Wooten JG, Blikslager AT, Marks SL, Law JM, Graeber EC, Lascelles BD. Gastrointestinal Biology Research Laboratory, College of Veterinary Medicine, North Carolina State University, Raleigh, NC 27606, USA. AJVR 2009 Oct 70 10 1243-9 OBJECTIVE: To assess in vivo effects of short-term administration of NSAIDs with varied cyclooxygenase (COX)-2 selectivity on pyloric and duodenal mucosa. ANIMALS: 8 healthy dogs. PROCEDURES: Each dog received deracoxib (2 mg/kg, PO, q 24 h for 3 days), firocoxib (5 mg/kg, PO, q 24 h for 3 days), meloxicam (0.2 mg/kg, PO, q 24 h for 1 day followed by 0.1 mg/kg, PO, q 24 h for 2 days), or placebo orally for 3 days; there was a 4-week interval between successive treatments. Prior to and on day 3 of drug administration, pyloric and duodenal mucosae were assessed endoscopically and biopsy specimens obtained for histologic examination. Cyclooxygenase-1 and -2 protein expressions were assessed (western blotting) and prostanoid concentrations measured (ELISAs). Data were analyzed by use of an ANOVA. RESULTS: Drug administration did not significantly affect endoscopic mucosal scores, histologic scores, or COX-1 or -2 protein expression. The COX-1 protein expression was significantly higher in the pylorus than in the duodenum. Total prostaglandin and thromboxane B(2) (TXB(2)) concentrations were significantly greater in pyloric than in duodenal mucosa. Drug administration had no effect on prostaglandin or TXB(2) concentrations. CONCLUSIONS AND CLINICAL RELEVANCE: Prostanoid concentrations in gastric and duodenal tissues, and gross and histologic appearances, were not significantly affected by drugs with varied COX-2 selectivity. These findings suggested that, for these experimental conditions, there were no differences among the preferential and selective COX-2 inhibitors with regard to adverse effects on the gastric and duodenal portions of the gastrointestinal tract of dogs.