Although the level of hyperglycemia is clearly a risk factor for microvascular complications in diabetic patients, its role in macrovascular complications remains controversial. We followed 4,875 subjects (65% Mexican-American) for 7–8 years to investigate the effects of diabetes and hyperglycemia on all-cause and cardiovascular disease (CVD) mortality. These end points were also analyzed according to quartiles of baseline fasting plasma glucose among diabetic participants.

The Cox proportional hazards model was used to estimate the relative risks (RRs) for all-cause and CVD mortality.

Diabetes was significantly associated with increased all-cause mortality (RR [95% CI] = 2.1 [1.3–3.5] in men; 3.2 [1.9–5.4] in women) and increased CVD mortality (3.2 [1.4–7.1] in men; 8.5 [2.8–25.2] in women). Among diabetic subjects, those in quartile 4 had a 4.2-fold greater risk of all-cause mortality (P < 0.001) and a 4.7-fold greater risk of CVD mortality (P = 0.01) than those in quartiles 1 and 2 combined. After further adjustment for other potential risk factors, subjects in quartile 4 had a 4.9-fold greater risk of all-cause mortality and a 4.9-fold greater risk of CVD mortality than those in quartiles 1 and 2. In addition, hypertension, current smoking, and cholesterol >6.2 mmol/1 were significant predictors of CVD mortality using Cox models.

We conclude that diabetes is a predictor of both all-cause and CVD mortality in the general population and that both hyperglycemia and common CVD risk factors are important predictors of all-cause and CVD mortality in diabetic subjects.