BASIC NUTRITIONAL INVESTIGATION

Influence of Fiber Fermentability on Nutrient

Digestion in the Dog

Jennifer Silvio, MS, David L. Harmon, PhD, Kathy L. Gross, PhD, and Kyle R. McLeod, PhD

From the Department of Animal Sciences, University of Kentucky, Lexington, Kentucky;

Hill’s Pet Nutrition, Topeka, Kansas; and the USDA, ARS, Beltsville, Maryland, USA

Eight mature dogs (17.2

⫾ 0.2 kg) surgically fitted with ileal T-cannulas were used in a replicated 4-⫻-4

Latin-square-design experiment to evaluate nutrient disappearance at the terminal ileum and through the
digestive tract. Two fiber types, cellulose, a crystalline, slowly fermented fiber, and pectin, a soluble,
rapidly fermented fiber, were fed in different increments, and the effects on nutrient availability were
assessed. Treatments included 1) 100% cellulose, 2) 66% cellulose and 33% pectin, 3) 66% pectin and
33% cellulose, and 4) 100% pectin. Fiber was added at 10% of diet dry matter (DM). Diets were fed at
100% of ME for maintenance and offered at 0730 and 1730 h. All periods were 21 d, which included 3 d
of diet transition and 7 d of adaptation. Daily DM intake was 210

⫾ 5 g. Total tract and large-intestine

DM digestibility increased linearly (P

⬍ 0.01) with increased pectin. These changes in DM digestion were

largely the result of changes in fiber digestion. Fermentation of total dietary fiber in the large intestine
went from less than zero to 39% of ileal flow (linear, P

⬍ 0.01). Total-tract crude-protein digestibility

decreased linearly (P

⬍ 0.01) with increased pectin. This study demonstrated that fiber fermentability

significantly affects digestion in the dog. Increasing fermentable fiber increased the digestion of DM and
energy. However, increased fiber fermentability inversely affects crude protein digestibility. The lower
crude-protein digestibility could be attributed to larger microbial protein excretion as a result of greater
fermentation of pectin versus cellulose.

Nutrition 2000;16:289 –295. ©Elsevier Science Inc. 2000

Key words: canine, fiber, fermentation, digestion, ileum

INTRODUCTION

The effects of fiber on nutrient availability and gastrointestinal
health and function have received considerable attention in recent
years. Much of the research has been performed using the human
as a model; however, the less complex digestive tract of the dog
makes direct comparisons difficult. The effect of different dietary
fiber sources on nutrient digestibility, gastrointestinal motility, and
stool characteristics in healthy dogs consuming adequately bal-
anced diets has been evaluated.

1–5

For optimum health, a good

fiber source should be moderately fermentable for the production
of short-chain fatty acids (SCFA) with a non-fermentable compo-
nent to provide bulk. The majority of previous research has fo-
cused on the effects of only one fiber type.

6

Therefore, the dietary

fiber included did not necessarily contain both soluble and insol-
uble components. An exception was seen in a previous study

3

where graded levels of beet pulp, a moderately fermentable source
containing both soluble and insoluble fiber, were fed to establish
the level of inclusion in the diet necessary for adequate nutrient
digestibility and acceptable stool characteristics.

The ever-increasing roles of dogs as both pets and experimental

models are necessitating further research in the area of both rapidly
fermentable and slowly fermentable fiber inclusions in the diet.

Examples of these fiber types include the rapidly fermented pectin
and the slowly fermented cellulose. The following experiment
focused on both of these fiber types, alone and in combination, and
on their effects on nutrient digestibilities and large-intestinal fer-
mentation characteristics in the dog.

MATERIALS AND METHODS

Dogs

Eight mature, female mongrel dogs with body weights of 17.2

⫾

0.2 kg were used throughout this experiment to evaluate nutrient
disappearance at the terminal ileum and through the digestive tract.
Approximately 20 mo before this experiment, a polyvinyl chloride
T-cannula was surgically fitted 6 –10 cm from the ileal-cecal
junction and exteriorized through the body wall.

7

This allowed for

the sampling of digesta and the determination of intestinal digest-
ibilities with minimal contribution of the post-small-intestinal
microflora.

The animals were located in the Division of Laboratory Animal

Resources at the University of Kentucky and were cared for in
accordance with IACUC protocols. All dogs were individually
housed in stainless-steel metabolism cages (1.2

⫻ 1.8 m), with

steps (1.01

⫻ 0.46 m) covered by raised, holed mats. These cages

were located in an environmentally controlled room at 22°C with
a 14-h:10-h light:dark schedule. Within their cages, the dogs had
unlimited access to nipple waterers and nylon chew toys. While
their cages were cleaned, all dogs were exercised daily in two
groups for approximately 25 min.

