You are on page 1of 7

J Vet Intern Med 2001;15:2632

Subnormal Concentrations of Serum Cobalamin (Vitamin B 1 2 ) in

Cats with Gastrointestinal Disease
Kenneth W. Simpson, John Fyfe, Angelyn Cornetta, Amy Sachs, Dalit Strauss-Ayali, Stephen V. Lamb, and
Thomas J. Reimers

The present study sought to determine the spectrum of diseases associated with subnormal concentrations of serum cobalamin in
cats undergoing investigation of suspected gastrointestinal problems. The solid-phase boil radioassay (RA) for cobalamin employed
in the present study was immunologically specific, precise, and accurate, with a sensitivity of 15 pg/mL. The RA yielded results
that strongly correlated with those obtained by bioassay (Spearmann rho 5 .805; P , .0001), although the absolute values were
lower for the RA. Forty-nine of 80 serum samples submitted during the period of January 1996January 1998 had cobalamin
concentrations below the reference range for healthy cats (range 9002,800 pg/mL; mean 6 SD, 1,775 6 535 pg/mL; n 5 33).
Cats with subnormal cobalamin concentrations (mean 6 SD; 384 6 272 pg/mL, range 3883 pg/mL) were middle-aged or older
and were presented for weight loss, diarrhea, vomiting, anorexia, and thickened intestines. Definitive diagnoses in 22 cats included
inflammatory bowel disease (IBD), intestinal lymphoma, cholangiohepatitis or cholangits, and pancreatic inflammation. Serum
concentrations of cobalamin were particularly low in cats with intestinal lymphoma, three-fifths of whom also had subnormal
serum concentrations of folate (,9 ng/mL). The simultaneous presence of disease in the intestines, pancreas, or hepatobiliary
system in many cats made it difficult to determine the cause of subnormal cobalamin concentrations. The circulating half-life of
parenteral cyanocobalamin was shorter in 2 cats with IBD (5 days) than in 4 healthy cats (12.75 days). The presence of subnormal
serum concentrations of cobalamin in 49 of 80 cats evaluated suggests that the measurement of serum cobalamin may be a useful
indirect indicator of enteric or pancreatic disease in cats. The rapid depletion of circulating cobalamin in cats suggests that cats
may be highly susceptible to cobalamin deficiency. However, the relationship of subnormal serum cobalamin concentrations to
cobalamin deficiency and the effect of cobalamin deficiency on cats remain to be determined.
Key words: Inflammatory bowel disease; Intestinal lymphoma; Folate; Methylmalonic acid; Pancreatic disease; Trypsinlike

D etermination of serum cobalamin concentration has

been useful in identifying and characterizing intesti-
nal and pancreatic disease in dogs.14 Cobalamin homeosta-
border membrane of ileal enterocytes.13,14 Cobalamin is then
transcytosed to the portal bloodstream and binds to a pro-
tein called transcobalamin 2, which mediates cobalamin ab-
sis is a complex, multistep process that involves participa- sorption by target cells.15,16 A portion of cobalamin taken
tion of the stomach, pancreas, intestines, and liver.5 After up by hepatocytes is rapidly (within an hour in the dog) re-
food is ingested, cobalamin is released from food in the excreted in bile bound to haptocorrin. Cobalamin of hepa-
stomach. It is then bound to a nonspecific cobalamin-bind- tobiliary origin, in common with dietary-derived cobala-
ing protein of salivary and gastric origin called haptocorrin. min, undergoes transfer to IF and receptor-mediated ab-
Intrinsic factor (IF), a cobalamin-binding protein that pro- sorption, thus establishing enterohepatic recirculation of the
motes cobalamin absorption in the ileum, is produced by vitamin.17
the stomach and pancreas in dogs68 and by the pancreas, Low serum cobalamin concentrations in dogs have been
but not the stomach, in the cat.9 The affinity of cobalamin associated with exocrine pancreatic insufficiency (EPI), se-
for haptocorrin is higher at acid pH than for IF, so most is vere intestinal disease, and putatively small intestinal bac-
bound to haptocorrin in the stomach.10 Upon entering the terial overgrowth.14,18 When low serum cobalamin is de-
duodenum, haptocorrin is degraded by pancreatic proteases, tected and EPI and small intestinal bacterial overgrowth
and cobalamin is transferred from haptocorrin to IF, a pro- have been excluded, localization of the problem to the il-
cess facilitated by the high affinity of IF for cobalamin at eum can be inferred.3 Selective cobalamin malabsorption
neutral pH.11,12 Cobalamin-IF complexes traverse the intes- and cobalamin deficiency have been recognized in Giant
tine until they bind to specific receptors (previously called Schnauzers with defective localization of the ileal cobala-
intrinsic factor cobalamin receptor, but recently dubbed minintrinsic factor receptor.13,19,20 Cobalamin is an essen-
cubilin) located in the microvillus pits of the apical brush- tial cofactor for the activity of methylmalonyl-CoA mutase
and methionine synthase.21 Reduced activity of these 2 en-
From the Departments of Clinical Science (Simpson, Cornetta, zymes causes the biochemical signatures of cobalamin de-
Sachs, Strauss-Ayali) and Population Medicine and Diagnostic Sci- ficiency: methylmalonicacidemia/methylmalonicaciduria
ences (Lamb, Reimers), College of Veterinary Medicine, Cornell Uni- and homocystinemia/homocystinuria, respectively. Affected
versity, Ithaca, NY; and the Department of Microbiology (Fyfe), Col- Schnauzers have inappetence, failure to thrive, anemia, leu-
lege of Veterinary Medicine, Michigan State University, East Lansing, kopenia, and methylmalonic aciduria. These are completely
MI. reversed by the parenteral administration of cobala-
Reprint requests: Kenneth W. Simpson, BVM&S, PhD, Department min.13,19,20
of Clinical Sciences, College of Veterinary Medicine, Cornell Univer-
Cobalamin deficiency has rarely been reported in the
sity, Ithaca, NY 14853; e-mail:
Submitted March 3, 2000; Revised June 30, 2000; Accepted August
cat.22,23 The use of serum cobalamin concentration as an
4, 2000. indirect indicator of enteric or pancreatic disease has not
Copyright q 2001 by the American College of Veterinary Internal been critically evaluated in cats, though preliminary ob-
Medicine servations suggest they are subnormal in cats with EPI.24
0891-6640/01/1501-0003/$3.00/0 The present study sought to determine the spectrum of
Subnormal Cobalamin in Cats 27

