J Vet Intern Med 2006;20:221–227
Perinuclear Antineutrophilic Cytoplasmic Antibody and Response to Treatment in Diarrheic Dogs with Food Responsive Disease or Inflammatory Bowel Disease Nicole Luckschander, Karin Allenspach, Jean Hall, Frank Seibold, Andrea Gro¨ne, Marcus G. Doherr, and Fre´de´ric Gaschen The goal of this study was to investigate the correlation between perinuclear antineutrophilic cytoplasmic antibody (pANCA) and clinical scores before and after treatment in diarrheic dogs with food-responsive disease (FRD) or inflammatory bowel disease (IBD). pANCA serology was evaluated prospectively by indirect immunofluorescence in 65 dogs with signs of gastrointestinal disease, and if positive, pANCA antibody titers were determined. Thirty-nine dogs with FRD responded to a novel diet, and 26 dogs with IBD were treated with corticosteroids. The severity of clinical signs was scored by means of a canine IBD activity index (CIBDAI). At initial examination, a significantly (P 5 .002) higher percentage of dogs were pANCA-positive in the FRD group (62%) compared with the IBD group (23%). pANCA titers were significantly higher (P 5 .003) before treatment in the FRD group (median titer 100) compared with the IBD group (median titer 1). However, there was no difference in pANCA titers between the groups after respective treatments because dogs in the IBD group had a significant increase in pANCA titer after treatment. The CIBDAI score decreased significantly (P , .001) after treatment in both groups (74% moderate to severe in FRD dogs before versus 8% after treatment; 85% moderate to severe in IBD dogs before versus 32% after treatment). There was no correlation between pANCA status in FRD or IBD dogs before treatment and scores for CIBDAI, endoscopy, or histopathology before or after treatment, except for the endoscopic duodenal score in dogs with FRD after treatment (P 5 .03). A positive pANCA test before therapy may aid in the diagnosis of FRD. Key words: Canine inflammatory bowel disease activity index; Chronic enteropathies; Endoscopy; Histopathology; Hypoallergenic diet; Steroids.
number of diseases cause chronic enteropathy in dogs, including chronic infection or infestation; diet responsive disease, such as food intolerance and food allergy; intestinal neoplasia; and other primary gastrointestinal diseases. Inflammatory bowel disease (IBD) in dogs is a common cause of chronic vomiting, with or without diarrhea, although the cause remains unknown.1 Diagnostic tools, such as routine hematologic investigations, parasitologic and bacteriologic fecal analyses, abdominal ultrasonographic examination, gastroduodenoscopy with or without colonoscopy, and histopathologic evaluation of intestinal mucosal biopsies, help to rule out known causes of chronic enteropathy. However, there remains a number of idiopathic enteropathies, grouped under the term idiopathic IBD, that encompass disorders characterized
A
From the Department of Clinical Veterinary Medicine, Faculty of Veterinary Medicine, University of Bern, Bern, Switzerland (Luckschander, Allenspach, Doherr, Gaschen); the Department of Biomedical Sciences, College of Veterinary Medicine, Oregon State University, Corvallis, OR (Hall); the Department of Gastrointestinal and Liver Diseases, Division of Gastrointestinal Diseases, Inselspital, Faculty of Medicine, University of Bern, Bern, Switzerland (Seibold); and the Department of Pathobiology, Faculty of Veterinary Medicine, Utrecht University, The Netherlands (Gro¨ne). The work was performed at the Department of Clinical Veterinary Medicine, University of Bern, Bern, Switzerland. Previously presented in part at the 14th ECVIM-CA Congress, Barcelona, Spain, September 2004. Reprint requests: Nicole Luckschander, Kleintierklinik der Universita¨t Bern, La¨nggass-Strasse 128, Postfach, 3001 Bern, Switzerland; e-mail:
[email protected]. Submitted March 10, 2005; Revised July 14, 2005, September 10, 2005; Accepted September 13, 2005. Copyright E 2006 by the American College of Veterinary Internal Medicine 0891-6640/06/2002-0001/$3.00/0
by persistent or recurrent gastrointestinal signs and histologic evidence of intestinal inflammation.1 Because of the frequency of IBD in dogs, and the time consuming, invasive, and expensive procedures required to obtain biopsies for diagnosis confirmation, new diagnostic approaches are desired. A noninvasive test that is helpful for establishing a diagnosis and, in addition, that might help predict the best course of treatment and prognosis, would be very useful. Serologic markers have been available in human medicine for several years and are useful for diagnosis and monitoring of therapy in patients with IBD.2 Serologic tests include antineutrophil cytoplasmic antibody with perinuclear staining (pANCA), antiSaccharomyces cerevisiae antibody (ASCA), and outer membrane porin C (Omp C).3 Two distinct disease entities are classically identified in human IBD: Crohn’s disease, which can affect both small and large intestines, and ulcerative colitis, which affects the large intestine exclusively.2 Antibodies to pANCA are present in the serum of approximately 50–80% of patients with ulcerative colitis.2–4 In one study, a pattern of pANCA-positive and ASCA-negative was 57% sensitive and 97% specific for a diagnosis of ulcerative colitis, whereas a pattern of pANCA-negative and ASCApositive was 49% sensitive and 97% specific for a diagnosis of Crohn’s disease.3 Although pANCAs are specific markers for IBD, they can also be detected in people with various non-IBD disorders, such as vasculitis or systemic lupus erythematosus.5,6 Titers for pANCA in humans are determined with an indirect immunofluorescence assay with granulocytes, whereby a characteristic pattern of perinuclear staining can be seen in pANCA-positive samples.7 Candidate antigens for pANCA have been identified in granules of neutrophils and include proteinase 3, lactoferrin, mye-
