Background: The prognostic significance of diabetic retinopathy (DR) for cardiovascular diseases (CVD) remained unclear. Therefore, we performed this meta-analysis to assess whether DR predicted CVD mortality in diabetic patients.Methods: We searched PubMed, Embase, Web of Science and Cochrane Library for cohort studies reporting the association of DR and CVD mortality. Then we pooled the data for analysis.Results: After screening the literature, 10 eligible studies with 11,239 diabetic subjects were finally included in quantitative synthesis. The pooled risk ratio (RR) of DR, mild DR, and severe DR for CVD mortality was 1.83 (95% confidence interval (CI): 1.42, 2.36; P < 0.001), 1.13 (95% CI: 0.81, 1.59; P = 0.46), and 2.26 (1.31, 3.91; P = 0.003), respectively, compared to those without DR. In type 2 DM, the patients with DR had a significantly higher CVD mortality (RR: 1.69; 95% CI: 1.27, 2.24; P < 0.001). Subgroup analysis also showed a significantly higher CVD mortality in DR according to various regions, study design, data source, and follow-up period (all RR > 1; all P values < 0.05). Data from 2 studies showed no significant correlation of DR and CVD mortality in diabetic patients receiving cardiovascular surgery (RR: 2.40; 95% CI: 0.63, 9.18; P = 0.200).Conclusions: DR is a risk marker of cardiovascular death, and severe DR predicts a doubled mortality of CVD in diabetes. These findings indicate the importance of early identification and management of diabetic patients with DR to reduce the risk of death.

Background: The prognostic significance of diabetic retinopathy (DR) for cardiovascular diseases (CVD) remained unclear. Therefore, we performed this meta-analysis to assess whether DR predicted CVD mortality in diabetic patients.Methods: We searched PubMed, Embase, Web of Science and Cochrane Library for cohort studies reporting the association of DR and CVD mortality. Then we pooled the data for analysis.Results: After screening the literature, 10 eligible studies with 11,239 diabetic subjects were finally included in quantitative synthesis. The pooled risk ratio (RR) of DR, mild DR, and severe DR for CVD mortality was 1.83 (95% confidence interval (CI): 1.42, 2.36; P < 0.001), 1.13 (95% CI: 0.81, 1.59; P = 0.46), and 2.26 (1.31, 3.91; P = 0.003), respectively, compared to those without DR. In type 2 DM, the patients with DR had a significantly higher CVD mortality (RR: 1.69; 95% CI: 1.27, 2.24; P < 0.001). Subgroup analysis also showed a significantly higher CVD mortality in DR according to various regions, study design, data source, and follow-up period (all RR > 1; all P values < 0.05). Data from 2 studies showed no significant correlation of DR and CVD mortality in diabetic patients receiving cardiovascular surgery (RR: 2.40; 95% CI: 0.63, 9.18; P = 0.200). Conclusions: DR is a risk marker of cardiovascular death, and severe DR predicts a doubled mortality of CVD in diabetes. These findings indicate the importance of early identification and management of diabetic patients with DR to reduce the risk of death.