Purpose: Postoperative delirium (POD) occurs in about one-third of patients and the incidence increases with age. The rate of missed delirium is 60%-70%, and there has recently been a shift from early detection and treatment to preventive strategies to reduce risk. While GABAA receptor agonists (GRAs) are a risk factor for delirium, discontinuation may also worsen delirium through withdrawal or worsening of insomnia. This study aimed to evaluate the effect of switching from daily preoperative GRA therapy to lemborexant monotherapy on the incidence of postoperative delirium.
Methods: A retrospective study was conducted in cancer patients aged ≥ 75 years who visited the Department of Psycho-Oncology at the National Cancer Center Hospital in Japan and were taking a GRA daily for insomnia before surgery under general anesthesia. Delirium was screened at least once daily with the Nursing Delirium Screening Scale on postoperative days 0-6 (positive if ≥ 2). We used a two-group intention-to-treat (ITT) framework, classifying patients according to a psycho-oncologist-verified preoperative switch to lemborexant.
Results: Fifty patients satisfied eligibility. POD occurred in 1/17 (5.9%) in the Switch group and 15/33 (45.5%) in the Continue group; RD -39.6 percentage points (95% CI -59.9 to -19.2), RR 0.13 (95% CI 0.02-0.90), OR 0.08 (95% CI 0.01-0.63), p = 0.005.
Conclusion: A preoperative switch policy to lemborexant monotherapy was associated with a lower observed risk of POD. Given the observational, unadjusted design, these associations are hypothesis-generating and cannot establish causality.
Keywords: Delirium prevention; GABAA receptor agonists; Insomnia; Lemborexant; Postoperative delirium.
© 2025. The Author(s).