Visual acuity after intravitreal triamcinolone for diabetic macular edema refractory to laser treatment: a meta-analysis.

摘要

OBJECTIVE: To quantify the effect on visual acuity of intravitreal triamcinolone for the treatment of laser-refractory diabetic macular edema (DME). STUDY DESIGN: Meta-analysis of eligible studies identified by searching MEDLINE, EMBASE, the Cochrane Library, and Google. Participants: 7 randomized controlled trials and 3 cohort studies. METHODS: A search of the literature between 1950 and September 2008 identified 540 articles. Studies that evaluated the efficacy of triamcinolone for the treatment of DME refractory to laser photocoagulation, reported visual acuity data, and compared the intervention with an appropriate control group were included. Exclusion criteria were studies of non-DME, triamcinolone used as an adjunct to another treatment, and triamcinolone delivery other than intravitreally. RESULTS: Using a random-effects model, there was a statistically significant summary mean difference in visual acuity of -0.313 logarithm of the minimum angle of resolution (logMAR) units (95% CI -0.551, -0.074) after 1 month of follow-up. This difference declined to -0.125 logMAR units (95% CI -0.181, -0.070) by 3 months and to -0.043 logMAR units (95% CI -0.090, 0.003) by 6 months. No evidence of publication bias was present. There was a high level of heterogeneity in this group of studies (meta-analysis of 1-month follow-up data: Q-statistic = 21.987, p< 0.001), attributable primarily to study design. CONCLUSIONS: These meta-analyses demonstrate that intravitreal triamcinolone results in a temporary improvement of visual acuity in patients with laser-refractory DME, with a peak benefit of approximately 3 lines of visual acuity 1 month postinjection.