine disease
described here thus stands as a potentially valuable model for the changes
occurring in the extraocular muscles in Graves’ ophthalmopathy, although,
clearly, substantial differences occur in presentation, response to treatment,

357

EXTRAOCULAR MYOSITIS IN THE DOG

and initial etiopathogenesis. It remains up to us veterinary ophthalmologists to
investigate this fascinating but rare disease in more detail.

SUMMARY

It is hoped that this case series and review are of value to those veterinary
ophthalmologists seeking to investigate the disease further and also to veteri-
narians outside the specialty, for whom this may be their first introduction
to the disease.

Acknowledgments

The case series that forms the center of this review would not have been possible
without the kind cooperation of several veterinary ophthalmologists who really
should have appeared as coauthors of this article. The author’s sincere thanks
go to Drs. Ellen Bjerkas, Jason Evans, Bradford Holmberg, Kerry Ketring, Na-
tasha Mitchell, Diane Shelton, Joe Wolfer, and Bernard Spiess, but the more
general opinions expressed here regarding the condition are entirely his own.

References

[1] Carpenter JL, Schmidt GM, Moore FM, et al. Canine bilateral extraocular polymyositis. Vet

Pathol 1989;26:510–2.

[2] Ramsey DT, Hamor RE, Gerding PA, et al. Clinical and immunohistochemical characteristics

of bilateral extraocular polymyositis of dogs. Proc Am Coll Vet Ophthalmol 1995;129–35.

[3] Mitra S. [Eosinophilic myositis of the extraocular muscles. A case report]. Proceedings of the

American College of Veterinary Ophthalmologists 1998;26:336–40 [in German].

[4] Dubey JP, Koestner A, Piper RC. Repeated transplacental transmission of Neospora caninum

in dogs. J Am Vet Med Assoc 1990;197:857–60.

[5] Allgoewer I, Blair M, Basher T, et al. Extraocular muscle myositis and restrictive strabismus in

10 dogs. Vet Ophthalmol 2000;3:21–6.

[6] Shelton GD. From dog to man: the broad spectrum of inflammatory myopathies. Neuromus-

cul Disord 2007;17:663–70.

[7] Evans J. Canine inflammatory myopathies: a clinicopathologic review of 200 cases. J Vet

Intern Med 2004;18:679–91.

[8] Shelton GD, Cardinet GH, Bandman E, et al. Fiber type-specific autoantibodies in a dog

with eosinophilic myositis. Muscle Nerve 1985;8:783–90.

[9] Wu X, Li ZF, Brooks R, et al. Autoantibodies in canine masticatory muscle myositis recognize

a novel myosin binding protein-C family member. J Immunol 2007;179:4939–44.

[10] Pumorola M, Moore PF, Shelton GD. Canine inflammatory myopathy: analysis of cellular

infiltrates. Muscle Nerve 2004;29:782–9.

[11] Neuman J, Biltzer T. Evidence for MHC I-restricted CD8þ T-cell-mediated immunopathology

in canine masticatory muscle myositis and polymyositis. Muscle Nerve 2006;33:215–24.

[12] Shelton GD, Hoffman EP, Ghimbovschi S, et al. Immunopathogenic pathways in canine

inflammatory myopathies resemble human myositis. Vet Immunol Immunopathol
2006;113:200–14.

[13] Bartalena L, Wiersinga WM, Pinchera A. Graves’ ophthalmopathy: state of the art and

perspectives. J Endocrinol Invest 2004;27:295–301.

[14] Mizokami T, Salvi M, Wall JR. Eye muscle antibodies in Graves’ ophthalmopathy: patho-

genic or secondary epiphenomenon? J Endocrinol Invest 2004;27:221–9.

[15] Tani J, Wall JR. Autoimmunity against eye-muscle antigens may explain thyroid-associated

ophthalmopathy. CMAJ 2006;175:239.

[16] Porter JD, Khanna S, Kaminski HJ, et al. Extraocular muscle is defined by a fundamentally

distinct gene expression profile. Proc Natl Acad Sci U S A 2001;98(21):12062–7.

358

WILLIAMS

[17] Gopinath B, Musselman R, Beard N, et al. Antibodies targeting the calcium binding skeletal

muscle protein calsequestrin are specific markers of ophthalmopathy and sensitive indica-
tors of ocular myopathy in patients with Graves’ disease. Clin Exp Immunol 2006;145:
56–62.

[18] Seetharamaih GS. Animal models of Graves’ hyperthyroidism. Autoimmunity 2003;36:

381–7.

[19] McLachlan SM, Nagayama Y, Rapoport B. Insight into Graves’ hyperthyroidism from

animal models. Endocr Rev 2005;26:800–32.

[20] Pettersson A, Wilson D, Daniels T, et al. Thyroiditis in the BB rat is associated with lympho-

penia but occurs independently of diabetes. J Autoimmun 1995;8:493–505.

[21] Barrett K, Liakata E, Rao PV, et al. Induction of hyperthyroidism in mice by intradermal

immunization with DNA encoding the thyrotropin receptor. Clin Exp Immunol 2004;136:
413–22.

359

EXTRAOCULAR MYOSITIS IN THE DOG

Antibody-Mediated Retinopathies
in Canine Patients: Mechanism,
Diagnosis, and Treatment Modalities

Sinisa D. Grozdanic, DVM, PhD*, Matthew M. Harper, PhD,
Helga Kecova, DVM, PhD

Department of Veterinary Clinical Sciences, College of Veterinary Medicine,
Iowa State University, Ames, IA 50011, USA

udden onset of blindness in canine patients with a normal ocular appear-
ance is considered an emergency condition, which is frequently related to
ocular and central nervous system (CNS) abnormalities. This condition

usually presents a diagnostic challenge to regular veterinary practitioners and
veterinary ophthalmology specialists, because a normal ocular appearance is
frequently suggestive of optic nerve, chiasmal, or brain pathologic conditions
causing blindness. The principal purpose of this review is to describe the mech-
anistic basis, clinical signs, diagnostic methods, and treatment options for reti-
nal diseases causing sudden onset of blindness with absence of typical signs of
intraocular inflammation or retinal degeneration—sudden acquired retinal
degeneration syndrome (SARDS) and immune-mediated retinitis (IMR).

SUDDEN ACQUIRED RETINAL DEGENERATION SYNDROME

SARDS has been recognized as a cause of sudden onset of blindness for more
than 2 decades

[1,2]

. Preliminary analysis of prevalence data revealed that most

veterinary ophthalmologists in North America diagnose between 10 and 30 pa-
tients with SARDS every year, which brings the estimated number of SARDS
patients close to 2000 patients annually (S.D. Grozdanic, CR:N, unpublished
observation, 2006). Despite large numbers of affected dogs, the etiology and
pathologic mechanisms causing this disease remain poorly understood.

Clinical Presentation

Patients that have SARDS are typically presented on an emergent basis because
of sudden onset of blindness, which results in lack of orientation, colliding with

This work was supported in part by the Department of Veterans Affairs, Veterans Health Administration,

Rehabilitation Research and Development Service grant C3919R and the Iowa State University
Biotechnology Fund.

*Corresponding author. E-mail address: sgrozdan@iastate.edu (S.D. Grozdanic).

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.12.003

Vet Clin Small Anim 38 (2008) 361–387

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

objects (even in a familiar environment), and depression

. Owners frequently

report polyphagia, polydipsia, and polyuria, which are usually detected several
days or even several weeks before the onset of blindness; however, in some
instances, these symptoms develop soon after the onset of blindness

[1,3]

. Ret-

rospective studies and the authors’ personal experience is that all symptoms of
abnormal metabolic activity (eg, polyphagia, polydipsia, polyuria) resolve
within 3 to 6 months after the onset of blindness in most patients

. The

results of the general physical examination are usually unremarkable, and
serum analysis frequently shows the presence of elevated liver enzyme values,
lipid abnormalities (predominantly increased cholesterol levels), increased
levels of vitamins A and E, increased serum protein fractions (predominantly
a

-globulin fraction), and increased levels of cortisol and sex hormones

[1,5]

On ophthalmic examination, patients that have SARDS display absent vision
and an absent menace response, whereas photopic blink response (dazzle
response) and pupil light reflex (PLR) are usually present

[1,6]

. Fundus evalu-

ation typically does not reveal any major abnormalities in the early stage of the
disease; however, all patients that have SARDS have a characteristic ‘‘pale optic
nerve’’ appearance because of the presence of optic nerve head vasculature
attenuation (

). Subtle hyperreflective spots can be observed in patients

with a longer duration of disease (more than 2 months). The authors observed
the presence of hyperreflective spots in patients presented to their service only
7 days after the onset of blindness, which may be suggestive of a chronic nature
of the disease, despite the predominant clinical symptom of a sudden onset of
visual loss

. Patients with a longer duration of disease (more than 1 year) fre-

quently may have advanced retinal degenerative changes. The authors had
a chance to evaluate patients in which the retina had a relatively normal ap-
pearance on funduscopic examination; however, structural analysis of retinal
thickness using optical coherence tomography (OCT) analysis showed signifi-
cant retinal thinning.

Fig. 1. Fundus evaluation of patients that have SARDS frequently reveals a relatively normal
retinal appearance; however, optic nerve arterioles always are decreased in size, which
results in the characteristic symptom of a ‘‘pale optic nerve head.’’ Patients that have SARDS
with 6 months’ duration of blindness (A), 5 weeks’ duration of blindness (B), and 10 months’
duration of blindness (C). Despite a 6-month-long duration of blindness, patient A did not have
detectable hyperreflective changes.

362

GROZDANIC, HARPER, & KECOVA

Colorimetric Pupil Light Reflex Analysis in Diagnosis of Sudden Acquired
Retinal Degeneration Syndrome

Pupils in patients that have SARDS have a large resting diameter (10 mm) and
respond poorly (or not at all) to white light illumination of 10 to 40 kcd/m

(standard illumination of Finoff transilluminator [Heine, Herrsching, Germany]
and Kowa SL-14 slit lamp biomicroscope [Kowa Inc, Tokyo, Japan])

[1,6]

. Pupils

respond with almost complete constriction when white light of high intensity
(1000 kcd/m

, Kowa SL-2 slit lamp biomicroscope fiberoptic light source) is

used, however. Regardless of the light source used, pupil constriction is slow,
with a delay of at least 1 second between the time of illumination and the start
of constriction

. A recent study described detailed spectral properties of the

PLR response in healthy canine eyes and in the eyes of dogs that have SARDS

. It has been demonstrated that dogs that have SARDS have strong pupil

responses when blue light of narrow wave length (480 nm) and of high light
intensity (200 kcd/m

) is used, most likely as a result of stimulation of a photosen-

sitive pigment, melanopsin

. Melanopsin is a vitamin A–based photosensitive

pigment located in a specific subpopulation of retinal ganglion cells that can drive
PLR responses even in the complete absence of photoreceptor activity, as is the
case with patients that have SARDS. If red light of a specific wave length
(630 nm) and high light intensity (200 kcd/m

) is used for pupil light stimulation

(red light of 630-nm wave length does not overlap with the melanopsin spectral
sensitivity and cannot activate melanopsin), however, the PLR response in
patients with severe or complete photoreceptor dysfunction is absent

. This

particular physiologic property of the PLR response can be effectively used to es-
tablish the fast diagnosis of SARDS. Because patients that have SARDS do not
have photoreceptor activity, activation of the photoreceptor-mediated pathway
(red light illumination) results in a fixed and dilated pupil, whereas activation
of the melanopsin pathway (blue light illumination) results in the complete pupil
constriction

. At this time, the authors have tested nearly 200 patients that have

different forms of retinal and optic nerve disease and did not notice any other
form of ocular disease in canine patients with a normal retinal appearance in
which such characteristic PLR properties (eg, no red response, good blue re-
sponse) could be detected. Currently, the only available instrument on the market
for colorimetric PLR testing is a Melan-100 unit (BioMed Vision Technologies,
Inc., Ames, Iowa). This particular unit was built to match spectral properties of
canine visual pigments; it is a portable instrument and has powerful diode-based
light sources with narrow wave lengths for blue and red light, which fit to spectral
sensitivity curves of melanopsin (480 nm) and rod-cone opsins (630 nm).

Electroretinography in Diagnosis of Sudden Acquired Retinal
Degeneration Syndrome

Electroretinography (ERG) is considered the ‘‘gold standard’’ in establishing
the diagnosis of SARDS, because patients that have SARDS do not have
detectable rod and cone–mediated electrical activity (

)

. The major

disadvantage of ERG remains the relatively high price of diagnostic equipment

363

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

and a need for an experienced operator with an excellent understanding of the
basic principles of ERG. Although this method remains an irreplaceable tool in
diagnosis of SARDS, colorimetric PLR evaluation may become an inexpensive,
fast, and simple alternative for establishing an early diagnosis of SARDS by vet-
erinary practitioners.

Optical Coherence Tomography in Patients That Have Sudden Acquired
Retinal Degeneration Syndrome

OCT of eyes in dogs that have SARDS shows primary damage of the nerve fiber
layer and decreased total retinal thickness of the inferior retina in the early stage
of disease (

). Dogs with prolonged duration of blindness do have signs of

thinning in all retinal layers (including the photoreceptor layer) in the superior
and inferior retinal quadrants (

Figs. 4 and 5

). The authors recently performed

a detailed analysis of OCT data in 6 patients that had SARDS with a history
of blindness for less than 6 weeks, which were presented to the Iowa State
University (ISU) for possible intravenous immunoglobulin (IVIg) treatment

. Linear scans through the area centralis (superior temporal retinal region in

the canine retina, which corresponds to the human macula) and corresponding
inferior retina showed significant loss of nerve fiber layer (NFL) thickness
(superior

NFL

: SARDS ¼ 76.7 2.4 lm, control ¼ 86.4 1.8 lm [mean

SEM; P ¼ .007, Student’s t test]; inferior

NFL

: SARDS ¼ 64 6 lm, control ¼

96.6 3 lm [P ¼ .0012, Student’s t test]). The photoreceptor layer thickness
was not significantly different compared with retinas in dogs that had SARDS
(SARDS

superior

¼ 102 4.7 versus control

superior

¼ 99.1 2.4 [P ¼ .73, Student’s

t test]; SARDS

inferior

¼ 89.1 5.9 versus control

inferior

¼ 95.9 3.2 lm [P ¼ .28,

Fig. 2. Patients that have SARDS do not have detectable rod (scotopic ERG) or cone (photopic
ERG) electrical activity.

364

GROZDANIC, HARPER, & KECOVA

Student’s t test]). Circular scan analysis showed significant NFL damage in all
quadrants in patients that had SARDS when compared with healthy control
eyes (superior, P ¼ .01; inferior, P ¼ .0065; nasal, P ¼ .0024; temporal, P ¼
.005; Student’s t test; see

). Total retinal thickness analysis showed

Fig. 4. (A) Dogs that have SARDS have significantly reduced total retinal thickness of the
inferior retina. (B) OCT scans from two dogs that have SARDS with different disease durations.
The top image is from a dog with 4 weeks’ duration of blindness, whereas the bottom image is
from a dog with a 7-month-long duration of blindness, which shows dramatic retinal thinning.
Linear scans were performed at the superior retinal regions, and they are suggestive of degen-
erative changes that may develop in the superior retina with the longer disease duration.
Because of the severe retinal structural loss, this particular patient (bottom OCT scan) was
not treated with intravenous immunoglobulin (IVIg) therapy. The arrowhead shows a thick white
line that represents the tapetum (structure present in the superior nonpigmented part of the
canine retina). The black vertical lines show total retinal thickness (**P < .001; normal superior
retina thickness in canine patients is in the range of 180–200 lm).

