A systematic review of endophthalmitis after microincisional versus 20-gauge vitrectomy.

Abstract

Endophthalmitis is a rare but severe complication of vitrectomy. Post-surgical endophthalmitis is suspected to be more frequent after microincisional (23- and 25-gauge) compared with standard (20-gauge) vitrectomy. We conducted a systematic review of studies that compared microincisional and standard vitrectomy by searching MEDLINE and EMBASE up to November 2012. We used the Bayesian meta-analysis method to compute the odds ratio (OR) of endophthalmitis. We conducted subgroup analyses to compare the effect of different incision types and use of perioperative antibiotics. We identified 3 small randomized and 18 nonrandomized studies that reported 68 cases of endophthalmitis in 148 643 participants. The overall OR of endophthalmitis for microincisional versus standard vitrectomy was 2.3 (95% credible interval [CrI], 0.8-5.8). We found an increased risk of endophthalmitis using a microincisional straight approach compared with standard vitrectomy (OR, 15.1; 95% CrI, 2.01-179), but not for a beveled approach (OR, 0.82; 95% CrI, 0.23-2.28). The OR of studies that reported on mixed microincision was between these 2 values (OR, 4.4; 95% CrI, 1.32-14.3). We estimated that the overall rate of endophthalmitis with 20-gauge vitrectomy was 3 cases in 10 000 procedures, and the probability that a beveled microincision increases the rate of endophthalmitis to more than 6 or 9 events was small (no more than 5% or 1%, respectively). We did not find an increased risk of endophthalmitis for microincisional vitrectomy compared with standard vitrectomy. The beveled approach seems to be safer than a straight approach, supporting the current recommendation of its adoption in microincisional vitrectomy. However, these findings must be interpreted cautiously because of the small number of endophthalmitis events reported from included studies.