Background & aims: In vitro studies have shown that selective serotonin reuptake inhibitors (SSRIs) inhibit platelet aggregation. It is controversial whether use of SSRIs is a cause of clinically important bleeding; results from observational studies have been equivocal.
Methods: A population-based case-control study was conducted in Denmark. The 3652 cases all had a first discharge diagnosis of serious upper gastrointestinal bleeding (UGB) from 1995 to 2006. Controls (n = 36,502), matched for age and sex, were selected by risk-set sampling. Data on drug exposure and medical history were retrieved from a prescription database and the county's patient register. Confounders were controlled for by conditional logistic regression and the case-crossover design.
Results: The adjusted odds ratio (OR) of UGB among current, recent, and past users of SSRIs was 1.67 (95% confidence interval [CI], 1.46-1.92), 1.88 (95% CI, 1.42-2.5), and 1.22 (95% CI, 1.07-1.39). The adjusted OR for concurrent use of SSRI and nonsteroidal anti-inflammatory drugs (NSAIDs) was 8.0 (95% CI, 4.8-13). The adjusted OR for the concurrent use of NSAID, aspirin, and SSRI was 28 (95% CI, 7.6-103). Of the UGB cases, 377 were current users of SSRI; the adjusted OR for UGB in the case crossover analysis was 2.8 (95% CI, 2.2-3.6). The adjusted OR among users of proton pump inhibitors was 0.96 (95% CI, 0.50-1.82).
Conclusions: Use of SSRI was associated with UGB, consistent with its antiplatelet effects. SSRIs should be prescribed with caution for patients at high risk for UGB.