Selective serotonin reuptake inhibitors and risk of cerebral bleeding.

S elective serotonin reuptake inhibitors (SSRIs) are used commonly to treat poststroke depression. They are also being evaluated in clinical trials for their effectiveness in facilitating functional recovery after stroke. 3,4 However, observational studies suggest that SSRIs are associated with several adverse outcomes in older people (aged >65 years), including an increased risk of all-cause mortality (10.61% per year with SSRI versus 7.04% per year with 5 Furthermore, a meta-analysis of 16 observational studies involving 506 411 participants reported that SSRIs are also associated with an increased risk of intracerebral hemorrhage 2 =29%). 6 Given an estimated global incidence of ICH of 24.6 per 100 000 person-years, these data suggested that SSRIs may realize 1 additional ICH per 10 000 persons (0.01%) treated for 1 year. 7 The validity of these observational data is supported to some extent by known biological effects of SSRIs, which include inhibition of the serotonin reuptake transporter (5-HTT) in neurons and other cells, including platelets. 8 Because SSRIs block reuptake of serotonin by platelets, the concentration of serotonin in platelets decreases, thus compromising hemosta-sis and predisposing to an increased risk of hemorrhagic events (but also a reduction in thromboembolic ischemic events). 8 However, a causal relationship between exposure to SSRIs and increased risk of ICH is yet to be established. The above observational studies could not exclude potential confounding by indication, residual confounding by alcohol abuse, smoking, diabetes mellitus, intracranial small-vessel disease (each of which may be associated with exposure to SSRIs and ICH), and potential reverse causality bias (ie, cerebral micro-bleeds may be a risk factor for depression and fatigue). 6,9–11 Moreover, a systematic review of 56 randomized controlled trials of SSRIs versus controls identified only 2 trials that recorded bleeding as an adverse effect and reported that there was a no significant increase in bleeding with exposure to SSRIs versus control among a total of 249 partici-3 In this issue of Stroke, Aarts et al 12 extend our knowledge by exploring the hypotheses that potent SSRIs may increase the prevalence of subclinical (ie, neurologically asymptomatic) intracerebral hemorrhage (ie, cerebral microbleeds) and reduce the prevalence of subclinical ischemic brain lesions because of small-vessel disease (ie, lacunes and white matter lesion volume). They undertook a cross-sectional epidemiological study of 4945 individuals (mean age, 64 years; 55% women) who were enrolled in the population-based Rotterdam cohort study in 1990, were not demented, and underwent a brain …