Fig. 3. OCT shows loss of the nerve fiber layer in all observed quadrants (circular scan
around the optic nerve head; Student’s t test: *P < .05, **P < .001).

365

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

significant retinal thinning of the inferior (nontapetal) retina (P ¼ .0008,
Student’s t test; see

), whereas the superior (tapetal) retina showed no

decrease in total thickness when compared with control (healthy) canine retinas
(P ¼ .96, Student’s t test).

Value of Optical Coherence Tomography for Predicting Therapeutic
Success

OCT in patients that have SARDS shows a significant decrease of retinal thick-
ness, which is an important prognostic parameter before pursuing IVIg-based
treatment options. In the authors’ experience, patients that have SARDS
with severe loss of retinal structure (total retinal thickness less than 150 lm)
do not respond well to the IVIg treatment. Loss of retinal thickness is fre-
quently associated with the presence of subtle hyperreflective spots on fundus
examination; however, the authors had a chance to perform scans in canine
patients without hyperreflective spots and determined that retinal thickness is
already decreased to less than 150 lm. Furthermore, it is virtually impossible
to make a decision about the quality of retinal thickness in patients that do not
have a well-developed tapetum without performing OCT analysis.

PATHOGENESIS OF SUDDEN ACQUIRED RETINAL
DEGENERATION SYNDROME

Although different hypotheses have been established as a possible explanation
for SARDS etiology (exposure of photoreceptors to unidentified toxins

, pho-

toreceptor degeneration attributable to hormonal or metabolic abnormalities

[1,5]

, and glutamate toxicity

), the clinical appearance of sudden and painless

onset of blindness is most similar to antibody-mediated retinopathies (eg, cancer-
associated retinopathy [CAR], autoimmune retinopathy in the absence of

Fig. 5. Patient that has SARDS (7 months’ duration of blindness). The lack of a tapetum causes
difficulties in determining the total retinal thickness before making a decision whether to pursue the
intravenous immunoglobulin (IVIg) treatment. This patient had a total retinal thickness of 120 lm
(inferior retina) and severe loss of the nerve fiber layer; thus, IVIg treatment was not pursued.

366

GROZDANIC, HARPER, & KECOVA

cancer) in human beings. Early work by Bellhorn and colleagues

showed the

possible presence of retinal autoantibodies in the serum of patients that had
SARDS; however, these results have been disputed by recent studies demon-
strating that patients that have SARDS do not have the detectable presence of
retinal autoantibodies in serum or the presence of systemic neoplasia

[10,11]

Furthermore, Miller and colleagues

demonstrated that retinas of dogs that

have SARDS have extensive numbers of photoreceptors undergoing apoptosis,
which is most likely responsible for the development of blindness. Because of the
presence of metabolic and hormonal abnormalities in dogs that have SARDS, the
most plausible hypothesis for photoreceptor damage in dogs that have SARDS
was the presence of abnormal levels of hormones, which can have a toxic effect
on photoreceptors. This hypothesis is not considered likely, however, because
there is no published evidence demonstrating that hormonal abnormalities can
have such a dramatic effect on retinal function. Furthermore, the authors
recently demonstrated that patients that have SARDS have abnormal PLR
responses, which require much higher light intensity for activation

. Because

melanopsin-containing retinal ganglion cells control the circadian regulation of
hormonal activity

[13–15]

, the authors speculate that decreased sensitivity of

these cells in patients that have SARDS (even before the onset of blindness)
may be responsible for abnormal regulation of hormonal status and the resulting
changes in metabolism frequently observed in patients that have SARDS.

The authors’ laboratory recently performed extensive analysis of tissue from

patients that had SARDS and demonstrated the presence of immunoglobulin-
producing plasma cells in the retinas of dogs that had SARDS, which may be
responsible for localized intraretinal production of autoantibodies (

Figs. 6 and

;

) and the development of antibody-mediated retinopathy

. Further-

more, the authors have demonstrated strong complement activity in the retinas
of dogs that have SARDS, which may be responsible for antibody-mediated
neuronal damage (

Figs. 8–14

; see

SUDDEN ACQUIRED RETINAL DEGENERATION SYNDROME
TREATMENT

SARDS has been considered an untreatable canine-blinding disease because of
the complete lack of therapeutic response to anti-inflammatory, antimicrobial,
or immunosuppressive medications. Antibody-mediated retinopathy in human
beings is frequently described as poorly responsive to medical treatment; how-
ever, high-dose steroids, plasmapheresis, and IVIg therapies have been
described to reverse symptoms of blindness partially

[16–20]

. The IVIg is a syn-

onym for intravenous human immunoglobulins; it is a mixture of different classes
of globulins found in healthy human circulation. IVIg is made by mixing globu-
lins of 5000 to 10,000 samples from healthy human donors; thus, it really repre-
sents a mosaic of normal circulating antibodies, which are controlling the immune
system so that it does not become excessively aggressive in contact with antigens
that the canine body and human body are regularly exposed to on a daily basis. It
is used extensively in human medicine for variety of immune-mediated diseases

367

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

that are poorly responsive (or nonresponsive) to steroids and other immunosup-
pressive medications. Generally, commercially available IVIg solutions contain
between 95% and 98% of IgG and traces of IgA and IgM fractions. It has been
recently demonstrated that most of the immunomodulatory properties of IVIg
are associated with a small fraction of fucose (Fc) fragment–sialylated immuno-
globulins within the IVIg, which offered the possibility of creating more clinically
effective IVIg solutions by increasing the Fc-sialylated IgG fraction using different
chemical procedures

[21,22]

. The production of IVIg is a rigorous and expensive

process, because multiple production steps are needed to eliminate bacterial,

Fig. 7. (A) Immunohistochemistry analysis shows the presence of T lymphocytes (CD3þ small
cells, small arrow) in the retinas of patients that have SARDS. Interestingly, strong CD3 immu-
noreactivity was detected in all neuronal cells, particularly in retinal ganglion cell bodies (open
arrows). (B) Sporadic presence of B lymphocytes (CD79þ cells) can be detected in the retina
of a patient that has SARDS (open arrows).

Fig. 6. Immunohistochemistry analysis using anti-IgG, anti-IgM, and anti-IgA antibodies shows
strong labeling in cells with morphologic characteristics of plasma cells (open arrows), which further
confirmed the authors’ microarray data demonstrating strong upregulation of genes related to im-
munoglobulin production. (A) Image from a patient with 2 months’ duration of blindness. (B) Image
from a patient with 12 months’ duration of blindness (severe panretinal degeneration is present).

368

GROZDANIC, HARPER, & KECOVA

viral, and prion particles completely from the final product by using cold ethanol
fractionation, heat inactivation, and virus filtration

[23,24]

. Although this is vital

for the prevention of transmission of human infectious diseases, it is likely that the
risk for infectious agent transmission across species is minimal (eg, human IVIg
administration to a canine patient).

Table 1
Microarray analysis of eyes with sudden acquired retinal degeneration syndrome showed
strong upregulation of genes responsible for immunoglobulin production and complement
activity

Gene symbol

Gene

Fold of increase

Immunoglobulin synthesis
IGHAC

IgA heavy chain constant region

147

LOC607467

Ig heavy chain V-III region VH26 precursor

127

LOC475754

Igj chain C region, B allele

110

Igc heavy chain B

LOC612122

Igk chain V-I region BL2 precursor

Igc heavy chain C

LOC486411

Igi chain preproprotein

LOC475166

IgJ chain

OC477699

-macroglobulin precursor (a

-M)

Complement activation
LOC476728

Complement C3 precursor

LOC474853

Complement C2 precursor

LOC481722

Complement C4 precursor

LOC487382

Complement C1q subcomponent, c-polypeptide

LOC478194

Complement C1q subcomponent, A chain precursor

LOC477707

Complement C1r subcomponent precursor

Fig. 8. Immunohistochemistry analysis shows significantly higher complement component
(C1q and C4) protein expression in the retinas of patients that have SARDS compared with
control (healthy) eyes (*P < .05; **P < .001).

369

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

Based on previous reports of the beneficial effect of IVIg in human patients

who have CAR, the IVIg treatment of canine patients that have SARDS (by
using human IVIg products) was initiated in April 2007

[17,19]

. At this time,

the authors have successfully restored limited visual behavior (evaluated by
improvement in visual maze testing) in 8 patients that had SARDS and were
treated at ISU (

Table 2

). An additional 11 patients were treated elsewhere using

the ISU protocol, and results are summarized in

Table 3

IOWA STATE UNIVERSITY SUDDEN ACQUIRED RETINAL
DEGENERATION SYNDROME INTRAVENOUS
IMMUNOGLOBULIN TREATMENT PROTOCOL

1. Perform a general physical examination, complete blood cell count, serum

chemistry, urine analysis, and blood pressure measurement.

2. Place two intravenous catheters (one for IVIg and one for safety).

Fig. 10. (A) Lack of menace response after treatment in a patient that has SARDS. (B) Excel-
lent visual maze navigation is present despite the lack of a menace response (this patient did
not have any detectable ERG activity before or after treatment).

Fig. 9. Evaluation of retinal electrical activity shows significant improvement after a single
IVIg treatment in one patient that has SARDS, which was sustained for 5 months (last examina-
tion time point). Arrows point to the b-wave of the electroretinogram.

370

GROZDANIC, HARPER, & KECOVA

3. Start administering IVIg as a slow infusion over 6 hours (administer 0.5 g/kg

of body weight on day 1 and day 3; the authors do not recommend faster
infusion and higher doses of IVIg because of the increased risk for systemic
hypertension development).

4. Monitor temperature, pulse, and respiration in addition to blood pressure

hourly during IVIg treatment, 3 hours after finishing the infusion, and every
4 hours after that.

5. Monitor the leg receiving IVIg for any signs of swelling (perivascular admin-

istration can cause it).

6. Monitor for any signs of possible anaphylactic reaction or systemic hyper-

tension; if noticed, decrease the speed of the IVIg infusion so that treat-
ment can be administered over a period of 12 hours. If problems are
still present despite decreasing the rate of infusion, treatment should be
discontinued for 1 to 2 hours and then continued at a decreased rate of
infusion.

7. Repeat IVIg treatment on day 3.
8. Hospitalize the patient for 1 day after the last treatment.

Fig. 11. (A) Fundus picture of patient that has IMR shows a ‘‘pale optic nerve head’’ appear-
ance because of attenuation of arteriolar vasculature. (B) Normal ERG amplitudes were
recorded in the same patient, despite complete blindness and an almost completely absent
photoreceptor-mediated (red light) PLR response. (C, D) Characteristic appearance of fundus
and lack of ERG activity in a patient that has SARDS.

371

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

Objective Evaluation of Functional Recovery After Intravenous
Immunoglobulin Treatment

The authors’ experimental data show that visual maze behavior and recovery
of the photoreceptor-mediated pupil response (illumination with red light) are
the most reliable parameters for evaluation of a possible IVIg therapeutic effect
in patients. Even after recovering good visual navigation skills, owners should
be warned that vision is crude and most likely completely absent in conditions
of dim light. After IVIg treatment, dogs that have SARDS have great difficulties

Fig. 12. Although patients that have IMR usually have only a detectable ‘‘pale optic nerve
head’’ appearance, the authors detected patients that had relatively subtle retinal changes.
(A) IMR patient 5 (

Tables 4 and 5

) had subtle retinal infiltrates in the nontapetal retina (white

arrows). (B) IMR patient 9 (see

) had a distinct hyperreflective line above the optic

nerve, with the presence of retinal edema at the area centralis.

Fig. 13. IMR patient 5 (see

; fundus image is presented in

Fig. 12

A). ERG analysis

shows predominantly decreased b-wave amplitudes. Extraction of oscillatory potentials
revealed almost complete absence of this ERG component (notice smooth slope of the b-waves
without characteristic oscillatory potential spikes, especially on the top of the b-wave). This
patient was presented with a 3-day-long history of blindness in the right eye, which progressed
to complete bilateral blindness within 48 hours. PLR evaluation showed absent direct and
indirect responses with white light stimuli (15 kcd/m

and 40 kcd/m

) both eyes and a barely

present response with a red light stimulus (200 kcd/m

) right eye. The PLR was slow and

delayed but almost complete when a blue light stimulus was used (200 kcd/m

372

GROZDANIC, HARPER, & KECOVA

in discriminating even large objects if adequate background contrast is not pres-
ent. The most misleading parameter in the evaluation of therapeutic outcome is
probably a lack of menace response, which does not correlate with visual func-
tion when animals are tested in the visual maze. The authors’ data are clearly
indicative that a lack of menace response recovery is not of diagnostic impor-
tance when evaluating visual function in IVIg-treated patients that have
SARDS (see

Fig. 8

The authors’ data demonstrated significant improvement in photoreceptor-

mediated PLR (pupil response to red light) responses in treated patients that
had SARDS and improvement of retinal electrical activity in four patients after
receiving IVIg treatment. These data are consistent with previous reports of the
beneficial effect of IVIg in human patients who had CAR

[17,19]

. Although the

mechanism of the therapeutic effect of IVIg in patients that have SARDS and
CAR is not known, it has been demonstrated that IVIg can regulate general
immune cell activity by modulating expression and activity of Fc-c receptors
as demonstrated in different models of autoimmune diseases

[21,22,25,26]

An even more plausible hypothesis is that IVIg can effectively eliminate

Fig. 14. Normal ERG amplitudes in patient 4 (see

) with a 7-month-long history of

intermittent blindness. The PLR evaluation showed absent direct and indirect responses with
white light stimuli (15 kcd/m

and 40 kcd/m

) and a barely present response with a red light

stimulus (200 kcd/m

; Melan-100 unit) both eyes. The PLR was slow and delayed but almost

complete (left eye and poor right eye) when a blue light stimulus (200 kcd/m

) was used. Pred-

nisone and doxycycline therapy recovered vision, the menace response, and a red PLR
response in both eyes 24 hours after treatment initiation.

Table 2
Summary of therapeutic outcome in Iowa State University–treated patients that had sudden
acquired retinal degeneration syndrome treated after intravenous immunoglobulin therapy

Pupil light
reflex

Dazzle Menace

Visual
maze
testing Electroretinography

Hyperreflective
spots in retina

Red
light

Blue
light

Before IVIg 1/8

8/8

1/8

0/8

After IVIg

7/8

8/8

1/8

8/8

4/8

0/8

All patients were examined using OCT before treatment and did not have any hyperreflective change.

373

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

pathogenic autoantibodies by stimulating their catabolism by means of activa-
tion of neonatal Fc receptor (FcRn)

. Because evidence from the authors’

laboratory is strongly suggestive of localized intraocular antibody production
as a possible mechanism of antibody-mediated retinopathy (which is responsive
to systemic IVIg treatment), the authors can speculate that more effective ther-
apeutic effect may be achieved by performing local (intraocular) IVIg adminis-
tration. This approach could provide a higher concentration of therapeutic
IVIg antibodies in the retinal tissue for a longer period and dramatically
decrease the dose and cost of treatment. Indeed, intraocular IVIg administra-
tion does not seem to affect retinal function and structure, as observed by
ERG and OCT (Sinisa D. Grozdanic, DVM, PhD, unpublished observation,
2007); however, the clinical efficacy of this approach still needs to be tested.
Although multiple reports have demonstrated the safety of systemic IVIg use
in canine patients

[28–31]

, numerous side effects (renal failure, cardiac failure,

hemolytic anemia, aseptic meningitis, and anaphylactic reaction) have been
reported in a small percentage of human patients receiving IVIg

[32–36]

The authors had five patients that developed temporary systemic hypertension
while receiving IVIg infusion and one patient that developed signs of an ana-
phylactic reaction (eg, hypotension, bradycardia). All symptoms resolved after
decreasing the rate of infusion (five patients with hypertension), and adminis-
tration of corticosteroids and antihistamine medications (one patient with an
anaphylactic reaction).