Feeding and Treatments

The diets included in this experiment were prepared by the Hills
Science and Technology Center (Topeka, KS, USA). Each diet

This study was approved by the Director of the Kentucky Agricultural
Experiment Station as Paper 99-07-33.

Correspondence to: David L. Harmon, PhD, Department of Animal Sci-
ences, University of Kentucky, 809 W.P. Garrigus Building, Lexington,
KY 40546 – 0215, USA. E-mail: [e-mail]

Date accepted: Dec. 6, 1999.

Nutrition 16:289 –295, 2000

0899-9007/00/$20.00

©Elsevier Science Inc., 2000. Printed in the United States. All rights reserved.

PII S0899-9007(99)00298-1

was formulated in accordance with the Association of American
Feed Control Officials

8

nutrient guide for dogs and balanced to

meet maintenance requirements (Table I). Any variabilities be-
tween diets were a result of the fiber inclusion, which was added
at 10% of diet dry matter (DM). Ten-percent dietary fiber was
chosen to maximize the effects of dietary fiber and maintain an
amount that would not be overly detrimental to energy intake or
stool quality. Two fiber types were combined in different incre-
ments: 100% cellulose, 66% cellulose and 33% pectin, 66% pectin
and 33% cellulose, and 100% pectin. All of the diets included the
addition of chromic oxide (0.20% of total diet DM), which served
as a marker for the determination of digestibility. Diets were
offered at 0730 and 1730 h and fed at 100% of ME for mainte-
nance. All meals were preweighed into stainless-steel bowls. An
exception occurred during the third and fourth periods of the
experiment, when one dog was fed at 125% of ME throughout
because of weight loss. Dogs were allowed 15 min to consume the
twice-daily meals, which was more than adequate for the complete
consumption of the diets. After the allotted time period, bowls
were collected and any orts were recorded. The measured DM
intake of the dogs was 210

⫾ 5 g/d. Throughout each of the four

periods, samples of the diets were pooled daily for analysis.

Sampling

The study was performed in four consecutive 21-d periods. During
the first 3 d of each period, dogs were fed a 1:1 blend of their
current diet and their next experimental diet to avoid any gastric
upset or meal refusal. A 7-d adaptation then followed, during
which the dogs consumed their respective diets. On the first day of

fecal collection, all feces were removed from the cages and dis-
carded before 0730 h. Fecal output from this point on for the next
3 d was collected at feeding and composited for each dog. During
fecal collection, the dogs were fitted with ruffed collars for in-
creasing increments of time between the 0730-h and the 1730-h
feedings to let them become adjusted to wearing the collars.

The ileal sampling period consisted of the 3 d after the fecal

collection. Dogs were fed at 0730 h and fitted with ruffed collars
upon removal of their bowls. These collars restricted the dogs from
removing the plastic collection bags attached to their ileal cannulas
during sampling hours while allowing them to drink from nipple
waterers and water bowls located outside of their cages. Ileal
sampling began at either 0800 h or 0900 h and occurred every
other hour during the time period between the morning and
evening meals, ending at 1700 h. The first 2 d included collecting
during the even hours and collection during the odd hours occurred
on the third day to allow for samples to be composited for all hours
between 0730 h and 1730 h. One-ounce Whirl-Pak collection bags
(Nasco, Fort Atkinson, WI, USA) were placed on the ileal cannu-
las and then removed after 1 h. The bags were observed closely in
case of leakage or the need for replacement when they became too
full. The content of the bags was weighed into tared collection
containers and stored frozen for the remainder of the collection
period.

Analyses

Ileal and fecal samples were both lyophilized. By weighing fecal
samples before and after lyophilization, DM was obtained. Ileal
and fecal samples were ground in a Waring 7010G laboratory
blender (New Hartford, CT, USA) and a Cyclotec 1093 Sample
Mill (Tecator, Hoganas, Sweden), respectively. Diet samples were
ground through a 1-mm screen in a Wiley mill. The dried and
ground samples were then stored in plastic bags at room temper-
ature. DM for all ground samples were obtained by drying at 75°C
under vacuum for 24 h.

