PURPOSE: A lower retinal arteriolar-to-venular ratio (AVR) has been suggested to reflect generalized arteriolar narrowing and to predict the risk of cardiovascular diseases. The contribution of the separate arteriolar and venular diameters to this AVR is unknown. Thus, associations between retinal arteriolar and venular diameters, and the AVR on the one hand and blood pressure, atherosclerosis, inflammation markers, and cholesterol levels on the other were examined in the Rotterdam Study. METHODS: In this cross-sectional population-based study, for one eye of each subject (> or =55 years; n = 5674), retinal arteriolar and venular diameters (in micrometers) of the blood columns were summed on digitized images. At baseline blood pressures, cholesterol levels, and markers of atherosclerosis and inflammation were also measured. RESULTS: With increasing blood and pulse pressures, retinal arteriolar and venular diameters and the AVR decreased significantly and linearly. Lower arteriolar diameters were associated with increased carotid intima-media thickness. Larger venular diameters were associated with higher carotid plaque score, more aortic calcifications, lower ankle-arm index, higher leukocyte count, higher erythrocyte sedimentation rate, higher total serum cholesterol, lower HDL, higher waist-to-hip ratio, and smoking. A lower AVR was related to increased carotid intima-media thickness, higher carotid plaque score, higher leukocyte count, lower HDL, higher body mass index, higher waist-to-hip ratio, and smoking. CONCLUSIONS: Because larger venular diameters are associated with atherosclerosis, inflammation, and cholesterol levels, the AVR does not depend only on generalized arteriolar narrowing due to the association between smaller arteriolar diameters and higher blood pressures. These data indicate that retinal venular diameters are variable and may play their own independent role in predicting cardiovascular disorders.