Risk factors for serious nonsteroidal-induced gastrointestinal complications: regression analysis of the MUCOSA trial.

Abstract

OBJECTIVES: This analysis evaluated the clinical and demographic risk factors for a suspected, serious nonsteroidal anti-inflammatory drug (NSAID)-induced gastrointestinal (GI) complication in everyday clinical practice and calculated the risk reduction associated with misoprostol therapy in these "at-risk" patients. METHODS: Using logistic regression analysis, the data set from a randomized, parallel, placebo-controlled trial of misoprostol in 8,843 rheumatoid arthritis patients taking NSAIDs (the Misoprostol Ulcer Complications Outcomes Safety Assessment trial) was modeled to identify risk factors for GI adverse events. The dependent variable was defined as a "suspected serious GI complication," and the independent variables included demographic features, level of functional disability, presence of co-morbid diseases, use of certain drugs, and treatment arm. RESULTS: Two hundred forty-two suspected serious GI complications were observed; 102 occurred in the misoprostol treatment group (risk: 2.32%) and 140 in the placebo group (risk: 3.15%). Overall risk reduction due to misoprostol therapy was 26.6% (confidence interval 5.5%-42.9%, P < .05). However, in patient groups with identified risk factors, misoprostol use decreased the risk for an adverse GI event by 38.3%-87.3%. Specifically, those who benefitted significantly from therapy with misoprostol were patients with a history of peptic ulcer disease (risk reduction 52.4%), history of previous GI bleeding (risk reduction 50%), history of significant cardiovascular disease (risk reduction 38.3%), significant functional disability (risk reduction 87.2%), and patients whose symptoms required concomitant antacid use (risk reduction 48.3%). CONCLUSION: We conclude that in everyday practice, patients who require chronic NSAID therapy and who have specific clinical risk factors may benefit from misoprostol co-therapy.