Visual and anatomical outcomes following vitrectomy for complications of diabetic retinopathy: the DRIVE UK study

End-stage diabetic eye disease is an important cause of severe visual impairment in the working-age group. With the increasing availability of refined surgical techniques as well as the early diagnosis of disease because of screening, one would predict that the prevalence of this condition is decreasing and the visual outcome is improving.

To study the prevalence and visual outcome following vitrectomy for complications of diabetic retinopathy.

This study identified the patients who underwent vitrectomy from January 2007 to December 2009 because of diabetes-related complications in South East London. Data collected included baseline demographics, best-corrected visual acuity, indication for the vitrectomy, complication, outcome, and duration of follow-up.

The prevalence of people requiring vitrectomy who are registered in the diabetes register of this region was 2 per 1000 people with diabetes. Vitrectomy was required in 185 eyes of 158 patients during this period. These included 83 Caucasians, 51 Afro-Caribbeans, 17 South Asians, and 7 from other ethnic groups. There were 58 patients with type I diabetes and 100 with type II, with a mean duration of diabetes of 23 and 16.5 years, respectively. The reason for vitrectomy included tractional retinal detachment (TRD) in 109 eyes, non-clearing vitreous haemorrhage (NCVH) in 68 eyes, and other causes in 8 eyes. In all, 50% of the eyes with TRD and NCVH, and 87% of the eyes with NCVH improved by at least three ETDRS lines at 12 months. Poor predictors of visual success included longer duration of diabetes (OR: 0.69), use of insulin (OR: 0.04), presence of ischaemic heart disease (OR: 0.04), delay in surgery (OR: 0.59), and the failure to attend clinic appointments (OR: 0.58). Preoperative use of intravitreal bevacizumab in eyes with TRD undergoing vitrectomy showed a marginal beneficial effect on co-existent maculopathy (P=0.08) and required less laser intervention post procedure, but did not affect the number of episodes of late-onset vitreous haemorrhage post vitrectomy (P=0.81).

Visual outcome has improved significantly in eyes with complications due to diabetic retinopathy compared with the previously reported Diabetic Vitrectomy Study.