Risk of bias assessment
Only two studies included in the review (Singh et al. and Stelter et al.) had a clear statement regarding the randomization and allocation process. This raises concerns about the possibility of bias in the interpretation of the review results. The complete risk of bias assessment is summarized in Table 4.
The strength of the evidence assessed by GRADE was low to moderate due to the small number of events, to the risk of bias of the included studies and to the high heterogeneity among the studies (Figure 4).