diseases associated with subnormal concentrations of se-

rum cobalamin in cats undergoing investigation of sus-
pected gastrointestinal disease.

Materials and Methods

Validation of a Radioassay to Measure Serum
Cobalamin in Cats
Serum samples were collected from 18 clinically healthy male col-
ony cats (aged 16 months) that were fed a commercial dieta and were
being bled for serological testing. Blood was collected after an over-
night fast, and serum was separated and stored frozen at 2308C until
Serum cobalamin concentrations were evaluated by radioassay (RA)
and bioassay. Radioassay was performed with a kitb that simultaneous-
ly determines serum cobalamin and folate concentrations. Bioassay of
serum cobalamin was performed with protozoa.25,c
The RA for cobalamin and folate was evaluated for immunological
specificity by serially diluting 4 pools of combined feline serum: 1: 1,
1:2, 1:4, and 1 :8. Intra-assay precision was assessed by running 4
feline pools at least 9 times within the same assay. Interassay precision Fig 1. Comparison of a bioassay and radioassay for the determina-
was evaluated by assaying 6 pooled specimens for at least 5 separate tion of serum cobalamin in cats.
assays over a 3-week period. Accuracy was verified by adding known
quantities of cobalamin and folic acid to feline samples.
The mean recovery of folic acid added to feline specimens (n 5 3)
ranging from 2 to 20 ng was 99%. The mean recovery when including
Serum Cobalamin
additions to kit zero matrix was 103% (n 5 5). This indicates that the
The mean percentage observed/expected concentration for the 4 kit is accurate in determining cobalamin concentrations in feline sam-
pools of serum (ranging from 535 to 1,070 pg/mL) serially diluted in ples. The sensitivity of the assay, as defined as the concentration at
kit zero matrix was 101%, with values ranging from 80 to 981 pg/ 95% B/B0, is approximately 0.1 ng/mL. Serum folate concentrations
mL. The mean coefficient of variation (CV) for the corrected results in 18 colony cats and 15 clinically healthy pet cats ranged from 9.4
for each dilution was 0.08, with a range from 0.03 to 0.16. These to 30 ng/mL (mean 6 SD; 20.5 6 6.8 ng/mL).
results indicate linearity with the standard curve. Three of the 4 pools
were tested 12 times within the same assay. The mean intra-assay CV Evaluation of Serum Cobalamin in Sick Cats
was 0.11 (range 0.090.14).
The concentration range for the 6 pools used to test interassay pre- The records of the Diagnostic Laboratory at Cornell University were
cision was 564 to 1,010 pg/mL. Four pools were tested in 5 assays, searched to identify cats that had serum cobalamin concentrations
whereas the other 2 were tested in 6 and 8 separate assays. The mean measured by RA during the period from January 1996 to January
CV was 0.08 (range 0.040.11). 1998. The clinical records of cats with subnormal concentrations of
The mean recovery of cobalamin added to feline specimens (n 5 cobalamin (,900 pg/mL) were examined to determine the age, gender,
5) ranging from 200 to 4,000 pg was 111%. The mean recovery when breed, historical complaints, and physical findings. The records of cats
including additions to kit zero matrix was 102% (n 5 8). This indi- that had undergone a thorough diagnostic investigation that included
cates the kit is accurate in determining cobalamin concentrations in clinical biochemistry and CBC, abdominal ultrasound, measurement
feline samples. of circulating trypsinlike activity,d and histopathological evaluation
There was a strong correlation (Spearmann rho 5 .805; P , .0001) were further evaluated to identify specific diseases associated with low
between the results of bioassay and RA (Fig 1), although the relative serum cobalamin.
values (mean 6 SD) were lower for the RA (2,030 6 646 pg/mL)
than the bioassay (3,460 6 1,430 pg/mL). The sensitivity of the assay, Evaluation of the Half-Life of Exogenous Cobalamin
defined as the concentration at which the antibody is 95% saturated in Sick Cats
with labeled cobalamin (B/B0), is approximately 15 pg/mL. Serum
cobalamin concentrations in 18 colony cats and 15 clinically healthy The half-life of exogenous cobalamin was determined in 2 cats with
pet cats ranged from 9002,800 pg/mL (mean 6 SD; 1,780 6 535 chronic diarrhea and weight loss (cat 1, 9 years old, spayed female;
pg/mL). cat 9, 12 years old, spayed female) associated with inflammatory bow-
el disease (IBD) and high circulating trypsinlike immunoreactivity
(TLI), and 4 clinically healthy 16-month-old male cats bled for a se-
Serum Folate rological study. Control cats were bled at 0, 7, 17, 21, 29, and 41 days
The mean percentage observed/expected concentration for the 4 se- after the injection of 1 mg of cyanocobalamin subcutaneously. The
rum pools (ranging from 7.4 to 16.6 ng/mL) serially diluted in kit zero sick cats were administered 1 mg of cyanocobalamin subcutaneously,
matrix was 99%, with values ranging from 0.64 to 13.8 ng/mL. The and serum samples were obtained at 0, 7, 14, and 28 days (cat 1) and
mean CV for the corrected results for each dilution was 0.06 (range at 0, 3, 11, and 17 days (cat 9). Cat 1 was also re-evaluated after 7
0.050.09). These results indicate linearity with the standard curve. months of prednisolone treatment (doses ranging from 5 mg PO q12h
Three of these pools were assayed 12 times within the same assay. to 5 mg PO q24h), during which time the cats weight had increased
The mean intra-assay CV was 0.04 (range 0.030.05). from 3.5 to 4.4 kg and diarrhea was in remission. Circulating TLI in
The concentration range for the 6 pools used for interassay precision this cat was persistently high over this 7-month period: 128, 130, 115,
was 7.0 to 15.5 ng/mL. Three pools were tested in 5 assays, whereas 156, 176, and 129 mg/L at 0, 28, 53, 117, 222, and 243 days after
the other 3 were tested in 6, 7, and 9 assays. The mean CV was 0.11 initial evaluation, respectively.
(range 0.060.14). Serum cobalamin was determined by RA in all cats and additionally
28 Simpson et al

