Anaesthetic drugs and survival: a Bayesian network meta-analysis of randomized trials in cardiac surgery.

Many studies have compared desflurane, isoflurane, sevoflurane, total i.v. anaesthesia (TIVA), or all in cardiac surgery to assess their effects on patient survival. We performed standard pairwise and Bayesian network meta-analyses; the latter allows indirect assessments if any of the anaesthetic agents were not compared in head-to-head trials. Pertinent studies were identified using BioMedCentral, MEDLINE/PubMed, Embase, and the Cochrane Library (last updated in June 2012). We identified 38 randomized trials with survival data published between 1991 and 2012, with most studies (63%) done in coronary artery bypass grafting (CABG) patients with standard cardiopulmonary bypass. Standard meta-analysis showed that the use of a volatile agent was associated with a reduction in mortality when compared with TIVA at the longest follow-up available [25/1994 (1.3%) in the volatile group vs 43/1648 (2.6%) in the TIVA arm, odds ratio (OR)=0.51, 95% confidence interval (CI) 0.33-0.81, P-value for effect=0.004, number needed to treat 74, I(2)=0%] with results confirmed in trials with low risk of bias, in large trials, and when including only CABG studies. Bayesian network meta-analysis showed that sevoflurane (OR=0.31, 95% credible interval 0.14-0.64) and desflurane (OR=0.43, 95% credible interval 0.21-0.82) were individually associated with a reduction in mortality when compared with TIVA. Anaesthesia with volatile agents appears to reduce mortality after cardiac surgery when compared with TIVA, especially when sevoflurane or desflurane is used. A large, multicentre trial is warranted to confirm that long-term survival is significantly affected by the choice of anaesthetic.