Serum cobalamin and methylmalonic acid concentrations in juvenile dogs with parvoviral enteritis or other acute enteropathies

Abstract Background Low serum cobalamin concentrations have been associated with ileal malabsorption in dogs with chronic enteropathy. Increased serum methylmalonic acid (MMA) concentrations indicate cobalamin deficiency on a cellular level. Few studies have evaluated serum cobalamin concentrations or methylmalonic acid concentrations in juvenile dogs with parvoviral enteritis or nonparvoviral acute enteropathies. Objectives Evaluate serum cobalamin and methylmalonic acid concentrations in juvenile dogs (6 weeks to 10 months old) with parvoviral enteritis or nonparvoviral acute enteropathy. Animals Thirty‐one juvenile dogs with parvoviral enteritis, 29 dogs with nonparvoviral acute diarrhea (NPVAD), and 40 healthy juvenile control dogs. Methods Single‐center, prospective, observational, cross‐sectional study. Serum cobalamin and, when sufficient serum was available, MMA concentrations were measured. Results Most serum cobalamin concentrations were within the adult reference interval. Serum cobalamin concentrations in healthy dogs (median, 848 ng/L; range, 293‐1912 ng/L) were significantly higher than in dogs with parvoviral enteritis (P = .0002; median, 463 ng/L; range, <150‐10 000 ng/L) or dogs with NPVAD (P = .02; median, 528 ng/L; range, 160‐8998 ng/L). Serum MMA concentrations were not significantly different between groups (healthy dogs: median, 796 nmol/L; range, 427‐1933 nmol/L; parvoviral enteritis: median, 858 nmol/L; range, 554‐3424 nmol/L; NPVAD: median, 764 nmol/L; range, 392‐1222 nmol/L; P = .1). Conclusions and Clinical Importance Juvenile dogs with parvoviral enteritis or NPVAD had lower serum cobalamin concentrations than healthy juvenile dogs. However, based on serum MMA concentrations cellular cobalamin deficiency was not apparent.


| INTRODUCTION
Cobalamin, also known as Vitamin B 12 , is a water-soluble vitamin required for DNA synthesis and cellular energy production. 1 Cobalamin deficiency is a well-characterized phenomenon in dogs with chronic enteropathy. 2,3 Measurement of serum cobalamin concentration commonly is used as a diagnostic marker for ileal malabsorption in small animal medicine 4 and generally is presumed to occur secondary to a chronic inflammatory enteropathy, but can also occur secondary to exocrine pancreatic insufficiency 5 or possibly small intestinal dysbiosis. 6 Oral or parenteral supplementation 7,8 often is recommended for patients with serum cobalamin concentrations <400 ng/L. 7,9 The reported prevalence of low serum cobalamin concentration in dogs with chronic enteropathies ranges from 19% to 38%. 2,10 Dogs affected by Imerslund-Gräsbeck syndrome (a mutation resulting in selective cobalamin malabsorption in the ileum) can develop blood cell dyscrasias (e.g., neutropenia, nonregenerative anemia), central and peripheral neurologic signs, and gastrointestinal (GI) signs such as anorexia, vomiting, diarrhea, lethargy, and failure to thrive. 11,12 Although some of these signs exist in dogs with chronic enteropathies, improvement generally is attributed to resolving the underlying condition as opposed to supplementation alone. Weight gain and improvement in clinical outcome have been described in cats with parenteral administration of cobalamin in addition to other treatments. 13 Cobalamin is an essential cofactor of the enzyme methylmalonyl-CoA mutase that converts L-methylmalonyl-CoA to succinyl-CoA. 2 Therefore, free MMA accumulates in patients with cobalamin deficiency at the cellular level. Measurement of serum MMA concentration is thought to be a more accurate indicator of the body's cobalamin status because serum cobalamin concentrations may not reflect cellular concentrations. 2,10,14 However, serum MMA concentration is not routinely determined because measurement of MMA in serum is very labor-intensive and expensive.
Enteritis caused by canine parvovirus type 2 is the world's most common infectious disease of dogs. 15 The virus typically is shed in young puppies from 6 weeks to 6 months of age, and the virus is difficult to eliminate from the environment. 15,16 The disease results in severe damage to the intestinal epithelial crypt cells and other rapidly dividing cells, such as hematopoietic progenitor cells. 17 Clinical signs are typically gastrointestinal in origin, although septicemia from bacterial translocation 18 and myocarditis 19 also can be seen.
Given the destructive effects of parvovirus on the intestine, a deficiency in cobalamin because of ileal malabsorption is possible. If so, cobalamin supplementation theoretically could be beneficial in affected dogs. At the same time, other acute processes can cause intestinal disease in puppies, and may or may not have a similar effect on cobalamin status. Although a previous study identified decreased serum cobalamin concentrations in dogs with parvoviral enteritis compared to healthy dogs, 20 serum MMA concentrations, and therefore cellular cobalamin status, were not investigated. Furthermore, the prevalence and severity of cobalamin deficiency in dogs of similar age with other acute enteropathies were not described.
Therefore, we designed a prospective study to investigate serum cobalamin and MMA concentrations in juvenile dogs with parvoviral enteritis, dogs with nonparvoviral acute diarrhea (NPVAD), and juvenile healthy control dogs. It was hypothesized that serum cobalamin concentrations in dogs with any enteropathy would be lower than in healthy dogs and would be associated with an increase in serum MMA concentrations, indicating cellular cobalamin deficiency.

