High HDL cholesterol: A risk factor for diabetic retinopathy? Findings from NO BLIND study. (April 2019)

• Record Type: Journal Article
• Authors: Sasso, Ferdinando Carlo; Pafundi, Pia Clara; Gelso, Aldo; Bono, Valeria; Costagliola, Ciro; Marfella, Raffaele; Sardu, Celestino; Rinaldi, Luca; Galiero, Raffaele; Acierno, Carlo; de Sio, Chiara; Caturano, Alfredo; Salvatore, Teresa; Adinolfi, Luigi Elio
• Is Part Of: Diabetes research and clinical practice. Volume 150(2019)
• Page Start: 236
• Page End: 244
• Publication Date: 2019-04
• Subjects: Diabetic retinopathy -- Risk factors -- HDL cholesterol -- Diabetes complications; Diabetes -- Periodicals; Diabetes Mellitus -- Periodicals; 616.462
• DOI: 10.1016/j.diabres.2019.03.028
• Languages: English
• ISSNs: 0168-8227

Abstract

Highlights: In a real life multicentre setting high HDL values are associated to DR. A ROC curve defined a cut-off of 40 mg/dl which significantly defined the risk of DR. DR was associated in particular to HDL >60 mg/dl. Association between DR and high HDL was observed only in proliferative DR. Correlation between micro and macroangiopathy could have new basis. Abstract: Aims: To assess the correlation between diabetic retinopathy (DR) and potential risk factors, as well as the relationship between DR and the other complications of diabetes, in a real-life population of type 2 diabetes patients recruited in several centres in Italy. Methods: The NO BLIND is a cross-sectional, multicentre, observational study, which involved nine public outpatient clinics in Italy. The patients were assessed for eligibility from November 2016 till November 2017. Those enrolled underwent standard fundus oculi exam. Clinical and laboratory data were also collected. Results: 2068 T2DM underwent fundus oculi exam. 435 received diagnosis of diabetic retinopathy (21%). Diabetic retinopathy was independently associated with HDL cholesterol (O.R.: 1.042; 95% C.I.: 1.012–1.109; p = 0.004), Albumin Excretion Rate (AER) (O.R.: 1.001; 95% C.I.: 1.000–1.002; p = 0.034) and GFR (O.R.: 1.159; 95% C.I.: 1.039–1.294; p = 0.008). HDL cholesterol values were hence split in two classes according to a potential cut-off (40 mg/dL), as defined by the ROC curve. Following analysis confirmed the association between DR and high HDL values (p = 0.032). Somatic neuropathy and diabetic ulcer were independently related with DR (p < 0.001 and p = 0.012, respectively). Conclusions: A novel relationship between high HDL cholesterol and DR was observed.