Abstract: Objective To evaluate the efficacy and safety in treatment of diabetic macular edema (DME) by intravitreal bevacizumab (IVB) with intravitreal triamcinolone acetonide (IVT) for first priority. Design Meta analysis. Participants Published literature in Medline, EMbase, Cochrane library and CBM which comparing IVB with IVT in treatment of DME for first priority. Methods According to evaluation guidelines of Cochrane collaboration, clinical controlled trials (CCTs) were searched using Medline, EMbase, the Cochrane Library and CBM. Methodology qualities of literatures were performed by experienced researchers according to the Jadad Score. RevMan 5.1 offered by Cochrane was used to do the meta-analysis. Main Outcome Measures central macular thickness, best corrected visual acuity and adverse events. Results Only 7 literatures (417 eyes) were selected into meta-analysis. The result of analysis showed that IVT group had a more reduced central macular thickness with -55.34 μm (95%CI: -104.51- -6.17 μm, P=0.03) than IVB group. However, no significant difference in visual acuity was observed between IVB and IVT group (P=0.06). Adverse event rates (mainly elevation of intraocular pressure) in IVT group was 14.3 times (95%CI: 4-50 times) more than IVB group (P=0.000).  Conclusion Meta-analysis shows that IVT group had a more reduced central macular thickness and equal visual acuity compared with IVB group in treatment of DME, while the adverse event rates were much more than IVB group. More high quality perspective studies are still required for further analysis. (Ophthalmol CHN, 2012, 21: 60-64)

Key words:  bevacizumab, triamcinolone, diabetic macular edema, meta-analysis