Purchase this article with an account.
Gavin S Tan, Carol Yim-lui Cheung, Ching-Yu Cheng, Tien Y Wong; Do retinal neurovascular changes predict the development of diabetes?. Invest. Ophthalmol. Vis. Sci. 2019;60(9):1559. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To evaluate if retinal vascular geometric variables and neuroretinal thickness are associated with incident diabetes at 6 years of follow-up.
Participants were recruited from the Singapore Epidemiology of Eye Diseases Study, a prospective population-based study among Chinese, Malays and Indians aged 40 years or older residing in Singapore with a 6 year followup. All participants underwent standardised systemic and ophthalmic examinations.The study population for this analysis included only those who did not have diabetes at baseline. Incidence diabetes was defined as random glucose ≥11.1mmol/L, glycated haemoglobin (HbA1c) ≥6.5%, use of diabetic medication or self-reported history of diabetes at the 6 year followup visit.At baseline, all patients had 2-field fundus photography graded for retinal vascular geometric variables including tortuosity, branching, fractal dimension and calibre [Central retinal arteriolar equivalent (CRAE), central retinal venular equivalent (CRVE) and arteriole to venous ratio (AVR)] using a computer-assisted program (Singapore I Vessel Assessment). Subjects of Chinese ethnicity had additional optical coherence tomography(OCT) [Cirrus OCT, Zeiss Meditec] done with the total neuroretinal thickness in each sector of the ETDRS grid recorded.A cox regression model was used to estimate the hazard ratio (HR) adjusting for age, gender, systolic blood pressure glycated hemogloblin(HbA1c) and body mass index.
There were 4992 participants without diabetes at baseline who returned for follow-up, of which 1835 had OCT. There were 461 incident cases of diabetes. Participants who developed diabetes were more likely to be Indian, and had higher systolic blood pressure, HbA1c, casual blood glucose and body mass index at baseline, as well as higher CRAE, CRVE and lower AVR. There was no significant difference in fractal dimension, vessel tortuosity or branching. After multivariate regression, those with greater arteriole calibre [HR: 1.03; p<0.001] and greater venous calibre [HR: 1.01; p=0.013] were more likely to develop diabetes. Retinal thickness in any sector of the ETDRS grid was not associated with incident diabetes.
Retinal arteriole and venous dilation are independently associated with the risk of diabetes however neuroretinal thickness on OCT was not, suggesting microvascular but not neuroretina changes have a role in the development of clinical diabetes.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.
This PDF is available to Subscribers Only