Ileal digesta was prepared for SCFA and ammonia analysis

before lyophilization. Digesta was thawed and thoroughly mixed
before 1 g was removed from each pooled sample (representing
individual dogs within each period) and centrifuged with 1 mL
25% metaphosphoric acid. The supernatant fluid was removed,
placed in a vial, and frozen until analysis. For the determination of
SCFA concentrations, samples were analyzed on a Hewlett Pack-
ard 5890 Gas Chromatograph (Avondale, PA, USA) with a 1.8-m

⫻ 4-mm glass column packed with 10% SP-1000/1%H

3

PO

4

on

100/120 Chromosorb W AW (Supelco, Bellefonte, PA, USA). The
remaining contents of the vials were then analyzed for ammonia
concentration with a Cobas Fara II (Roche Diagnostic Systems,
Branchburg, NJ, USA). The ammonia procedure

9

was modified by

replacing phenol with sodium salicylate.

The ileal and fecal samples, along with those of the diet, were

wet ashed

10

and stored in amber-colored glass bottles with screw-

top lids at room temperature. The resulting liquid was analyzed for
chromium by atomic absorption spectroscopy (Unicam 929 Spec-
trometer, Thermo Jarell Ash, Franklin, MA, USA). Diet, ileal, and
fecal samples were analyzed for energy with a 1264 Isoperibol
Bomb Calorimeter (Parr Instrument Co., Moline, IL, USA). Crude
protein content of the samples were obtained by using a LECO
CNS2000 nitrogen analyzer (St. Joseph, MI, USA).

Acid detergent fiber (ADF) was analyzed by using the proce-

dure of Robertson and Van Soest.

11

This procedure allowed for a

measure of primarily the insoluble fiber components, cellulose, in
the diet, ileal, and fecal samples. Total dietary fiber (TDF) was
determined as described previously.

12

TDF is a measure of both

the water-soluble and water-insoluble fiber components. Upon
completion of the procedure, the resulting residue contained both
the soluble pectin and insoluble cellulose consumed throughout
this experiment. The TDF procedure requires correction for ash

TABLE I.

INGREDIENT COMPOSITION OF THE DIETS*

Treatments

Cellulose

66%

Cellulose

66%

Pectin

Pectin

Ingredient

Rice

56.0

56.0

56.0

56.0

Poultry byproduct meal

21.0

21.0

21.0

21.0

Pectin

—

3.33

6.67

10.00

Cellulose

10.00

6.67

3.33

—

Animal fat

8.5

8.5

8.5

8.5

Natural flavor

2.0

2.0

2.0

2.0

Soybean oil

1.0

1.0

1.0

1.0

Potassium chloride

0.70

0.70

0.70

0.70

Salt

0.28

0.28

0.28

0.28

Chromic oxide

0.20

0.20

0.20

0.20

Choline chloride

0.20

0.20

0.20

0.20

Vitamin mix

0.06

0.06

0.06

0.06

Mineral mix

0.04

0.04

0.04

0.04

Ethoxyquin

0.02

0.02

0.02

0.02

Nutrient composition

Protein

21

22

21

22

Fat

14

14

15

14

Nitrogen-free extract

52

54

56

58

Crude fiber

8.5

5.7

3.0

1.7

Neutral detergent fiber

9.5

8.7

5.6

5.6

Total dietary fiber

11.6

12.3

11.6

8.9

Calcium

0.8

0.7

0.8

0.8

Phosphorus

0.6

0.6

0.6

0.6

Chromium

0.15

0.14

0.14

0.15

* Ingredient and nutrient compositions on a dry-matter basis.

FIBER AND DIGESTION IN THE DOG

290

and protein content in the resulting residue. Whereas corrections to
ash content occurred, corrections for protein content did not occur
because nitrogen analysis was unsuccessful.

Calculations and Statistics

Nutrient digestibilities were calculated as described previously

13

with chromium as a marker. The quantity of marker ingested was
adjusted to the percentage recovered in the feces.

The data were analyzed statistically as a replicated Latin-square

design by using the SAS Proc GLM.

14

The experimental unit was

dog. The statistical model consisted of square, periods within
square, dogs within square, and treatments. Treatment responses
were separated by using linear, quadratic, and cubic contrasts.
Treatment responses were considered different when P

⬍ 0.05.

RESULTS AND DISCUSSION

Dry Matter

There was a linear decrease (P

⬍ 0.04) in DM intake (Table II)

with increasing pectin. This difference was small, less than 12 g/d,
and may have little biological significance. Muir et al.

15

fed similar

combinations of pectin and cellulose and reported no changes in
DM intake; however, their fiber was limited to 7.5% of the diet.

No differences (P

⫽ 0.78) were seen in ileal flow (g DM/d), but

fecal DM and fecal output (g DM/d) were highest for the cellulose
diets, decreasing linearly (P

⬍ 0.0001 and P ⬍ 0.0003, respec-

tively) as pectin content increased (Table II). This result can be
related to the water-holding capacity

16

observed for both fiber

types (with pectin exhibiting the greatest water-holding capacity)
and also to the fibers’ respective fermentabilities.

