23‐gauge versus 20‐gauge vitrectomy for diabetic vitreous hemorrhage

Abstract

Abstract Purpose To compare 23‐gauge and 20‐gauge vitrectomy for diabetic vitreous hemorrhage (VH) regarding postoperative VH (PVH), re‐operation rates, visual acuity (VA) and safety profiles. Methods This was a retrospective chart review of diabetic patients who underwent 20‐gauge or 23‐gauge vitrectomy for persistent VH at our unit between January 2004 and June 2009. Exclusion criteria were silicone oil or gas tamponade, prior vitrectomies, perioperative injection of anti‐vascular endothelial growth factor and a follow up less than 12 months. PVH was classified as early and late PVH occurring within and >2 months after surgery. Samples were compared with independent t test and Mann‐Whitney‐U test depending on distribution. Categorical data were compared with Fisher exact test. Results Hundred and six eyes met our inclusion criteria. Sixty one eyes underwent 20‐gauge and forty six 23‐gauge vitrectomy. Early PVH was observed in 27.8% and 20% in the 20‐gauge and 23‐gauge group, respectively (p=0.44). Late PVH was noted in 24.6% and 22.2% using 20‐gauge and 23‐gauge system (0.83). Re‐operation rates due to early and late PVH were similar among the groups. Final VA averaged 0.71 and 0.68 logMAR in the 20‐gauge and 23‐gauge group (p=0.81). Postoperative hypotony was observed in 8.9% of cases in the 23‐gauge group. Rhegmatogeneous retinal detachment occurred in one eye in each group. Conclusion 23‐gauge vitrectomy is a viable method for diabetic VH providing similar anatomical and functional outcome like after conventional 20‐gauge vitrectomy. The safety profile is acceptable since postoperative hypotony resolved in all cases without intervention.