Nine trials were included in the review (n=667 eyes). Four trials were judged to be at low risk of selection bias; the rest were rated as moderate risk. Three trials used sham interventions which accounted for performance bias; five trials used blinded outcome assessors; three trials were judged to be at high risk of detection bias. All trials were rated as low risk of attrition bias.
Best corrected visual acuity: One trial demonstrated significant mean differences in visual acuity in favour of bevacizumab versus sham treatment (WMD -0.18, 95% CI -0.28 to -0.08), and bevacizumab plus triamcinolone versus sham treatment (WMD -0.15, 95% CI -0.26 to-0.04). One trial also reported significant benefits for the bevacizumab alone versus photodynamic therapy (WMD -0.09, 95% CI -0.13 to -0.06), bevacizumab plus photodynamic therapy versus bevacizumab alone (WMD -0.14, 95% CI -0.18 to -0.11) and bevacizumab plus photodynamic therapy versus bevacizumab alone (WMD -0.24, 95% CI -0.27 to -0.20). There were no other statistically significant differences for the other comparisons.
Improvements in best corrected visual acuity (all definitions): Patients receiving bevacizumab were significantly more likely to present improved visual acuity than those receiving photodynamic therapy alone or with triamcinolone (RR 0.49, 95% CI 0.31 to 0.78; NNT=4, 95% CI 1 to 4).
Adverse events: The most common adverse events associated with bevacizumab were moderate anterior chamber reaction, transient anterior chamber reaction, iris neovascularisation, subconjunctival haemorrhage, posterior vitreous detachment, and foreign body sensation. Further details were listed in the paper.