Table 1. Clinical signs and physical findings in 49 cats dependent diabetes mellitus. Serum thyroxine was subnor-
with subnormal concentrations of serum cobalamin.a mal in 8 of 20 cats. One of these cats had a previous thy-
roidectomy and was suffering from lymphoma and active
Finding 49 Cats (%) 22 Cats (%)
pancreatitis. The other cats were diagnosed as having IBD
Weight loss 34 (69) 15 (68) (3), lymphoma (1), lymphoma and lymphocytic pancreatic
Diarrhea 21 (43) 12 (55) infiltrates (1), and gall bladder carcinoma and cholangio-
Anorexia 21 (43) 11 (50) hepatitis (1); 1 was undiagnosed. Serum concentrations of
Vomiting 19 (39) 10 (45)
folate were subnormal in 3 cats with gastrointestinal lym-
Thinness 12 (24) 10 (45)
phoma. Eight cats had high serum concentrations of folate
Lethargy 15 (31) 8 (36)
Thick gut 17 (35) 8 (36) (5 IBD, 2 hepatobiliary disease, 1 hepatobiliary disease and
Polyuria or polydipsia 13 (26) 6 (27) intestinal disease). Both cats with subnormal concentrations
Depression 7 (14) 4 (18) of TLI had normal concentrations of folate. All cats eval-
Dehydration 13 (26) 4 (18) uated tested negative for feline leukemia virus and feline
Icterus 5 (10) 2 (9) immunodeficiency virus.
Overweight 6 (12) 2 (9) The results of abdominal ultrasonography, endoscopic bi-
Hepatomegaly 10 (10) 1 (5) opsy, or exploratory laparotomy and serum TLI are detailed
Polyphagia 2 (4) 0 in Table 3. The most frequent histological diagnoses were
22 cats had a thorough diagnostic investigation (see Table 3). IBD (12), intestinal lymphoma (6), cholangiohepatitis or
cholangitis (5), and pancreatic inflammation (3). Carcinoma
of the gall bladder (1) or liver and intestine (1) and lym-
by bioassay in healthy cats. The half-life of circulating cobalamin was phoma or lymphocytic infiltration of the pancreas (2) were
determined by measuring the time taken for serum cobalamin concen-
also documented. Many cats had simultaneous abnormali-
trations to drop by 50% between 7 and 21 days after parenteral ad-
ties in multiple organ systems. For example, of the 5 cats
ministration of cyanocobalamin.
To determine if subnormal cobalamin concentration was associated with cholangiohepatitis or cholangitis, 3 also had diabetes
with cobalamin deficiency at the tissue level,21 methylmalonic acid was mellitus, IBD, or both, and 4 had pancreatic abnormalities
measured in the urine of cat 1 by gas-liquid chromatography as pre- (on histology or ultrasound; Table 3).
viously described,13 before and after parenteral cobalamin administra- The most frequent ultrasonographic abnormalities were
tion. abnormal pancreatic size or echogenicity, thickened intes-
tines, mesenteric lymphadenopathy, and abnormalities of
Results the hepatobiliary system. In cats with ultrasonographic ab-
normalities of the pancreas, concomitant ultrasonographic
Clinical Findings in Cats with Subnormal
abnormalities of the hepatobiliary system or intestines were
Concentrations of Serum Cobalamin
observed in 8 of 10 and 4 of 10 cats, respectively.
Forty-nine of 80 serum samples submitted for cobalamin A number of tests were employed to evaluate pancreatic
determination during the period of January 1996January structure and function (Table 4). Serum TLI was high in 14
1998 had serum concentrations ,900 pg/mL (mean 6 SD; of 22 cats, suggesting pancreatitis, and decreased in 2 of
384 6 272 pg/mL; range, 3883 pg/mL). Review of the 22 cats, suggesting exocrine pancreatic insufficiency (Ta-
medical records of those 49 cats indicated there were 27 bles 2, 4). Ultrasonographic abnormalities of the pancreas
castrated males, 18 spayed females, 3 females, and 1 male were observed in 6 of 14 cats with increased TLI and 1 of
aged 11.25 6 4.2 years (mean 6 SD; range, 0.518 years). 2 cats with decreased TLI. Pancreatic biopsy was not per-
Thirty-two cats were domestic short hairs, 4 were domestic formed in the 2 cats with subnormal TLI to confirm the
long hairs, 3 were Himalayans, 3 were Abyssinians, 3 were presence of EPI. Of the 5 cats with histologically confirmed
Burmese, 2 were Siamese, 1 was a Persian, and 1 was an pancreatic inflammation, 5 had an abnormal ultrasono-
American short hair. The majority of cats were presented graphic appearance, and 4 had high concentrations of TLI.
for investigation of weight loss, anorexia, diarrhea, and One cat with active pancreatitis had a normal TLI concen-
vomiting (Table 1). Thickened intestines were palpated in tration. Three cats with normal pancreatic histology and
17 of 49 cats. ultrasound had increased serum TLI.
The clinical records of 22 of 49 cats that included the
assessment of CBC, biochemistry, serum TLI, abdominal Evaluation of the Half-Life of Exogenous Cobalamin
ultrasonography, and intestinal biopsy were examined in in Cats
more detail. In this group of 22 cats, the serum cobalamin
concentration (mean 6 SD; 372 6 284 pg/mL; range, 34 The mean concentration of serum cobalamin in 4 healthy
828), age (11.4 6 3.8 years), gender (13 castrated males, cats remained over 1,000 pg/mL 41 days after the parenteral
7 spayed females, 1 female, and 1 male), breed, and phys- administration of cyanocobalamin (Fig 2). It took 11, 12,
ical and historical features were broadly similar to the group 13, and 14 days for cobalamin concentrations to drop by
of 49 cats (Table 1). Icterus was present in 2 cats. Hema- 50% (the half-life) in those healthy cats that were measured
tological abnormalities included lymphopenia, erythrocyte between 7 and 21 days after parenteral cyanocobalamin. In
macrocytosis, monocytosis, and a left shift in neutrophils contrast, serum concentrations of cobalamin in the 2 sick
(Table 2). Elevated AST and ALT, hyperglycemia, hypo- cats were subnormal (,900 pg/mL) at 7 and 11 days after
albuminemia, and hypochloremia were the most common supplementation, with estimated half-lives of 4.5 and 5.5
biochemical abnormalities (Table 2). Four cats had insulin- days.
Subnormal Cobalamin in Cats 29

