RESUMO
OBJECTIVE: To evaluate the effects of carprofen and meloxicam on conductance and permeability to mannitol and on the histologic appearance of sections of canine gastric mucosa. SAMPLE: Gastric mucosa from 6 mature mixed-breed dogs. PROCEDURES: Sections of gastric mucosa were mounted in Ussing chambers, and carprofen (40 or 400µg/mL [CAR40 and CAR400, respectively]), meloxicam (8 or 80µg/mL [MEL8 and MEL80, respectively]), or no drug (controls) was added to the bathing solution. For all sections, conductance was calculated every 15 minutes for 240 minutes and flux of mannitol was calculated for 3 consecutive 1-hour periods; histologic examination was performed after the experiment. The area under the conductance-time curve for each chamber was calculated. Values of conductance × time, flux of mannitol, and the frequency distribution of histologic findings were analyzed for treatment effects. RESULTS: For CAR400- and MEL80-treated sections, conductance X time was significantly higher than that for control and MEL8-treated sections. The effect of CAR40 treatment was not different from that of any other treatment. Over the three 1-hour periods, mannitol flux increased significantly in MEL80-, CAR40-, and CAR400-treated sections but not in MEL8- treated or control sections. Major histologic changes including epithelial cell sloughing were limited to the CAR400-treated sections. CONCLUSIONS AND CLINICAL RELEVANCE: In the gastric mucosa of dogs, carprofen and meloxicam increased in vitro conductance and permeability to mannitol. At a concentration of 400 µg/mL, carprofen caused sloughing of epithelial cells. Carprofen and meloxicam appear to compromise gastric mucosal integrity and barrier function in dogs.