Outcomes of Surgically Managed Proliferative Diabetic Retinopathy

Abstract

Purpose: The purpose of this article is to conduct a retrospective chart review of the results of vitrectomy for diabetic vitreous hemorrhage (VH) and diabetic traction retinal detachment (TRD) using small-gauge instruments. Methods: We retrospectively reviewed medical records of all diabetic vitrectomies performed at the University of Louisville from 2012 to 2016 that had at least 6 months of follow-up. Patients included in this study underwent pars plana vitrectomy (PPV) for proliferative diabetic retinopathy complications. We analyzed the preoperative and 6- and 12-month postoperative visual acuities (VAs) in patients who had vitrectomy for their diabetic retinopathy. We also determined the proportion of patients who had best-corrected visual acuity (BCVA) of 20/40 or better and 20/80 or worse. In the patients who had BCVA of 20/80 or worse, we identified the reasons for the decreased VA. We also report intraocular pressure, demographic characteristics, operative techniques, and complications. Results: We identified 93 eyes that underwent diabetic vitrectomy; 81 eyes of 63 patients had at least 6 months’ follow-up. Of those 81 eyes, 40 eyes had VH only; the average duration of vision loss before surgery was 4.7 months. Forty-one eyes had TRDs with average duration of vision loss before surgery of 7 months. The presenting BCVA was 1.14 logMAR (20/300) in the VH-only group and 1.49 logMAR (20/600) in the TRD group ( P < .09). Six months after surgery, the BCVA was 0.59 logMAR (20/80) in the VH-only group but still 1.37 logMAR (20/500) in the TRD group ( P < .001). By 12 months after surgery (63 eyes), the BCVA was 0.60 logMAR (20/80) vs 1.09 logMAR (20/250), respectively ( P < .02). Technical success was achieved in 79 of 81 eyes. At 12 months after surgery, 22 eyes (55%) in the VH-only group and 19 eyes (46%) in the TRD group gained 15 letters or more (doubling the VA) compared with their presenting VA. In the VH-only group 7 eyes (18%) were 23-gauge (G) PPV, 22 (58%) were 25G PPV, and 9 (24%) were 27G PPV, while in the TRD group 15 (38%) were 20G PPV, 11 (27%) were 23G PPV, and 14 (35%) were 25G PPV. There was no difference in outcome with different gauges. Conclusions: While vitrectomy for complications of diabetic retinopathy was technically successful with small-gauge surgery, visual results were disappointing. Further work is needed to clarify the best timing of surgery to improve visual outcomes.