Purchase this article with an account.
Sangeetha Srinivasan, Cirous Dehghani, Nicola Pritchard, Katie Edwards, Anthony W Russell, Rayaz A Malik, Nathan Efron; Ophthalmic and clinical factors that predict four-year incidence and progression of diabetic retinopathy in Type 1 diabetes. Invest. Ophthalmol. Vis. Sci. 2017;58(8):2887.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To investigate the role of retinal thickness measures and corneal nerve morphology in predicting four-year incidence and progression of diabetic retinopathy in Type 1 diabetes
Data of 126 eyes of 126 participants were examined at baseline and after four years. Diabetic retinopathy (DR) at baseline and at follow-up was graded according to Early Treatment Diabetic Retinopathy Study scale. The following measures were assessed at baseline: HbA1c, nephropathy, neuropathy, cardiovascular measures, retinal thickness measures (using optical coherence tomography), and corneal nerve fiber length (CNFL) using corneal confocal microscopy. Incident DR was defined as newly developed DR when examined after four years. Progression of DR was defined as worsening of baseline DR to any level. Baseline factors were assessed by univariate and step-wise multiple logistic regression and examined for their diagnostic capabilities.
Four-year incident DR was 19% (13 of 68 without DR at baseline developed DR - very mild and mild non-proliferative DR). Progression of DR was 43% (25 of 58 with DR at baseline showed progression at follow-up). The group with ‘incident DR’ had prolonged duration of diabetes (p=0.025) and lower peroneal nerve conduction velocity (p=0.006) compared with ‘no DR’ group. The ganglion cell complex global loss volume (GCC GLV) was higher and CNFL was lower in the progressed group. When adjusted for confounding factors, a lower CNFL (AUC=0.637, p=0.040, 64% sensitivity and 64% specificity at 14.9 mm/mm2 cut-off), higher triglycerides (AUC=0.669, p=0.012, 64% sens, 62% spec at 0.85 mmol/L cut-off) and elevated vibration perception threshold (AUC=0.708, p=0.002, 96% sens, 40% spec at 3.55 Hertz cut-off) were significant predictors for DR progression. For incident or progression of DR compared with ‘stable /no DR at follow-up’, higher HbA1c (AUC=0.638, p=0.017, 66% sens, 57% spec at 7.75% cut-off) and prolonged diabetes duration (AUC=0.659, p=0.001, 95% sens, 43% spec at 10.6 years cut-off) were the only predictors.
A lower CNFL, an elevated vibration perception threshold and higher serum triglyceride levels can predict DR progression; higher glycemic levels and prolonged diabetes duration are independent predictors of incidence or progression of DR. Knowledge of these risk factors may serve as targets for medical management or clinical trials.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.
This PDF is available to Subscribers Only