Table 2. Clinicopathological findings in cats with subnormal concentrations of serum cobalamin.

No. of Cats Outside Range
Variable Reference Range Below (%) Above (%)
Hematocrit (%) 2545 2/22 (9)
Hemoglobin (g/dL) 8.415 2/22 (9)
Red blood cells (millions per mL) 5.510.3 2/22 (9)
MCV (fL) 4151 6/22 (27)
MCH (pg) 1318 1/22 (5)
MCHC (g/dL) 3236 3/22 (13)
RDW (%) 14.820 5/22 (23)
White blood cells (thousands per mL) 6.121.1 4/22 (18)
Neutrophils (thousands per mL) 2.613.6 4/22 (18)
Band neutrophils (thousands per mL) 00 5/22 (23)
Lymphocytes (thousands per mL) 1.39.1 10/22 (45) 1/22 (5)
Monocytes (thousands per mL) 00.7 9/22 (41)
Eosinophils (thousands per mL) 0.24.3 4/22 (18)
Basophils (thousands per mL) 00.2 2/22 (9)
Platelets (thousands per mL) 215760
Sodium (mEq/L) 148161 3/22 (13) 1/22 (5)
Potassium (mEq/L) 3.35.2 2/22 (9)
Chloride (mEq/L) 116130 5/22 (23)
Bicarb (mEq/L) 1324 3/22 (13)
A gap (mEq/L) 12.024 5/22 (23)
Urea (mg/dL) 1735 2/22 (9) 3/22 (13)
Creatinine (mg/dL) 0.62.3
Calcium (mg/dL) 8.211.5 1/22 (5) 1/22 (5)
Phosphorous (mg/dL) 2.76.5 1/22 (5)
Magnesium (mEq/L) 1.62.4 2/13 (15)
Total protein (g/dL) 6.58.9 1/22 (5) 1/22 (5)
Albumin (g/dL) 3.24.7 6/22 (27)
Globulin (g/dL) 2.84.8 2/22 (9)
Glucose (mg/dL) 63140 9/22 (41)
ALT (U/L) 35134 3/22 (13) 7/22 (32)
AST (U/L) 1346 1/22 (5) 12/22 (55)
ALP (U/L) 1596 1/22 (5) 4/22 (18)
GGT (U/L) 02 4/22 (18)
Bilirubin (mg/dL) 00.4 3/21 (14)
Amylase (U/L) 5152210 1/22 (5)
Cholesterol (mg/dL) 73265 1/22 (5) 3/22 (13)
CK (U/L) 71502 1/22 (5) 1/22 (5)
T4 (mg/dL) 1.54.0 8/20 (40)
FELV 15/15 negative
FIV 14/14 negative
TLI (mg/dL) 17.148.6 2/22 (9) 14/22 (64)
Folate (ng/mL) 12.020 3/22 (13) 8/22 (36)

