The Use of Antipsychotic Medications in Cardiac Surgery: Prevalence, Trajectory, Risk Factors and Outcomes

Background: Patients undergoing cardiac surgery are at risk for postoperative delirium and antipsychotic medication (APM) therapy. However, the prevalence, risk factors, and outcomes associated with these therapies remain unclear. Methods: We retrospectively studied patients who underwent cardiac surgery between 2012 and 2021 using electronic databases. We obtained data on the use of APMs. We used cardinality matching to identify the patient and surgical characteristics associated with postoperative APM therapy and to study whether APM therapy was associated with outcomes. Results: We studied 3,188 patients. Of these, 13.6% (433/3,188) received APMs in the form of quetiapine (85.0%), haloperidol (41.1%), olanzapine (22.2%), or risperidone (2.1%). APM therapy lasted a median of 3 days at a median daily dose of 2.5 mg olanzapine equivalents, most commonly first administered on day 2 (interquartile range 1.7–3.9), after extubation, and before discharge from the intensive care unit (ICU). Haloperidol use decreased during the study period. Cardinality matching revealed that longer cardiopulmonary bypass time, mechanical ventilation, longer ICU stay, and postoperative acute kidney injury were associated with APM therapy. After matching, there was no difference between APM and control patients for mortality (absolute risk reduction [aRR] 0.94; 95% confidence interval [CI] 0.46–1.89; p=0.85) or hospital readmission rates (aRR 0.95; 95% CI 0.74–1.20; p=0.66). APM therapy, however, was associated with discharge to a destination other than home (aRR 1.39; 95% CI 1.11–1.66; p<0.001). Only a minority (12.4%) of those who received postoperative APM received further prescriptions upon discharge. Conclusions: APM therapy was common in cardiac surgery patients. APM therapy was an average of 3 days. The risk factors included longer cardiopulmonary bypass and ventilation, longer ICU stay, and acute kidney injury. APM therapy was not associated with mortality or hospital readmission. However, this was associated with decreased discharge to the home. Prescriptions upon discharge and at the 6 months following were rare.