Long-term survival rates of patients undergoing vitrectomy for diabetic retinopathy in an Australian population: a population-based audit

Importance: Five-year survival rates in patients undergoing vitrectomy for diabetic retinopathy (DR) vary from 68% to 95%. No study has been conducted in an Australian population. Background: We aimed to determine the survival rates of patients undergoing diabetic vitrectomy in an Australian population. Design: Retrospective audit, tertiary centre hospitals and private practices. Participants: All individuals in South Australia and the Northern Territory who underwent their first vitrectomy for diabetic complications between January 1, 2007 and December 31, 2011. Methods: An audit of all eligible participants has been completed previously. Survival status as of July 6, 2018 and cause of death were obtained using SA/NT DataLink. Kaplan-Meier survival curves and multivariate cox-regressions were used to analyse survival rates and identify risk factors for mortality. Main outcome measures: Five-, seven- and nine-year survival rates. Results: The 5-, 7- and 9-year survival rates were 84.4%, 77.9% and 74.7%, respectively. The most common cause of death was cardiovascular disease. Associated with increased mortality independent of age were Indigenous ethnicity (HR = 2.04, 95% confidence interval [CI]: 1.17-3.57, P = 0.012), chronic renal failure (HR = 1.76, 95% CI: 1.07-2.89, P = 0.026) and renal failure requiring dialysis (HR = 2.32, 95% CI: 1.25-4.32, P = 0.008). Conclusions and relevance: Long-term survival rates after diabetic vitrectomy in Australia are similar to rates reported in other populations. Indigenous ethnicity and chronic renal failure were the most significant factors associated with long-term mortality. This information can guide allocation of future resources to improve the prognosis of these high risk groups.