A Bayesian network meta-analysis on the effect of anesthetic drugs in cardiac surgery

We carried out both a pair-wise and a Bayesian network meta-analysis, on 38 randomised trials, to assess how volatile-based anaesthesia (desflurane, isoflurane or sevoflurane) and total intravenous anaesthesia (TIVA) influence patients’ survival after cardiac surgery. A network meta-analysis allow to compare different treatments that were never properly compared. On the basis of statistical inference, it is possible to establish which treatment is superior reaching, through indirect comparison, reliable conclusions otherwise difficult to achieve. The standard meta-analysis showed that the use of a volatile agent was associated with a reduction in mortality when compared to TIVA at the longest follow-up available (25/1994 [1.3%] in the volatile group versus 43/1648 [2.6%] in the TIVA group, odds ratio=0.51, 95% confidence interval 0.33-0.81, p for effect=0.004). The Bayesian network meta-analysis showed that sevoflurane (posterior mean of odds ratio =0.31, 95% credible interval 0.14-0.64) and desflurane (posterior mean of odds ratio =0.43, 95% credible interval 0.21-0.82) were individually associated with a reduction in mortality when compared to TIVA. Anaesthesia with volatile agents appears to reduce mortality after cardiac surgery when compared to TIVA, especially when sevoflurane or desflurane are used.