2.5 Statistical analysis
We performed this systematic review and meta-analysis following the
Cochrane Collaboration and the Preferred Reporting Items for Systematic
Reviews and Meta-Analysis (PRISMA) statement
guidelines.13,14 Treatment effects were compared using
mean differences (MD) and 95% confidence intervals (CI) for continuous
outcomes, and risk ratio (RR) with 95% CI for binary endpoints using
the DerSimonian and Laird random-effects model. We used the Cochran Q
test and I2 statistics to assess for heterogeneity; P
values inferior to 0.01 were considered significant for heterogeneity.
We considered I2 < 25%; 25% <
I2 < 75%, and I2> 75% as low, medium and high heterogeneity, respectively.
Review Manager 5.1.7 (Cochrane Centre, The Cochrane Collaboration,
Denmark) was used for statistical analysis.
Sensitivity analysis
To explore the robustness of the results and identify possible outliers,
we performed a leave-one-out sensitivity analysis for outcomes with high
heterogeneity, by systematically removing each study from the pooled
result. Additionally, another sensitivity analysis was performed by
removing studies with a high risk of bias.