FELV, feline leukemia virus; FIV, feline immunodeficiency virus; TLI, trypsinlike immunoreactivity; MCH, mean corpuscular hemoglobin;
MCHC, mean corpuscular hemoglobin concentration; RDW, red cell distribution width; ALT, alanine aminotransferase; AST, aspartate amino-
transferase; ALP, alkaline phosphatase; GGT, g-glutamyl transferase; CK, creatine kinase.

The successful treatment of IBD in cat 1 was accompa- intestinal disease in cats. The radioassay employed in the
nied by a substantial increase in the persistence of cobala- present study yielded results that were specific, precise, and
min in serum at 7 days (after treatment, 2,165 pg/mL; be- accurate and strongly correlated with those obtained by bio-
fore treatment, 610 pg/mL) and 14 days (after treatment, assay. Further evidence that the RA was an accurate indi-
796 pg/mL; before treatment, 144 pg/mL) after administra- cator of serum cobalamin concentration was provided by
tion. Half-life increased to 8.5 days (Fig 2). the detection of MMA in a cat with low serum concentra-
Urinary methylmalonic acid (MMA) excretion was de- tions of cobalamin and by the increase in cobalamin after
tectable before, but not after, parenteral cobalamin supple- parenteral administration of cyanocobalamin to healthy and
mentation, indicating that before supplementation the cat sick cats.
(cat 1) was suffering from cobalamin deficiency. Review of clinicopathologic data indicated that the serum
concentration of cobalamin was subnormal (,900 pg/mL)
Discussion in 49 of 80 cats being evaluated for weight loss, diarrhea,
The present study sought to evaluate the utility of mea- vomiting, anorexia, and thickened intestines. Definitive di-
suring serum cobalamin as an indirect indicator of gastro- agnosis in these cats included IBD, intestinal lymphoma,
30 Simpson et al

Table 3. Histopathological, ultrasonographic, and selected clinicopathological findings in 22 cats with subnormal con-
centrations of serum cobalamin.
Ultrasonographic Abnormalities
Pancreatic Mesen- Hepat- Diabetes
Histopathological Findings Lesions Intestines teric LN obiliary Pancreas .49 mg/L ,17 mg/L Mellitus
Inflammatory bowel disease (n 5 9) 1a 2 3 2 1 5 1
Lymphosarcoma (n 5 6) 2b, 1c 4 3 2 4 5 1
Hepatobiliary disease (n 5 3) 1c 1 1 2 2 2 2
Hepatobiliary or intestinal disease (n 5 3) 1d 1 3 3 1 1 1
Undiagnosed (n 5 1) 1 1

LN 5 lymph node; TLI 5 trypsinlike immunoreactivity.

Nodular hyperplasia.
Lymphocytic infiltrates.
Fibrosis adenoma.

cholangiohepatitis or cholangitis, and pancreatic inflam- normal histological and ultrasonographic findings is un-
mation. Many cats had disease present in several organ sys- clear.
tems, and it was difficult to determine whether their sub- Treatment of 1 cat with lymphoplasmacytic enteritis and
normal cobalamin concentration was attributable to exo- a high TLI concentration with prednisolone was associated
crine pancreatic disease, intestinal disease, biliary disease, with remission from diarrhea and weight gain and also with
or a combination of those. For example, 4 of 5 cats with less rapid depletion of serum cobalamin. This cat was ini-
gastrointestinal lymphoma also had evidence of pancreatic tially supplemented with cobalamin alone, but weight gain
disease, and 8 cats had a combination of IBD and high TLI and clinical improvement did not occur until prednisolone
concentration; these are suggestive of pancreatitis.2628 The therapy. It is noteworthy that this cat maintained a high TLI
simultaneous presence of pancreatic, intestinal, and hepatic over a 222-day period despite having a normal gross and
disease has been described previously27,29,30 and emphasizes histological structure of the pancreas. Thus the reduced co-
the complex nature of gastrointestinal disease in cats. balamin concentration in this cat seemed to reflect intestinal
Because pancreatic disease is particularly difficult to di- malabsorption secondary to intramural inflammatory
agnose in cats, the present study used ultrasonography, his- changes, rather than exocrine pancreatic disease.
tology, and TLI measurements to generate a detailed picture Serum concentrations of cobalamin were particularly low
of pancreatic structure and function. Ultrasonographic ab- in cats with intestinal lymphoma, 3 of whom also had sub-
normalities of the pancreas were observed in 6 of the 14 normal concentrations of folate consistent with severe, dif-
cats with increased TLI suggestive of pancreatitis2628 and
in 1 of 2 cats with decreased TLI suggestive of EPI.24 Of
the 5 cats with histologically confirmed pancreatic inflam-
mation, all had an abnormal ultrasonographic appearance,
and 4 had high concentrations of TLI. The ultrasonographic
findings were similar to those previously reported in cats
with acute pancreatitis.31 Interestingly, 1 cat with active
pancreatitis had a normal TLI concentration, and 3 cats with
normal pancreatic histology and ultrasound had increased
serum TLI concentrations. Azotemia was not present in any
cats, suggesting that decreased renal filtration or degrada-
tion of TLI was not the cause of high serum concentrations.
Thus the reason for high TLI concentrations in the face of

