No Impact of Preadmission Anti-Inflammatory Drug Use on Risk of Depression and Anxiety After Critical Illness*

Objectives: Risk of depression and anxiety is elevated after intensive care. Drugs with anti-inflammatory properties may have antidepressant and anxiolytic effects. The aim of this study was to investigate the association between preadmission use of drugs with anti-inflammatory effects and risk of new-onset depression and anxiety among adult patients admitted to an ICU. Design: Propensity score–matched, population-based cohort study. Setting: All ICUs in Denmark from 2005 to 2013. Patients: Adults receiving mechanical ventilation in an ICU. Interventions: None. Measurements and Main Results: A total of 48,207 ICU patients were included. Exposures were preadmission single-agent or combined use of statins, nonsteroidal anti-inflammatory drugs, or glucocorticoids. Outcomes were cumulative incidence (risk) and risk ratio of new-onset psychiatrist-diagnosed depression or anxiety or prescriptions for antidepressants or anxiolytics. Propensity score matching yielded 6,088 statin user pairs, 2,886 nonsteroidal anti-inflammatory drug user pairs, 1,440 glucocorticoid user pairs, and 1,743 combination drug user pairs. The cumulative incidence of anxiety and depression during the 3 years following intensive care was 18.0% (95% CI, 17.0–19.0%) for statin users, 21.3% (95% CI, 19.8–22.9%) for nonsteroidal anti-inflammatory drug users, 17.4% (95% CI, 15.4–19.5%) for glucocorticoid users, and 19.0% (95% CI, 16.3–20.2%) for combination users. The cumulative incidence was similar in nonusers compared with users in all drug groups. The risk ratio of depression and anxiety 3 years after admission to ICU was 1.04 (95% CI, 0.96–1.13) for statin users, 1.00 (95% CI, 0.90–1.11) for nonsteroidal anti-inflammatory drug users, 0.97 (95% CI, 0.82–1.14) for glucocorticoid users, and 1.05 (95% CI, 0.90–1.21) for combination users, compared with nonusers. Results were consistent across subgroups (gender, age, preadmission diseases, type of admission) and sensitivity analyses (depression and anxiety separately). Conclusions: Preadmission use of statins, nonsteroidal anti-inflammatory drugs, glucocorticoids, or combinations did not alter the risk of depression and anxiety after critical illness.