Quality assessment:
Two reviewers separately assessed the quality of included trials by Cochrane Collaboration’s tool (39). Following items are included in this tool for risk of bias assessment: I) selection bias: A) random sequence generation, B) allocation concealment; II) reporting bias; III) performance bias; IV) detection bias; IV) attrition bias. Each item was judged as “unclear risk of bias”, “high risk of bias,” or “low risk of bias”. We considered included RCTs as good quality if it had at least low risk of bias for three criteria, “fair” if it had low risk for two criteria, and “weak” if it had low risk for less than two criteria (39).