Systematic review with meta-analysis of relative risk of prolonged times to tracheal extubation with desflurane versus sevoflurane or isoflurane

The objective of this systematic review was to estimate the relative risk of prolonged times to tracheal extubation with desflurane versus sevoflurane or isoflurane. Prolonged times are defined as ≥15 min from end of surgery (or anesthetic discontinuation) to extubation in the operating room. They are associated with reintubations, naloxone and flumazenil administration, longer times from procedure end to operating room exit, greater differences between actual and scheduled operating room times, longer times from operating room exit to next case start, longer durations of the workday, and more operating room personnel idle while waiting for extubation. Published randomized clinical trials of humans were included. Generalized pivotal methods were used to estimate the relative risk of prolonged extubation for each study from reported means and standard deviations of extubation times. The relative risks were combined using DerSimonian-Laird random effects meta-analysis with Knapp-Hartung adjustment. From 67 papers, there were 78 two-drug comparisons, including 5167 patients. Studies were of high quality (23/78) or moderate quality (55/78), the latter due to lack of blinding of observers to group assignment and/or patient attrition because patients were extubated after operating room exit. Desflurane resulted in a 65% relative reduction in the incidence of prolonged extubation compared with sevoflurane (95% confidence interval 49% to 76%, P < .0001) and in a 78% relative reduction compared with isoflurane (58% to 89%, P = .0001). There were no significant associations between studies' relative risks and quality, industry funding, or year of publication (all six meta-regressions P ≥ .35). In conclusion, when emergence from general anesthesia with different drugs are compared with sevoflurane or isoflurane, suitable benchmarks quantifying rapidity of emergence are reductions in the incidence of prolonged extubation achieved by desflurane, approximately 65% and 78%, respectively. These estimates give realistic context for interpretation of results of future studies that compare new anesthetic agents to current anesthetics.