Gastrointestinal safety profile of nabumetone: a meta-analysis.

Individual comparative studies suggest that nabumetone has a gastrointestinal (GI) safety profile superior to comparator NSAIDs but lack power to show a statistical difference. The aim of this study was to evaluate systematically the difference in GI adverse events--especially the rate of perforations, ulcers, and bleeds (PUBs)-- between studies, meta-analyses of comparative trials of nabumetone and conventional NSAIDs, and postmarketing, open-label studies of nabumetone meeting predefined inclusion and exclusion criteria. A fully recursive literature search identified 13 studies consisting of 29 treatment arms and 49,501 patients that met the predefined criteria. Tests for heterogeneity found no significant difference between studies of each subgroup. Overall, the dyspeptic symptoms flatulence, constipation, and diarrhea were the most commonly reported adverse events accounting for 98.6% of the total GI adverse events. Significantly more patients treated with a comparator NSAID experienced GI adverse events than did those taking nabumetone (P = 0.007). After adjustment for patient-exposure years, PUBs were 10 to 36 times more likely to develop in patients treated with a comparator NSAID than with nabumetone. This was consistently seen in patients in nonendoscopic (n = 7,468) and endoscopic studies (n = 244). In the analysis of postmarketing or open-label studies of nabumetone, only one PUB was reported per 500 patient-exposure years over 17,502 treatment years (n = 39,389). GI adverse event-related dropouts and hospitalizations were increased by 1.3- and 3.7-fold if patients were treated with a comparator NSAID than with nabumetone. Significantly fewer treatment-related GI adverse events, especially PUBs, are seen in patients treated with nabumetone than with a comparator NSAID. Nabumetone is very safe for the GI tract.