Because of the potential systemic health risks of IVIg therapy and the nega-

tive correlation between improvement in visual outcome and the presence of
retinal degenerative changes (ie, hyperreflective spots; see

Table 3

), the

authors’ current recommendations for the selection of patients for IVIg treat-
ment are as follows:

1. Patients have to be in good systemic health. The authors do not recommend

treatment of patients that have systemic hypertension, acute or chronic renal
disease, liver insufficiency, or cardiac insufficiency. If a heart murmur is
detected, a detailed cardiology examination needs to be performed to eval-
uate whether cardiac condition is sufficiently good for IVIg treatment.

2. Patients that have cataracts or controlled glaucoma should not be treated

with IVIg, because visual recovery is limited and preexisting ocular disease
may negate the possible positive effect of IVIg therapy.

Table 3
Summary of the therapeutic outcome in patients that had sudden acquired retinal degeneration
syndrome treated outside of Iowa State University

Pupil light reflex

Dazzle Menace

Visual
maze
testing Electroretinography

Hyperreflective
spots in retina

Slow

Fast

Before IVIg 11/11

0/11

2/11 1/11

1/11 0/11

7/11

After IVIg

1/11 10/11 11/11 1/11

5/11 2/11

7/11

No OCT was performed before treatment in any of patients, so retinal thickness data are not available.

374

GROZDANIC, HARPER, & KECOVA

3. Dogs that have retinal degenerative changes (hyperreflectivity on fundus

examination) should not be treated. The authors recommend that OCT be
performed in all patients that have SARDS and are candidates for treatment,
because it is their experience that even normal-appearing retinas may have
severe retinal damage, which can be confirmed by OCT analysis. All pa-
tients that had hyperreflective lesions or a long duration of disease (>6
months) had a total retinal thickness less than 150 lm; thus, at this time,
the authors do not recommend treatment of patients that have retinal thick-
ness less than that value. If an OCT examination cannot be performed, the
authors advise that the IVIg treatment be pursued only in patients that
have SARDS that do not have hyperreflective lesions on fundic
examination, have a history of blindness for less than 2 months, and have a
good PLR response when blue light is used (Melan-100 unit; 200-kcd/m

light intensity, 480-nm wave length).

4. Patients that have detectable PLR responses with red light (Melan-100 unit;

200-kcd/m

light intensity, 630-nm wave length) or even minimal ERG

amplitudes should be treated for IMR first (with high-dose steroids and doxy-
cycline) before pursuing the IVIg treatment.

IMMUNE-MEDIATED RETINITIS

IMR is a potentially blinding disease in canine patients, which is characterized
by the sudden and painless onset of complete blindness or night blindness only.
This particular disease has been recognized relatively recently and shares many
similar features with SARDS. Despite many similarities, there are essential
differences between SARDS and IMR, which are described in this article.

Clinical Presentation

Patients that have IMR are usually presented to the veterinary practitioner (or
ophthalmology specialist) with a history of sudden and painless onset of blind-
ness. A careful review of patient history is important in these animals, because
owners frequently report sporadic and temporary episodes of decreased vision
(usually night vision) even months or years before the development of com-
plete blindness. One of the frequently reported syndromes in patients that
have IMR is an abnormal pupil appearance (dilated pupils even in bright light
or prominent anisocoria because of unilateral abnormal pupil dilatation). Most
patients do not have a history of any other health problems; however, 20% of
all dogs presented to the authors’ service had other systemic abnormalities (eg,
neoplasia, neurologic problems [eg, ataxia, stiff gait]; see

). On ophthal-

mologic examination, patients that have IMR have lack of vision and an absent
menace response, whereas the photopic blink response (dazzle response) is usu-
ally present. Fundus evaluation usually does not reveal any abnormalities;
however, similar to the case in patients that have SARDS, a ‘‘pale optic nerve’’
appearance (attributable to the presence of attenuation of optic nerve head
arterioles) may be detected. The most important diagnostic parameter is almost
complete absence of the PLR response when red light is used and a normal
response when blue light is used (

Figs. 15 and 16

). Compared with patients

375

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

Table 4
List of patients diagnosed with immune-mediated retinitis and treated at Iowa State University

Breed

Age
(years) Gender Electroretinography/vision

Other abnormalities

Treatment

Response

Mixed

Decreased OD, absent OS/

blind

None

Standard poodle

Normal ERG/night blindness

Stiff gait, exercise intolerance

None

Golden retriever

Normal OD, decreased OS/

night blindness OD,
complete blindness OS

None

Prednisone,

doxycycline

Recovery of day and

night vision OU/decline
with dose reduction

Shih tsu

Normal ERG/ blind OU

Previous cervical

decompression sx

Prednisone,

doxycycline

Recovery of day and night

vision

Golden retriever

Decreased ERG (b-wave)

OU/blind OU

Positive for Bartonella (1:256)

and Rickettsia (1:64)

Prednisone,

doxycycline

Recovery of day and night

vision

Golden retriever

Normal ERG OU/blind OU

None

Prednisone,

doxycycline

Recovery of day and night

vision/decline when
dose reduced

Mixed breed

Normal ERG OU/night blind

None

Springer spaniel

Supernormal ERG/blind OU

Hypophysis tumor

Prednisone/

doxycycline

None

376

GROZDANIC,

HARPER,

KECOV

Eskimo dog

Night blindness OU (sudden

onset)

Chronic frontal bone

squamous cell carcinoma

Prednisone

Improved on high dose,

decline when dose
reduced

Standard poodle

Night blindness OU

Insulinoma

None

Bernese mountain

dog

Normal ERG OU/blindness

None

Prednisone/

doxycycline

None

Maltese

Severely decreased ERG OU/

minimal vision present

None

Prednisone/

doxycycline

None, IVIg treatment

improved ERG amplitudes

Miniature schnauzer

Severely decreased ERG/

blind OU

None

Prednisone/

doxycycline

ERG improvement, still blind;

IVIg resulted in ERG
improvement and minimal
vision recovery

Diagnosis was established by historical data (intermittent loss of vision or sudden onset of blindness), presence of recordable ERG activity (normal, supernormal or decreased),
and, most importantly, characteristic pupil light reflex responses (almost complete absence of pupil constriction with red light, normal constriction with blue light; Melan-100 unit).
Abbreviations: CM, castrated male; M, male; OD, right eye; OS, left eye; OU, both eyes; SF, spayed female; sx, surgical procedure.

377

ANTIBODY

-MEDIA

TED

RETINOP

THIES

IN
CANINE

TIENTS

that have SARDS, patients that have IMR always have detectable ERG ampli-
tudes, which can be normal, supernormal, or decreased. It is not unusual to
detect completely extinguished ERG amplitudes in one eye and barely detect-
able or normal amplitudes in the opposite eye.

Objective Evaluation of Functional Recovery After Therapy in Patients
That Have Immune-Mediated Retinitis

Recovery of the rod and cone–mediated PLRs (response of the pupil to red
light) and presence of visual navigation behavior in the maze test are the
only reliable parameters for objective evaluation of therapeutic success in
patients that have IMR. Similar to the case in patients that have SARDS,
dogs that have IMR can completely recover visual behavior but still have an
absent menace response, which may be misleading when evaluating the

Table 5
Summary of serum analysis for presence of retinal autoantibodies from healthy dogs and dogs
that had immune-mediated retinitis

Patient

Molecular weight of retinal
autoantigen (Western blot)

Anti-a-enolase
antibody (ELISA)

Antirecoverin
antibody (ELISA)

IMR a

22 kDa

Negative

IMR b

Negative

IMR 3

33 kDa, 22 kDa, 65 kDa,

82 kDa, 120 kDa

Negative

Healthy control beagle 1

Negative

Healthy control beagle 2

Negative

Fig. 15. PLR testing in a patient that has SARDS using a Melan-100 unit. Characteristic
responses are present: no red response (A) and good blue response (B). The presence of
a relatively normal retinal appearance on examination, complete absence of pupil constriction
when red light is used, and good constriction when blue light is used are characteristic of
SARDS.

378

GROZDANIC, HARPER, & KECOVA

presence of visual function (

Figs. 17 and 18

). Monitoring of the ERG ampli-

tudes can also be a useful indicator in some patients, because progression of
disease or lack of disease control is frequently associated with a decrease in ret-
inal electrical activity (

Figs. 19–21

It is the authors’ clinical impression that most patients that have IMR and are

responsive to doxycycline and steroid therapy need long-term treatment,
because a decrease in medication dose (especially steroids) can rapidly result
in severe visual disturbances within 24 hours. Interestingly, an increase in
the dose of steroids reverses symptoms of blindness within 1 to 2 days; thus,
it is relatively easy to titrate the dose of needed medications to maintain
adequate visual function.

Fig. 16. Graph shows data from seven patients that have SARDS. Red light could not induce
pupil constriction in any of the patients, whereas blue light induced strong pupil constriction.

Fig. 17. This patient that has IMR (patient 6, see

) did not have a detectable menace

response after systemic steroid and doxycycline therapy (A); however, the patient had excel-
lent navigational skills in the visual maze (B).

379

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

In the cases in which blindness is not complete (night vision affected only),

the most reliable parameter for evaluating therapeutic success is recovery of the
PLR response after stimulation with the red light, because evaluation of vision
in dim light conditions may not be always consistent with anticipated results
after changing medication dosage.

Pathogenesis of Immune-Mediated Retinitis

Although the exact pathogenesis of IMR remains unknown, clinical symptoms
are most similar to different forms of antibody-mediated retinopathies in hu-
man patients. The authors performed an analysis of serum samples from three
patients that had IMR and detected the presence of several different classes of
retinal autoantibodies in one patient (IMR patient 3), whereas one patient did
not have detectable autoantibodies (IMRb) and one patient had only the pres-
ence of an antibody reacting against 22 kDa protein (most likely, light IgG
chain). These limited results are similar to findings from human patients
who have different forms of antibody-mediated retinopathies, in which almost
40% to 60% of patients do not have detectable retinal autoantibodies in the
serum despite specific clinical signs of disease

SUMMARY

Autoantibodies against retinal antigens are considered the primary cause of
pathologic change in CAR and different types of immune-mediated retinopa-
thies in human patients

[37–39]

. Furthermore retinal autoantibodies have

been implicated in the pathologic changes of macular degeneration

[40–42]

Fig. 18. Proposed mechanisms of antibody-mediated retinopathies that may be responsible
for clinical symptoms of SARDS and IMR in canine patients.

380

GROZDANIC, HARPER, & KECOVA

nonhereditary forms of retinitis pigmentosa

[43–45]

, and glaucoma

[46,47]

. In

these diseases, however, it is not clear if the antibody production precedes the
retinal disease or if the immune reactivity is a consequence of the retinal degen-
erative process. Multiple lines of evidence demonstrated that autoantibodies are
usually generated outside of the eye as a result of an immune response against

Fig. 20. ERG tracings of a 5-year-old, castrated male, mixed-breed dog (case 1) with a 7-day-
long history of sudden onset of bilateral blindness. The owner reported polyuria, polydipsia,
and weight gain in the past 6 months. The owner also reported several episodes of decreased
vision or blindness, which were noticed during that period. The owner elected not to pursue
any additional diagnostic work and medical treatment. This patient spontaneously recovered
vision several times after being diagnosed with IMR and developed complete blindness 5
months after the initial examination.

Fig. 19. ERG tracings from patient 3 (see

). This dog had a 1-month-long history of

sudden onset of blindness; however, the owner perceived that vision was present intermittently
some days. On ophthalmic examination, the menace response was absent in dim light in both
eyes, although intermittently present left eye in bright light. The menace response was absent
right eye in bright light. The dog did not track objects in scotopic or photopic conditions but
could navigate the scotopic maze test with some difficulties (would intermittently hit objects
on the right side, likely attributable to the completely absent vision right eye). Prednisone
and doxycycline therapy was initiated, and vision was completely recovered 48 hours after
treatment initiation. The owner discontinued treatment 7 days after initiation because of exces-
sive polydipsia and polyuria. Two months after the initial examination, the owner again re-
ported visual problems. Ophthalmic examination revealed the presence of vision in the right
eye and blindness in the left eye. ERG showed a further decline in retinal electrical activity
in both eyes. Vision was recovered in both eyes again after prednisone and doxycycline treat-
ment was reinitiated.

381

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

tumor cells bearing retinal antigens or the exposure of degenerating retinal
elements to the immune system

. Although a recent study demonstrated

the presence of antiretinal antibodies in the circulation of 60% of patients
with symptoms of CAR, it still remains unexplained why almost 40% of
patients who have CAR and numerous autoimmune retinopathies develop
signs of disease in the absence of serum retinal autoantibodies

. SARDS

is a blinding disease in dogs characterized by a near-normal retinal appearance,
painless and absolute loss of vision, and electrical retinal activity

. Despite

striking similarities between antibody-mediated retinopathies in human patients
and SARDS, recent studies did not demonstrate the presence of serum retinal
autoantibodies and systemic neoplastic disease in patients that had SARDS

[10,11]

. IMR is a canine syndrome that shares may clinical similarities with

SARDS; however, the presence of electrical retinal activity, sporadic presence
of other systemic symptoms, and likely presence of circulating antiretinal auto-
antibodies makes this syndrome easily distinguished from SARDS.

Antibody-mediated retinopathy in human beings is frequently associated

with the presence of systemic neoplasia (CAR), which results in an immune
response against cancer cells bearing retinal proteins and, ultimately, the
production of circulating retinal autoantibodies

[19,37,38,48–54]

. A significant

percentage of patients who have CAR and autoimmune retinopathies do not
have detectable retinal autoantibodies in systemic circulation, however, which
has resulted in difficulties in understanding how retinal damage occurs in these

Fig. 21. Border collie (7 years old) diagnosed with IMR using the Melan-100 unit. ERG activity
is shown before (A, B) and after (C, D) systemic steroid therapy, which completely reversed the
symptoms of blindness. (Courtesy of Peter McElroy, BVSc, CertVOpthal, MRCVS, Cheshire, UK.)

382

GROZDANIC, HARPER, & KECOVA

patients and what diagnostic parameters should be used to establish the correct
diagnosis

[37,55–57]

. The authors’ studies demonstrated the presence of intra-

retinal plasma cells producing different classes of immunoglobulins and strong
upregulation of genes controlling globulin synthesis in dogs that have SARDS,
which is strongly suggestive of the intraretinal production of antibodies. Fur-
thermore, the authors have recognized and described a new syndrome
(IMR) that can be associated with the presence of systemic neoplasia in some
patients and shares similarities with CAR in human patients. Localized intrare-
tinal antibody production potentially offers an explanation for the development
of clinical symptoms of retinopathy in the absence of circulating antiretinal
antibodies, because concentration of antibodies in the retinal tissue may be suf-
ficiently high to create retinal damage but too low (or completely absent) in
serum to be detected with readily available tests (ELISA and Western blot).
Although numerous studies demonstrated a direct intracellular effect of anti-
bodies on retinal neurons

[58–62]

, the authors’ data are suggestive of possible

additional mechanisms, such as complement-mediated and T-cell–mediated
retinal damage in patients that have SARDS.

