Pharmacologic prevention of postoperative delirium after on-pump cardiac surgery A meta-analysis of randomized trials

Background: Postoperative delirium is a prevalent and disabling mental disorder in patients undergoing on-pump cardiac surgery. There is some evidence that the use of pharmacological interventions may reduce the risk of developing of postoperative delirium. Therefore, the aim of this meta-analysis was to determine the effect of pharmacologic agents for the prevention postoperative delirium after cardiac surgery. Methods: Randomized controlled trials (RCTs) were identified through a systematic literature search of electronic databases and article references up to October 2016. End points included incidence of postoperative delirium, severity of postoperative delirium, cognitive disturbances of postoperative delirium, duration of postoperative delirium, length of stay in intensive care unit (ICU) and hospital, and short-term mortality. Results: A total of 14 RCTs with an aggregate of 14,139 patients were included. The results of the present meta-analysis show that pharmacologic agents significantly decrease postoperative delirium [relative risk (RR), 0.83; 95% confidence interval (95% CI), 0.75-0.91, P < .00001] and duration of postoperative delirium (RR = -0.37, 95% CI = -0.47 to -0.27, P < .00001) after on-pump cardiac surgery. In addition, subgroup analysis shows that dexamethasone and dexamethasone were associated with a trend toward a reduction in postoperative delirium (RR, 0.45; 95% CI, 0.30-0.66, P < .0001; RR, 0.80; 95% CI, 0.68-0.93, P = .003, respectively). However, our results fail to support the assumption that pharmacologic prophylaxis is associated with a positively reduction in short-term mortality, length of ICU, or hospital stay. Conclusion: This meta-analysis suggests that the perioperative use of pharmacologic agents can prevent postoperative delirium development in patients undergoing cardiac surgery. However, there remain important gaps in the evidence base on a few small studies with multiple limitations. Further large-scale, high-quality RCTs are needed in this area.