Comparative Safety Analysis of Oral Antipsychotics for In-Hospital Adverse Clinical Events in Older Adults After Major Surgery : A Nationwide Cohort Study

Abstract

Background: Antipsychotics are commonly used to manage postoperative delirium. Recent studies reported that haloperidol use has declined, and atypical antipsychotic use has increased over time.

Objective: To compare the risk for in-hospital adverse events associated with oral haloperidol, olanzapine, quetiapine, and risperidone in older patients after major surgery.

Design: Retrospective cohort study.

Setting: U.S. hospitals in the Premier Healthcare Database.

Patients: 17 115 patients aged 65 years and older without psychiatric disorders who were prescribed an oral antipsychotic drug after major surgery from 2009 to 2018.

Interventions: Haloperidol (≤4 mg on the day of initiation), olanzapine (≤10 mg), quetiapine (≤150 mg), and risperidone (≤4 mg).

Measurements: The risk ratios (RRs) for in-hospital death, cardiac arrhythmia events, pneumonia, and stroke or transient ischemic attack (TIA) were estimated after propensity score overlap weighting.

Results: The weighted population had a mean age of 79.6 years, was 60.5% female, and had in-hospital death of 3.1%. Among the 4 antipsychotics, quetiapine was the most prescribed (53.0% of total exposure). There was no statistically significant difference in the risk for in-hospital death among patients treated with haloperidol (3.7%, reference group), olanzapine (2.8%; RR, 0.74 [95% CI, 0.42 to 1.27]), quetiapine (2.6%; RR, 0.70 [CI, 0.47 to 1.04]), and risperidone (3.3%; RR, 0.90 [CI, 0.53 to 1.41]). The risk for nonfatal clinical events ranged from 2.0% to 2.6% for a cardiac arrhythmia event, 4.2% to 4.6% for pneumonia, and 0.6% to 1.2% for stroke or TIA, with no statistically significant differences by treatment group.

Limitation: Residual confounding by delirium severity; lack of untreated group; restriction to oral low-to-moderate dose treatment.

Conclusion: These results suggest that atypical antipsychotics and haloperidol have similar rates of in-hospital adverse clinical events in older patients with postoperative delirium who receive an oral low-to-moderate dose antipsychotic drug.

Primary funding source: National Institute on Aging.

Figure 1.. Selection of Study Population

Figure 2.. Postoperative Oral Antipsychotic Drug Use and In-Hospital Adverse Events

The propensity score overlap-weighted analysis was performed to evaluate the association of antipsychotics with in-hospital adverse clinical events according to the intention-to-treat analysis (follow-up truncated at 14 days from the drug initiation). Abbreviations: CI, confidence interval; RD, risk difference; RR, risk ratio; TIA, transient ischemic attack.

Figure 3.. Subgroup Analysis of Postoperative Oral Antipsychotic Drug Use and In-Hospital Death

The propensity score overlap-weighted analysis was performed to evaluate the association of antipsychotics with in-hospital death according to the intention-to-treat analysis (follow-up truncated at 14 days from the drug initiation) with antipsychotic dose adjustment. Abbreviations: CI, confidence interval; RD, risk difference; RR, risk ratio.