Antiretinal antibodies can be produced as a result of presentation of tumor

cells bearing retinal proteins or retina-like antigen to the immune system in
the following ways:

1. Autoantibodies are produced at periphery; they penetrate the blood-retinal

barrier and inflict direct damage to retinal neurons by intracellular penetration.

2. T cells and antibodies penetrate the blood-retinal barrier and mediate cell-

dependent destruction of retinal neurons.

3. Plasma cells penetrate the blood-retinal barrier and produce autoanti-

bodies within retinal tissue (proposed mechanism in patients that have
SARDS and in human patients who have clinical symptoms of disease
but without the presence of circulating autoantibodies). Complement-
mediated retinal neuronal damage can occur in the presence of autoanti-
bodies, regardless of the place of production (extraocular or intraocular
antibody production).

PUPIL LIGHT REFLEX–BASED DIAGNOSTIC MODALITIES

The authors previously demonstrated that colorimetric PLR analysis is a sensi-
tive test for rapid diagnosis of SARDS in canine patients with sudden onset of
visual loss and a relatively normal fundus appearance

. In a recent study

and this review (see

Tables 2 and 3

), the authors demonstrate that the therapeu-

tic response to IVIg therapy can be objectively evaluated by measuring PLR
amplitudes when red light of a specific wave length (630 nm) is used, which
strictly activates the photoreceptor-mediated component of the PLR

http://www.biomedcentral.com/1471-2415/

. Color-

imetric PLR analysis is a rapid test, which can be useful in clinical settings to
establish an early diagnosis of antibody-mediated retinopathies. Severe PLR
deficits with a red light stimulus and relatively normal PLRs with a blue light
stimulus in patients with sudden onset of visual loss are immediate indicators
of disrupted photoreceptor activity or synaptic transmission between

383

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

photoreceptors and retinal ganglion cells driving pupil responses, which is the
usually observed clinical situation in antibody-mediated retinopathies.

INTRAVENOUS IMMUNOGLOBULIN–BASED TREATMENT
MODALITIES

SARDS has been considered an untreatable canine blinding disease because of
the complete lack of therapeutic response to anti-inflammatory, antimicrobial,
or immunosuppressive medications. Antibody-mediated retinopathies in
human beings are frequently described as poorly responsive to medical treat-
ment; however, a high dose of steroids, plasmapheresis, and IVIg therapy
have been described to reverse symptoms of blindness partially

[16–20]

. The

authors’ data demonstrated significant improvement in visual behavior and
photoreceptor-mediated PLR responses in all treated patients that had SARDS
and in two patients that had IMR and were treated with IVIg.

SUMMARY

Antibody-mediated retinopathies may be widely present among the canine
population. Early diagnosis and appropriate treatment are essential for visual
preservation and reversal of blindness in these patients.

Acknowledgments

The authors thank Dr. Grazyna Adamus for analyzing serum samples for
retinal autoantibodies, Drs. Richard Dubielzig and Tanja Nushbaum for pro-
viding SARDS tissue for histologic and molecular analysis, and Drs. Nelms
and Korsh for providing a majority of clinical data from IVIg treated patients
outside the ISU.

References

[1] Vainisi SJ, Schmidt GM, West CS. Metabolic toxic retinopathy—preliminary report. In:

Proceedings 14th Annu Meet Am Coll Vet Ophthalmol; 1983.

[2] Vainisi SJ, Font RL, Anderson R. Idiopathic photoreceptor cell degeneration in dogs

[abstract]. Invest Ophthalmol Vis Sci 1985;26(Suppl 3).

[3] Abrams KL, Gareen IF, Marchand KN. Factors associated with canine sudden acquired

retinal degeneration syndrome (SARDS)—350 cases [abstract]. Vet Ophthalmol
2001;4(4):289–300.

[4] Gilmour MA. Clinical follow-up on 15 dogs with sudden acquired retinal degeneration

syndrome. Vet Ophthalmol 2004;7(6):437–53.

[5] Carter RT, Bentley E, Oliver JW, et al. Elevations in adrenal sex hormones in canine sudden

acquired retinal degeneration syndrome (SARDS). Vet Ophthalmol 2004;7(6):437–53.

[6] Grozdanic SD, Matic M, Sakaguchi DS, et al. Evaluation of retinal status using chromatic

pupil light reflex activity in healthy and diseased canine eyes. Invest Ophthalmol Vis Sci
2007;48:5178–83.

[7] Grozdanic SD, Harper MH, Boll AE, et al. Antobody-mediated retinopathy–new mecha-

nisms and treatment strategies. Invest Opthalmol Vis Sci, submitted for publication.

[8] Abrams KL, Dreyer EB. Excitotoxicity in canine sudden acquired retinal degeneration.

Proceedings of the Association for Research in Vision and Ophthalmology 1993;34:3297.

[9] Bellhorn RW, Murphy CJ, Thirkill CE. Anti-retinal immunoglobulins in canine ocular

diseases. Semin Vet Med Surg (Small Anim) 1988;3:28–32.

384

GROZDANIC, HARPER, & KECOVA

[10] Keller RL, Kania SA, Hendrix DV, et al. Evaluation of canine serum for the presence of

antiretinal autoantibodies in sudden acquired retinal degeneration syndrome. Vet Ophthal-
mol 2006;9:195–200.

[11] Gilmour MA, Cardenas MR, Blaik MA, et al. Evaluation of a comparative pathogenesis

between cancer-associated retinopathy in humans and sudden acquired retinal degenera-
tion syndrome in dogs via diagnostic imaging and western blot analysis. Am J Vet Res
2006;67:877–81.

[12] Miller PE, Galbreath EJ, Kehren JC, et al. Photoreceptor cell death by apoptosis in dogs with

sudden acquired retinal degeneration syndrome. Am J Vet Res 1998;59:149–52.

[13] Van Gelder RN. Nonvisual ocular photoreception in the mammal. Methods Enzymol

2005;393:746–55.

[14] Ruby NF, Brennan TJ, Xie X, et al. Role of melanopsin in circadian responses to light. Science

2002;298:2211–3.

[15] Hanifin JP, Brainard GC. Photoreception for circadian, neuroendocrine, and neurobehavio-

ral regulation. J Physiol Anthropol 2007;26:87–94.

[16] Klingele TG, Burde RM, Rappazzo JA, et al. Paraneoplastic retinopathy. J Clin Neurooph-

thalmol 1984;4:239–45.

[17] Guy J, Aptsiauri N. Treatment of paraneoplastic visual loss with intravenous immunoglobu-

lin: report of 3 cases. Arch Ophthalmol 1999;117:471–7.

[18] Keltner JL, Thirkill CE, Tyler NK, et al. Management and monitoring of cancer-associated

retinopathy. Arch Ophthalmol 1992;110:48–53.

[19] Keltner JL, Thirkill CE, Yip PT. Clinical and immunologic characteristics of melanoma-

associated retinopathy syndrome: eleven new cases and a review of 51 previously
published cases. J Neuroophthalmol 2001;21:173–87.

[20] Murphy MA, Thirkill CE, Hart WM Jr. Paraneoplastic retinopathy: a novel autoantibody

reaction associated with small-cell lung carcinoma. J Neuroophthalmol 1997;17:77–83.

[21] Kaneko Y, Nimmerjahn F, Ravetch JV. Anti-inflammatory activity of immunoglobulin G

resulting from Fc sialylation. Science 2006;313:670–3.

[22] Nimmerjahn F, Ravetch JV. The antiinflammatory activity of IgG: the intravenous IgG

paradox. J Exp Med 2007;204:11–5.

[23] Kempf C, Stucki M, Boschetti N. Pathogen inactivation and removal procedures used in the

production of intravenous immunoglobulins. Biologicals 2007;35:35–42.

[24] Boschetti N, Stucki M, Spath PJ, et al. Virus safety of intravenous immunoglobulin: future

challenges. Clin Rev Allergy Immunol 2005;29:333–44.

[25] Kaneko Y, Nimmerjahn F, Madaio MP, et al. Pathology and protection in nephrotoxic nephri-

tis is determined by selective engagement of specific Fc receptors. J Exp Med 2006;203:
789–97.

[26] Samuelsson A, Towers TL, Ravetch JV. Anti-inflammatory activity of IVIG mediated through

the inhibitory Fc receptor. Science 2001;291:484–6.

[27] Vaccaro C, Zhou J, Ober RJ, et al. Engineering the Fc region of immunoglobulin G to

modulate in vivo antibody levels. Nat Biotechnol 2005;23:1283–8.

[28] Bianco D, Armstrong PJ, Washabau RJ. Treatment of severe immune-mediated thrombocyto-

penia with human IV immunoglobulin in 5 dogs. J Vet Intern Med 2007;21:694–9.

[29] Nuttall TJ, Malham T. Successful intravenous human immunoglobulin treatment of drug-

induced Stevens-Johnson syndrome in a dog. J Small Anim Pract 2004;45:357–61.

[30] Rahilly LJ, Keating JH, O’Toole TE. The use of intravenous human immunoglobulin in

treatment of severe pemphigus foliaceus in a dog. J Vet Intern Med 2006;20:1483–6.

[31] Trotman TK, Phillips H, Fordyce H, et al. Treatment of severe adverse cutaneous drug

reactions with human intravenous immunoglobulin in two dogs. J Am Anim Hosp Assoc
2006;42:312–20.

[32] Schiavotto C, Ruggeri M, Rodeghiero F. Adverse reactions after high-dose intravenous

immunoglobulin: incidence in 83 patients treated for idiopathic thrombocytopenic purpura
(ITP) and review of the literature. Haematologica 1993;78:35–40.

385

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

[33] Stenton SB, Dalen D, Wilbur K. Myocardial infarction associated with intravenous immune

globulin. Ann Pharmacother 2005;39:2114–8.

[34] Orbach H, Tishler M, Shoenfeld Y. Intravenous immunoglobulin and the kidney—a two-

edged sword. Semin Arthritis Rheum 2004;34:593–601.

[35] Jolles S, Sewell WA, Leighton C. Drug-induced aseptic meningitis: diagnosis and manage-

ment. Drug Saf 2000;22:215–26.

[36] Gaines AR. Acute onset hemoglobinemia and/or hemoglobinuria and sequelae following

Rh(o)(D) immune globulin intravenous administration in immune thrombocytopenic purpura
patients. Blood 2000;95:2523–9.

[37] Adamus G, Ren G, Weleber RG. Autoantibodies against retinal proteins in paraneoplastic

and autoimmune retinopathy. Available at:

4/5

[38] Thirkill CE, FitzGerald P, Sergott RC, et al. Cancer-associated retinopathy (CAR syndrome)

with antibodies reacting with retinal, optic-nerve, and cancer cells. N Engl J Med
1989;321:1589–94.

[39] Thirkill CE. Experimental, cancer-induced retinopathy. Ocul Immunol Inflamm 1997;5:

55–65.

[40] Patel N, Ohbayashi M, Nugent AK, et al. Circulating anti-retinal antibodies as immune

markers in age-related macular degeneration. Immunology 2005;115:422–30.

[41] Cherepanoff S, Mitchell P, Wang JJ, et al. Retinal autoantibody profile in early age-related

macular degeneration: preliminary findings from the Blue Mountains Eye Study. Clin
Experiment Ophthalmol 2006;34:590–5.

[42] Umeda S, Suzuki MT, Okamoto H, et al. Molecular composition of drusen and possible

involvement of anti-retinal autoimmunity in two different forms of macular degeneration in
cynomolgus monkey (Macaca fascicularis). FASEB J 2005;19:1683–5.

[43] Heckenlively JR, Aptsiauri N, Nusinowitz S, et al. Investigations of antiretinal antibodies in

pigmentary retinopathy and other retinal degenerations. Trans Am Ophthalmol Soc
1996;94:179–200 [discussion: 200–6].

[44] Heckenlively JR, Jordan BL, Aptsiauri N. Association of antiretinal antibodies and cystoid

macular edema in patients with retinitis pigmentosa. Am J Ophthalmol 1999;127:565–73.

[45] Heckenlively JR, Solish AM, Chant SM, et al. Autoimmunity in hereditary retinal degenera-

tions. II. Clinical studies: antiretinal antibodies and fluorescein angiogram findings.
Br J Ophthalmol 1985;69:758–64.

[46] Romano C, Barrett DA, Li Z, et al. Anti-rhodopsin antibodies in sera from patients with

normal-pressure glaucoma. Invest Ophthalmol Vis Sci 1995;36:1968–75.

[47] Tezel G, Edward DP, Wax MB. Serum autoantibodies to optic nerve head glycosaminogly-

cans in patients with glaucoma. Arch Ophthalmol 1999;117:917–24.

[48] Hooks JJ, Tso MO, Detrick B. Retinopathies associated with antiretinal antibodies.

Clin Diagn Lab Immunol 2001;8:853–8.

[49] Adamus G, Aptsiauri N, Guy J, et al. The occurrence of serum autoantibodies against

enolase in cancer-associated retinopathy. Clin Immunol Immunopathol 1996;78:120–9.

[50] Adamus G, Amundson D, MacKay C, et al. Long-term persistence of antirecoverin anti-

bodies in endometrial cancer-associated retinopathy. Arch Ophthalmol 1998;116:251–3.

[51] Adamus G, Guy J, Schmied JL, et al. Role of anti-recoverin autoantibodies in cancer-

associated retinopathy. Invest Ophthalmol Vis Sci 1993;34:2626–33.

[52] Thirkill CE, Keltner JL, Tyler NK, et al. Antibody reactions with retina and cancer-associated

antigens in 10 patients with cancer-associated retinopathy. Arch Ophthalmol 1993;111:
931–7.

[53] Thirkill CE, Roth AM, Keltner JL. Cancer-associated retinopathy. Arch Ophthalmol

1987;105:372–5.

[54] Thirkill CE, Tait RC, Tyler NK, et al. The cancer-associated retinopathy antigen is a recoverin-

like protein. Invest Ophthalmol Vis Sci 1992;33:2768–72.

386

GROZDANIC, HARPER, & KECOVA

[55] Weleber RG, Watzke RC, Shults WT, et al. Clinical and electrophysiologic characterization

of paraneoplastic and autoimmune retinopathies associated with antienolase antibodies.
Am J Ophthalmol 2005;139:780–94.

[56] Heckenlively JR, Fawzi AA, Oversier J, et al. Autoimmune retinopathy: patients with

antirecoverin immunoreactivity and panretinal degeneration. Arch Ophthalmol
2000;118:1525–33.

[57] Mizener JB, Kimura AE, Adamus G, et al. Autoimmune retinopathy in the absence of cancer.

Am J Ophthalmol 1997;123:607–18.

[58] Adamus G. Autoantibody-induced apoptosis as a possible mechanism of autoimmune

retinopathy. Autoimmun Rev 2003;2:63–8.

[59] Shiraga S, Adamus G. Mechanism of CAR syndrome: anti-recoverin antibodies are the

inducers of retinal cell apoptotic death via the caspase 9- and caspase 3-dependent
pathway. J Neuroimmunol 2002;132:72–82.

[60] Adamus G, Machnicki M, Elerding H, et al. Antibodies to recoverin induce apoptosis of

photoreceptor and bipolar cells in vivo. J Autoimmun 1998;11:523–33.

