Dose-dependent relationship between intra-operative dexmedetomidine and delirium after non-cardiac surgery: a retrospective cohort study.

Abstract

Dexmedetomidine can attenuate delirium in patients who are critically ill, but evidence with regards to its preventive effect on postoperative delirium remains equivocal. We hypothesised that the risk of delirium after intra-operative dexmedetomidine administration varies depending on the dose administered and aimed to identify the optimum dose to mitigate delirium. We included 114,786 adults undergoing general anaesthesia for non-cardiac, non-transplant surgery. Primary exposure was intra-operative dexmedetomidine dose in cumulative μg.kg^-1 body weight, dichotomised into high vs. low dose based on the cohort median (0.49 μg.kg^-1). Primary outcome was delirium within 7 days, identified from discharge notes, Confusion Assessment Method assessments and diagnostic codes. A total of 4804 (4.2%) patients received dexmedetomidine, with a median (IQR [range]) cumulative dose of 0.49 (0.28-0.84 [0.01-2.50]) μg.kg^-1. Postoperative delirium occurred in 3227 (2.8%) patients. Compared with no dexmedetomidine, the risk of delirium was lower in patients receiving low doses (≤ 0.49 μg.kg^-1) of dexmedetomidine (adjusted odds ratio 0.61, 95%CI 0.44-0.85, p = 0.004), but not among those receiving high doses (> 0.49 μg.kg^-1) (adjusted odds ratio 1.06, 95%CI 0.84-1.34, p = 0.62). Fractional polynomial regression analyses suggested that doses between 0.25 μg.kg^-1 and 0.35 μg.kg^-1 were associated with the lowest delirium risk. Threshold regression and restricted cubic splines confirmed these findings. Low, but not high, dose dexmedetomidine administration was associated with lower risks of delirium, with optimal doses ranging between 0.25 μg.kg^-1 and 0.35 μg.kg^-1.