17

Dogs consum-

ing the high-pectin diet had a more frequent, looser stool than did
the dogs on the cellulose diet, which may indicate that additions of
pectin to dog food may not be acceptable by pet owners because of
its effect on stool quality. Because cellulose is only partly digested
in the dog,

2,18

it adds to fecal weight, whereas pectin, the rapidly

fermented fiber, contributes less to fecal output. The DM digest-
ibilities reported in Table II supported enhanced fermentability as
pectin increased. As pectin increased in the diet, digestibilities
were higher in both the large intestine (linear, P

⬍ 0.0001; qua-

dratic, P

⬍ 0.008) and total tract (linear, P ⬍ 0.001), whereas ileal

DM digestibility was unaffected by treatment. Muir et al.

15

also

reported decreased total-tract digestibility with the addition of
cellulose that did not affect ileal DM digestibility.

Starch

Starch intake (Table III) decreased linearly (P

⬍ 0.02) as pectin

increased in the diet; however, the greatest difference was only 13
g/d. These differences were related to the differences in DM
intake. Ileal flow of starch increased (linear, P

⬍ 0.0001) as pectin

increased in the diet. This increased ileal flow of starch was not
accompanied by an increased fecal loss of starch because fecal loss
was unaffected. This observation indicates that increasing the
fermentable fiber, pectin, not only increases the fermentable fiber
presented to the large intestine but also increases starch fermented
in the large intestine. These changes coincided with decreased ileal
starch digestibility (linear, P

⬍ 0.0001) and increased large-

intestinal starch digestibility (linear, P

⬍ 0.0001). Because of the

increased digestibility in the large intestine, total-tract digestibility
was unaffected by treatment and nearly complete, averaging
99.64%. It is unclear why pectin feeding decreased starch diges-
tion in the small intestine. Increasing pectin in the diet may have
decreased intestinal transit time, thereby decreasing time for di-
gestion, as was suggested previously.

5

This may indicate that more

fermentable fibers such as pectin may be best suited for a reduced-
calorie diet. However, the advantages gained from decreased in-
testinal digestion may be more than offset by the increased large-
intestinal fermentation. These differences may also contribute
greatly to the poor stool quality obtained. Pectin may also affect
digesta viscosity, resulting in altered enzyme access to substrate
and slightly lower starch digestion in the small intestine.

Crude Protein

There was a minor decrease in crude-protein intake (P

⬍ 0.05),

largely the result of the differences in DM intake (Table IV). Ileal
flow of crude protein was unaffected by treatment, whereas fecal
loss of crude protein (g/d) increased linearly (P

⬍ 0.009) with

pectin in the diet. No change was seen in ileal protein digestibility
(P

⫽ 0.60); however, large-intestinal crude protein digestibility

decreased curvilinearly (linear, P

⬍ 0.0001; quadratic, P ⬍ 0.003)

with increasing inclusion of pectin in the diet. Total-tract digest-

TABLE II.

INFLUENCE OF FIBER SOURCE ON DRY MATTER (DM) DIGESTIBILITY

Item

Treatments

SEM†

Contrasts*

Celluose

66%

Cellulose

66%

Pectin

Pectin

Linear

Quadratic

Cubic

Body weight (kg)

17.2

17.3

17.2

17.2

0.2

0.64

0.74

0.54

DMI (g/d)

214.2

219.2

202.2

202.9

5.1

0.04

0.69

0.10

Fecal DM (%)

49.4

44.9

38.0

35.0

0.01

0.0001

0.57

0.30

Feces (g DM/d)

40.0

34.5

29.7

23.7

2.7

0.0003

0.93

0.87

Ileal flow (g DM/d)

45.5

39.4

42.5

42.7

4.2

0.78

0.46

0.53

DM digestibility

Ileal (%)

78.9

81.7

79.1

79.1

2.0

0.84

0.49

0.38

Large intestine (%)‡

10.4

12.1

28.4

43.3

2.2

0.0001

0.008

0.13

Total tract (%)

81.3

84.0

85.4

88.3

1.3

0.001

0.94

0.60

* Probability of a greater F value.
† Standard error of the mean, n

⫽ 8.

‡ Percentage of ileal flow.
DM, dry matter; DMI, dry matter intake.

