OBJECTIVE: Remimazolam tosylate is an ultra‑short‑acting benzodiazepine sedative. This study aimed to determine a suitable intraoperative remimazolam dose for older adults undergoing hip arthroplasty with neuraxial anesthesia and to explore its association with postoperative delirium (POD) and perioperative stress responses. METHODS: In this randomized study, 153 older adults undergoing hip arthroplasty with neuraxial anesthesia were allocated to four groups: saline control (N), remimazolam 0.1 mg/(kg•h) (R1), 0.2 mg/(kg•h) (R2), and 0.3 mg/(kg•h) (R3). Primary outcomes were intraoperative sedation success (modified Observer's Assessment of Alertness/Sedation scale), plasma cortisol and norepinephrine concentrations at operating room entry and exit, and incidence of POD. RESULTS: In the R2 group, sedation success was 100%, and the mean awakening time was 6.82 ± 0.64 min, shorter than in the R3 group (8.16 ± 0.93 min), which exhibited deeper sedation and longer recovery. No statistically significant differences were observed in POD incidence across groups (2.86-6.00%; p = 0.922), and the study was underpowered to detect modest between‑group differences. Compared with saline, the remimazolam groups had lower cortisol and norepinephrine concentrations at operating room exit, whereas within‑group changes from entry to exit did not reach statistical significance (p > 0.05). CONCLUSION: Compared with saline, the remimazolam groups had lower cortisol and norepinephrine concentrations at operating room exit, whereas within‑group changes from entry to exit did not reach statistical significance (p > 0.05). Within the observed POD incidence range (2.86-6.00%), remimazolam did not appear to increase POD over the first 7 postoperative days compared with saline, although the study was not powered to detect modest differences.
