Pharmacological Agents That Prevent Postoperative Cognitive Dysfunction in Patients With General Anesthesia: A Network Meta-analysis

摘要

Background: Postoperative cognitive dysfunction (POCD) is associated with prolonged hospital stays, increased mortality, and negative socioeconomic consequences. Dexmedetomidine, ketamine, dexamethasone, and lidocaine have previously been reported to be effective for preventing POCD.Study question: In this network meta-analysis, we apply direct and indirect comparisons to rank these pharmacological agents in terms of their effect on POCD, through which we seek to provide evidence for future clinical medication.Data sources: A comprehensive literature search of PubMed, EMBASE, the Cochrane Library, and Web of Science was conducted to identify randomized controlled trials that examined the effects of dexmedetomidine, ketamine, dexamethasone, or lidocaine on POCD induced by general anesthesia.Study design: For eligible studies, 2 reviewers independently extracted data and assessed the respective risk of bias. Bayesian network meta-analysis was conducted using R-3.4.1 software. A total of 30 articles were included in this meta-analysis.Results: Direct comparison showed that dexmedetomidine [odds ratio (OR) 5 0.42, 95% confidence interval (CI): 0.32-0.55] may decrease the incidence of POCD for noncardiac surgery; dexmedetomidine (OR 5 0.08, 95% CI: 0.01-0.63) and ketamine (OR 5 0.09, 95% CI: 0.02-0.32) were found to decrease POCD for cardiac surgery. The R-software ranking result for prevention of POCD was dexmedetomidine, lidocaine, ketamine, placebo, and dexamethasone, respectively, in noncardiac surgery, and dexmedetomidine, ketamine, lidocaine, placebo, and dexamethasone in cardiac surgery.Conclusion: Dexmedetomidine exhibited obvious superiority to other agents for noncardiac surgery; dexmedetomidine and ketamine exhibited a significantly better performance than other agents for cardiac surgery. Dexamethasone did not show better efficacy than the placebo. However, more rigorously designed studies comprising larger sample sizes are needed to confirm our findings.