222
Luckschander et al
loperoxidase, and lysozyme. These proteins are also present in cell membranes of bacteria and have recently been found in intestinal mast cells and neuroendocrine cells.8–11 Titers for pANCA may be helpful for defining patients who have different response patterns to conventional and immunomodulatory therapies.12–14 On the other hand, even though pANCA titers are highly correlated with ulcerative colitis in humans, there is no evidence that pANCA correlates with clinical disease activity indices, endoscopic evidence of inflammation, or histologic grading of biopsies.15 We have previously reported the presence of pANCA antibodies in serum from dogs with chronic intestinal inflammation.16 In the former study, 31 dogs with chronic intestinal inflammation were compared with 29 dogs with acute or chronic diarrhea of known origin and with 42 healthy dogs.16 The results of pANCA serology were highly specific for dogs with chronic intestinal inflammation. This is in agreement with reports from human medicine that reveal a specificity of up to 94% for pANCA when distinguishing among IBD and healthy controls and patients with non-IBD related diarrhea.2 The purpose of this 2nd study was to prospectively evaluate the utility of pANCA for diagnostic purposes and to assess response to treatment in groups of dogs with IBD or food responsive disease (FRD). Our goal was to find out if there was a correlation between pANCA and canine IBD activity index (CIBDAI), intestinal endoscopic score, or histopathologic grade, before and after treatment, in dogs with IBD or FRD.
Materials and Methods Dogs Sixty-five dogs with chronic gastrointestinal signs were entered into this prospective study over a 3-year period. Criteria for the selection of cases included dogs with a history of clinical signs of chronic diarrhea with or without vomiting of at least 6-weeks duration, exclusion of identifiable underlying disorders, and histopathologic evidence of intestinal inflammatory cellular infiltrates. Dogs with hypoalbuminemia and clinical signs of proteinlosing enteropathy were not excluded as long as there was histopathologic evidence of intestinal inflammatory cellular infiltrates. Owners of dogs signed a letter of consent in which they agreed to participate in initial and follow-up diagnostic evaluations. The experimental protocol was reviewed and approved by the Ethics Committee of the Faculty of Veterinary Medicine, University of Bern, and a permit was obtained from the Canton of Bern Committee regulating animal experimentation. None of the dogs had been treated with antibiotics, corticosteroids, or antacids in the 2 weeks before sample collection. Diagnostics performed to exclude other disorders included CBC, serum biochemical analysis, urinalysis, parasitic and bacterial analysis of fecal samples, abdominal ultrasonography, and assessment of serum concentration of trypsin-like immunoreactivity. Fecal examinations for endoparasite ova and Giardia sp. cysts were also performed. No underlying infectious, pancreatic, or neoplastic diseases were identified in these dogs. Dogs were classified according to their predominant clinical signs as having upper or lower gastrointestinal disease or both. Duodenoscopy and colonoscopy were recommended for all dog, except those with severe hypoalbuminemia. Colonoscopy was not performed in the latter dogs because a 36-hour fast was considered detrimental in these patients.
All dogs were assessed and given a clinical score by means of the CIBDAI scoring system established by Jergens et al,17 which is based on 6 gastrointestinal parameters that are routinely observed in affected dogs. The dogs were also assigned an endoscopic score. Mucosal biopsy specimens from stomach, duodenum, and colon were examined histologically by an ACVP board-certified pathologist (AG) and graded using previously published guidelines.18 A blood sample for pANCA was taken on the day of initial work up. All dogs were treated initially with an elimination dieta for 10 days. Although recommendations usually state that dogs should be fed an appropriate formulation for at least 4 to 6 weeks,19 for owner-compliance reasons, we limited the elimination diet trial to 10 days. Dogs that did not respond, as evidenced by clinical signs that persisted while on this elimination diet, were given PO prednisolone (2 mg/kg, q24h) for 10 days followed by a tapering dosage over 10 weeks (IBD dogs). Dogs that responded to the elimination diet in the 1st 10 days (clinical signs improved or resolved) were called FRD. Thus, dogs were separated into 2 groups according to their initial response to treatment with elimination diet. The FRD group was rechecked 4 weeks after beginning the elimination diet. At this time, the CIBDAI score was reassigned and endoscopy was repeated to determine the endoscopic score and histopathologic grade after treatment. The IBD group was rechecked in the same manner at the end of the 10-week treatment period, which was 4 weeks after reaching the lowest or maintenance prednisolone dosage. Serum was collected from dogs in both groups at the time of reassessment for measuring pANCA titer. All dogs were fed the elimination diet exclusively for 14 weeks.