| MATERIALS AND METHODS
Ours was a single-center, prospective, observational cross-sectional study. Dogs between 6 weeks and 10 months of age that presented Eligible dogs had to weigh at least 2 kg, have had at least 2 episodes of diarrhea with or without vomiting within the last 12 hours or diarrhea with or without vomiting or inappetence lasting over 24 hours, and a commercial, rapid, patient-side canine parvovirus (CPV) SNAP ELISA (Idexx Laboratories, USA) performed at the facility or at a referring veterinarian's practice. These dogs then were divided into 2 groups. One group included dogs that tested positive on ELISA, called the CPV group. The other group consisted of dogs that tested negative on ELISA, called the nonparvovirus acute diarrhea (NPVAD) group. Fecal samples were required for inclusion into the latter group for fecal CPV PCR testing. Dogs that tested positive on PCR were considered part of the CPV group. Dogs that were diagnosed with a foreign body or intussusception after the study were excluded. Dogs with diarrhea lasting >3 weeks were excluded from the study. Age, breed, sex, reproductive status, vaccination status, and duration of reported clinical signs were recorded for each dog.
Clinically healthy dogs >2 kg between the ages of 6 weeks and 10 months that presented for routine vaccination or an elective surgical procedure were included as controls. Dogs in this group were considered healthy based on history and physical examination findings with no evidence of vomiting or diarrhea within the 72 hours before presentation. After sample collection, each dog was treated for its primary reason for presentation at the discretion of the attending clinician.

| Sample processing
Whole blood was collected from each dog and stored in no-additive collection tubes at 4 C within 24 hours of presentation. If not processed for cobalamin measurement within 8 hours, whole blood was centrifuged within 24 hours and serum stored at À20 C.
When required, fecal samples were collected using swab, digital rectal examination, or floor collection and stored in plastic containers at 4 C. Real-time PCR was performed on the samples for CPV DNA as previously described. 21

| Assays
Serum cobalamin measurements were performed in all dogs within 48 hours of blood collection using a solid-phase competitive chemiluminescent immunoassay (Vitamin B 12 , Immulite 2000; Siemens Healthcare Diagnostics). The working range was from 150 to 1000 ng/L, with a reference interval (RI) of 251 to 908 ng/L. When the cobalamin concentration exceeded the upper detection limit, the serum sample was diluted and rerun to obtain a numerical value.
Results >10 000 ng/L were entered as 10 000 ng/L for statistical analysis. The surplus serum for MMA measurement was stored at À80 C. Serum MMA concentration was measured using the stable isotope dilution gas chromatography-mass spectrometry method in batches as described previously. 2 The RI for serum MMA in dogs previously was determined to be 415 to 1193 nmol/L. 2 The lower and upper detection limits for the assay were 63 and 16 000 nmol/L, respectively.

