The review question and inclusion criteria were generally clear. The authors searched a range of relevant sources. The search was limited to studies in English, so some relevant studies may have been missed. Publication bias was not assessed and the risk of this bias in the review was uncertain.
Initial study selection was done by a single reviewer, so there was a risk of reviewer errors and bias at this stage. Appropriate methods were used to minimise errors and bias in other review processes.
Validity of the included studies was assessed using a standard checklist. The authors distinguished between evidence from RCTs and non-randomised studies in the synthesis. It was unclear whether the included non-randomised studies were experimental or observational, but all were at high risk of bias. Most of the included studies had HRQoL as a secondary outcome and the studies may have been too small to detect any differences. A narrative synthesis was appropriate in view of the heterogeneity of outcome measures in the included studies. Heterogeneity in interventions was difficult to assess as no details of the ERAS or conventional protocols were reported.
The authors' conclusions broadly reflect the evidence presented, but the limitations of the evidence suggest that the conclusions should be regarded as provisional. The authors' recommendations for further research seem appropriate.