Canine IBD Activity Index (CIBDAI) The guidelines for assigning CIBDAI scores have been published17 and are based on attitude and activity, vomiting, stool consistency, appetite, stool frequency, and weight loss. The scores are summed for a total cumulative CIBDAI score. The disease is classified as clinically insignificant (CIBDAI score 0–3), mild (score 4–5), moderate (score 6–8), or severe (score 9 or greater).
Endoscopy Sampling Dogs were prepared for endoscopy by withholding food for 48 hours and administering a colonic lavage solutionb by gastric intubation (2 doses of 30 mL/kg of body weight 6–8 hours apart). Multiple mucosal biopsy specimens were routinely obtained by 1 of 2 investigators (KA, FG) from the stomach (fundus, antrum, pylorus), duodenum (10 cm below the flexura duodeni), and middle portion of the descending colon, or from where lesions were visible. Biopsies were collected with flexible fiberoptic endoscopesc with a standard 2.8-mm endoscopic biopsy forceps. An endoscopic score was assigned based on mucosal appearance.18 Findings included mucosal erythema, tissue friability, mucosal granularity, erosions and ulcers, and inadequate insufflation. An endoscopic score (0 5 normal, 1 5 mild, 2 5 moderate, and 3 5 severe) was assigned based on the severity of changes and subjective opinions of the endoscopists. Samples for subsequent histopathologic evaluation were placed in 4% neutral-buffered formalin for at least 48 hours before paraffin embedding.
Histopathology Formalin-fixed, paraffin-embedded biopsy samples were sectioned at 4 mm, stained with hematoxylin and eosin (H&E), and histologically assessed. Blinded qualitative evaluation of the degree of inflammation and overall cellular infiltrate was performed at 403 objective by an ACVP board-certified pathologist (AG), who assigned a grade (normal 5 0, mild 5 1, moderate 5 2, and severe 5 3) based on previously published guidelines.18 A minimum of 5
Chronic Enteropathy in Dogs biopsy specimens from each site were examined histologically.20 Mild IBD lesions were those with cellular infiltrates but without architectural distortion or mucosal epithelial immaturity. Moderate lesions had cellular infiltrates accompanied by mucosal epithelial immaturity or solitary epithelial necrosis or both. Severe IBD lesions consisted of cellular infiltrates accompanied by multifocal epithelial necrosis or extensive architectural distortion with epithelial immaturity.
Assay for pANCA Two milliliters of whole blood were collected from each dog before treatment at the time of initial work up and again after respective dietary or prednisolone treatment on the recheck day. The blood was allowed to clot and then centrifuged at 450 3 g for 20 minutes to obtain serum, which was stored at 220uC until further use. The pANCA status was assessed as previously described.16 Briefly, serum (1 : 10 dilution) was placed over a canine granulocyte-mounted slide and allowed to incubate for 1 hour. Cells were washed, a secondary fluorescein isothiocyanate-labeled antibody (sheep anticanine immunoglobulin G [IgG] antibodyd ) was added, and cells were allowed to incubate for 1 hour in a humid chamber at room temperature. After the final wash, slides were dried and mounted with fluorescence mounting medium.e Slides were evaluated by fluorescence microscopyf by 2 independent observers (KA, NL, JH). Sera from dogs that resulted in a specific pattern of perinuclear staining in granulocytes were considered pANCA-positive, whereas an atypical staining pattern of granulocytes (including evidence of intranuclear staining) was considered pANCA-negative. If serum was assessed to be pANCA-positive, further dilution steps were performed (1 : 100 to 1 : 1,600) to determine pANCA antibody titer.
Statistical Analyses Because none of the interval-measured variables were normally distributed, both they and the ordinal (score) variables are reported as median (range) unless specified otherwise. A nonparametric 2sample test (Mann-Whitney U-test or Wilcoxon rank-sum test) was used to evaluate data between the 2 groups of dogs, and a paired ttest (Wilcoxon signed-rank test) was used to evaluate values before and after treatment within a group of dogs. Cross-tabulations for ordinal or categorical variables were performed by a Fisher’s exact or chi-squared test. When data were paired, the McNemar’s test was used instead. Correlations between interval and score variables were evaluated with the Spearman’s rank-correlation test. Statistical significance was set at P , .05. Statistical analyses were performed with the Number Cruncher Statistical Systemg (NCSS).
