Association between body mass index and diabetic retinopathy in Asians: the Asian Eye Epidemiology Consortium (AEEC) study. Issue 7 (23rd February 2021)

• Record Type: Journal Article
• Authors: Sabanayagam, Charumathi; Sultana, Rehena; Banu, Riswana; Rim, Tyler; Tham, Yih Chung; Mohan, Sunita; Chee, Miao Li; Wang, Ya Xing; Nangia, Vinay; Fujiwara, Kohta; Tai, E Shyong; Song, Su Jeong; Bikbov, Mukharram M; Raman, Rajiv; Cheng, Ching Yu; Wong, Tien Yin; Jonas, Jost B
• Is Part Of: British journal of ophthalmology. Volume 106:Issue 7(2022)
• Page Start: 980
• Page End: 986
• Publication Date: 2021-02-23
• Subjects: epidemiology -- retina; Ophthalmology -- Periodicals; 617.7
• DOI: 10.1136/bjophthalmol-2020-318208
• Languages: English
• ISSNs: 0007-1161

Abstract

Abstract : Background/aims: Obesity is a well-known risk factor for diabetes, but its association with diabetic retinopathy (DR) is inconclusive, in particular in Asians. We aimed to assess whether body mass index (BMI) is associated with the presence and severity of DR in Asian populations with diabetes. Methods: Pooled analysis of individual-level cross-sectional data from 10 010 adults with diabetes who participated in 12 population-based studies conducted in China, India, Japan, Russia (Asian), Singapore and South Korea that were part of the Asian Eye Epidemiology Consortium (AEEC). BMI was calculated as weight in kilograms divided by height in square metres and categorised into normal (<25 kg/m 2, reference), overweight (25–29.9 kg/m 2 ) and obese (≥30 kg/m 2 ). Any-DR (n=1669) and vision-threatening DR (VTDR, n=489) were assessed from digital retinal photographs and graded according to standard protocols. Each study was analysed separately using multivariable logistic regression models adjusted for age, sex, haemoglobin A1c%, systolic blood pressure and diabetes duration, and the estimated odds ratios (ORs) and 95% confidence interval (CIs) from all studies were then combined using random-effects models. Results: In multivariable models, obesity showed a significant inverse association with any-DR (pooled OR (95% CI) =0.74 (0.59 to 0.91)) and VTDR (0.75 (0.60 to 0.93)). Similarly, in continuous analysis, BMI showed a significant inverse association with both any-DR (0.93 (0.87 to 0.99)) and VTDR (0.79 (0.68 to 0.92) per SD increase). Overweight did not show a significant association with any-DR. Conclusions: Among Asian adults with diabetes, both BMI and obesity showed an inverse association with DR. These findings warrant confirmation in further longitudinal studies.