Purchase this article with an account.
Wong Chee Wai, Tien Y Wong, Wee Boon Teo, Ecosse L Lamoureux, Mohammad Kamran Ikram, Paul Mitchell, Jie Jin Wang, E SHyong Tai, Sunil Sethi, Charumathi Sabanayagam; Serum Cystatin C, Markers of Chronic Kidney Disease and Retinopathy in Indians with Diabetes. Invest. Ophthalmol. Vis. Sci. 2014;55(13):5363.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Chronic kidney disease (CKD) defined by triple markers including serum creatinine, serum cystatin C and microalbuminuria have been shown to improve prediction of renal failure and mortality in recent studies. We examined the association of CKD defined by triple markers with moderate diabetic retinopathy (DR) in a population-based sample of adults.
We examined 1,119 Indian adults with diabetes, aged 40-80 years who participated in the Singapore Indian Eye Study (2007-2009). CKD was defined as 1) serum creatinine-based estimated glomerular filtration rate (eGFRcr) <60 ml/min/1.73m2 (n=128) 2) serum cystatin C-based eGFR <60 (eGFRcys, n=225), and 3) microalbuminuria (urine albumin to creatinine ratio of ≥30 mg/g, n=54). Moderate retinopathy defined as an ETDRS severity score of level >43, was assessed from digital fundus photographs of both eyes by trained graders. The associations of CKD defined by each of the three markers alone and in combination (triple markers, eGFRcr<60+eGFRcys<60+microalbuminuria, n=22) with moderate DR were examined using logistic regression models adjusted for potential confounding factors including duration of diabetes, smoking, body mass index, systolic blood pressure and fellow markers.
The prevalence of moderate DR among those with CKD defined by creatinine, cystatin C, and microalbuminuria were 26.6%, 20.9%, and 44.4% respectively. In separate models, CKD-eGFRcr, CKD-eGFRcys and microalbuminuria were independently associated with DR (odds ratio [OR] (95% confidence interval [CI]) = 3.81 (2.09-6.95), 4.19 (2.39-7.33) and 3.47 (1.61-7.49). In models including all 3 markers, compared to those without CKD by any marker, those with CKD defined by triple markers was associated with a twenty-fold odds of having moderate DR (OR [95% CI] =19.45 [6.54-57.85]).
CKD defined by a triple marker panel was strongly associated with moderate DR in Indian adults with diabetes. If confirmed by future prospective studies, a triple marker approach may improve risk stratification of DR in persons with CKD.
This PDF is available to Subscribers Only