Background and objective: Selective serotonin reuptake inhibitor (SSRI) antidepressants can inhibit uptake of serotonin by platelets, and their use may predispose patients to bleeding. Case reports and observational studies from databases have suggested an association between the use of SSRIs and gastrointestinal bleeding. Their risk appears to be increased if they are concurrently used with aspirin (acetylsalicylic acid) or with other NSAIDs. With the aim of establishing the risk of major upper gastrointestinal bleeding associated with various groups of drugs, we performed a multicentre case-control study. We present the results related to the use of antidepressants by the degree of serotonin reuptake inhibition they induce, the selectivity at monoamine transporters and the dose.
Methods: A population-based multicentre case-control study in 18 hospitals in Spain and in Italy, including 2813 incident cases of upper gastrointestinal bleeding and 7193 matched controls. Regression analyses are based on 2783 cases and 7058 controls because of missing variable data. Odds ratios (ORs) of upper gastrointestinal bleeding for antidepressant drugs grouped by affinity for the serotonin transporter, selectivity and dose, with adjustment for potential confounders were estimated.
Results: Overall, 84 (3.0%) cases and 160 (2.2%) controls had used a high-affinity serotonin reuptake inhibitor (SRI) antidepressant. Their use in the 7 days prior to the index day was not associated with a substantially increased risk of upper gastrointestinal bleeding (OR = 1.24; 95% CI 0.88, 1.76). Forty-one (1.5%) cases and 26 (0.4%) controls had concurrently used a high-affinity SRI antidepressant and an NSAID. The OR of upper gastrointestinal bleeding among these concurrent users (8.32; 95% CI 4.69, 14.76) did not differ from that in users of NSAIDs only (7.82; 95% CI 6.79, 9.00). No significant association was found between the use of SSRIs and the risk of upper gastrointestinal bleeding, neither with the degree of affinity for the serotonin transporter, by the selectivity of each individual agent (101 cases [3.6%] vs 192 controls [2.7%]; OR = 1.23; 95% CI 0.90, 1.68), nor by dose.
Conclusions: The risk of upper gastrointestinal bleeding is not increased by the use of SRIs. An interaction with coadministered NSAIDs was not observed. If there is a risk associated to these drugs, it seems to be low and not an important cause of hospital admission due to upper gastrointestinal bleeding. However, additional studies may be warranted in subgroup populations at potentially increased risk of bleeding, such as older adults and men.