Results A total of 65 dogs were examined. The FRD group consisted of 39 dogs; the IBD group consisted of 26 dogs. In the FRD group, there were 16 females and 23 males of which 9 and 5, respectively, were neutered. Dog ages ranged from 1.0 to 10.6 years old (median 3.3 years). Breeds included mixed breeds (n 5 7), Golden Retriever (n 5 6), Bernese Mountain Dog (n 5 4), Labrador Retriever (n 5 3), German Shepherd Dog (n 5 3), Belgian Malinois (n 5 2), and 1 each Leonberger, Great Dane, Boxer, Alaskan Malamute, English Setter, Border Collie, Whippet, Border Terrier, Cairn Terrier, Parson Russell Terrier, Dachshund, Chihuahua, West Highland White Terrier, and Shih Tzu. Body weights in the FRD group ranged from 1.5 to
223
61.0 kg (median 27.0 kg). Thirty-nine dogs had the full diagnostic work up at the initial examination, except for 1 dog that did not receive a colonoscopy (owner declined, but duodenoscopy was performed). Thirtyseven dogs were available for follow-up 4 weeks after beginning the elimination diet. At that time, 6 additional owners declined duodenoscopy, and 7 owners declined colonoscopy. Thus, 30/39 (77%) dogs had the full diagnostic work up at the 4-week follow-up examination. The IBD group consisted of 11 females and 15 males, of which 11 and 5, respectively, were neutered. The dogs were between 2.1 to 11.3 years old (median 8.6 years). Breeds in this group included Yorkshire Terrier (n 5 5), Dachshund (n 5 3), mixed breeds (n 5 3), Labrador Retriever (n 5 2), German Shepherd Dog (n 5 2) and 1 each Bernese Mountain Dog, Golden Retriever, Toy Poodle, Mastiff, Boxer, Chinese Shar-Pei, Rottweiler, Papillon, Coton de Tulear, West Highland White Terrier, and Dalmatian. Body weights in the IBD group ranged from 1.8 to 71.5 kg (median 17.6 kg). Fifteen of 26 dogs (58%) had the full diagnostic work up at the time of initial examination. One dog did not undergo duodenoscopy (owner declined), and 10 were not examined with colonoscopy (6 were severely hypoalbuminemic; 4 owners declined). Twenty-two dogs were available for follow up 10 weeks after beginning prednisolone therapy. Four owners declined duodenoscopy, and 10 dogs did not receive colonoscopy (4 were severely hypoalbuminemic; 6 owners declined). Thus, 11/26 (42%) of the dogs had a full diagnostic work up at the 10-week follow-up examination. At the time of initial examination, the IBD group was significantly older (P , .001) and weighed significantly less (P 5 .006) than the FRD group. There was no sex difference between the 2 groups. In the FRD group, 28/39 (72%) of the dogs had clinical signs consistent with both upper gastrointestinal disease and lower gastrointestinal disease, 5/39 (13%) had clinical signs of upper gastrointestinal disease only, and 6/39 (15%) had clinical signs of lower gastrointestinal disease only. In the IBD group, 14/26 (54%) of the dogs had clinical signs consistent with both upper gastrointestinal disease and lower gastrointestinal disease, 10/26 (38%) had clinical signs of upper gastrointestinal disease only, and 2/26 (8%) had clinical signs of lower gastrointestinal disease only. There was no difference between the 2 groups for clinical signs of upper and lower gastrointestinal disease or both (P 5 .18). In the FRD group 24 (62%) dogs were pANCApositive, and 15 (38%) dogs were pANCA-negative before treatment; in the IBD group 6 (23%) dogs were pANCA-positive and 20 (77%) dogs were pANCAnegative before treatment. There were significantly more pANCA-positive dogs in the FRD group than in the IBD group before treatment (P 5 .002). The pANCA titers were also significantly higher in the FRD group before treatment: median titer 100 (range, 1–1,600) compared with the IBD group: median titer 1 (range, 1– 1,600) (P 5 .003). For this analysis, all titers that were
224
Luckschander et al
Table 1. Clinical scores before and after treatment in dogs with FRD and IBD. FRD Dogs Insignificant to Mild Scores CIBDAIa Endoscopy Duodenumb Endoscopy Colonb Histopathology Stomachc Histopathology Duodenumc Histopathology Colonc
Moderate to Severe
FRD Dogs Insignificant to Mild
IBD Dogs
Moderate to Severe
Insignificant to Mild
Moderate to Severe
Insignificant to Mild
Moderate to Severe
Before Treatment
Before Treatment
After Treatment
After Treatment
22/26 (85%) 10/24 (42%)
15/22* (68%) 9/18* (50%)
7/22* (32%) 9/18* (50%)
Before Treatment
After Treatment
10/39 (26%) 28/39 (72%)
29/39 (74%) 11/39 (28%)
34/37* (92%) 28/31* (90%)
3/37* (8%) 4/26 (15%) 3/31* (10%) 14/24 (58%)
31/38 (82%)
7/38 (18%)
30/30* (100%)
0/30* (0%)
11/16 (69%)
5/16 (31%)
8/12* (67%)
4/12* (33%)
30/39* (77%)
9/39* (23%)
28/31 (90%)
3/31 (10%) 12/24* (50%)
12/24* (50%)
15/18 (83%)
3/18 (17%)