Table 4. Evaluation of pancreatic disease in cats with

subnormal serum concentrations of cobalamin.
Pancreatic Pancreatic
Ultrasonography Histopathology
TLI Normal Abnormal Normal Abnormal
Normal (1749 mg/L) 3 3 0 2 Fig 2. Serum cobalamin concentrations after the parenteral injection
Increased (.49 mg/L) 8 6 3 4 of cyanocobalamin (1 mg subcutaneously) in healthy cats and cats with
Decreased (,17 mg/L) 1 1 inflammatory bowel disease. Healthy cats, open box; n 5 4; mean 6
SD. Cat 1 before (open circles) and after (closed circles) treatment
TLI 5 trypsinlike immunoreactivity. with prednisolone. Cat 9 (open triangles).
Subnormal Cobalamin in Cats 31

fuse intestinal disease.3 The low folate concentrations in balamin deficiency do not arise for several years after the
these cats with lymphoma may be an important consider- onset of cobalamin malabsorption.38 The rapidity of cobal-
ation for treatment planning, as drugs such as methotrexate amin turnover in cats is consistent with absence of the hap-
could exacerbate the folate deficiency, and alkylating agents tocorrin in cat and dog plasma16 and suggests that after
could cause increased immunological and genetic damage absorption of the cobalamin-intrinsic factor complex in the
in the face of folate deficiency.32,33 High folate concentra- ileum, cobalamin in cats, as in dogs, is bound to transcob-
tions were detected in 5 cats, all of whom had IBD. The alamin 2 and partially excreted in bile.17 The presence of
reasons for the high folate concentrations are unclear, and substantial amounts of intrinsic factor in the pancreas, but
hemolysis, which can substantially elevate serum folate, not the stomach, of the cat (and dog) would be advanta-
could not be excluded as the cause. geous in this situation and would facilitate optimal transfer
Although serum concentrations of cobalamin are thought of cobalamin to intrinsic factor.
to reflect tissue stores accurately, it is not clear at what The precise cobalamin requirements of cats are not
serum concentration overt cobalamin deficiency develops. known. Current Association of American Feed Control Of-
Cobalamin is an essential cofactor for the activity of meth- ficials and National Research Council recommendations39,40
ylmalonyl-CoA mutase and methionine synthase.21 Reduced are based on the results of a study that examined the effects
activity of these 2 enzymes causes the biochemical signa- of feeding a cobalamin deficient diet to 5 weanling kit-
tures of cobalamin deficiency: methylmalonicacidemia/ tens.41,42 In this study, kittens fed a cobalamin deficient diet
methylmalonicaciduria and homocystinemia, respectively. showed poor growth at 34 months and increased urinary
The presence of methylmalonicacidemia/methylmalonica- excretion of methylmalonic acid when compared with cats
ciduria in the urine of 1 cat with IBD that was undetectable fed a diet containing 50 mg/kg cyanocobalamin. Weight
after parenteral cyanocobalamin indicates that gastrointes- gain and decreased methylmalonic acid excretion occurred
tinal disease can be associated with cobalamin deficiency after cobalamin-deficient cats were given 10 mg cyanoco-
in the cat.
balamin per day parenterally. Further observations39 suggest
Presumed cobalamin deficiency has been previously re-
that queens and kittens are clinically healthy when fed a
ported in a cat with anorexia and failure to thrive,22,23 but
20-mg/kg diet. Our findings suggest that cats with subnor-
it is unclear whether this cats cobalamin deficiency was
mal cobalamin absorption may require parenteral supple-
primary or the result of some other undiagnosed abnor-
mentation (1 mg) every 2 weeks to maintain a normal se-
mality. A group of cats with hepatic lipidosis has also been
reported to be suffering from cobalamin deficiency on the rum cobalamin concentration. Determination of MMA in
basis of hepatic cobalt estimation.34 In these lipidotic cats serum or urine would be required to determine the dosing
it was hypothesized that cobalamin deficiency may have interval that prevents cobalamin deficiency, but cyanoco-
been important in the pathogenesis of fatty liver, as has balamin is entirely nontoxic and can be used with a wide
been noted in cobalt-deficient sheep. Interestingly, high margin of safety.
plasma betaine, with no increase in homocysteine, has been In conclusion, a solid-phase boil radioassay was an ac-
observed in a cobalamin-deficient cat (Fyfe, personal com- curate means of measuring serum cobalamin in cats. The
munication). This observation suggests that methionine high prevalence (61%) of subnormal concentrations of se-
synthesis in cats may be betaine rather than methylfolate rum cobalamin in this study suggests that the measurement
dependent.35 of serum cobalamin may be a useful indirect indicator of
Cobalamin deficiency indirectly inhibits nucleic acid syn- gastrointestinal disease in cats. The rapid depletion of co-
thesis, and thus the tissues that are most affected are those balamin in cats suggests that cats may be very susceptible
in which cells divide most rapidly. In the adult, these in- to cobalamin deficiency. However, the relationship of sub-
clude hematopoietic tissue, testes, and intestinal epithelium normal cobalamin concentrations to cobalamin deficiency,
but involve most tissues in a developing fetus or rapidly and the effect of cobalamin deficiency on cats, remain to
growing newborn. be determined.
Inhibition of nuclear maturation during hematopoiesis
leads to the hematological hallmarks of cobalamin deficien-
cy in people: nonregenerative anemia, neutropenia with me-
galoblastosis of erythroid, myeloid precursor cells in the Footnotes
bone marrow, and release of macrocytic erythrocytes and
hypersegmented neutrophils.13,36,37 The effects of cobalamin a
Teklad, lab cat diet, Madison, WI
deficiency on routine CBC, biochemistry, serum cobalamin b
Dualcount solid-phase boil assay, Diagnostics Products Corporation,
concentration, and organ pathology have not been reported Los Angeles, CA
in cats. Interestingly, macrocytosis (mean corpuscular vol-
Vitamin Diagnostics Inc, Cliffwood Beach, NJ
ume .52 fL) was observed in 6 of 22 cats in the present
TLI: GI Laboratory, Texas A&M, College Station, TX
study and was not associated with feline leukemia virus,
hyperthyroidism, or erythroid regeneration.
The rapid decline in serum cobalamin concentrations af- Acknowledgments
ter parenteral administration in 2 cats with gastrointestinal
disease indicates that cats have a very rapid turnover of This study was supported in part by a grant from the Cor-
cobalamin compared with humans. The biological half-life nell Feline Health Center. We thank Drs Susan Wade and
of cobalamin in humans is about 1 year, and signs of co- Nancy Allen for their assistance.
32 Simpson et al

