Purchase this article with an account.
Francesca Menchini, Andrea Govetto, Gianni Virgili, Paolo Lanzetta; Endophthalmitis Following Microincisional Vs 20-Gauge Vitrectomy: A Systematic Review And Meta-analysis. Invest. Ophthalmol. Vis. Sci. 2011;52(14):6133.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Endophthalmitis is an uncommon severe complication of vitrectomy and is thought to be more frequent after microincisional (23- and 25- gauge) compared to standard (20-gauge) vitrectomy. Summarizing data on the risk of endophthalmitis using either vitrectomy technique is important to patients and treating physicians, because a single study is unlikely to be conclusive given the rarity of the event.
We conducted a systematic review of randomized and non-randomized studies comparing microincisional and standard vitrectomy, searching Medline and EMBASE up to November 2010 and handsearching references of retrieved studies. The rate of endophthalmitis was investigated by a direct comparison between microincisional and standard pars plana vitrectomy and by subgroup analyses, exploring whether differences in endophthalmitis rate could be explained by different surgical approaches in microincisional technique (straight vs beveled incision) and use of prophylactic topical antibiotics. Bayesian meta-analysis was used to compute a pooled odds ratio (OR) of endophthalmitis for microincisional vs standard vitrectomy. We decided to use Bayesian methods due to the infrequency of the event investigated and the observational nature of most of the studies included in our review. Bayesian methods can indeed manage rare events and heterogeneity with a minimum of statistical assumption.
The search found 3 small randomized and 15 non-randomized studies recording 60 cases of endophthalmitis in 112,663 participants. The meta-analysis found the overall OR of endophthalmitis for microincisional vs standard vitrectomy was 2.5 (95% credible interval 0.71 to 7.0). Subgroup analyses including incision type found an increased risk of endophthalmitis using a microincisional straight approach compared to standard vitrectomy (OR=16, 1.9 to 200), but no increase for a beveled approach (OR=0.76, 0.20 to 2.5). The post-operative use of subconjunctival antibiotics tended to reduce the risk of endophthalmis.
The results of our systematic review support current recommendation that a beveled incision should be used in microincisional vitrectomy because there was no significant increase of endophthalmitis compared to standard vitrectomy using this approach.
This PDF is available to Subscribers Only