10/39 (26%)
29/39 (74%)
9/31 (29%)
22/31 (71%)
4/25 (16%)
21/25 (84%)
28/38 (74%)
10/38 (26%)
23/30 (77%)
7/30 (23%)
9/16 (56%)
7/16 (44%)
Before Treatment
After Treatment
IBD Dogs
3/18 (17%) 15/18 (83%) 6/12 (50%)
6/12 (50%)
FRD, food responsive disease; IBD, inflammatory bowel disease; CIBDAI, canine inflammatory bowel disease activity index. a Dogs were regrouped by combining those with insignificant (CIBDAI score 0–3 ) and mild clinical signs (CIBDAI score 4–5) versus those with moderate (CIBDAI score 6–8) and severe clinical signs (CIBDAI score .9). b Dogs were regrouped by combining those with normal or mild changes (original endoscopic scores 0–1) versus those with moderate to severe changes (original endoscopic scores 2–3). c Dogs were regrouped by combining those with normal or mild changes (original histopathologic scores 0–1) versus those with moderate to severe changes (original histopathologic scores 2–3). * Indicates a significant difference (P , .05) between the FRD dogs and the IBD dogs before or after treatment.
negative at a 1 : 10 dilution were assigned a (titer) value of 1. In the FRD group, serum albumin ranged from 2.6 to 4.0 g/dL (median 3.6 g/dL). In the IBD group, serum albumin ranged from 1.1 to 4.2 g/dL (median 2.9 g/dL). Before treatment, serum albumin was significantly lower in the IBD group compared with the FRD group (P , .001). Other variables compared between the 2 groups of dogs before treatment included the scores for CIBDAI, endoscopy, and histopathology (Table 1). The FRD group had less severe histopathologic stomach scores when compared with the IBD group (P 5 .03). All other measured scores did not differ significantly between the 2 groups of dogs before treatment. Body weights of the IBD group significantly increased from 17.6 kg median (range, 1.8–71.5 kg) before treatment to 18.5 kg median (range, 2.6–80.0 kg) after treatment (P 5 .001); however, the IBD dogs weighed less than the FRD dogs (P 5 .007). Body weights within the FRD group were 27.0 kg median (range, 1.5– 61.0 kg) before treatment and 29.4 kg median (range, 1.5–64.0 kg) after treatment (P 5 0.058). The number of pANCA-positive dogs in the IBD group increased significantly after prednisolone treatment from 6/26 (23%) before treatment to 17/20 (85%) after treatment (P , .001), and the pANCA titers were also significantly higher in the IBD group after treatment (median titer 1 [range, 1–1,600] versus median titer 400 [range, 1–1,600] before and after treatment, respectively; P 5 .002). In the FRD group, there was also a change in pANCA status (24/39 [62%] versus 29/ 36 [81%] positive before and after treatment, respectively; P 5 .03) but not in pANCA titers before and after
treatment. After therapy, there were no differences in the pANCA status or pANCA titers between the 2 groups of dogs. Serum albumin in the IBD group increased significantly from 2.9 g/dL median (range, 1.1–4.2 g/dL) versus 3.0 g/dL median (range, 1.3–4.3 g/dL) before and after treatment, respectively (P 5 .007); however, the IBD dogs had lower serum albumin concentrations after treatment compared with the FRD dogs (P , .001). Serum albumin concentrations within the FRD group were not significantly different after treatment, 3.6 g/dL median (range, 2.6–4.0 g/dL) versus 3.7 g/dL median (range, 2.9–4.6 g/dL) before and after treatment, respectively (P 5 .06). Other variables compared between the 2 groups of dogs after treatment included the scores for CIBDAI, endoscopy, and histopathology (Table 1). The FRD group had lower CIBDAI scores (P 5 .03) and lower endoscopic scores for duodenum (P 5 .002) and colon (P 5 .004) compared with the IBD group. All other measured scores were similar between the 2 groups of dogs after treatment. Within each group of dogs, the differences in scores before and after treatment were compared with expected frequencies, assuming randomness of change versus no change. Only the CIBDAI scores changed significantly (lower) after treatment (P , .001 for both FRD and IBD groups.) Before treatment, there were no significant correlations between the serologic marker pANCA and the CIBDAI score, intestinal endoscopic scores, and histopathologic grades for stomach, duodenum, and colon in dogs with FRD or IBD disease. All comparisons had |r| , .28 and were not significant. After treatment, only the
Chronic Enteropathy in Dogs
difference in endoscopic duodenum score in dogs with FRD disease was significantly correlated to the pANCA titer before treatment (r 5 .42; P 5 .03). Those dogs in the FRD group with a higher pANCA titer had a lower endoscopic duodenum score after treatment. All other comparisons had |r| , .35 and were not significant.