FIBER AND DIGESTION IN THE DOG

291

ibilities of crude protein also decreased linearly (P

⬍ 0.003) with

added pectin. These differences demonstrate the profound influ-
ence that fiber fermentability can have on estimates of total-tract
digestibility. Total-tract apparent-digestibility calculations do not
take into account the nitrogen sink that occurs when fiber is
fermented in the large intestine. It is believed that the higher
fermentative properties of pectin increases the microbial popula-
tion, whereas cellulose, because of the reduced fermentability, has
little impact on fecal N-excretion. This was confirmed in a previ-
ous study

15

where N-digestibility was similar for cellulose and

control diets. Other studies have reported increased fecal nitrogen
excretion when fermentable fibers are fed.

6,15,17

Using rats,

Younes et al.

19

demonstrated that increased fiber fermentation in

the cecum coincides with an increased flux of blood urea nitrogen
to the cecum and an increased fecal nitrogen excretion. These
results were later

20

confirmed with nephrectomized rats. Feeding

fermentable fiber to nephrectomized rats increased fecal nitrogen
excretion, decreased urinary nitrogen excretion, and lowered
plasma urea by 30%. Similar responses have also been described
in humans.

21

Wheat and oat bran were added to the diet of adult

males, and both fiber sources increased fecal excretion; however,
oat bran, which contains more soluble fiber, increased fecal nitro-
gen excretion more than wheat bran.

Energy

Energy intake decreased slightly (linear, P

⫽ 0.08) as pectin

increased (Table V). Ileal energy flow (P

⫽ 0.42) was unaffected

by treatment. However, fecal energy flow decreased linearly (P

⫽

0.003) as pectin increased. Ileal energy digestibility was unaf-
fected by treatment (P

⫽ 0.70), whereas large-intestinal energy

digestibility increased curvilinearly (linear, P

⫽ 0.0001; quadratic,

P

⫽ 0.006) as pectin increased. These changes in large-intestinal

digestion corresponded to a linear increase (P

⫽ 0.007) in total-

tract energy digestibility. Fiber exerts its effects on nutrient digest-
ibilities and stool characteristics in the hind gut, where fermenta-
tion occurs. The increase in digestibility seen in the pectin diets
relates to its higher fermentative capacities when compared with
that of cellulose.

17

Acid Detergent Fiber

The ADF is primarily a measurement of the insoluble component
of fiber, consisting of cellulose in this study. This serves as an
explanation for effects presented in Table VI. It was expected that
ADF intake would increase (linear, P

⬍ 0.0001) with greater

amounts of dietary cellulose. Ileal ADF flow also increased (P

⬍

TABLE III.

INFLUENCE OF FIBER SOURCE ON STARCH DIGESTIBILITY IN DOGS

Item

Treatments

SEM*

Contrasts†

Celluose

66%

Cellulose

66%

Pectin

Pectin

Linear

Quadratic

Cubic

Intake (g/d)

112.2

118.1

106.4

105.3

2.7

0.02

0.22

0.03

Ileal flow (g/d)

1.57

1.41

3.08

4.00

0.39

0.0001

0.18

0.15

Fecal loss (g/d)

0.42

0.34

0.38

0.46

0.07

0.64

0.31

0.80

Starch digestibility

Ileal (%)

98.60

98.78

97.12

96.24

0.35

0.0001

0.14

0.11

Large intestine (%)‡

72.16

76.49

86.42

89.38

1.96

0.0001

0.73

0.17

Total tract (%)

99.63

99.71

99.65

99.58

0.06

0.51

0.27

0.61

* Standard error of the mean, n

⫽ 8.

† Probability of a greater F value.
‡ Percentage of ileal flow.

TABLE IV.

INFLUENCE OF FIBER SOURCE ON CRUDE PROTEIN DIGESTIBILITY IN DOGS

Item

Treatments

SEM*

Contrasts†

Cellulose

66%

Cellulose

66%

Pectin

Pectin

Linear

Quadratic

Cubic

Intake (g/d)

44.9

45.6

42.3

42.6

1.1

0.05

0.86

0.13

Ileal flow (g/d)

14.2

11.5

12.9

14.0

1.3

0.92

0.17

0.48

Fecal loss (g/d)

7.7

8.7

10.5

11.3

1.0

0.009

0.87

0.66

Protein digestibility

Ileal (%)

68.5

74.2

69.8

67.5

3.0

0.60

0.21

0.38

Large intestine (%)‡

43.6

24.4

17.0

18.9

3.1

0.0001

0.003

0.86

Total tract (%)

83.0

80.7

75.3

73.6

2.2

0.003

0.89

0.51

* Standard error of the mean, n

⫽ 8.

† Probability of a greater F value.
‡ Percentage of ileal flow.

