2.7. Data Analysis

Statistical analysis of data was carried out using the RevMan 5.3 software. Firstly, the existence of the clinical heterogeneities among different studies was assessed by the χ²-test and I²-judgment. If a small heterogeneity was present (P > 0.1, I² < 50%), the fixed effects model (FEM) was used for the analysis. If clinical heterogeneity was present (P ≤ 0.1, I² ≥ 50%), the heterogeneity source was discussed and the subgroup analysis was performed under the necessary conditions [19]. For various test results of the evident clinical heterogeneity, a random-effects model (REM) was employed for the analysis. The continuous variables used the mean difference (MD) or weight mean difference (WMD) as the effect analysis. The odds risk was determined for the classified variables, and a 95% confidence interval (CI) was calculated by the effect analysis. The test standard was α = 0.05. A funnel graph was drawn to evaluate the potential publication bias.