Graphical summary

What is known? 

Mental disorders are common in critically ill patients and are associated with unfavorable outcomes. Mental disorders are typically treated with antidepressants and/or antipsychotics. Of these, selective serotonin reuptake inhibitors (SSRIs) are the first-line pharmacological treatments, with approximately one in five critically ill patients receiving them before ICU admission. In 2014, a retrospective study found that the use of SSRIs prior to ICU admission was associated with increased in-hospital mortality in critically ill patients [Chest 2014;145:745-752].

What did we do?

In the present study, we employed a large-sample retrospective cohort design to determine the association between pre-ICU use of SSRIs and in-hospital mortality in critically ill adults with mental disorders. The study is based on the Medical Information Mart in Intensive Care-IV version 2.0 database, containing electronic health records from over 50000 patients admitted to the ICUs at the Beth Israel Deaconess Medical Center (Boston, Massachusetts, USA) from 2008 to 2019. All critically ill adults (age ≥18 years) with mental disorders were selected from the database. Mental disorders, including depression, anxiety, schizophrenia, bipolar disorders, and other unspecified mental disorders, were identified by the International Classification of Diseases codes. The exposure was the use of SSRIs during the period after hospital admission and before ICU admission and extracted from the prescriptions in the database. This study included six common SSRIs prescriptions: citalopram, escitalopram, fluoxetine, sertraline, paroxetine, and fluvoxamine. The outcome was in-hospital mortality. Time-dependent Cox proportional hazards regression models were used to estimate the adjusted hazard ratio (aHR) with 95% confidence interval (CI). To further test the robustness of the results, we performed propensity score matching and marginal structural Cox model estimated by inverse probability of treatment weighting.

What did we find?

The original cohort identified 16601 patients. Of those, 2232 (13.4%) received pre-ICU SSRIs, and 14369 (86.6%) did not. Matched cohort obtained 4406 patients, with 2203 patients in each group (SSRIs users vs. non-users). In the original cohort, pre-ICU use of SSRIs was associated with a 24% increase in the hazard for in-hospital mortality (aHR, 1.24; 95% CI, 1.05-1.46; P = 0.010). The results were robust in the matched cohort (aHR, 1.26; 95% CI, 1.02-1.57; P = 0.032) and the weighted cohort (aHR, 1.43; 95% CI, 1.32-1.54; P < 0.001). In summary, pre-ICU use of SSRIs is associated with an increase in the hazard for in-hospital mortality in critically ill adults with mental disorders.