FIBER AND DIGESTION IN THE DOG

292

0.0001) with increasing ADF intake. Fecal ADF loss decreased
curvilinearly (linear, P

⬍ 0.0001; quadratic, P ⬍ 0.0009; cubic,

P

⬍ 0.0007, respectively) with added pectin. Ileal, large-intestinal,

and total-tract ADF digestibilities all increased (P

⬍ 0.01) curvi-

linearly with increasing pectin. There was an increase in small-
intestinal, large-intestinal, and total-tract digestibilities with addi-
tions of pectin to the diet. This may be caused by higher microbial
activity resulting from greater fermentability of pectin. However,
ADF does a poor job of measuring the soluble fiber of pectin, as
indicated by the decreasing intake of ADF as pectin increased in
the diet. Because of the decreasing ADF intake, the actual quan-
tities of ADF disappearing in the small intestine (g/d) were similar
for all diets (range

⫽ 2.6–4.0 g/d). This was in sharp contrast to

the amounts of ADF digested in the large intestine, which was
9 g/d for the cellulose diet versus only 1 g/d for the pectin diet.

Total Dietary Fiber

The TDF procedure measures both the insoluble and soluble
components in dietary fiber. Therefore, the values observed in
Table VII do a better job of accounting for both pectin and
cellulose. A decrease (linear, P

⬍ 0.0003; quadratic, P ⬍ 0.0001;

cubic, P

⬍ 0.002) in TDF intake occurred with increased pectin in

the diet, but this statistical decrease occurred largely because of a
lower value for the 66% pectin treatment. This difference was the
result of a lower TDF content for this diet and lower DM intakes.
No effect (P

⫽ 0.21) was observed for ileal TDF flow (g/d);

however, fecal TDF loss decreased (linear, P

⬍ 0.0001) with

added pectin. This could be attributed to the higher fermentability
of pectin. Fecal TDF loss decreased with higher pectin in the diet
because of pectin’s greater disappearance in the large intestine. We
saw no differences (P

⫽ 0.82) in ileal TDF digestibility; however,

both large-intestinal and total-tract TDF digestibilities decreased
linearly (P

⬍ 0.0001) with increasing pectin additions to the diet.

Ammonia and SCFA

Analysis of collected ileal samples showed that ammonia concen-
trations (Table VIII) were highest for the cellulose diet and de-
creased linearly (P

⬍ 0.002) with increasing pectin. Because these

are only concentrations in ileal digesta, they have limited interpre-
tation with regard to protein degradation and ammonia use. The
increased concentrations of ammonia may be indicative of more
protein being fermented in the cellulose diets because the cellulose
is less fermentable or of an enhanced rate of ammonia use in the
pectin diets because of the greater fermentability. Ileal protein

TABLE V.

INFLUENCE OF FIBER SOURCE ON ENERGY DIGESTIBILITY IN DOGS

Item

Treatments

SEM*

Contrasts†

Cellulose

66%

Cellulose

66%

Pectin

Pectin

Linear

Quadratic

Cubic

Intake (kcal/d)

1021

1055

974

980

24.5

0.08

0.58

0.08

Ileal flow (kcal/d)

189

157

168

166

15.6

0.42

0.34

0.42

Fecal loss (kcal/d)

161

141

126

106

11.5

0.003

1.0

0.83

Energy digestibility

Ileal (%)

81.6

84.9

82.8

83.1

1.6

0.70

0.35

0.27

Large intestine (%)‡

13.0

9.9

23.7

36.0

2.5

0.0001

0.006

0.11

Total tract (%)

84.3

86.5

87.1

89.2

1.1

0.007

0.95

0.56

* Standard error of the mean, n

⫽ 8.

† Probability of a greater F value.
‡ Percentage of ileal flow.

TABLE VI.

INFLUENCE OF FIBER SOURCE ON ACID DETERGENT FIBER (ADF) DIGESTIBILITY IN DOGS

Item

Treatments

SEM*

Contrasts†

Cellulose

66%

Cellulose

66%

Pectin

Pectin

Linear

Quadratic

Cubic

Intake (g/d)

24.9

19.8

10.8

3.9

0.5

0.0001

0.10

0.03

Ileal flow (g/d)

21.7

16.6

6.5

1.2

0.4

0.0001

0.78

0.0001

Fecal loss (g/d)

12.7

8.2

2.2

0.21

0.3

0.0001

0.0009

0.0007

ADF digestibility

Ileal (%)

12.4

14.6

38.6

66.0

2.57

0.0001

0.0001

0.13

Large intestine (%)‡

41.6

50.4

65.9

84.5

0.73

0.0001

0.0001

0.27

Total tract (%)

48.8

57.8

78.8

94.1

1.18

0.0001

0.01

0.004

* Standard error of the mean, n

⫽ 8.

