Comparative effectiveness of pharmacological interventions to prevent postoperative delirium: a network meta-analysis

Many pharmacologic agents were investigated for the effect to prevent delirium. We aimed to comprehensively compare the effect of the pharmacological interventions to prevent postoperative delirium. A Bayesian network meta-analysis of randomized trials was performed using random effects model. PubMed, the Cochrane Central Register of Controlled Trials, and Embase were searched on 20 January 2021. Randomized trials comparing the effect of a drug to prevent postoperative delirium with another drug or placebo in adult patients undergoing any kind of surgery were included. Primary outcome was the postoperative incidence of delirium. Eighty-six trials with 26,992 participants were included. Dexmedetomidine, haloperidol, and atypical antipsychotics significantly decreased the incidence of delirium than placebo [dexmedetomidine: odds ratio 0.51, 95% credible interval (CrI) 0.40-0.66, moderate quality of evidence (QOE); haloperidol: odds ratio 0.59, 95% CrI 0.37-0.95, moderate QOE; atypical antipsychotics: odds ratio 0.27, 95% CrI 0.14-0.51, moderate QOE]. Dexmedetomidine and atypical antipsychotics had the highest-ranking probabilities to be the best. However, significant heterogeneity regarding diagnostic time window as well as small study effects precludes firm conclusion.