Midazolam Premedication Immediately Before Surgery Is Not Associated With Early Postoperative Delirium

BACKGROUND: Postoperative delirium is common among older surgical patients and may be associated with anesthetic management during the perioperative period. The aim of this study is to assess whether intravenous midazolam, a short-acting benzodiazepine used frequently as premedication, increased the incidence of postoperative delirium. METHODS: Analyses of existing data were conducted using a database created from 3 prospective studies in patients aged 65 years or older who underwent elective major noncardiac surgery. Postoperative delirium occurring on the first postoperative day was measured using the confusion assessment method. We assessed the association between the use or nonuse of premedication with midazolam and postoperative delirium using a χ2 test, using propensity scores to match up with 3 midazolam patients for each control patient who did not receive midazolam. RESULTS: A total of 1266 patients were included in this study. Intravenous midazolam was administered as premedication in 909 patients (72%), and 357 patients did not receive midazolam. Those who did and did not receive midazolam significantly differed in age, Charlson comorbidity scores, preoperative cognitive status, preoperative use of benzodiazepines, type of surgery, and year of surgery. Propensity score matching for these variables and American Society of Anesthesiology physical status scores resulted in propensity score-matched samples with 1-3 patients who used midazolam (N = 749) for each patient who did not receive midazolam (N = 357). After propensity score matching, all standardized differences in preoperative patient characteristics ranged from -0.07 to 0.06, indicating good balance on baseline variables between the 2 exposure groups. No association was found between premedication with midazolam and incident delirium on the morning of the first postoperative day in the matched dataset, with odds ratio (95% confidence interval) of 0.91 (0.65-1.29), P = .67. CONCLUSIONS: Premedication using midazolam was not associated with higher incidence of delirium on the first postoperative day in older patients undergoing major noncardiac surgery.