† Probability of a greater F value.
‡ Percentage of ileal flow.

FIBER AND DIGESTION IN THE DOG

293

digestibility was unaffected by fiber type (Table IV); thus, if fiber
type affected protein fermentation, the effect must have been
small.

The results for total ileal SCFA concentration (Table VIII)

show that higher concentrations are present in the cellulose diets
(linear, P

⬍ 0.0001; quadratic, P ⬍ 0.002). This would indicate

that more fermentation of the cellulose-containing diets was oc-
curring at the terminal ileum. However, concentrations could be
misleading because these are expressed as per gram of ileal di-
gesta. Differences in ileal DM content and perhaps fiber water-
holding capacity could influence these results, but these values are
unavailable. Pectin diets have a greater water-holding capacity

16

and should have had a lower ileal DM content. The differences in
ileal SCFA concentrations could also have been the result of
greater fermentation of the cellulose, perhaps a result of pectin
decreasing substrate degradation in the small intestine, as was seen
for starch.

The proportions of SCFA produced indicate differences in the

substrates being fermented. Ileal acetate proportion increased (lin-
ear, P

⬍ 0.004; quadratic, P ⬍ 0.02) and propionate proportion

decreased (linear, P

⬍ 0.0002) as pectin increased in the diet. The

proportions of isobutyrate and isovalerate in the ileal samples
decreased (linear and quadratic, P

⬍ 0.02) as pectin increased.

These values coincide with the higher ammonia concentrations for
the cellulose diets, indicative of greater protein fermentation as
cellulose increased. Butyrate and valerate proportions were unaf-
fected by treatment.

CONCLUSION

Large-intestinal fermentation has a significant impact on total
digestibility in the dog. Altering fiber type and subsequent fer-
mentability follows the trend that the more fermentable the fiber,
the higher the digestibilty. This can be observed in most nutrient
digestibilities described in this study. An exception occurred in
apparent nitrogen digestibility where there was an inverse effect.
The increased excretion of colonic microbes when feeding a more
fermentable fiber results in lower nitrogen digestibility. The
present results imply that combinations of pectin and cellulose,

TABLE VII.

INFLUENCE OF FIBER SOURCE ON TOTAL DIETARY FIBER (TDF) DIGESTIBILITY IN DOGS

Item

Treatments

SEM*

Contrasts†

Cellulose

66%

Cellulose

66%

Pectin

Pectin

Linear

Quadratic

Cubic

Intake (g/d)

53.2

49.3

41.1

48.2

1.2

0.0003

0.0001

0.002

Ileal flow (g/d)

30.2

27.6

26.7

25.4

2.7

0.21

0.80

0.86

Fecal loss (g/d)

33.2

27.2

20.9

14.9

2.0

0.0001

0.99

0.95

TDF digestibility

Ileal (%)

43.3

43.0

35.6

47.8

6.0

0.82

0.31

0.33

Large intestine (%)‡

⫺10.4

0.41

18.8

38.7

3.4

0.0001

0.19

0.69

Total tract (%)

37.7

44.1

49.4

69.0

4.1

0.0001

0.12

0.42

* Standard error of the mean, n

⫽ 8.

† Probability of a greater F value.
‡ Percentage of ileal flow.

TABLE VIII.

INFLUENCE OF FIBER SOURCE ON ILEAL AMMONIA AND SHORT CHAIN FATTY ACID (SCFA) CONCENTRATIONS

Item

Treatments

SEM*

Contrasts†

Cellulose

66%

Cellulose

66%

Pectin

Pectin

Linear

Quadratic

Cubic

Ileal ammonia (

␮mol/g)‡

34.0

32.3

26.6

29.0

1.3

0.002

0.12

0.05

Ileal SCFA (mol/100 mol)

Acetate

72.8

77.7

80.2

78.5

1.3

0.004

0.02

0.75

Propionate

15.6

13.3

12.5

11.8

0.6

0.0002

0.20

0.61

Isobutyrate

2.0

.05

ND

ND

0.1

0.0001

0.0001

0.57

Butyrate

7.2

6.9

14.9

9.1

3.8

0.43

0.49

0.21

Isovalerate

2.2

1.6

0.2

ND

0.1

0.0001

0.02

0.0002

Valerate

0.2

ND

ND

0.6

0.2

0.25

0.10

0.70

Total (

␮mol/g)‡

109.0

74.2

60.4

64.5

5.4

0.0001

0.002

0.90

* Standard error of the mean, n

⫽ 8.

