Meta-analysis
In the light of the debate in the literature regarding the association between macrolides and CHD, we performed a meta-analysis using RevMan 5.3. We updated the literature review described above to February 2020. For all macrolides (J01FA), we combined our results with that of a recent meta-analysis of CHD and macrolides29 and one new study13 not included in the published meta-analysis. We took into consideration the heterogeneity result from the recent CHD-macrolide meta-analysis and also investigated heterogeneity among studies included in our meta-analysis. Assuming OR provide an unbiased estimate of relative risk in rare outcomes as CA,38 we combined case-control and cohort studies to obtained pooled OR. Statistical significance level was set at 0∙05 and results were reported as OR and 95% CI.