RESUMO
OBJECTIVE: To estimate the risk ratio (RR) of thromboembolic events in chloroquine and hydroxychloroquine users compared to non-users. METHODS: We systematically reviewed randomized controlled trials (RCTs), using MEDLINE and EMBASE databases from inception to the present, reporting thromboembolic events in chloroquine and hydroxychloroquine users compared to non-users. Four authors independently screened all the records obtained through our search strategy and later revised the selected full-text articles for eligibility, according to our inclusion criteria. The same four authors independently extracted relevant data through a customized data collection form while two other authors assessed the quality of the included RCTs using the Cochrane risk-of-bias tool (Version 2.0). All the disagreements were resolved through discussions among the authors. We calculated the risk ratio (RR) and its respective standard error of developing thromboembolic events in hydroxychloroquine users and non-users for each individual study and pooled the results using a random effects model meta-analysis. We assessed Heterogeneity using the Tau2 and I2, and publication bias using funnel plotting and Egger's regression. The protocol for this systematic review is registered at the PROSPERO database (CRD42021247902). RESULTS: Thirteen RCTs met our eligibility criteria and were included in our analysis (2663 patients). We found that hydroxychloroquine-no study on chloroquine was found-reduced the risk of thromboembolic events by 49% (RR 0.51[IC 95% 0.31-0.84]) with a medium heterogeneity (I2 = 67% and T2 = 0.4948). We did find some asymmetry in the inspection of the funnel plot, which was ruled out through an Egger's regression (p-value = 0.1025). CONCLUSION: Our data reinforce the idea that hydroxychloroquine reduces the risk of thromboembolic events.