† Probability of a greater F value.
‡ Concentrations are per gram ileal digesta.
ND, Non-detectable concentration.

FIBER AND DIGESTION IN THE DOG

294

considered fermentable and non-fermentable, respectively, can be
used as fiber sources by the dog without any extreme reductions in
nutrient availability.

REFERENCES

1. Bueno L, Praddaude F, Fioramonti J, Ruckebusch Y. Effect of dietary fiber on

gastrointestinal motility and jejunal transit time in dogs. Gastroenterology 1981;
80:701

2. Burrows CF, Kronfeld DS, Banta CA, Merritt AM. Effects of fiber on digest-

ibility and transit time in dogs. J Nutr 1982;112:1726

3. Fahey GC Jr, Merchen NR, Corbin JE, et al. Dietary fiber for dogs: I. Effects of

graded levels of dietary beet pulp on nutrient intake digestibility, metabolizable
energy and digesta mean retention time. J Anim Sci 1990;68:4221

4. Fahey GC Jr, Merchen NR, Corbin JE, et al. Dietary fiber for dogs: III. Effects

of beet pulp and oat fiber additions to dog diets on nutrient intake, digestibility,
metabolizable energy, and digesta mean retention time. J Anim Sci 1992;70:1169

5. Lewis LD, Magerkurth JH, Roudebush P, et al. Stool characteristics, gastroin-

testinal transit time and nutrient digestibility in dogs fed different fiber sources.
J Nutr 1994;124:2716S

6. Harmon DL, Walker JA, Silvio JM, Jamikorm AM, Gross KL. Nutrient digest-

ibility in dogs fed fiber-containing diets. Vet Clin Nutr 1999;6:6

7. Walker JA, Harmon DL, Gross KL, Collings GF. Evaluation of nutrient utiliza-

tion in the canine using the ileal cannulation technique. J Nutr 1994;124:S2672

8. American Association of Feed Control Officials. Petfood regulations. Atlanta:

AAFCO, 1994

9. Broderick GA, Kang JH. Automated simultaneous determination of ammonia and

total amino acids in ruminal fluid and in vitro media. J Dairy Sci 1980;63:64

10. Kimura FT, Miller VL. Improved determination of chromic oxide in cow feed

and feces. J Agri Food Chem 1957;5:216

11. Robertson JB, Van Soest PJ. The detergent system of analysis and its application

to human food. In: James WPT, Theander O, eds. The analysis of dietary fiber.
New York: Marcel Dekker, 1981:123

12. Prosky L, Asp N, Furda I, et al. Determination of total dietary fiber in foods and

food products: collaborative study. J Assoc Anal Chem 1985;68:677

13. Merchen NR. Digestion, absorption and excretion in ruminants. In: Church DC,

ed. The ruminant animal: digestive physiology and nutrition. Englewood Cliffs,
NJ: Prentice-Hall, 1988:188

14. SAS Institute Inc. SAS language guide for personal computers (release 6.03).

Cary, NC: SAS Institute Inc, 1988

15. Muir HE, Murray SM, Fahey GC Jr, Merchen NR, Reinhart GA. Nutrient

digestion by ileal cannulated dogs as affected by dietary fibers with various
fermentation characteristics. J Anim Sci 1996;74:1641

16. Stephen AM, Cummings JH. Water-holding capacity by dietary fiber in vitro and

its relationship to faecal output in man. Gut 1979;20:722

17. Sunvold GD, Fahey GC Jr, Merchen NR, et al. Dietary fiber for dogs: IV. In vitro

fermentation of selected fiber sources by dog fecal inoculum and in vivo digestion
and metabolism of fiber-supplemented diets. J Anim Sci 1995;73:1099

18. Moore ML, Fottler ML, Fahey GC Jr, Corbin JE. Utilization of corn-soybean

meal-substituted diets by dogs. J Anim Sci 1980;50:892

19. Younes H, Garleb KA, Behr S, Re´me´sy C, Demigne´ C. Fermentable fibers or

oligosaccharides reduce urinary nitrogen excretion by increasing urea disposal in
the rat cecum. J Nutr 1995;125:1010

20. Younes H, Re´me´sy C, Behr S, Demigne´ C. Fermentable carbohydrate exerts a

urea-lowering effect in normal and nephrectomized rats. Am J Physiol 1997;272:
G515

21. Chen HL, Haack VS, Janecky CW, Vollendorf NW, Marlett JA. Mechanisms by

which wheat bran and oat bran increase stool weight in humans. Am J Clin Nutr
1998;68:711

FIBER AND DIGESTION IN THE DOG

295