| Statistical methods
A power calculation showed that at least 22 dogs per group were needed to detect a difference of 200 ng/L between groups. This determination was made using a SD of 236 ng/L for serum cobalamin concentration, 22 power of 0.8, and an alpha of 0.05.
Continuous data were assessed for normality using D'Agostino-Pearson tests and visual inspection of q-q plots. The data was nonparametric and is expressed as median (range). Differences in serum cobalamin and MMA concentrations between groups were assessed using Kruskal-Wallis tests followed by Dunn's post-test as appropriate. The correlation between serum cobalamin and MMA concentrations was assessed using Spearman's rank correlation coefficient.
Statistical analysis was performed using a commercial software package (Prism v8, GraphPad, San Diego, California). Values of P < .05 were considered significant.

| RESULTS
During this period, 102 dogs with acute GI signs were screened. One dog was excluded because it tested positive on SNAP CPV, but was negative on concurrent fecal PCR. This patient was treated for CPV and recovered but was excluded from the analysis because of the discordant test results. Another dog was excluded for diarrhea lasting >3 weeks. One hundred dogs were included in the analysis. No dogs were diagnosed with foreign bodies or intussusception. Forty healthy dogs also were enrolled.

| NPVAD group
Twenty-nine dogs were enrolled in the NPVAD group. Three dogs presented with signs of inappetence in addition to reported diarrhea.
Twelve dogs had signs of vomiting, diarrhea, or inappetence lasting for >1 day before presentation, with the longest duration of signs being 3 weeks. This latter dog was diagnosed with a whipworm infection.

| Healthy group
Forty dogs were enrolled in the healthy group. Mixed breed dogs (7) and Golden Retrievers (6)

| Comparison among groups
No significant association between sex and groups of dogs was observed (P = .7). A significant difference in age among groups was found; healthy dogs were older than the CPV group (P = .008; Figure 1).

| Serum cobalamin concentrations
Serum cobalamin concentrations were compared between groups ( Figure 2). The median serum concentration of cobalamin in the CPV group was 463 ng/L (range, <150-10 000 ng/L). One dog in the CPV group had a serum cobalamin concentration > 10 000 ng/L. Thirteen dogs had serum cobalamin concentrations <400 ng/L. Of these, 5 dogs had cobalamin concentrations below the lower limit of the reference interval.
The median serum cobalamin concentration in the NPVAD group was 528 ng/L (range, 160-8998 ng/L). One dog had a serum cobalamin concentration >4000 ng/L, but an exact value could not be deter-  CPV group (P = .0002) and the NPVAD group (P = .02). However, no significant difference was found between the CPV and NPVAD groups (P = .8).