[61] Magrys A, Anekonda T, Ren G, et al. The role of anti-alpha-enolase autoantibodies in

pathogenicity of autoimmune-mediated retinopathy. J Clin Immunol 2007;27:181–92.

[62] Maeda T, Maeda A, Maruyama I, et al. Mechanisms of photoreceptor cell death in cancer-

associated retinopathy. Invest Ophthalmol Vis Sci 2001;42:705–12.

387

ANTIBODY-MEDIATED RETINOPATHIES IN CANINE PATIENTS

Orbital Inflammatory Disease
and Pseudotumor in Dogs and Cats

Alexandra van der Woerdt, DVM, MS

The Animal Medical Center, 510 East 62nd Street, New York, NY 10065, USA

he orbit is a conical cavity, which contains the eyeball and the ocular
adnexa

. The orbital margin is incomplete in the dog and cat. It is lined

with bony tissue for approximately 80% of its circumference. The orbital

ligament forms the remaining 20% of the orbital margin. Only the medial wall
and part of the orbital roof are osseous in the dog. Soft tissue forms the lateral
wall and the orbital floor. The dorsal surface of the zygomatic salivary gland
forms a large part of the orbital floor. This salivary gland rests on the dorsolat-
eral surface of the medial pterygoid muscle. The lateral and dorsolateral part of
the orbit is formed by the medial surface of the temporalis muscle and the
orbital ligament. The ramus of the mandible is embedded in the masseter
and temporal muscles immediately caudal to the orbit. The ramus of the man-
dible compresses the orbital contents on opening of the mouth. The orbital
fascia is a tough connective tissue liner that envelops all the structures within
the orbit

. It can be subdivided into the periorbita (lines the orbit); Tenon’s

capsule (also known as ‘‘fascia bulbi’’); and the fascial sheaths of the extraocu-
lar muscles. Orbital fat separates these three sheets of orbital fascia.

Clinical signs of orbital disease can be variable, depending on whether the

disease process is associated with a loss of orbital tissue or a space-occupying
lesion in the orbit. Diseases associated with space-occupying lesions in the orbit
are more common than diseases associated with orbital tissue loss. Examples of
the latter include resorption of retrobulbar fat in aging or systemically ill
animals, and fractures of orbital bones resulting in loss of support for the ocular
contents. Orbital neoplasia in cats can result in destruction of the ventral floor
of the orbit resulting in enophthalmos

[3,4]

. Clinical signs associated with

space-occupying lesions in the orbit include exophthalmos, protrusion of the
third eyelid, strabismus, dislocation of the globe within the orbit, and resistance
to retropulsion. Associated signs in the globe may include conjunctival hyper-
emia, (ulcerative) keratitis secondary to lagophthalmos and exposure, pupillary
light reflex abnormalities, and loss of vision. Abnormalities of the posterior seg-
ment may be caused by impaired venous drainage from the eye, or indentation

E-mail address: sandra.vanderwoerdt@amcny.org

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.11.006

Vet Clin Small Anim 38 (2008) 389–401

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

of the globe by the space-occupying lesion. Space-occupying lesions in the orbit
can be congenital, inflammatory, or neoplastic in nature. Although many con-
genital lesions may manifest soon after birth, they cannot be excluded based on
the age of the animal. Dogs and cats with inflammatory orbital disease often
have a history of acute onset of clinical signs. General malaise may be present,
and the animal may be febrile. The area may be warm and tender to touch.
Pain is often present on opening of the mouth and the animals often resist
manipulation of the mouth. They may refuse to eat hard foods and instead
prefer to eat soft foods. Evidence of inflammation is often present in a complete
blood count. Dogs and cats with neoplastic orbital disease often have a more
insidious onset of clinical signs. The owner may present the animal for an eval-
uation of a slowly progressive protrusion of the third eyelid, or protrusion of
the globe, or deviation of the globe. Fever and general malaise are usually
absent and pain is typically absent on opening of the mouth. Weight loss
may have been noticed. Response to antibiotic therapy is absent or incomplete.
Associated clinical signs of orbital neoplasia depend on whether the neoplastic
process is primary orbital neoplasia (most common in dogs), or secondary
neoplasia by extension or metastasis (most common in cats).

This article discusses diagnostic work-up for dogs and cats with signs sug-

gestive of orbital disease. Diagnosis and treatment of nonneoplastic orbital
diseases follow. Congenital diseases are discussed first, followed by inflamma-
tory diseases of the orbit including orbital pseudotumor. Lastly, other less
common orbital conditions are addressed. Orbital neoplasia is not the focus
of this article, and the reader is referred to the reference list for additional read-
ing on orbital neoplasia

[3,5–17]

DIAGNOSTIC WORK-UP

Diagnostic work-up for exophthalmos starts with a thorough physical examina-
tion with careful examination of the teeth and the mouth. Routine blood work
(complete blood count and serum biochemistry profile) and urinalysis are indi-
cated. Radiographs of the chest cavity to look for evidence of metastatic disease
are especially important if neoplastic disease is suspected. Advanced imaging
may include orbital ultrasonography, skull radiographs, CT scan, or MRI

. Orbital ultrasonography has the advantage of a quick, easy procedure

that does not require sedation or general anesthesia. It is often used as the first
imaging technique in orbital disease. Abnormalities were detected on ultrasono-
graphic examination in 86% of dogs studied in a retrospective study involving
50 dogs with signs of orbital disease

. Cavitary lesions were recognized

ultrasonographically in 50% of dogs with a retrobulbar abscess, 75% of dogs
with a salivary mucocele, and in 12% of dogs with retrobulbar neoplasia.
The presence of lesions of orbital bones or a mass on the medial aspect of
the orbit was highly suggestive of neoplasia in this study. Other ultrasono-
graphic signs were nonspecific. Although ultrasound had limited ability to
distinguish between neoplastic and inflammatory tissues in this study, it can
be a useful tool to guide needle placement for aspiration of tissue for cytologic

390

VAN DER WOERDT

examination

. Skull radiographs are of limited use in the diagnostic work-up

of orbital disease. In a study evaluating 20 animals with orbital disease, skull
radiographs were found to be helpful only in those animals in which the disease
process had extended significantly beyond the orbit into the nasal cavity and
sinuses

. CT eliminates the problem of superimposition of body parts

Compared with radiography, CT has a higher sensitivity for detecting disease
and allows more accurate assessment of the extension of lesions

. A recent

study used both CT and ultrasonography to try to predict whether orbital lesions
were neoplastic or nonneoplastic

. Both ultrasonography and CT accurately

detected orbital lesions in a large percentage of the animals examined. Both CT
and ultrasonography depicted size, location, and margins of the lesion similarly.
Neoplastic lesions were more often clearly defined, were more often focal, and
were more likely to indent the globe than inflammatory lesions

. CT detected

extraorbital extension more frequently than ultrasonography. MRI produces
detailed images of the globe and associated structures, such as the optic nerve.
It is superior to CT in characterization of lesions and is more likely than CT
accurately to determine intracranial extension of lesions

[20,22]

. All four imaging

modalities were used in a cat with exophthalmos

. Orbital ultrasonography

and skull radiographs failed to delineate the mass. A CT scan was performed
and revealed lysis of the calvarium, but the intracranial extent was not well
defined. MRI was the only of the four techniques performed that accurately
delineated the border of the neoplasm. A tissue diagnosis is obtained through
fine-needle aspirate of the lesion, or preferably, by obtaining a surgical biopsy.

CONGENITAL ORBITAL ABNORMALITIES

A congenital orbital cyst has been reported in a 4-year-old wire-haired Dachs-
hund

. The dog was presented for an evaluation of protrusion of the left

eye of 1-week duration. Ophthalmic examination revealed exophthalmos, third
eyelid protrusion, and medial strabismus. Fundus examination revealed a protru-
sion of the left optic disk and tapetal fundus adjacent to the dorsal large retinal
vessels. A nonhomogeneous, echogenic, well-demarcated mass was visible in
the caudomedial orbit using ultrasonography. A CT scan was performed and
showed a cystic lesion, which was removed surgically through a lateral orbito-
tomy. Surgical removal was curative.

A congenital cystic eye in a 3-month-old Cocker Spaniel, and a colobomatous

cyst with extreme microphthalmos in a 6-month-old Miniature Poodle have
also been reported

Congenital or acquired arteriovenous fistulas in the orbit or orbital varix are

rare

[26,27]

. Intermittent exophthalmos that may vary with position of the head

or pressure on the jugular veins may be present. Auscultation of the orbit may
reveal a systolic murmur.

ORBITAL INFLAMMATORY DISEASES

Many diseases and events can result in infection or inflammation of the orbital
tissues. Foreign bodies may penetrate the conjunctiva and migrate into the

391

ORBITAL INFLAMMATORY DISEASE AND PSEUDOTUMOR

orbital space. A foreign body located in the oral cavity may penetrate the soft
palate and lodge behind the eye. Severe dental disease can result in the forma-
tion of a retrobulbar abscess. Penetrating bite wounds and hematogenous
spread of infectious organisms are other causes of retrobulbar disease. Lastly,
diseases of the tissues in the orbit (zygomatic and lacrimal gland) and the
tissues that form the orbital wall (bone, muscle, fat) may result in orbital
disease.

Foreign bodies are often plant material, wood splinters, glass, gunshot pellets,

or metallic fragments

. They can migrate from the mouth or from periocular

tissues. Foreign bodies that are difficult to access, and are composed of inert
material, are often left in situ. Lead is well tolerated by the orbital tissues. Ferrous
and copper ions destroy retinal photoreceptors and retinal pigment epithelium
and foreign bodies of these materials should be removed. Bacteria or fungi often
contaminate plant material. Orbital phaeohyphomycosis has been reported in
a cat, resulting in excenteration of the orbit. A grass awn was found within the
mass on histopathologic examination

. Five cases of intraocular or orbital

disease secondary to porcupine quills have been reported in the dog

. The

quills have sharp points that can penetrate the eye or orbit directly and destroy
the eye. Porcupine quills are not inert, can be irritating, and are often contami-
nated with bacteria. Surgical removal is indicated. Ocular ultrasonography
was used to identify the location of the porcupine quills in the dogs in the study.
A characteristic double-banded, linear, hyperechoic object was seen in the orbital
space or the globe

. Surgical removal of the quill with additional medical

management cured four out of five dogs. Perforation of the globe by a quill
had resulted in phacoclastic uveitis that required enucleation of the painful eye
in the fifth dog. It can be challenging to detect the presence of plant material
in the orbit. Failure to respond to antibiotic therapy with or without drainage
through the pterygopalatine fossa may indicate the presence of a persistent
foreign body in the orbit. A migrating grass awn caused an intraocular abscess
and orbital cellulitis in a cat that ultimately resulted in enucleation of the affected
eye. Histopathologic examination of the globe revealed the presence of a grass
foreign body

Dental disease is very common in middle-aged to older dogs and cats. The

posterior maxillary teeth have a very close proximity to the orbit. The floor of
the orbit is composed of soft tissue structures. The fourth premolar and the first
and second molar of the maxilla are the teeth that are most commonly the
cause of orbital disease

. Periapical abscess formation of those teeth may

result in orbital inflammation. An acute onset of clinical signs associated with
the inflammation may be the reason that the owner presents their dog to the
veterinarian. Careful inspection of the teeth is indicated in any animal with
acute onset of orbital disease. Dental radiographs are often required to properly
diagnose and treat dental disease. Fractures of teeth may result in periapical
abscess formation that may not be visible externally. Surgical removal of
affected teeth is indicated with surgical drainage and systemic antibiotic
therapy. Care should be taken to avoid injury to the orbital tissues and globe

392

VAN DER WOERDT

when removing affected teeth. Inadvertent penetration of the orbit and globe
with a root elevator with subsequent loss of the eye has been reported

Hematogenous spread of infectious organisms or spread from adjacent

tissues is another cause of orbital inflammation in dogs and cats. The source
of the infection is sometimes known, but the source cannot always be located.
Infectious agents include aerobic and anaerobic bacteria

, Ehrlichia canis

, Toxocara canis

, and fungi

. Orbital disease can be the result

of an extension from sinusitis

[41,42]

or osteomyelitis

[43,44]

. Two cases of

osteomyelitis with orbital disease have been reported in young dogs. One
was a 3-month-old Great Dane puppy with osteomyelitis of the frontal sinus.
Aggressive surgical removal with reconstructive surgery resulted in a favorable
long-term outcome in this dog. A 5-month-old Miniature Pincher developed
osteomyelitis of the orbital part of the frontal bone. Despite surgical removal
of the affected tissues, symptoms deteriorated and the dog was euthanized.
Sulfur granules typical of actinomycetes were found on histopathologic or
cytologic examination in both dogs. Inflammation of the zygomatic salivary
gland with or without cyst formation has been described in 10 dogs

. Clin-

ical signs indicated orbital disease in combination with increased salivation or
increased swallowing. Physical examination revealed a distended excretory
duct of the zygomatic gland in the region of the upper fourth premolar tooth.
MRI confirmed diagnosis in eight dogs. Treatment included systemic broad-
spectrum antibiotics and nonsteroidal anti-inflammatory drugs. Retrograde
flushing of the duct with acetylcysteine was also performed. Response to treat-
ment was rapid with only one recurrence noticed. A modified lateral orbitotomy
was performed to treat a zygomatic mucocele in another dog

Masticatory muscle myositis is a disease that occurs mainly in young, large-

breed dogs

. Inflammation of the masticatory muscles (masseter, tempora-

lis, and pterygoid muscles) is associated with production of antibodies against
a unique myosin isoform present in type 2 M fibers in the masticatory muscles.
The antibodies detected in dogs with masticatory muscle myositis are directed
against muscles derived from the mesoderm (masticatory muscles), but not
against antigen of skeletal muscles. In this disease, early muscle fiber damage
is initiated by CD8

cytotoxic T cells, which then leads to production of anti-

bodies against muscle fiber protein

. Dogs with acute masticatory muscle

myositis may have trismus, swelling of the masticatory muscles, painful facial
muscles, and ocular abnormalities

. Ocular abnormalities described include

exophthalmos and conjunctival hyperemia. Vision loss is possible secondary to
optic nerve compression

. The disease has a recurrent nature and may

result in extensive atrophy and fibrosis of the masticatory muscles. Inability
to open the jaw may lead to malnourishment

. The diagnosis of masticatory

muscle myositis is based on clinical signs and the presence of antibodies against
type 2 M fibers. Chronic cases are best diagnosed using a muscle biopsy. Treat-
ment consists of immunosuppressive corticosteroid therapy.

