Supplementary Material for: Early Antidepressant Treatment and All-Cause 30-Day Mortality in Patients with Ischemic Stroke

Background: Antidepressants, in particular selective serotonin reuptake inhibitors, have been associated with antithrombotic and neuroprotective properties and their more widespread use has been suggested in stroke recovery. However, data are sparse on their effects on the clinical outcome, including mortality, associated with early antidepressant treatment after stroke. We aimed to study all-cause 30-day mortality related to early antidepressant treatment in patients with ischemic stroke. Methods: We did a population-based follow-up study identifying patients from the Danish Stroke Registry admitted in the former Aarhus County from 2003 to 2010. During this time, initiation of antidepressant treatment during admission was registered in the Danish Stoke Registry. The registry also holds clinical information including stroke type, stroke severity and quality of in-hospital stroke care. Information on vital status and covariates including comorbidities and co-medication was obtained from the following population-based medical registries: the Danish Civil Registration System, Danish Medicines Agency's Medical Register and The Danish National Patients Registry. Information was linked using the unique civil registration number assigned to all Danish residents. Multivariable logistic regression was used to compute the adjusted odds ratio (OR) of 30-day mortality in patients treated with antidepressants during admission as compared to patients not treated. In addition, we did stratified analyses on sex, age, stroke severity and propensity score-matched analyses as well as multiple imputation. Results: Among 5,070 consecutive first-ever stroke patients without prior antidepressant treatment, 955 (18.8%) started antidepressant treatment during admission with a median time from admission until treatment of 5 days (interquartile range 2-11). The proportion of patients with severe stroke was higher among treated patients as compared to that among non-treated patients. The adjusted OR of 30-day mortality was 0.28 (95% confidence interval (CI) 0.18-0.43) for patients treated during admission as compared to patients not treated during admission. Stratification by stroke severity showed signs of effect modification, stratification by sex and age did not. Included in the propensity score-matched analyses were 1,908 patients matched 1:1. The propensity score-matched adjusted OR of death within 30 days was 0.31 (95% CI 0.19-0.49). Conclusion: Although early antidepressant treatment was more often started in patients with severe stroke, treatment was associated with significantly lower mortality. This result requires replication in randomized trials; however, it indicates that early start of antidepressant treatment after stroke may be safe and a more routine use may be feasible.