Risk of upper gastrointestinal events with the use of various NSAIDs: a case-control study in a general population

OBJECTIVE: The gastrointestinal (GI) safety of different non-steroidal anti-inflammatory drugs (NSAIDs) in a real-life setting remains ill defined. The aim of this study was to examine the risk of upper GI events associated with various NSAIDs in a general population. MATERIAL AND METHODS: A nationwide, register-based, matched case-control study was carried out in outpatient residents of Finland in 2000-04. Cases with upper GI events (n=9191) were drawn from the Hospital Discharge Register and individually matched to controls (n=41,780) from the Population Register. RESULTS: The semi-selective NSAIDs (nimesulide, nabumetone, meloxicam, etodolac) had the highest odds ratio for upper GI events even after adjusting for various potential confounders (adjusted odds ratio (AOR) 3.63; 95% CI 3.08-4.28), followed by non-selective (2.98; 2.70-3.29) and COX-2 selective NSAIDs (2.53; 2.09-3.07). When the current use of semi-selective NSAIDs was compared with that of non-selective and COX-2 selective NSAIDs, the AORs were 1.54 (1.13-2.09) and 1.67 (1.10-2.53), respectively. The AORs for the use of COX-2 selective NSAIDs did not differ statistically from the non-selective NSAIDs (AOR 0.92; 0.65-1.31). The AORs for individual NSAIDs varied across and within categories. CONCLUSIONS: As a group, the GI safety of the COX-2 selective NSAIDs was not demonstrated as definitively superior to non-selective NSAIDs. Semi-selective NSAIDs do not seem to offer any GI advantage over other NSAIDs.