The safety of bevacizumab and ranibizumab in clinical studies.

In comparing the safety of ranibizumab versus bevacizumab in age-related macular degeneration, the meta-analyses published thus far have given conflicting results, particularly about the risk of venous thrombotic events and ocular inflammation. From the comparison of the design and the findings of these meta-analysis, we tried to identify the potential explanations to account for these discrepancies. We separately evaluated the incidence of ocular inflammation with the two agents between randomized studies and non-randomized studies. While no increase in risk was found in randomized studies, non-randomized studies showed an increase in risk for bevacizumab versus ranibizumab. One interpretation of these findings is that bevacizumab itself does not represent any increase in risk of ocular inflammation and/or cardiovascular events under the rigorous conditions of a randomized study, but this agent can be at the origin of an increase in risk when administered in the "real world"; this setting could in fact leave space for less strictly controlled preparation of aliquots for intravitreal injection.