Extraocular polymyositis is an inflammatory disease of the extraocular

muscles. Although it has been reported in many breeds, a predisposition for

393

ORBITAL INFLAMMATORY DISEASE AND PSEUDOTUMOR

the Golden Retriever exists

[50,51]

. Young female dogs are most commonly

affected and a stressful episode is often present immediately before the onset
of clinical signs. Clinical signs include exophthalmos (often bilateral) and chemo-
sis limited to the bulbar conjunctiva. Show of the superior sclera with or without
upper eyelid retraction is also present in a large percentage of these dogs. Mild
ocular hypertension, visual impairment, and fundoscopic abnormalities are
less common. Treatment consists of immunosuppressive therapy. Recurrences
are common and may lead to strabismus and enophthalmos

Treatment of a retrobulbar abscess consists of establishing drainage and

treatment of the underlying disease. Drainage of a retrobulbar abscess can be
performed through the pterygopalatine fossa. A small stab incision is made
in the mucosa behind the last upper molar tooth. A pair of small hemostats
is then carefully inserted into the wound and slowly advanced through the
tissues. Repeated opening and closing of the hemostat allows penetration of
the tissue and opening of a retrobulbar abscess. Drainage can then be estab-
lished through the mouth. Improvement of clinical signs following this proce-
dure can be immediate and dramatic (

Figs. 1 and 2

). Introduction of sharp

instruments in the area should be avoided because large vessels traverse the
area and extensive hemorrhage can occur if these vessels are lacerated. Drain-
age through the mouth is not always successful. Removal of a foreign body, if
identified, may necessitate more aggressive surgical therapy including orbitot-
omy. Ultrasound-guided surgical removal of intraconal orbital grass awns
through a small conjunctival incision has been described

Aftercare consists of systemic broad-spectrum antibiotic therapy and nonste-

roidal anti-inflammatory therapy to reduce inflammation and to alleviate pain.
Topical therapy for the eye is often necessary and may include frequent lubri-
cation of the eye if the corneal epithelium is intact. Topical antibiotic therapy

Fig. 1. One-year-old domestic shorthair cat with a retrobulbar abscess behind the left eye.
Note the exophthalmos, protrusion of the third eyelid, conjunctival hyperemia, and exposure
keratitis in the left eye.

394

VAN DER WOERDT

with or without topical atropine is indicated if corneal ulceration secondary to
exposure is present.

ORBITAL PSEUDOTUMOR

A rare sclerosing orbital disease has recently been described in a few reports in
cats

[54–56]

. The onset of clinical signs is gradual. Clinical signs include exoph-

thalmos with progressive lack of motility of the globe and surrounding tissues.
Resistance to retropulsion and protrusion of the third eyelid is often present.
Progressive lack of eyelid function may lead to exposure keratitis and entro-
pion. Corneal perforation necessitating enucleation is common. CT reveals
thickening of the sclera and adjacent tissues in most cases (

). A distinct

mass can be seen in other cases (

). Fine-needle aspirate of the abnormal

tissues is generally nondiagnostic. The diagnosis requires a biopsy of the

Fig. 2. The same cat as in

, approximately 1 hour after the abscess was drained through

the mouth under general anesthesia. Note the remarkable improvement in clinical signs.

Fig. 3. CT scan image of the orbits of a middle-aged female spayed Persian cat. Note the
increased soft tissue density involving the sclera and surrounding tissues in the left orbit.

395

ORBITAL INFLAMMATORY DISEASE AND PSEUDOTUMOR

abnormal tissue. This diagnosis is often made after the eye has been enucleated
because of poor response to therapy. Histopathologic examination revealed
similar findings in all seven cases in one study

. Macroscopically, extraocular

muscles and orbital fat were entrapped by tissue. A mixed inflammatory
response (lymphocytes and plasma cells) was present in the orbital tissues on
microscopic examination. A perivascular cellular infiltrate was present in orbital
vessels. Fibrous tissue was present posterior to the globe in all cases, infiltrating
between extraocular muscles and encapsulating muscles and orbital fat. Fibrous
tissue was present in all eyelids examined. Treatment has been attempted using
antibiotic therapy, systemic corticosteroid therapy at immunosuppressive dos-
ing, cyclophosphamide, and radiation therapy. Progression of the disease was
noted despite treatment in all reported cases. Involvement of the contralateral
eye is common. The largest study to date

evaluated globes retrospectively

that were submitted for histopathologic examination. A bias may exist in this
study toward those cases that failed to respond to treatment. The author has
diagnosed and treated several cats over the past few years with mixed results.
Quick progression of clinical signs with spread to the contralateral eye was pres-
ent in a few cats treated with immunosuppressive corticosteroid therapy. Finan-
cial constraints prevented enucleation of an affected eye in one cat. This cat was
treated with immunosuppressive corticosteroid therapy and did remarkably well
for 1.5 years, at which time the other eye became involved and euthanasia was
elected. The cat in

lost its affected eye because of corneal perforation, but

has been doing well without signs in the contralateral eye for over a year. Radi-
ation therapy quickly reduced clinical signs in the cat in

. This is a recent

case and it is too soon to evaluate whether radiation therapy will result in long-
term management of the disease in this cat.

Idiopathic inflammatory orbital pseudotumor is the third most common

orbital disease in human beings

[57–62]

. Thyroid-associated orbitopathy and

lymphoproliferative disorders are the most common orbital diseases in human

Fig. 4. CT scan image of the orbits in an older female spayed domestic shorthaired cat.
A focal mass indenting the globe can be seen in the left orbit.

396

VAN DER WOERDT

beings. Orbital pseudotumor is defined as nonspecific benign orbital inflamma-
tion without evidence of specific local or systemic disease. It is considered a diag-
nosis of exclusion after all other causes have been eliminated. Although orbital
pseudotumor has no potential for metastatic disease or cytologic malignancy, it
can run a clinically malignant course with severe vision loss and oculomotor
dysfunction. It is usually unilateral, although bilateral involvement has been
reported, especially in children

. It can be subdivided by chronicity (acute,

subacute, chronic); location; and histologic subtype. Clinical signs include prop-
tosis, motility disturbance, optic nerve compression, pain, redness, and edema

. A focal or diffuse mass, usually poorly demarcated that enhances with

contrast is seen on CT. Sinus involvement or intracranial extension is rarely
seen. Fine-needle aspiration biopsy is not helpful in the diagnosis of orbital pseu-
dotumor

. Fine-needle aspiration was not helpful in the diagnosis of pseudo-

tumor in the cats in the literature either. Prompt resolution of clinical signs with
corticosteroid therapy is reported in the literature in human beings. Several stud-
ies have also shown, however, that a large percentage of patients may experience
recurrences, or may fail to respond to steroid therapy

. Additional therapeu-

tic options include nonsteroidal anti-inflammatory agents, cyclophosphamide,
chlorambucil, methotrexate, intravenous immunoglobulin, surgical debulking,
and radiation therapy. A favorable response to tumor necrosis factor-a inhibitors
has been reported in patients who were nonresponsive to other treatment modal-
ities

. An increase in connective tissue in the pseudotumor lesion is always

present on histopathologic examination

. Histopathologic subtypes of orbital

pseudotumor include granulomatous orbital pseudotumor, vasculitic orbital
pseudotumor, eosinophilic orbital pseudotumor, and sclerosing orbital pseudo-
tumor. The presence of granulomatous inflammation is uncommon and should
result in searching for an underlying systemic granulomatous disorder

Examples include sarcoidosis, Wegener’s granulomatosis, and tuberculosis.
Vasculitic orbital pseudotumor is also rare. Systemic vasculitis should be ruled
out

. Predominant tissue eosinophilia without vasculitis is common in

children with orbital pseudotumor. Peripheral blood eosinophilia is present in
some of these children

. An eosinophilic infiltrate was present in a cat with

retrobulbar disease. This cat responded quickly to corticosteroid therapy with
no recurrence 2 years later

. If the amount of connective tissue is great and

the inflammatory component is minimal, the pseudotumor is called fibrotic, or
sclerotic. Although it has been suggested that the fibrosis presents a late stage
of the disease, others have suggested that this is a separate entity with a poor
prognosis. The amount of fibrosis is not always related to duration of the disease,
and extensive fibrosis can be present early in the course of the disease. Sclerosing
orbital inflammation seems to be associated with a poor response to treatment
and poor outcome in human beings

[59,61]

. A retrospective study involving

31 patients with idiopathic sclerosing orbital inflammation found bilateral
disease in two patients

. Proptosis was the most common clinical sign,

followed by restriction of extraocular movements. Retraction of the lower
eyelid, decreased visual acuity, optic atrophy, ptosis, and palpable mass were

397

ORBITAL INFLAMMATORY DISEASE AND PSEUDOTUMOR

other clinical signs. Intracranial extension was seen in one patient. Response
to systemic corticosteroid therapy was good in 9 of 27 patients treated, poor
in 7 patients, and partial in 11 patients. Cyclophosphamide and azathioprine
therapy was tried with mixed results. Response to radiation therapy was poor
in all six patients treated. Clinical signs, histopathologic results, and poor
response to treatment in the cats in the studies reported to date are comparable
with sclerosing orbital inflammation as seen in human beings.

The pathogenesis of idiopathic orbital inflammation is still unclear. Infection

, immune-mediated disease

, and aberrant immune-mediated produc-

tion of fibrogenic cytokines have been suggested in the pathogenesis of this
disease

. The disease has been suggested to be an autoimmune reaction

to orbital tissues following an unknown immunologic trigger

. Sclerosing

pseudotumors of the orbit resemble idiopathic retroperitoneal fibrosis immuno-
histologically and are considered to be a primary fibrosing disease

[59,62]

OTHER ORBITAL DISEASES

Prolapse of orbital fat through a hernia in the orbital fascia is uncommonly
reported in dogs and cats

[66,67]

. Clinical signs include the presence of

a pain-free, soft, fluctuant, subconjunctival mass. The diagnosis is confirmed
by fine-needle aspirate or biopsy. Surgical removal is optional. These lesions
are nonprogressive in nature. This needs to be differentiated from an orbital
lipoma. An orbital lipoma has been reported in a 10-year-old Cairn terrier. A
progressively enlarging subconjunctival swelling was present in the right eye
of this dog. Ultrasound of the orbit showed that this mass had a distinct border
and extended posteriorly. The mass was removed through a conjunctival inci-
sion

. The progressive nature of this mass suggested a neoplastic disease,

rather than displacement of normal orbital fat.

Emphysema of the orbit is a late and rare complication after enucleation.

Retrograde air movement is present through an intact nasolacrimal system

[69,70]

or connection with a sinus

. This is more common in brachycephalic

dogs than in dogs with other skull types. Increased respiratory effort in these
dogs is suggested as a probable reason for the predisposition in these dogs.
Fine-needle aspirate of an orbital swelling confirms the presence of air in the
orbit. Treatment consists of injection with sclerosing agents (eg, tetracycline),
or scarification or ligation of the patent airway

. Another potential compli-

cation after enucleation is the formation of an orbital mucocele. This is a fluid-
filled cyst that may form secondary to the presence of lacrimal gland or goblet
cell secretions from conjunctival tissue that was not removed at the time of
surgery. History indicates a progressive nonpainful swelling of the orbit after
enucleation. Fine-needle aspirate of the orbit may yield a thick tenacious fluid.
Treatment consists of removal of the offending tissues from the orbit.

SUMMARY

Orbital disease is common in dogs and cats. Clinical signs include exophthal-
mos, protrusion of the third eyelid, and resistance to retropulsion of the globe.

398

VAN DER WOERDT

Retrobulbar disease is either inflammatory or neoplastic in nature. This article
describes the diagnosis and treatment of the most common causes of inflamma-
tory orbital disease in dogs and cats. In addition, a rare orbital disease in cats,
orbital pseudotumor, is discussed.

References

[1] Pollock RVH. The eye. In: Evans HE, Christensen GC, editors. Anatomy of the dog. 2nd

edition. Philadelphia: WB Saunders; 1979. p. 1093–7.

[2] Samuelson DA. Ophthalmic anatomy. In: Gelatt KN, editor. Veterinary ophthalmology.

4th edition. Ames: Blackwell Publishing; 2007. p. 37–41.

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

1990). J Am Vet Med Assoc 1992;201(7):1083–6.

[4] Pentlarge VW, Powell-Johnson G, Martin CL, et al. Orbital neoplasia with enophthalmos in

a cat. J Am Vet Med Assoc 1989;195(9):1249–51.

[5] Hartmann A. Feline exophthalmia. Vet Forum 2005;4:33–7.
[6] Mauldin EA, Deehr AJ, Hertzke D, et al. Canine orbital meningiomas: a review of 22 cases.

Vet Ophthalmol 2000;3:11–6.

[7] Wolfer J, Grahn B. Diagnostic ophthalmology. Amelanotic melanoma in an 8-year-eold cat.

Can Vet J 1995;36(8):518–9.

[8] Willis M, Stiles J, Martin C, et al. What’s your diagnosis. J Am Vet Med Assoc 1997;210(2):

177–8.

[9] Giudice C, Marco R, Mirko R, et al. Zygomatic gland adenoma in a dog: histochemical and

immunohistochemical evaluation. Vet Ophthalmol 2005;8(1):13–6.

[10] Lassaline ME, Gelatt KN, Brooks DE, et al. Orbitotomy for retrobulbar malignant fibrous his-

tiocytoma in a dog. Vet Ophthalmol 2005;8(1):1–6.

[11] Headrick JF, Bentley E, Dubielzig RR. Canine lobular orbital adenoma: a report of 15 cases

with distinctive features. Vet Ophthalmol 2004;7(1):47–51.

[12] Attali-Soussay K, Jegou JP, Clerc B. Retrobulbar tumors in dogs and cats: 25 cases. Vet

Ophthalmol 2001;4:19–27.

[13] Andrew SE. Orbital neurofibrosarcoma in a dog. Vet Ophthalmol 1999;2(3):141–5.
[14] Jones JC, Smith MM, Sponenberg DP, et al. Orbital multilobular tumor of bone in a Basset

Hound. Vet Comp Ophthalmol 1997;7(2):111–6.

[15] Peiffer RL, Spencer C, Popp JA. Nasal squamous cell carcinoma with periocular extension

and metastasis in a cat. Fel Pract 1978;8(6):43–6.

[16] Cottrill NB, Carter JD, Pechman RD, et al. Bilateral orbital parosteal osteoma in a cat. J Am

Anim Hosp Assoc 1987;23:405–8.

[17] Hendrix DV, Gelatt KN. Diagnosis, treatment and outcome of orbital neoplasia in dogs: a ret-

rospective study of 44 cases. J Small Anim Pract 2000;41:105–8.

[18] Daniel GB, Mitchell SK. The eye and orbit. Clin Tech Small Anim Pract 1999;14(3):

160–9.

[19] Mason DR, Lamb CR, McLellan GJ. Ultrasonographic findings in 50 dogs with retrobulbar

disease. J Am Anim Hosp Assoc 2001;37:557–62.

[20] Dennis R. Use of magnetic resonance imaging for the investigation of orbital disease in small

animals. J Small Anim Pract 2000;41:145–55.

[21] Whatmough C, Lamb CR. Computed tomography: principles and applications. Compen-

dium 2006;28(11):789–800.

[22] Boroffka SAEB, Verbruggen AM, Grinwis GCM, et al. Assessment of ultrasonography and

computed tomography for the evaluation of unilateral orbital disease in dogs. J Am Vet Med
Assoc 2007;230:671–80.

[23] Ramsey DT, Gerding PA, Losonsky JM, et al. Comparative value of diagnostic imaging tech-

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

[24] Walde I, Hittmair K, Henninger W, et al. Retrobulbar dermoid cyst in a dachshund. Vet

Comp Ophthalmol 1997;7:239–44.

399

ORBITAL INFLAMMATORY DISEASE AND PSEUDOTUMOR

[25] Regnier A, Raymond-Letron I. Orbital cyst of neural tissue related to ocular maldevelopment:

report of two cases in the dog. Proceedings Ann Meeting Eur Coll Vet Ophthalmol; 2005.
p. 26.

[26] Millichamp NJ, Spencer CP. Orbital varix in a dog. J Am Anim Hosp Assoc 1991;27:

56–60.

