Safety Profile of Non-Steroidal Anti-Inflammatory Drugs in Clinical Practice: New Insights from Electronic Health Care and Spontaneous Reporting Databases

Abstract

Non-steroidal anti-inflammatory drugs (NSAIDs) are widely used to treat condition with pain, inflammation, and/or fever. The relative inhibitory potency towards cyclooxygenase (COX) enzymes and the chemical groups of NSAIDs, including the presence or absence of a sulfonamide functional group seem to contribute to the adverse events of these drugs that vary from mild to severe and life-threatening conditions. Even though many studies have been performed to investigate the adverse events of NSAID use, still much is unknown and need further investigation. Our study showed that either current use of selective COX-2 inhibitors or conventional NSAIDs were not associated with an increased risk of ventricular tachycardia/ventricular fibrillation (VT/VF-OHCA) compared to nonuse. Stratification for VT/VF-OHCA with the presence or absence of acute myocardial infarction did not change these results. Our study also indicated that conventional NSAID use, mainly attributable to the acetic acid derivative (AAD) and propionic acid derivative (PAD) groups, is associated with an increased risk of nephrotic syndrome (NS) diagnosis starting from 2 weeks of exposure. Current use with duration 15-28 and >28 days of conventional NSAIDs was associated with an increased risk of NS diagnosis. Recent use (discontinuation 1-2 months before the index date) and past use (discontinuation 2 months-2 years) of conventional NSAIDs were also associated with an increased risk of NS diagnosis. These increased risks seem to disappear after 2 years of discontinuation. In contrast, selective COX-2 inhibitors did not seem to increase the risk of NS diagnosis. We compared the risk of gastrointestinal toxicity (perforation, ulcers, bleeding) among the use of selective COX-2 inhibitors either alone or combined with proton pump inhibitors (PPIs), and conventional NSAIDs with PPIs compared to conventional NSAIDs alone. We found that selective COX-2 inhibitors with PPIs had the lowest risk, followed by selective COX-2 inhibitors alone, and conventional NSAIDs with PPIs compared to conventional NSAIDs alone. Within 5 years after market approval, NSAIDs with poor COX selectivity were associated with the highest reporting odds ratios (ROR) of hypersensitivity reactions compared to coxibs. AAD, fenamate, and PAD groups belonging to NSAIDs were associated with the highest RORs, compared to coxibs. Sulfonamide NSAIDs were also associated with a higher ROR compared to non-sulfonamide NSAIDs. After the 1st five years of marketing, most of the RORs returned to approximately 1. Compared to non-use, current use for <1 month, 3-6 months, 6-12 months, and >12 months of conventional NSAIDs were associated with higher risk of revision surgery of lower joint replacement. Likewise, compared to non-use, current use for <1 month, 1-3 month(s), 3-6 months, and >12 months of selective COX-2 inhibitors were also associated with higher risk. A higher risk was also found for recent and past use of these drugs. However, current use for 1-3 month(s) of conventional NSAIDs and for 6-12 months for selective COX-2 inhibitors was associated with a similar risk compared to non-use.