Published online Jan 7, 2015. doi: 10.3748/wjg.v21.i1.377
Peer-review started: July 8, 2014
First decision: August 6, 2014
Revised: August 23, 2014
Accepted: September 29, 2014
Article in press: September 30, 2014
Published online: January 7, 2015
Twaij and colleagues have carried out a meta-analysis of cohort studies to review the currently available literature comparing laparoscopic to open resection of hepatocellular carcinoma in patients with known liver cirrhosis. We congratulate and applaud their important work, but several important issues should be noted.
Core tip: The combination of effect sizes in the meta-analysis is based on several factors: type of design of included studies, type of data, heterogeneity among studies, numerical stability and accuracy and understandability of results. For continuous variables, the standard mean difference is recommended when different measurement scales in the studies are used to reflect the outcomes. However, the weighted mean difference is more suitable when studies included use the same scale to report the outcomes.