[27] Adkins EA, Ward DA, Daniel GB, et al. Coil embolization of a congenital orbital varix in

a dog. J Am Vet Med Assoc 2005;227(12):1952–4.

[28] Grahn BH, Szentimrey D, Pharr JW, et al. Ocular and orbital porcupine quills in the dog:

a review and case series. Can Vet J 1995;36:488–93.

[29] Bernays ME, Peiffer RL. Ocular infections with dematiaceous fungi in two cats and a dog.

J Am Vet Med Assoc 1998;213:507–9.

[30] Tovar MC, Huguet E, Gomezi MA. Orbital cellulites and intraocular abscess caused by

migrating grass in a cat. Vet Ophthalmol 2005;8:353–6.

[31] Ramsey DT, Marretta SM, Hamor RE, et al. Ophthalmic manifestations and complications of

dental disease in dogs and cats. J Am Anim Hosp Assoc 1996;32:215–24.

[32] Smith MM, Smith EM, La Croix N, et al. Orbital penetration associated with tooth extraction.

J Vet Dent 2003;20:8–17.

[33] Wolfer J, Grahn B. Diagnostic ophthalmology. Periocular discharge and swelling around

the eye in a dog. Can Vet J 1995;36(10):661–2.

[34] Breitschwerdt EB, Menard MR, Cullen JM, et al. Acute membranous glomerulopathy in

a dog with retrobulbar cellulitis. J Am Anim Hosp Assoc 1993;29:536–40.

[35] Homma K, Schoster JV. Anaerobit orbital abscess/cellulitis in a Yorkshire Terrier dog. J Vet

Med Sci 2000;62(10):1105–7.

[36] Komnenou A, Mylonakis ME, Kouti V, et al. Ocular manifestations of canine monocytic

ehrlichiosis (Ehrlichia canis): a retrospective study of 90 cases in Greece. Proceedings
Ann Meeting Eur Coll Vet Ophthalmol; 2004. p. 65.

[37] Laus JL, Canola JC, Mamede FV, et al. Orbital cellulitis associated with Toxocara canis in

a dog. Vet Ophthalmol 2003;6(4):333–6.

[38] Hamilton HL, Whitley RD, McLaughlin SA. Exophthalmos secondary to aspergillosis in

a cat. J Am Anim Hosp Assoc 2000;36:343–7.

[39] Rebhun WC, Edwards NJ. Cryptococcus involving the orbit of a dog. Vet Med Small Anim

Clin 1977;72(9):1447–50.

[40] De Lorimier LP, Lacoste HI, Linnetz EH, et al. Blastomycosis: buddy’s budding problem. Vet

forum 2005;12(8):29–38.

[41] Peiffer RL, Belkin PV, Janke GH. Orbital cellulitis, sinusitis, and pneumonitis caused by

Penicillium spp. in a cat. J Am Vet Med Assoc 1980;176:449–51.

[42] Halenda RM, Reed AL. Fungal sinusitis and retrobulbar myofascitis in a cat. Vet Radiol

Ultrasound 1997;38:208–10.

[43] Grahn BH, Szentimrey D, Battison A, et al. Exophthalmos associated with frontal sinus

osteomyelitis in a puppy. J Am Anim Hosp Assoc 1995;31:397–401.

[44] Venter IJ, Petrick SW, Williams MC. Unilateral exophthalmos associated with bacterial

osteomyelitis of the cranium in a dog. Vet Comp Ophthalmol 1996;6(1):56–8.

[45] Allgoewer I, Jurina K, Stockhaus C. Canine zygomatic adenitis. Proceedings Ann Meeting

Eur Coll Vet Ophthalmol; 2004. p. 40.

[46] Bartoe JT, Brightman AH, Davidson HJ. Modified lateral orbitotomy for vision-sparing exci-

sion of a zygomatic mucocele in a dog. Vet Ophthalmol 2007;10:127–31.

[47] Neumann J, Bilzer T. Evidence for MHC-1 restricted CD8þ T-cell-mediated immunopathol-

ogy in canine masticatory muscle myositis and polymyositis. Muscle Nerve 2006;33(2):
215–24.

[48] Podell M. Inflammatory myopathies. Vet Clin North Am Small Anim Pract 2002;32(1):

147–67.

[49] Gilmore MA, Morgan RV. Masticatory myopathy in the dog: a retrospective study of 18

cases. J Am Anim Hosp Assoc 1992;28:300–5.

400

VAN DER WOERDT

[50] Carpenter JL, Schmidt GM, Moore FM, et al. Canine bilateral extraocular polymyositis. Vet

Pathol 1989;26:510–2.

[51] Ramsey DT, Hamor RE, Gerding PA, et al. Clinical and immunohistochemical characteristics

of bilateral extraocular polymyositis ofdogs. Proceedings Ann Meeting Am Coll Vet
Ophthalmol; 1995. p.130.

[52] Allgoewer I, Blair M, Basher T, et al. Extraocular muscle myositis and restrictive strabismus in

10 dogs. Vet Ophthalmol 2000;3(1):21–6.

[53] Stades FC, Djajadiningrat-Laanen SC, Boroffka SAEB, et al. Suprascleral removal of a for-

eign body from the retrobulbar muscle cone in two dogs. J Small Anim Pract 2003;44:
17–20.

[54] Miller SA, van der Woerdt A, Bartick TE. Retrobulbar pseudotumor of the orbit in a cat. J Am

Vet Med Assoc 2000;216:356–8.

[55] Miller-Michau T, Nadelstein B, Gilger BC, et al. Idiopathic inflammatory orbital pseudotu-

mor in 2 cats. 33rd Ann Meeting Am Coll Vet Ophthalmol, 2002.

[56] Billson FM, Miller-Michau T, Mould JRB, et al. Idiopathic sclerosing orbital pseudotumor in

seven cats. Vet Ophthalmol 2006;9:45–51.

[57] Chiu CS, Rubin PAD. Pharmacotherapies and nonpharmacotherapies for orbital inflamma-

tory diseases. Int Ophthalmol Clin 2004;44(3):165–85.

[58] Kapadia MK, Rubin PAD. The emerging use of TNF-alpha inhibitors in orbital inflammatory

disease. Int Ophthalmol Clin 2006;46(2):165–81.

[59] Hsuan JD, Selva D, McNab AA, et al. Idiopathic sclerosing orbital inflammation. Arch

Ophthalmol 2006;124(9):1244–50.

[60] Rubin PAD, Foster CS. Etiology and management of idiopathic orbital inflammation. Am J

Ophthalmol 2004;138(6):1041–3.

[61] Yuen SJA, Rubin PAD. Idiopathic orbital inflammation; distribution, clinical features and

treatment outcome. Arch Ophthalmol 2003;121(4):491–9.

[62] Mombaerts I, Goldschmeding R, Schlingemann RO, et al. What is orbital pseudotumor?

Surv Ophthalmol 1996;41(1):66–78.

[63] Dziezyc J, Barton CL. Exophthalmia in a cat caused by an eosinophilic infiltrate. Progr Vet

Comp Ophthalmol 1992;2(2):91–3.

[64] Carvounis PE, Mehta AP, Geist CE. Orbital myositis associated with Borrelia burgdorferi

(Lyme disease) infection. Ophthalmol 2004;111(5):1023–8.

[65] Mahdaviani S, Higgins GC, Kerr NC. Orbital pseudotumor in a child with juvenile rheuma-

toid arthritis. J Pediatr Ophthalmol Strabismus 2005;42(3):185–8.

[66] Crotti A. Orbital fatprolapse in cats: two clinical cases. Proceedings. Ann Meeting Eur Coll

Vet Ophthalmol; 2004. p.113.

[67] Boydell P, Paterson S, Pike R. Orbital fat prolapse in the dog. J Small Anim Pract 1996;37:

61–3.

[68] Williams DL, Haggett E. Surgical removal of a canine orbital lipoma. J Small Anim Pract

2006;47:35–7.

[69] Swinger RL, Schmidt KA. Clinical snapshot. Compendium 2007;3:120–3.
[70] Martin CL. Orbital emphysema: a complication of ocular enucleation in the dog. Vet Med

Small Anim Clin 1971;66:986–9.

[71] Wolfer J, Grahn B. Diagnostic ophthalmology. Orbital emphysema. Can Vet J 1995;36(3):

186–7.

401

ORBITAL INFLAMMATORY DISEASE AND PSEUDOTUMOR

Optic Neuritis in Dogs and Cats

Barbara Nell, MD, PhD (Habilitation)

Clinic for Surgery and Ophthalmology, Department for Small Animals and Horses,
Veterinary University of Vienna, Veterinaerplatz 1, 1210 Vienna, Austria

he term ‘‘optic neuritis’’ compromises all diseases of the optic nerve that
cause primary demyelination and usually manifest themselves as a sudden
visual field defect or total loss of vision in one or both eyes. On ophthal-

mologic examination, changes in pupillary light reflexes can be observed, and
in case the optic nerve head is involved (papillitis) it becomes hyperemic, the
edge blurred and indistinct, the veins dilated, and hemorrhage and exudates
may appear in the tissue of the swollen disc. A spread to the retina is possible,
causing neuroretinitis. In retro bulbar neuritis the same profound functional
symptoms occur but without ophthalmoscopic changes. The optic disc swelling
in cases of papillitis must be differentiated from papilledema, which is almost
always bilateral and is not associated with visual deficits, at least not in the
beginning. It is associated with increased intracranial pressure and is primarily
a mechanical, not a vascular phenomenon. Optic nerve fibers are compressed
by elevated cerebrospinal fluid (CSF) pressure in the subarachnoid space of the
intraorbital portion of the optic nerve, which results in reduction of the axoplas-
matic flow and swelling of the axons. In human beings the elevated pressure is
transmitted via the central retinal artery and vein, which do not exist in com-
panion animals. In dogs papilledema is described in association with brain
tumors. Optic nerve atrophy and vision loss occur after weeks, a result of nerve
fiber attrition

Diagnosis is made by clinical presentation, CSF studies, CT or MRI, electro-

encephalogram, and electroretinogram to exclude primary chorio-retinal dis-
ease. Hematologic and blood chemistry tests are usually not very helpful.
Described causes for optic neuritis in men and animals are demyelinating
diseases, infections caused by an intraocular or orbital process or spread from
encephalitis, systemic inflammatory conditions, ischemic, nutritional, toxic,
and cancer related neuropathies, endocrine disturbances, and isolated neuro-
logic symptoms

[2,3]

. Histopathology depends on the etiology and is rarely

accomplished in acute cases. End stage signs—whatever cause—are optic atrophy
and proliferation of astrozytes.

E-mail address: barbara.nell@vu-wien.ac.at

0195-5616/08/$ – see front matter

doi:10.1016/j.cvsm.2007.11.005

Vet Clin Small Anim 38 (2008) 403–415

VETERINARY CLINICS

SMALL ANIMAL PRACTICE

IMMUNE MEDIATED OPTIC NEURITIS IN COMPANION
ANIMALS

As in man, the cause of optic neuritis is often difficult to determine in the living
animal. Neurologic examination, CSF analysis, and laboratory tests are nor-
mal. Optic neuritis affects dogs far more frequently than other animal species.

Dogs

Histiozytic proliferative disorders in the canine central nervous system were
previously referred as ‘‘reticulosis.’’ For the following three histologic differen-
tiations, the old term (for the sake of comparison) is given in brackets, but it not
used further on in the article:

1. Granulomatous meningoencephalitis (GME) (inflammatory reticulosis) is the

most common form

Foci of the cells are found in chronic or granulomatous inflammation
(monocytes, histiocytes, lymphocytes, and plasma cells) with granulo-
cytes and multinucleated giant cells present in varying numbers

Proliferating cells are present primarily as perivascular infiltrations

Good cellular differentiation is seen, with no pleomorphism or polyploidy

Mitotic figures are either absent or observed rarely

Reticulin fiber proliferation networks are present

Both nodular granulomatous (focal) and disseminated (multifocal) forms of the

disease can be observed [4]. Although the disseminated form is usually
encountered, the focal form, which may result from confluence of the lesions,
has been recognized. The focal form of GME my have been previously
included in neoplastic reticulosis [5].

2. Lymphoma and malignant histiocytosis (neoplastic reticulosis) and
3. Microgliomatosis both show a high mitotic index and form solid tumor

masses or distinct parallel rows between nerve bundles, respectively [5].

The underlying cause of GME remains unclear, but an autoimmune disor-

der of delayed type hypersensitivity reaction is suspected, as on immunomor-
phologic studies a heterogeneous population of major histocompatibility
complex class II antigen-positive macrophages and predominantly CD3 antigen
positive lymphocytes were found

. The lesions affect the white matter to

a greater extent than the gray matter and are found throughout the brain
and cervical spine

[4,7–9]

. The distribution pattern suggests that the antigen

would associate with the central white matters; optic nerve involvement would
also support this hypothesis

Optic nerve involvement in GME can be part of the disseminated

[5,7,10–

16]

or the focal form

. Cases might present first with ophthalmologic signs

only; disseminated and focal lesions in the CNS may develop later

. GME

may afflict all breeds and ages but is more common in middle aged, small-breed
dogs

[7,13,17]

. Frankhauser and colleagues

and Braund

noted a predi-

lection for terriers and Russo

, for poodles.

Cerebrospinal fluid examination is extremely variable. Cases might show

a marginal to marked increase in protein or cellularity (mononuclear cells

404

NELL

and macrophages), depending on the potential of exfoliation and on break-
down of the blood-brain barrier

[4,9]

. The general pattern with the cells men-

tioned above is different from the abnormalities commonly seen with viral,
bacterial, or mycotic encephalitis

. MRI signs in the disseminated form

are contrast material enhancement with diffuse infiltration and bad delineation,
but that could not be considered disease-specific

[14,19]

. A definitive ante mor-

tem diagnosis is only possible by CT- or MRI-guided biopsies, as it is done in
human beings. Munana and Luttgen

describe seven cases with mass lesion

that underwent brain biopsies with no influence on mortality.

