Increased odds of diabetic retinopathy with combined suboptimal parameters of diabetes control compared to glycaemic control alone

Abstract

Purpose: While the association of poor glycaemic control (GC) with diabetic retinopathy (DR) and diabetic macular edema (DME) is well established, the incremental rise in the risk of DR and DME associated with poor GC combined with suboptimal blood pressure (BPC) and/or lipid control (LPC) is not well understood. We examined the cross-sectional association of suboptimal parameters of diabetes control with DR and DME.

Methods: Patients with type 2 diabetes at a tertiary eye hospital (March 2009 to December 2010) in Melbourne (Australia) underwent a comprehensive eye and clinical examination including BP measurement. Fasting blood samples were assessed for HbA_1c and serum lipids. Dilated fundus and OCT images were graded for the presence of DR [ETDRS] and DME [American Academy of Ophthalmology], respectively. Suboptimal parameters examined were: (GC [Hb_A1c≥7%]; BPC [SBP≥130 and DBP≥80]; LPC [total cholesterol (CHO) to HDL ratio≥4.0]); The association of suboptimal parameters of diabetes control alone and in combination (GC& BPC; GC&LPC; BPC &LPC; GC&BPC&LPC) with any DR and DME were examined using multivariate logistic regression models adjusted for traditional risk factors including duration of diabetes and medication.

Results: Of the 613 patients (mean±SD age 64.9± 11.6 years and 57% male), 372 had any DR and 183 any DME, respectively. In adjusted models, compared to those with good diabetes control (n=123; Hb_A1c<7%, SBP<130 and DBP<80 and CHO to HDL ratio<4.0), the odds ratio (OR) (95% confidence interval [CI]) of DR was 2.44 (1.34-4.46), 3.75 (1.75-8.07), 4.64 (2.13-10.12) and 2.28 (1.01-5.16) for poor GC only; GC& BPC; GC&LPC; and GC&BPC&LPC. Corresponding estimates for any DME were 3.19 (1.55-6.59); 3.60 (1.58-8.22); 2.76 (1.18-6.44); and 3.01 (1.18-7.67), respectively.

Conclusions: While the odds of any DME were approximately three fold regardless of poor diabetes control parameter combination, the odds of any DR were substantially higher when suboptimal GC is combined with either poor BPC or LPC. The odds of poor GC, BPC and LPC combined were not substantially higher than the other combinations, perhaps due to a small sample size. Targeting optimal combined diabetes control parameters, rather than GC alone, may be critical to reduce the burden of DR.