Karissa C. Arthur, MD
Depression is common in patients after a stroke. While depression can be treated effectively with selective serotonin reuptake inhibitors (SSRIs), several recent randomized controlled trials (RCTs), as well as previous observational cohort studies, have reported an increased risk of falls and fractures in patients treated with SSRIs. This risk is especially important considering that stroke patients are already at a higher risk of falls and fractures. Jones and colleagues aimed to determine the risk of fractures in stroke patients treated with SSRIs via a systematic review and meta-analysis.
A study was eligible for inclusion if it met the following criteria: 1) RCT, 2) included adult patients with previous hemorrhagic or ischemic stroke, 3) compared an SSRI to either placebo or no intervention, and 4) included incident fractures as an outcome. The authors searched for studies using combinations of key terms and subject headings without language restrictions in several databases and clinical trial registers, as well as used the references of included studies to identify other studies for possible inclusion. The results were independently evaluated by two reviewers.
Four unique trials of 6549 participants were included in the analysis. All studies were multicenter, parallel, double-blind RCTs. Three studies used fluoxetine 20 mg daily for six months, and the other used citalopram 10 to 20 mg daily for six months as the intervention arm. The primary aim in all studies was functional outcome after stroke, but all four reported safety data over the six months after randomization and were, therefore, included. Pooled analysis showed that treatment with an SSRI for six months after stroke, doubled the risk of fracture compared to placebo (3.02% fluoxetine versus 1.25% placebo; RR, 2.36 [95% CI, 1.64–3.39]; absolute risk increase, 1.77%). The analysis did not show an increased risk of falls, seizures, or recurrent stroke in patients treated with fluoxetine compared to placebo over 6 months.
Interestingly, however, the authors did not find an increased risk of falls, seizures, or recurrent stroke. The main benefit of this study was the inclusion of double-blind RCTs and systematic searches. There are several limitations to this study, however. Fracture was not a primary outcome in any of the RCTs included and was not pre-specified as a secondary outcome in one of the studies included. Only two SSRIs were studied in the four trials included, and, therefore, results may not be generalizable to other SSRIs.
Further research is needed to understand the mechanism by which SSRIs increase risk of fracture, and to determine whether other SSRIs carry this risk. Considering the high incidence of post-stroke depression, SSRIs may still be beneficial in select patients. However, these results highlight the need for a benefit-risk discussion and a patient-tailored approach when considering SSRI treatment in patients with stroke.