References nine inherited intestinal cobalamin malabsorption. J Biol Chem 1991;

1. Batt RM, Morgan JO. Role of serum folate and vitamin B12 con- 21. Allen RH, Stabler SP, Savage DG, Lindenbaum J. Metabolic
centrations in differentiation of small intestinal abnormalities in the abnormalities in cobalamin (vitamin B12) and folate deficiency. FASEB
dog. Res Vet Sci 1982;32:1722. J 1993;7:13441353.
2. Batt RM, McLean L. Comparison of the biochemical changes in 22. Vaden SL, Wood PL, Ledley FD, et al. Cobalamin deficiency
the jejunal mucosa of dogs with aerobic and anaerobic bacterial over- associated with methylmalonic acidemia in a cat. J Am Vet Med Assoc
growth. Gastroenterology 1987;93:986993.. 1992;200:11011103.
3. Williams DA. New tests of pancreatic and small intestinal func- 23. Fyfe JC, Giger U, Jezyk PF. Cobalamin metabolism (Letters). J
tion. Comp Cont Educ Pract Vet 1987;9:11671174. Am Vet Med Assoc 1992;201:202203.
4. Batt RM, McLean L, Rutgers HC, Hall EJ. Validation of a ra- 24. Steiner JM, Williams DA. Validation of a radioassay for feline
dioassay for the determination of serum folate and cobalamin concen- trypsin-like immunoreactivity (fTLI) and serum cobalamin and folate
trations in dogs. J Small Anim Pract 1991;32:221224. concentrations in cats with exocrine pancreatic insufficiency. J Vet
5. Seetharam B. Gastrointestinal absorption and transport of cobal- Intern Med 1995;9:193.
amin (vitamin B12). In: Johnson LR, ed. Physiology of the Gastroin- 25. Baker H, Schor SM, Murphy BD, et al. Blood vitamin and
testinal Tract, 3rd ed. New York, NY: Raven; 1994:19972026. choline concentrations in healthy domestic cats, dogs and horses. Am
6. Batt RM, Horadagoda NU, McLean L, et al. Identification and J Vet Res 1986;47:14681471.
characterization of a pancreatic intrinsic factor in the dog. Am J Phy- 26. Steiner JM, Medinger TL, Williams DA. Development and val-
siol 1989;256:G517G523. idation of a radioassay for feline trypsin-like immunoreactivity. Am J
7. Batt RM, Horadagoda NU. Gastric and pancreatic intrinsic factor Vet Res 1996;57:14171420.
mediated absorption of cobalamin in the dog. Am J Physiol 1989;257: 27. Simpson, KW. Acute pancreatitis. In: August JR, ed. Consul-
G344G349. tations in Feline Internal Medicine III. Philadelphia, PA: WB Saun-
8. Simpson KW, Alpers DH, DeWille J, et al. Cellular localization ders; 1997:9198.
and hormonal regulation of pancreatic intrinsic factor secretion in 28. Bruner JM, Steiner JM, Williams DA, et al. High feline trypsin-
dogs. Am J Physiol 1993;265:G178188. like immunoreactivity in a cat with pancreatitis and hepatic lipidosis.
9. Fyfe JC. Feline intrinsic factor (IF) is pancreatic in origin and J Am Vet Med Assoc 1997;210:17571760.
mediates ileal cobalamin (CBL) absorption. J Vet Intern Med 1993;7: 29. Weiss DJ, Gagne JM, Armstrong PJ. Relationship between in-
133. flammatory hepatic disease and inflammatory bowel disease, pancre-
10. Allen RH, Seetharam B, Podell ER, Alpers DH. Effect of pro- atitis, and nephritis in cats. J Am Vet Med Assoc 1996;209:1114
teolytic enzymes on the binding of cobalamin to R protein and intrinsic 1118.
factor: In vitro evidence that a failure to partially degrade R protein 30. Hill RC, Van Winkle TJ. Acute necrotizing and acute suppu-
is responsible for cobalamin malabsorption in pancreatic insufficiency. rative pancreatitis in the cat: A retrospective study of 40 cases (1976
J Clin Invest 1978;61:4754. 1989). J Vet Intern Med 1993;7:2533.
11. Allen RH, Seetharam B, Allen NC, et al. Correction of cobal-
31. Simpson KW, Shiroma JT, Biller DS, et al. Antemortem diag-