| DISCUSSION
We showed that healthy juvenile dogs had higher serum cobalamin concentrations than dogs with CPV enteritis or NPVAD. However, serum MMA concentrations did not show any significant differences among groups, suggesting no alteration in cellular cobalamin status in dogs with CPV or NPVAD.
Cobalamin is absorbed primarily by the ileal enterocytes via the cobalamin-intrinsic factor complex. 23,24 As such, mucosal disease, 25 lack of intrinsic factor, 5 and small intestinal dysbiosis 6 can be responsible for cobalamin malabsorption and resultant hypocobalaminemia.
Canine parvoviral enteritis is known to cause malabsorption, increased intestinal permeability, 26 and intestinal dysbiosis. 27,28 An acute enteropathy could cause similar changes, as observed in acute hemorrhagic diarrhea syndrome. 29 Any of these aforementioned causes, on their own or together, could result in decreased serum cobalamin concentrations. However, substantial body stores of cobalamin are found in the liver, and thus serum concentrations of cobalamin may not reflect acute cobalamin malabsorption. Therefore, whether acute gastrointestinal disease could cause cobalamin deficiency is uncertain.
Our results mirror a previous study in which dogs with CPV had lower serum cobalamin concentrations than healthy dogs. 20  The previous study did not measure MMA or any other markers for cellular cobalamin deficiency, such as homocysteine. In our study, no significant difference in serum MMA concentrations was found among groups, and no correlation between serum concentrations of cobalamin and MMA was found. When present, increases in serum MMA concentrations generally were mild in comparison to the reference interval for adult dogs, with the exception of 1 dog from the CPV group. Six dogs with serum MMA concentrations above the upper limit of the reference interval did not have a serum cobalamin concentration <400 ng/L. In studies of dogs with chronic enteropathies, a negative correlation between serum cobalamin and MMA concentrations has been described, suggesting that cellular cobalamin deficiency is related to serum concentrations. 2 The lack of correlation between serum cobalamin and MMA concentrations in these juvenile dogs with acute enteropathies suggests that decreased serum cobalamin concentrations do not reflect cellular cobalamin deficiency. One potential explanation is that in dogs with severe acute enteropathies, hypocobalaminemia may occur secondary to the loss of transcobalamin and other cobalamin-binding proteins from the blood through a "leaky" gut wall, causing rapid depletion of cobalamin from the serum. This situation could lead to an acute decrease in serum cobalamin concentration in the absence of the depletion of hepatic stores. It is also possible that as the acute enteropathy resolves, serum cobalamin concentrations will naturally return to normal without supplementation or a cellular deficiency ever developing. Therefore, we do not rec- Our study had some limitations. First, although we recruited healthy dogs from the same general age group (6 weeks to 10 months) as the diseased dogs, a small but significant difference in age was found between normal dogs and dogs with CPV. Serum cobalamin concentrations in human neonates are different than those of adults. This observation could explain the difference in serum cobalamin concentrations between the healthy dogs and dogs with CPV. A significant difference however was not found between NPVAD dogs and other groups, nor was an overall difference noted between serum cobalamin concentrations in juvenile dogs and adult reference ranges noted in a previous study. 20 Also, we did not distinguish among types of diarrhea. Dogs with colitis would be assumed not to have a problem with cobalamin absorption and this difference could have affected the serum cobalamin concentrations seen in the NPVAD group. Dogs with colitis were not excluded from the study, because dogs had a limited evaluation and the diagnosis for most of these dogs was nonspecific. A detailed history of diarrhea was not required for enrollment and thus localizing clinical signs reported by the client were not always recorded. Other contributing causes for hypocobalaminemia were not excluded. Additional theoretical causes for hypocobalaminemia could involve exocrine pancreatic insufficiency, 9 small intestinal bacterial overgrowth, 6 or a vegetarian diet. 31 Characterization of the fecal microbiome was not performed in our study. A thorough dietary history was not a criterion for enrollment, and was not available for all dogs. However, no owners reported feeding a vegetarian diet when a dietary history was available (58/100 dogs). Exocrine pancreatic insufficiency also was not excluded because it would be unlikely given our patient demographic.
Also, we only evaluated serum MMA concentrations as a marker of cellular cobalamin deficiency. Methylmalonic acid is excreted by the kidneys, and serum MMA concentrations are thought to be more stable and have a 40-fold higher concentration in urine. 22 Urine sample acquisition would have been difficult given the inconsistency in dogs urinating in the hospital and because clients may be unwilling to have their healthy animal undergo cystocentesis without a clinical indication. Other studies also have evaluated serum MMA concentrations in dogs. 9 Other causes for increased serum MMA concentrations include hypovolemia, renal insufficiency, and genetic disorders. 10