Discussion The pANCA status is commonly used to differentiate forms of IBD in human patients; for example, ulcerative colitis patients are 50–80% positive and Crohn’s disease patients are 70–90% negative.3,21 In this study, the principal finding was that the percentage of dogs that were pANCA-positive (62%) was significantly higher in the FRD group before treatment. Thus, it appears that this serologic marker may aid in the diagnosis of FRD in dogs. The number of pANCA-positive dogs in the IBD group was significantly lower (23%) before treatment. However, there was no difference in pANCA titers between the groups after respective treatments because dogs with IBD had a significant increase in pANCA titer after treatment. The pANCA-positive status usually remains stable in human patients, even with clinical improvement.15 In a study of ulcerative colitis patients by Patel et al.,22 pANCA-positive status was stable for 2 years after initiation of therapy. It would be unusual for the pANCA status in a patient with Crohn’s disease to change from negative to positive.h It is unclear why the IBD dogs in our study that were pANCA-negative initially became pANCA-positive after treatment. It is well known in humans that IBD is a multifactorial disease caused by an interplay of external factors (infectious agents, dietary factors, vascular factors) and internal factors (genetic).23 The internal environment may have been changed by external factors, such as dietary or prednisolone therapy, thus changing the balance of the mucosa-related flora. There may be multiple antigens involved in the production of pANCA.21 Mucosal antigens also lead to local production of pANCA in the intestinal tract.21 pANCA antibodies in ulcerative colitis patients have been shown to be produced by B-cell clones in lamina propria of mucosal lesions.24 It seems clear that B cells under appropriate conditions in chronic inflammation may be triggered to express pANCA, and the presence of antibody in serum may reflect a spillover of that production.24 The seroconverted dogs in the IBD group may represent a distinct subgroup of IBD dogs with more severe intestinal disease (reflected by hypoalbuminemia in some but also higher CIBDAI scores and more severe endoscopic and histopathologic changes as an entire group) that become pANCA-positive over time. Follow-up studies are needed to determine whether these dogs remain pANCA positive over a longer period of time or whether more of the pANCA negative dogs with IBD convert to pANCA-positive with time. We also showed that titers of pANCA do not correlate with CIBDAI scores before or after treatment, endoscopic scores, or histopathologic grades in biopsies from stomach, duodenum, or colon of dogs with FRD
225
or IBD. This is similar to what has been reported in people.15 Only the difference in endoscopic duodenum score in dogs with FRD after treatment was significantly correlated to pANCA titer before treatment. This was because those dogs with a higher pANCA titer had a lower endoscopic duodenum score after treatment. Clinically, it is difficult to interpret this finding because improvement was not substantiated by a similar change in the histopathologic duodenum score. In humans, although pANCA markers are specific for the diagnosis of IBD, the role that these locally produced antibodies play in the pathogenic process in not yet clear.24 It is known that the pANCA pattern is the result of positively charged protein molecules that migrate to the edge of the nuclei of neutrophils. The cytoplasmic granules redistribute around the nuclei, resulting in a typical pANCA pattern in the case of antibodies to elastase, lactoferrin, cathepsin G, and myeloperoxidase.21 These antibodies are probably not directly pathogenic; instead, they serve as markers for their particular disease and as useful tools for diagnostic purposes.24 Dogs with FRD were significantly younger than those with IBD. This is consistent with previous reports, where up to one third of cases occurred in dogs less than 1 year of age.1 Also, dogs in the FRD group were significantly heavier. Although both groups of dogs gained weight after their respective treatments, the IBD dogs still weighed significantly less. This may reflect the higher