amin malabsorption in pancreatic insufficiency with a cobalamin an-
nosis of pancreatitis in 4 cats. J Small Anim Pract 1994;35:9399.
alogue that binds with high affinity to R protein but not to intrinsic
32. Morgan SL, Baggot JE, Lee JY, Alarcon GS. Folic acid sup-
factor: In vivo evidence that a failure to partially degrade R protein is
plementation prevents deficient blood folate levels and hyperhomo-
responsible for cobalamin malabsorption in pancreatic insufficiency. J
cysteinemia during longterm, low dose methpotrexate therapy for
Clin Invest 1978;61:16281634.
rheumatoid arthritis. J Rheumatol 1998;25:441446.
12. Kanazawa S, Herbert V. Mechanism of enterohepatic circulation
33. Branda RF, Hacker M, Lafayette M, et al. Nutritional folate
of vitamin B12: Movement of vitamin B12 from bile R-binder to intrin-
deficiency augments the in vivo mutagenic and lymphocytotoxic ac-
sic factor due to the action of pancreatic trypsin. Trans Assoc Am
tivities of alkylating agents. Environ Mol Mutagen 1998:32:3338.
Physiol 1983;96:336343.
34. Frank A, Feinstein RE. Hepatic lipidosis associated with severe
13. Fyfe JC, Giger U, Hall CA, et al. Inherited selective intestinal
vitamin B12 deficiency recognized by liver cobalt status in three cats.
cobalamin malabsorption and cobalamin deficiency in dogs. Pediatr
Feline Pract 1991;19:1620.
Res 1991;29:2431.
35. Allen RH, Stabler SP, Lindenbaum J. Serum betaine, N,N-di-
14. Kozyraki R, Kristiansen M, Silahtaroglu A, et al. The human
intrinsic factorvitamin B12 receptor, cubilin: Molecular characteriza- methylglycine and N-methylglycine levels in patients with cobalamin
tion and chromosomal mapping of the gene to 10p within the auto- and folate deficiency and related inborn errors of metabolism. Metab-
somal recessive megaloblastic anemia (MGA1) region. Blood 1998; olism 1993;42:14481460.
91:35933600. 36. Sullivan LW. Vitamin B12 metabolism and megaloblastic ane-
15. Hall CA. Transcobalamins I and II as natural transport proteins mia. Seminars Hematol 1970;7:622.
of vitamin B12. J Clin Invest 1975;56:11251131. 37. Babior BM. The megaloblastic anemias. In: Williams WJ, Beu-
16. Linnell JC, Collings L, Down MC, England JM. Distribution tler E, Erslev AJ, Lichtman MA, eds. Hematology, 4th ed. New York,
of endogenous cobalamin between the transcobalamins in various NY: McGraw-Hill; 1990.
mammals. Clin Sci 1979;57:139144. 38. Schloesser LL, Deshpande P, Schilling RF. Biologic turnover
17. Willigan DA, Cronkite EP, Meyer LM, Noto SL. Biliary excre- rate of cyanocobalamin (vitamin B12) in human liver. Arch Intern Med
tion of Co60 labeled vitamin B12 in dogs. Proc Soc Exp Biol Med 1958; 1958;101:306309.
99:8184. 39. National Research Council. Nutrient Requirements of Cats.
18. Simpson KW, Morton DB, Batt RM. Effect of exocrine pan- Washington, DC: National Academy Press; 1986:28.
creatic insufficiency on cobalamin absorption in dogs. Am J Vet Res 40. Phillips T. New Profiles: Review of the AAFCO Feline Nutri-
1989;50:12331236. tion Expert Subcommittee report. Petfood Ind 1992; Sept/Oct:2630.
19. Fyfe JC, Jezyk PF, Giger U, Patterson DF. Inherited selective 41. Morris JG. The essentiality of biotin and vitamin B12 for the
malabsorption of vitamin B12 in Giant Schnauzers. J Am Anim Hosp cat. In: Proceedings of The Kal Kan Symposium for the Treatment of
Assoc 1989;25:533539. Dog and Cat Diseases 1977;1518.
20. Fyfe JC, Ramanujam KS, Ramaswamy K, et al. Defective 42. Keesling PT, Morris JG. Vitamin B12 deficiency in the cat. J
brush-border expression of intrinsic factorcobalamin receptor in ca- Anim Sci 1975;41:317 (abstract 289).