Effect of sedative premedication with oral midazolam on postanesthesia care unit delirium in older adults: a secondary analysis following an uncontrolled before-after design

Abstract

Abstract Background Sedative premedication with benzodiazepines has been linked with prolonged recovery and inadequate emergence during the immediate postoperative period. We aimed to analyze the association between postanesthesia care unit (PACU) delirium and sedative premedication with oral midazolam. Methods We performed a secondary analysis of prospectively collected data before (midazolam cohort) and after (non-midazolam cohort) implementation of a restrictive strategy for oral premedication with midazolam. From March 2015 until July 2018, we included patients 60 years and older, who underwent elective radical prostatectomy for prostate cancer. Exclusion criteria were contraindications to premedication with midazolam, preoperative anxiety, and a history of neurological disorders. Patients, who were scheduled for postoperative admission to the intensive care unit, were excluded. Between 2015 and 2016, patients received 7.5 mg oral midazolam preoperatively (midazolam cohort). Patients included between 2017 and 2018 did not receive any sedative medication preoperatively (non-midazolam cohort). The primary endpoint was the incidence of PACU delirium. Results PACU delirium rates were 49% in the midazolam cohort (n = 214) and 33% in the non-midazolam cohort (n = 218). This difference was not statistically significant on multivariable logistic regression analysis (OR 0.847 [95% CI 0.164; 4.367]; P = 0.842). Age (OR 1.102 [95% CI 1.050; 1.156]; P < 0.001), the cumulative dose of sufentanil (OR 1.014 [95% CI 1.005; 1.024]; P = 0.005), and propofol-sufentanil for anesthesia maintenance (OR 2.805 [95% CI 1.497; 5.256]; P = 0.001) were significantly associated with PACU delirium. Conclusion Midazolam for sedative premedication was not significantly associated with PACU delirium. The reduction in the incidence of PACU delirium throughout the study period may be attributable to improvements in perioperative management other than a more restrictive preoperative benzodiazepine administration.