Twenty-four healthy, mixed breed dogs were divided into four groups. Group I received a placebo PO BID, group II received an average 16.5 (range, 15.1-17.8) mg/kg buffered aspirin PO BID, group III received an average 2.2 (range, 2.0-2.4) mg/kg carprofen PO BID, and group IV received an average 12.8 (range, 11.7-13.8) mg/kg etodolac PO QD (with a placebo in the P.M.). All treatments continued for 28 consecutive days. Gastroduodenal endoscopy was performed on days – 9, 0, 5, 14 and 28. Multiple gastric biopsies were obtained endoscopically on day – 9 to determine each dog's Helicobacter spp. status.

Five areas, consisting of four regions in the stomach and one in the proximal duodenum, were evaluated endoscopically, and each was assigned a score from 1 to 11 based on qualitative assessment of submucosal hemorrhage, erosion, or ulceration. These scores for each region were then summed to give a total score for each endoscopic evaluation.

Erosions and submucosal hemorrhages were seen in all dogs receiving aspirin. Only minor gastric lesions were observed in the carprofen, etodolac, and control groups. No adverse clinical signs were noted in any dog given any treatment during the course of the study. There was no predilection site for lesion development in any group. Median total score on days 0, 5, 14, and 28 were as follows: group I, 5.0, 5.0, 5.0, 5.0; group II, 5.0, 27.0, 26.0, 27.5; group III, 5.0, 5.0, 6.0, 5.0; group IV, 5.0, 7.0, 5.0, 5.0, respectively.

There was no significant difference between dogs receiving carprofen, etodolac, or placebo. The administration of carprofen, etodolac, or placebo to healthy dogs resulted in significantly less gastroduodenal lesion development than in dogs receiving buffered aspirin.

Thirty healthy, random source, dogs were evaluated to determine the prevalence of Helicobacter spp., and to compare the ‘Campylobacter-like organism’ test (CLOtest®) to histopathologic identification of Helicobacter spp. organisms. Gastric mucosal biopsies from each of four gastric regions (cardia, pyloric antrum, greater curvature, and angularis incisura) were obtained endoscopically for use in the CLOtest® and for histopathologic evaluation. Twenty-seven of 30 dogs (90%) were positive for spiral bacteria suspected to be Helicobacter spp. by histopathologic evaluation in at least one of the four gastric regions. Three dogs (10%) were negative for Helicobacter spp. in all gastric regions by histopathologic evaluation. The CLOtest® was found to have a sensitivity, specificity, and positive predictive value of 84%, 81%, and 92%, respectively, when compared to histopathologic evaluation. When only the angularis incisura was evaluated, the sensitivity, specificity, and positive predictive value increased to 92%, 94%, and 96%, respectively. The angularis incisura had the highest, whereas the pyloric antrum had the lowest, prevalence of positive test results when compared to dogs determined to be overall Helicobacter spp. positive (histopathologic positive in at least one gastric region). The results of this study suggest the prevalence of Helicobacter spp. in apparently healthy dogs is high. For accurate and economical detection of Helicobacter spp. in a dog undergoing upper gastrointestinal endoscopy, a tissue sample should be taken from the angularis incisura for CLOtest® sampling.