representation of small breeds in the IBD group. There were more breeds weighing less than10 kg in the IBD group (12/26, 46%) and more large-breed dogs heavier than 10 kg in the FRD group (33/39, 85%). The serum albumin concentration in the FRD dogs was significantly higher before and after treatment compared with the IBD dogs (P , .001 for both), in spite of a significant increase in albumin concentration within the IBD group after treatment (P 5 .007). This may reflect inclusion of protein-losing enteropathy dogs in the IBD group. Fifty-four percent (14/26) of the IBD dogs had albumin concentrations below the normal range. Edema, ascites, or pleural effusion or combinations of these were noted in 8/14 (57 %) of these dogs, 4 of which were Yorkshire Terriers, which are known to be at increased risk for development of protein-losing enteropathy.25 However, there was still a significant difference in the albumin concentration between the 2 groups after removing the dogs with protein-losing enteropathy from the IBD group (P 5 .02, medians, 3.6 and 3.2 g/dL, FRD and IBD groups, respectively), and there were still significantly less dogs in the IBD group that were pANCA-positive before treatment (4/18; 22%; P 5 .009). There were no major differences at initial presentation between gastrointestinal signs of upper and lower disease and scores for CIBDAI, endoscopy, and histopathology between the FRD and IBD groups of dogs. This finding reflects that dogs with chronic intestinal inflammation may present with many clinical signs and none are pathognomonic for FRD or IBD in dogs.1
226
Luckschander et al
Although in both FRD and IBD dogs, the CIBDAI scores decreased significantly after treatment, the FRD group still had lower CIBDAI scores and lower endoscopic scores for duodenum and colon compared with the IBD group. These findings are consistent with other reports revealing a positive response of foodsensitive dogs to restricted-protein source diets.26 This observation also reflects the difficulty of treating idiopathic IBD and a suboptimal response of these dogs to prednisolone therapy.27 In conclusion, pANCA status may be helpful to distinguish between dogs with FRD and IBD before treatment. In human medicine, pANCA expression allows ulcerative colitis patients to be stratified into clinical subsets. Those with positive pANCA titers have a higher probability of left-sided ulcerative colitis, which is more resistant to treatment than other forms; a higher probability of more aggressive disease; and a higher probability of requiring surgery early in the course of disease.12,21 The use of a panel of serologic markers in human medicine has revealed that combining 2 or more tests, such as pANCA; Omp C, which is an outer membrane porin antigen purified from Escherichia coli; and ASCA, which tests for antibodies to Saccharomyces cerevisiae, increases the accuracy of diagnosing either ulcerative colitis or Crohn’s disease.3 Other surrogate markers of inflammation that have been investigated in dogs in other studies before and after IBD treatment include C-reactive protein (CRP) and haptoglobin (HAP). Values of CRP decreased and HAP concentrations increased after medical therapy.17 The findings in this study demonstrate the need for more prospective studies of serologic markers in chronic canine idiopathic enteropathies to establish an accurate serologic marker panel to better define subsets of IBD dogs, which may have different treatment requirements and prognoses.
Footnotes a
Purina Veterinary Diets LA Salmon and Rice, Socie´te´ des Produits Nestle´ SA, Vevey, Switzerland b Each liter solution contains 60.0 g Polyethyleneglycol (PEG) 4000, 1.46 g sodium chloride, 0.745 g potassium chloride, 1.68 g sodium bicarbonate, and 5.68 g sodium sulfate c Olympus GIF-P140 and CF-Q145L endoscopes, Olympus Schweiz AG, Volketswil, Switzerland d Serotec No. AA132, Serotec Ltd, Kidlington, Oxford, UK e Immu-mount, Thermo Shandon, Pittsburgh, PA f Nikon Eclipse E 600W, Ku¨snacht, Switzerland g Number Cruncher Statistical System, version 2004 (www. ncss.com), Kaysville, UT h Seibold F, unpublished observations, 2005
Acknowledgements Funded in part by a grant from the ACVIM Foundation, 1997 Wadsworth Blvd, Suite A, Lakewood, CO 80214-5293.
