Background: Postoperative delirium (POD) is a serious and common complication. The aim of present study is to investigate the diurnal variation of POD and the effects of esketamine in elderly patients.

Methods: A randomized, double-blind, placebo-controlled clinical trial with factorial design was conducted. Patients (aged 65 to 85 years) with normal Mini-Mental State Examination (MMSE) score were stratified by age (≤70 vs. >70) and American Society of Anesthesiologists physical status classification (Ⅱ vs. Ⅲ), then randomly assigned to either morning (08:00-12:00) or afternoon (14:00-18:00) noncardiac operation under general anesthesia with or without esketamine administration (0.2 mg/kg). The primary outcome was the incidence of POD (3-Minute Diagnostic Interview for Confusion Assessment Method-defined Delirium, 3D-CAM) on postoperative days 1, 3, and 7. The secondary outcomes were the scores of MMSE and Hospital Anxiety and Depression Scale. The intention-to-treat analysis of the outcomes were performed by generalized estimating equation.

Results: Six patients who did not receive an intervention because of canceled operation were excluded after randomization. The datasets containing 426 cases were analyzed following the intention-to-treat principle after handling missing data via multiple imputation method. The incidence of POD declined from about 55% on postoperative day 1 to 31 and 18% on postoperative days 3 and 7, respectively. Afternoon operation [B=-0.583, OR (95% CI) 0.558 (0.319-0.976); P=0.041], but not esketamine, significantly decreased the incidence of POD. Both esketamine and operation time failed to significantly affect MMSE, HAD, and NRS score. There was no interaction among operation time, esketamine, and follow up time.

Conclusion: Elderly patients undergoing elective noncardiac surgery in the afternoon displayed lower POD incidence than those operated in the morning. A single low-dose of esketamine before general anesthesia induction failed to significantly decrease the risk of POD but decrease the risk of intraoperative hypotension and emergence agitation.