Purchase this article with an account.
Mohammad Ikram, Jing Xie, Mary Frances Cotch, Barbara Klein, Paul Mitchell, Jonathan Shaw, Rohit Varma, Ecosse Lamoureux, Tien Wong; Vision-Threatening Diabetic Retinopathy and Incident Cardiovascular Disease: A Systematic Review and Meta-Analysis. Invest. Ophthalmol. Vis. Sci. 2013;54(15):2174.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To perform a meta-analysis on the association between baseline vision-threatening diabetic retinopathy (VTDR) and incident cardiovascular diseases (CVD) using individual participant data.
Following an extensive electronic and manual literature search and personal correspondence with investigators, high quality prospective population-based studies were included with data on participants with photographic-based DR grading, adequate follow-up visits, and well-defined cardiovascular endpoints, namely coronary heart disease (CHD) and/or stroke. VTDR was defined as the presence of proliferative DR (PDR, ETDRS level ≥ 60) and/or any diabetic macular edema (DME) in the worst eye.
Of 1904 identified search titles, 8 studies met our inclusion criteria and comprised of 6,343 subjects with diabetes mellitus type 2. At baseline the prevalence was 10.6% for VTDR, 7.9% for PDR and 5.0% for DME. After a mean follow-up time of 5.9 years [range 3.2 to 10.1 years across studies], there were 1100 incident cases of first-ever CVD events. The presence of VTDR was associated with risk of incident CVD (age-sex-race adjusted incidence rate ratio [IRR] 1.39; 95% confidence interval [CI]: 1.16-1.67) and CHD (IRR 1.47; 95% CI: 1.20-1.79). For fatal CVD and fatal CHD, the corresponding IRRs were 2.33 (95% CI: 1.49-3.67) and 2.74 (95% CI: 1.74-4.32), respectively. These associations persisted after multivariate adjustment for vascular risk factors, including smoking, systolic blood pressure, hypertension medication, triglyceride, body mass index, duration of diabetes, treatment of diabetes and Hb1Ac. When examining PDR and DME separately, both were related to an increased risk of first-ever CVD (for PDR IRR: 1.28; 95% CI: 1.03-1.58 and for DME IRR 1.59; 95% CI: 1.20-2.10). The corresponding HRs for fatal CVD were 1.85 (95% CI: 1.04-3.28) and 2.85 (95% CI: 1.43-5.68).
Persons with VTDR are at increased risk of incident CVD including CHD compared to those without the condition. These data suggest that persons with type 2 diabetes and VTDR should be followed more closely by their physicians in an effort to prevent CVD complications from occurring.
This PDF is available to Subscribers Only