Impact of Remimazolam on Postoperative Delirium in Elderly Non-Cardiac Surgery Patients Admitted to the ICU: A Retrospective Propensity Score Matched Study

Postoperative delirium (POD) is a common complication. Remimazolam, a novel ultra-short-acting benzodiazepine, has been introduced in anesthesia practice, but its impact on the incidence of POD remains unclear. This study aims to evaluate the association between intraoperative use of remimazolam and the incidence of POD. In this retrospective cohort study, patients aged 65 years and older who underwent elective non-cardiac surgery and were directly transferred to the ICU postoperatively were included. Based on intraoperative use of remimazolam, patients were categorized into the remimazolam group or the non-remimazolam group. The primary outcome was the incidence of POD within seven days postoperatively or until discharge (whichever occurred first), assessed using the CAM-ICU or the 3D-CAM. Propensity score matching (PSM) and multivariable logistic regression analyses were performed to adjust for potential confounders. After PSM, 826 pairs of patients were analyzed. The overall incidence of POD was 10.29%. The incidence of POD was 9.32% in the remimazolam group and 11.26% in the non-remimazolam group (P = 0.225). Multivariable logistic regression showed that remimazolam use was not significantly associated with POD incidence (aOR = 0.810, 95% CI = 0.588-1.113, P = 0.196). Intraoperative use of dexmedetomidine was associated with a reduced incidence of POD (aOR = 0.345, 95% CI = 0.236-0.511, P < 0.001), while postoperative administration of esketamine was an independent risk factor for POD (aOR = 2.644, 95% CI = 1.269-5.156, P = 0.006). In elderly patients undergoing non-cardiac surgery, intraoperative use of remimazolam was not significantly associated with the incidence of postoperative delirium.