References 1. Hall EJ, German AJ. Diseases of the small intestine. In: Ettinger SJ, Feldman ED, eds. Textbook of Veterinary Internal Medicine, 6th ed. Philadelphia, PA: WB Saunders Co; 2005:1332– 1378. 2. Nakamura RM, Barry M. Serologic markers in inflammatory bowel disease (IBD). MLO Med Lab Obs 2001;33:8–15. 3. Sandborn WJ. Serologic markers in inflammatory bowel disease: State of the art. Rev Gastroenterol Disord 2004;4:167–174. 4. Linskens RK, Mallant-Hent RC, Groothuismink ZM, et al. Evaluation of serological markers to differentiate between ulcerative colitis and Crohn’s disease: pANCA, ASCA and agglutinating antibodies to anaerobic coccoid rods. Eur J Gastroenterol Hepatol 2002;14:1013–1018. 5. Tidman M, Olander R, Svalander C, Danielsson D. Patients hospitalized because of small vessel vasculitides with renal involvement in the period 1975–95: Organ involvement, antineutrophil cytoplasmic antibodies patterns, seasonal attack rates and fluctuation of annual frequencies. J Intern Med 1998;244:133– 141. 6. Molna´r K, Kova´cs L, Kiss M, et al. Antineutrophil cytoplasmic antibodies in patients with systemic lupus erythematosus. Clin Exp Dermatol 2002;27:59–61. 7. Rump JA, Scholmerich J, Gross V, et al. A new type of perinuclear anti-neutrophil cytoplasmic antibody (p-ANCA) in active ulcerative-colitis but not in Crohn’s-disease. Immunobiology 1990;181:406–413. 8. Gordon LK, Eggena M, Targan SR, Braun J. Mast cell and neuroendocrine cytoplasmic autoantigen(s) detected by monoclonal pANCA antibodies. Clin Immunol 2000;94:42–50. 9. Cohavy O, Bruckner D, Gordon LK, et al. Colonic bacteria express an ulcerative colitis pANCA-related protein epitope. Infect Immun 2000;68:1542–1548. 10. Wei B, Dalwadi H, Gordon LK, et al. Molecular cloning of a Bacteroides caccae TonB-linked outer membrane protein identified by an inflammatory bowel disease marker antibody. Infect Immun 2001;69:6044–6054. 11. Seibold F, Brandwein S, Simpson S, et al. pANCA represents a cross-reactivity to enteric bacterial antigens. J Clin Immunol 1998;18:153–160. 12. Fleshner PR, Vasiliauskas EA, Kam LY, et al. High level perinuclear antineutrophil cytoplasmic antibody (pANCA) in ulcerative colitis patients before colectomy predicts the development of chronic pouchitis after ileal pouch-anal anastomosis. Gut 2001;49:671–677. 13. Taylor KD, Plevy SE, Yang H, et al. ANCA pattern and LTA haplotype relationship to clinical responses to anti-TNF antibody treatment in Crohn’s disease. Gastroenterology 2001;120: 1347–1355. 14. Vasiliauskas EA, Kam LY, Karp LC, et al. Marker antibody expression stratifies Crohn’s disease into immunologically homogeneous subgroups with distinct clinical characteristics. Gut 2000;47:487–496. 15. Ruemmele FM, Targan SR, Levy G, et al. Diagnostic accuracy of serological assays in pediatric inflammatory bowel disease. Gastroenterology 1998;115:822–829. 16. Allenspach K, Luckschander N, Styner M, et al. Evaluation of assays for perinuclear antineutrophilic cytoplasmic antibodies and antibodies to Saccharomyces cerevisiae in dogs with inflammatory bowel disease. Am J Vet Res 2004;65:1279–1283. 17. Jergens AE, Schreiner CA, Frank DE, et al. A scoring index for disease activity in canine inflammatory bowel disease. J Vet Intern Med 2003;17:291–297. 18. Jergens AE, Moore FM, Haynes JS, Miles KG. Idiopathic inflammatory bowel disease in dogs and cats: 84 cases (1987–1990). Am J Vet Med Assoc 1992;201:1603–1608.
Chronic Enteropathy in Dogs 19. Roudebush P. Adverse reactions to foods: Allergies versus intolerance. In: Ettinger SJ, Feldman ED, eds. Textbook of Veterinary Internal Medicine, 6th ed. Philadelphia, PA: WB Saunders; 2005:566–570. 20. Willard MD, Jergens AE, Duncan RB, et al. Interobserver variation among histopathologic evaluations of intestinal tissues from dogs and cats. J Am Vet Med Assoc 2002;220:1177– 1182. 21. Nakamura RM, Matsutani M, Barry M. Advances in clinical laboratory tests for inflammatory bowel disease. Clin Chim Acta 2003;335:9–20. 22. Patel RT, Stokes R, Birch D, et al. Influence of total colectomy on serum antineutrophil cytoplasmic antibodies in inflammatory bowel disease. Br J Surg 1994;81:724–726. 23. Jarnerot G. Future aspects on inflammatory bowel disease. Scand J Gastroenterol Suppl. 1996;220:87–90.
227
24. Targan SR, Landers CJ, Cobb L, et al. Perinuclear antineutrophil cytoplasmic antibodies are spontaneously produced by mucosal B cells of ulcerative colitis patients. J Immunol 1995;155: 3262–3267. 25. Kimmel SE, Waddell LS, Michel KE. Hypomagnesemia and hypocalcemia associated with protein-losing enteropathy in Yorkshire terriers: Five cases (1992–1998). J Am Vet Med Assoc 2000;217:703–706. 26. Leistra MH, Markwell PJ, Willemse T. Evaluation of selected-protein-source diets for management of dogs with adverse reactions to food. J Am Vet Med Assoc 2001;219:1411– 1414. 27. Craven M, Simpson JW, Ridyard AE, Chandler ML. Canine inflammatory bowel disease: retrospective analysis of diagnosis and outcome in 80 cases (1995–2002). J Small Animal Pract 2004;45:336–342.
