Gastrointestinal tolerability of nimesulide, a selective cyclooxygenase-2 inhibitor, in experimental animals and man

These studies assessed the gastrointestinal tolerability and cyclooxygeanse (COX) selectivity of nimesulide in animals and man. Nimesulide (15 and 30 mg/kg) had no significant 'topical' toxicity in animals as assessed by in vivo mitochondrial morphological studies and no increase in small intestinal permeability or inflammation in rats. Similarly, these doses did not affect mucosal prostaglandins significantly and there were no ulcers. At the very high does of 60 mg/kg, nimesulide had some 'topical' toxicity and lost some of its COX-2 selectivity, but this was not associated with inflammation or ulcers. Indomethacin (10 mg/kg, equivalent to 15 mg/kg of nimesulide) altered all of the above parameters and led to small bowel ulcers. This suggests that the gastrointestinal tolerability of nimesulide is not only due to its COX-2 selectivity, but that it is also due to its lack of 'topical' toxicity. Thirty six healthy subjects participated in a randomised, double-blind, double-dummy crossover study (14 day treatment with a washout) of the effect of nimesulide (100 mg bid) and naproxen (500 mg bid) on the gastroduodenal mucosa and small intestine (absorption-permeability and inflammation). An assessment was also made on their COX selectivity. Nimesulide caused significantly less gastric (p = 0.0001) and duodenal (p = 0.0086) damage than naproxen. Naproxen increased intestinal permeability significantly (p = 0.004) and caused intestinal inflammation while nimesulide (p > 0.40) did not. Nimesulide had no significant (p > 0.1) effect on COX-1 dependant platelet aggregatino, reduced serum thromboxane B2 levels moderately by 24-34% (p < 0.005) and decreased prostanoid generation in gastric biopsies by 9-30% (p > 0.5). By comparison, naproxen had significantly (p < 0.0001) greater effects on these functions: abolished platelet aggregation to arachidonic acid, suppressed serum thromboxane B2 levels by 98% and gastric prostaglandin production by 76-82%. Nimesulide inhibited lipopolysaccharide-induced PGE2 formation in whole blood to a significantly (p < 0.01) greater extent than naproxen. This study shows that nimesulide has a preferential selectivity for COX-2 over COX-1 in vivo at full therapeutic doses. This selectivity is associated with minimal gastrointestinal damage in the short term.