Intravitreal Bevacizumab and Cardiovascular Risk in Patients with Age-Related Macular Degeneration: Systematic Review and Meta-Analysis of Randomized Controlled Trials and Observational Studies.

INTRODUCTION: Intravitreal bevacizumab (IVTB) is used to treat age-related macular degeneration (ARMD), although its use is off-label and its cardiovascular safety has not been unequivocally established.
OBJECTIVES: Our objective was to assess the cardiovascular safety of IVTB in patients with ARMD.
METHODS: We conducted a systematic review and meta-analysis of published randomized controlled trials (RCTs) and observational studies.
RESULTS: Of the 2028 non-duplicate records, five RCTs versus ranibizumab (N = 3038, 12/24 months), four RCTs comparing different regimens (N = 809, 12/23 months), one RCT versus pegaptanib, photodynamic therapy (PDT), or sham (N = 131, 12 months), and three observational studies versus PDT, ranibizumab, or pegaptanib (~150,000 or 1666 patients/12 months and 317 patients/1-2 years, respectively) had a low risk of bias/high quality and ≥20 patients per arm with ≥6 months and ≥3 injections of treatment. RCT-based comparisons with PDT or pegaptanib are negligible. Observational data have not demonstrated differences [all-cause mortality, myocardial infarction (MI), stroke], but the level of evidence is "very low" (imprecise, indirect). RCT-based comparisons with ranibizumab did not demonstrate differences regarding some outcomes, although certain point estimates were at the level of a relevant harm/benefit [all-cause mortality odds ratio (OR) 1.103, 95 % confidence interval (CI) 0.641-1.898; vascular mortality OR 1.380, 95 % CI 0.476-3.997; MI OR 0.551, 95 % CI 0.265-1.146; stroke OR 0.657, 95 % CI 0.260-1.660; transitory ischemic attack OR 1.536, 95 % CI 0.444-5.313; atherothrombotic events (ATEs) OR 1.007, 95 % CI 0.641-1.593; venous thromboembolism OR 2.325, 95 % CI 0.963-5.612] or suggested a higher risk with bevacizumab (hypertension OR 7.512, 95 % CI 1.056-52.3), but estimates were based on sparse data, were extremely imprecise, and commonly exhibited considerable heterogeneity/inconsistency. The level of evidence per outcome was "low" or "very low". Observational data did not demonstrate difference (all-cause mortality, MI, stroke), or suggested a higher risk with bevacizumab (ATE), but were imprecise and indirect (level of evidence "very low"). RCT-based comparisons of different IVTB regimens suffered from the same limitations.
CONCLUSION: Published data on IVTB in AMRD provide only a low level of evidence on its cardiovascular safety and do not support any finite conclusions.