Therapy consists of 1 mg/kg to 2 mg/kg per day of oral prednisolone until

remission of signs, and a subsequent reduction to 2.5 mg to 5 mg, given on
alternate days

. Besides the treatment with corticosteroids, Adamo and

O’Brian

describe the successful use of 6 mg/kg of cyclosporine every 12

hours in two dogs with optic nerve involvement, with a follow up time of 2
and 5 months. Three dogs with brain and spinal cord lesions received
10 mg/kg of cyclosporine once daily for at least 6 weeks, then dosage was re-
duced to 5 mg/kg daily, after a variable period of glucocorticoid treatment, and
complete resolution of signs was achieved for 3, 5, and 6 months

. Nuhs-

baum and colleagues

chose a combination of prednisone (2 mg/kg, by

mouth, twice a day, slow taper over 4 weeks) and cytosine arabinoside (Cytar-
abine 100 mg per vial, Bedford Laboratories) (50 mg/m

, subcutaneous, twice

a day for 2 days, repeated every 3 weeks) and had a visual dog at the 12 month
follow-up on the cytosine arabinoside only. Zarfoss and colleagues

report

a survival time in 10 dogs with meningoencephalitis of unknown cause, of 46 to
1,025 days on a combination therapy of prednisolone and cytosine arabinoside.
Granger and colleagues

suggest a combination of 0.1 mg/kg prednisolone

with 60 mg/m

lomustine per month (six dogs), or cyclosporin 5 mg/kg twice

a day, tapered to 1 mg/kg twice a day (three dogs), or azathioprin 2 mg/kg once
a day, tapered to 2 mg/kg twice a week (one dog), with a survival time of
233 days in a case series of optic nerve involvement. In a case series study
of the focal and multifocal form and locations within the forebrain, brain,
cerebellum, spinal cord, and eye procarbazine was used in combination with
prednisolone. Procarbazine is a potent monoamine oxidase inhibitor, T cell
specific, and crosses the blood-brain barrier. Median survival time of dogs
treated with 25–50 mg/m

orally every 24 hours was 14 months, and 0.73

months for untreated dogs

. Radiation therapy, with a 6-mV linear acceler-

ator or a cobalt 60 teletherapy unit, and total doses ranging from 40 gray (Gy)
to 49.5 Gy, divided in 2.4 Gy to 4.0 Gy fractions, was successful in the focal
brain form in four of seven dogs

Dogs with disseminated disease without mass lesion have the shortest time

course

[5,7,13]

, and cessation of treatment is in any case associated with prompt

detoriation

[5,7,10,24]

. Twenty-one dogs with focal GME had a mean survival

time of 114 days, and 21 dogs with multifocal GME a mean survival time of 8
days, not taking into account whether they were treated or not

. Given

follow-up times in cases with optic neuritis and successful treatment are

405

OPTIC NEURITIS IN DOGS AND CATS

15 months on alternate-day corticosteroid therapy (dosage not given)

, 12

months on cytosine arabinoside

, 5 months on cyclosporine

, and

a mean of 233 days for 10 dogs on combination therapy

Distemper (paramyxovirus) and tick borne encephalitis (flavivirus) are viral

diseases that manifest themselves on the optic nerve and have the potential to
cause immune mediated reactions.

The paramyxovirus reaches the optic nerve by circulating in the CSF or

through infected mononuclear cells that have crossed the blood-brain barrier.
Often both optic nerves and the whole optic tracts are involved

[25–28]

. Optic

nerve and central nervous system (CNS) involvement without systemic signs
occurs in middle-aged dogs

[28,29]

. Adams and colleagues

call it ‘‘old dog

encephalitis.’’ The initial demyelination starts about 3 weeks after the
infection during the period of immunosuppression

. Depending on the strain

of the virus and the immune status, the dog can die, recover, or develop a chronic
disease that improves and relapses. With these cases an immunopathologic
reaction, 6 to 8 weeks after infection, causes progression of signs

[30,31]

. The

chronic inflammatory demyelination results from interactions between macro-
phages and antiviral antibodies

. Vandevelde and Zubriggen

and Cer-

ruti-Sola and colleagues

detected antimyelin antibodies in the serum of

diseased dogs, but could not prove a correlation to the course of the disease.

Histologic lesions in the acute stage are characterized by ballooning of mye-

lin sheets with vacuolation of the white matter, myelin phagocytosis, and astro-
cytic swelling

. Six to seven weeks following infection, prominent

perivascular infiltration with lymphocytes, plasma cells, and monocytes occurs

[26,28–30]

. The inflammatory reaction in the demyelinating lesions can lead to

progression of tissue damage

. Chronic changes are increases in astrocytic

glia and proliferation of fibroblasts

. Intranuclear

[25,26,28,29]

and intracy-

toplasmic eosinophilic inclusion bodies

[26,28]

in the astrozytes are a good

diagnostic criterion, but immunhistochemical demonstration of the viral anti-
gen proved to be superior

. Diagnosis in the living animal is made by

antibody testing of CSF, as an intrathecal production occurs

. Jozwik and

colleagues

propose nested polymerase chain reaction from white blood

cells as the most sensitive method in subacute and chronic form. CSF protein
might be elevated

Therapy can only be supportive and symptomatic (2-mg/kg prednisone at

a tapering dose

) as there is no treatment for canine distemper. Prognosis

for recovery is very poor

. Sarfaty and colleagues

report a clinical

course of 5 to 44 days in dogs with CNS symptoms without systemic signs.

Measles virus is a morbilli virus like the paramyxovirus virus. Human

patients with optic neuritis, that eventually developed multiple sclerosis
(MS), showed local measles virus antibody response. The pathologic changes
are similar to those in distemper

. Fischer

reports about cases with optic

neuritis and positive measles antibody titres in CSF. Hutchinson and Haire

found that individuals with MS after an episode of optic neuritis have signifi-
cantly higher titres of measles-virus-specific IgG in their serum than patients

406

NELL

with optic neuritis but no MS. In subacute sclerosing panencephalitis, antimea-
sles antibodies are detected in the CSF; cells of the retina and optic nerve con-
tain intranuclear and intracytoplasmatic inclusions and measles virus. The
neuropathic latency in distemper virus is similar to subacute sclerosing panen-
cephalitis. Canine distemper might be a model for measles-associated encepha-
lopathies and have a possible relationship to MS

. Optic neuritis after

rubella-measles vaccination is reported as well

Vandevelde and Zubriggen

state that there is no conclusive evidence in

the animal models for a virus-induced autoimmune disease, which continues to
progress despite complete clearance of the infectious agent, as proposed in MS.
Rather common, as in distemper, is virus persistence in the CNS, which
appears to be the key to the progression of the disease. The persistent virus pre-
cipitates recurrent immune reactions even though the infection load may be
very small and difficult to detect.

Tick-borne encephalitis virus (TBEV) is an arbovirus and belongs to the

family Flaviviridae, genus flavivirus. In central Europe, the vector of this virus
is the tick, Ixodes ricinus, which limits the infectious period because of its seasonal
activity. Only very few publications report clinical cases of TBEV in dogs

[39–

46]

. After an estimated incubation period of 5 to 9 days, clinical symptoms such

as fever, apathy, and neurologic symptoms may occur

. TBEV is mainly

described as a multifocal disease of the CNS, where the forebrain, brain
stem, cerebellum, meninges and spinal cord are involved

. The case re-

ported by Stadtba¨umer and colleagues

had bilateral optic neuritis and lym-

phocytic meningoencephalitis, and course and response to therapy suggest an
immune mediated mechanism besides the immediate effect of the virus.

Four different courses of TBEV infections in the dog (seroconversion but

asymptomatic, peracute, acute, chronic) have been described

. In the major-

ity of cases, the dog shows seroconversion without any clinical symptoms of
TBEV

. In the peracute course, the dog dies within the first 3 to 7 days

because of progression of multifocal neurologic symptoms

. In the acute

course, clinical symptoms can improve and often disappear completely after
1 to 3 weeks. In the chronic course neurologic symptoms improve after 1 to
6 months

[43,46]

. In the early stages of the disease, no antibody titres can be

detected and only pathohistologic examination provides a diagnosis. Charac-
teristic neuropathologic findings are severe meningoencephalomyelitis with
necrosis of neurons and glial cells, neuronophagia, glial nodules, perivascular
cuffs, and diffuse infiltration of leptomeninges. The lesions are most prominent
in the brainstem and cerebellum. Because of the rapid clearance mechanism,
which is known to occur in other forms of flavivirus-encephalitis, only in
a few cases could TBEV antigen be found with a polyclonal antibody

In human beings, macrophages, microglia, and granzyme B-expressing
cytotoxic T cells contribute to tissue destruction

. Because no causal ther-

apy for TBEV exists, only symptomatic therapy can be performed

The dog in the Stadtba¨umer and colleagues case

had elevated serum

(1:800) and cerebrospinal fluid (CSF, 1:200) IgG titres, and the left eye

407

OPTIC NEURITIS IN DOGS AND CATS

regained vision with prednisone (2 mg/kg every 24 hours); the right eye re-
mained blind. Analysis of CSF showed elevated protein and an increased
amount of cells with mononuclear pleocytosis with predominantly lympho-
cytes. When the anti-inflammatory therapy was stopped on day 18, the dog ex-
hibited hyperesthesia in the cerebral region and neck 4 days later. By then,
analysis of the CSF showed no abnormalities, as there would have been in
meningitis. Extrameningeal tissue damage might have been the cause for the
transient hyperesthesia. Because of the relapse, the dog was kept on a dose
of 0.2 mg/kg per day of prednisone. Although previously not described, this
dog had a recurrent TBEV infection within 1 year, with IgG titer in the CSF
(1:800) and in the serum (1:400), suggesting that no efficient immunity had
been developed after the first infection. Because of the rapid clearance mecha-
nism, the high antibody titres represented a second infection. The investigators
concluded that during the first period of TBEV infection, when therapy was
started after 2 weeks of ongoing symptoms, the ophthalmic immune-mediated
inflammatory response had enough time to cause damage to the optic nerve. In
contrast, following the second period of meningitis of only 2-days duration un-
til therapy was initiated, no additional ocular damage occurred

. Six months

later the liquor TBEV titre was negative again and the dog was kept on 25-mg
azathioprin every other day and did fine for almost 3 years, when hind limb
weakness and pain occurred. Before therapy could be initiated again the dog
developed tonic-clonic seizures, an inner body temperature of 44.8

C, and

was euthanized at the local veterinarian. A pathohistologic examination was
not done (Stadtba¨umer, Leschnik, Nell, unpublished data, 2007). As discussed
regarding distemper by Vandefelde and Zubriggen

, the symptoms might

have been caused by virus persistence in the CNS as well, or through virus-
induced autoimmune disease.

Since then the investigators have had two more cases that presented with

visual deficits that were diagnosed with TBEV. On ophthalmic examination,
negative menace response and reduction of pupillary light reflexes, when the
medial parts of the retinas in both eyes were illuminated, were noticed. Suspi-
cion of a lesion involving the chiasm was expressed and proven on CT. The
lesions showed contrast medium uptake, and increased vascularisation was
noted in the region of the brain stem and basal parts of the cerebrum. Liquor
analysis revealed elevated mononuclear cell counts with predominantly lym-
phocytes, and titres were positive for TBEV. One dog was treated with predni-
sone and azathioprin for 5 months, and has been off treatment for a year now,
and the other one was treated with prednisone for 1 month and than switched
to azathioprin for 5 months. The second dog has been doing fine without treat-
ment for 2 years now (Leschnik, Stadtba¨umer, Nell, unpublished data, 2007).

Leiva and colleagues

describe a case of optic neuritis as the single sign

caused by Ehrlichia canis infection. CSF analysis demonstrated reduced glucose
concentration, mild increase in protein concentration and lymphocytes, and
the serologic titre was 1:320. Response to doxycycline 5 mg/kg orally every
12 hours for 2 to 3 months, and oral prednisone (0.5 mg/kg twice a day for

408

NELL

10 days, 0.5 mg/kg per day for 10 days, and 0.5 mg/kg every other day) was
poor, and the dog never regained vision. Canine monocytic ehrlichiosis is asso-
ciated with immune complex formation, deposition in the vascular walls, and
antinuclear body formation, which might have caused the lesions in the optic
nerve as they do in the uvea. This might as well explain the correlation of sever-
ity of ocular sign and serum titres, according to Leiva and colleagues

Possible other etiologies for optic neuritis are the spread of infections from

the eye and orbit

, extraocular or masticatory muscle myositis

, infec-

tions with pseudorabies virus

, Toxoplasma gondii

, Hepatozoon canis

and fungi like cryptococcus

[55–57]

, blastomyces

[58–60]

, and histoplasma

, as well as intoxication with closantel

The diagnosis of idiopathic immune mediated optic neuritis is made by ruling

out other causes. An acute idiopathic optic neuritis that might have been im-
mune-mediated is described by Nafe and Carter

. Fischer and Jones

re-

port about 12 cases. They did not find a cause in 7 of 12 cases, but they may have
done so if all the diagnostic methods were available at that time. The same ap-
plies for the three cases described by Walde and Swoboda

. Four gained par-

tial vision with corticosteroid treatment and might have had an immune
mediated cause. Treatment consists of 2-mg/kg prednisolone daily, divided twice
a day for 10 days to 2 weeks. With signs of improvement the dosage is reduced to
1 mg/kg daily for 2 weeks, followed by a gradual reduction to maintenance
therapy every other day for 1 year

. Fischer and Jones

also started

treatment with 2-mg/kg prednisolone, which they gradually decreased over
3 to 5 weeks. Walde and colleagues

start with a dosage of 10-mg/kg pred-

nisolone, and go on with 2 mg/kg for 4 days and gradually reduce the dosage.
Reduction of dosage before signs abate, or premature withdrawal of corticoste-
roids, may result in exacerbation of symptoms. The course is difficult to predict.
Vision may return within 24 to 48 hours, but prognosis is always guarded
(

)

. In a retrospective study of 50 cases, Davidson and colleagues

had 11 cases of idiopathic optic neuritis, as neither CT scans, nor CSF anal-

ysis, nor serology yielded a diagnosis, which improved on prednisone.

Cats

In the cat no etiologies for immune mediated optic neuritis or an idiopathic
immune mediated optic neuritis are described. Reported causes are spread of
orbital and nasal cavity infections

[65,66]

, feline infectious peritonitis

[67,68]

toxoplasmosis

[69,70]

, mycotic infections like cryptoccosis

[71–73]

, histplasmo-

sis

, systemic hypertension

and lymphoma

[50,72]

IMMUNE MEDIATED OPTIC NEURITIS IN MAN

In man, immune mediated optic neuritis is described in connection with
systemic immune mediated diseases, such as lupus erythematosus

and

atopic dermatitis

, and after systemic infections

. In the demyelinating

neuropathies like multiple sclerosis

, neuromyelitis optica

, and heredi-

tary Leber’s disease

[81]

an immunologic component has long been suspected.

409

OPTIC NEURITIS IN DOGS AND CATS

In 2004, Lennon and colleagues

identified an IgG autoantibody in patients

with neuromyelitis optica that binds selectively to the aquaporin-4 water chan-
nel, a component of the dystroglycan protein complex located in astrocytic foot
processes at the blood-brain barrier that might be the causative agent, as it
serves as a marker for the disease. Subsequently, the marker was found in pa-
tients with opticospinal MS and in some, but not all cases, with classic MS

Isolated neurologic syndromes, such as an acute idiopathic

and autoim-

mune optic neuropathy

, are described. Spoors’

diagnosis is based on

exclusion and the response to corticosteroid treatment. Within the retrospec-
tive study group of Winer and colleagues

of 100 subjects, 67% responded

410

NELL

to treatment and their symptoms might have been caused by an autoimmune
disease.

The development of optic neuritis is a known complication after vaccinations

for influenza

[84]

, anthrax

[85]

, or rubella

[86]

. Based on detection of immune

complexes in the CSF, or retinal or optic nerve autoantibodies, it is assumed
that optic neuritis is caused through immune-mediated demyelinisation

[85,86]

Whether systemic immune mediated disease, isolated neurologic syndrome,

or post vaccination complication, the course leads to demyelisation of the optic
nerve. The discovery of the heterogeneity in demyelinating lesions has
suggested that different mechanisms may be involved in the pathogenesis

[87]

. The failure of CNS regeneration may be in part a result of the presence

of myelin-associated growth inhibitory molecules in MS

LABORATORY ANIMAL MODELS

Experimental optic neuritis was created in adult guinea pigs by immunization
with an isogenic spinal cord emulsion. Fourteen days after inoculation the
animals developed a retrobulbar optic neuritis or a neuroretinits. The histologic
changes were monocuclear cell infiltrations in the brain and retrobulbar part of
the optic nerve, and multiple foci of axial and periaxial demyelination. The
eyes showed papilledema with marked swelling of the nerve fibers of the optic
disc, but no inflammatory cell infiltration in the disc anterior to the lamina cri-
brosa