Background and purpose:
Poor mental health and depression are well-recognized sequelae of stroke; however, the association between stroke and subsequent risk of suicide is unknown.

Methods:
We systematically searched MEDLINE, Embase, PsycINFO, and Google Scholar from their inception to September 15, 2020, using keywords and database-specific subjects. We independently adjudicated and selected observational studies that reported suicide attempts or death by suicide in stroke survivors and a comparison group, consisting either of people without a history of stroke or the general population. We evaluated study quality using the Newcastle Ottawa scale. Using random-effects meta-analysis, we calculated the pooled adjusted risk ratio (RR) of suicide in stroke survivors and separately calculated the pooled adjusted RR of suicide attempt and death by suicide. Using prespecified analyses, we explored study-level factors to explain heterogeneity.

Results:
We screened 4093 articles and included 23 studies of fair quality, totaling over 2 million stroke survivors, of whom 5563 attempted suicide or died by suicide. Compared to the non-stroke group, the pooled adjusted RR of suicide in stroke survivors was 1.73 (95% CI, 1.53–1.96, I 2=93%), with a significantly (P=0.03) higher adjusted risk of suicide attempt (RR, 2.11 [1.73–2.56]) than of death by suicide (RR, 1.61 [1.41–1.84]). A longer follow-up time in cohort studies was associated with a lower risk of suicide (RR, 0.97 [0.95–0.99] for every 1-year increase).

Conclusions:
Stroke should be considered as a risk factor for suicide. Comprehensive strategies to screen and treat depression and suicidal ideation in stroke survivors should be developed to reduce the burden of suicide in stroke survivors.