It is unclear if anaesthesia maintenance with propofol is advantageous or beneficial over inhalational agents. This study is intended to compare the effects of propofol vs. inhalational agents in maintaining general anesthesia on patient-relevant outcomes and patient satisfaction.

Studies were identified by electronic database searches in PubMed ™, EMBASE ™ and the Cochrane ™ library between 01/01/1985 and 01/08/2016. Randomized controlled trials (RCTs) of peer-reviewed journals were studied. Of 6688 studies identified, 229 RCTs were included with a total of 20,991 patients. Quality control, risk assessment of bias, meta-bias, meta-regression and certainty in evidence were performed according to Cochrane.

Common estimates were derived from fixed or random-effects models depending on the presence of heterogeneity. Post-operative nausea and vomiting (PONV) was the primary outcome. Post-operative pain, emergence agitation, time to recovery, hospital length of stay, post-anesthetic shivering and haemodynamic instability were considered secondary outcomes.


The risk for PONV was lower with propofol than with inhalational agents (relative risk (RR) 0.61 [0.53, 0.69],  p  <0.00001). Additionally, pain score after extubation and time in the post-operative anesthesia care unit (PACU) were reduced with propofol (mean difference (MD) – 0.51 [- 0.81, – 0.20],  p  = 0.001, MD – 2.91 min [- 5.47 , – 0.35],  p  = 0.03).

In turn, time to respiratory recovery and tracheal extubation were longer with propofol than with inhalational agents (MD 0.82 min [0.20, 1.45],  p  = 0.01, MD 0.70 min [0.03, 1.38],  p = 0.04, respectively). Notably, patient satisfaction was higher with propofol (RR 1.06 [1.01, 1.10],  p  = 0.02, MD 0.13 [0.00, 0.26],  p  = 0.05). Secondary analyzes supported the primary results.

Based on the present meta-analysis there are several advantages of anesthesia maintenance with propofol over inhalational agents. While these benefits resulted in an increased patient satisfaction, the clinical and economic relevance of these findings is still needed in adequately powered prospective clinical trials.