The relative importance of different blood pressure indices on cardiovascular risk in type 1 diabetes (T1D) has not been established. We thus compared the strengths of associations between different baseline and most recent blood pressure measures (systolic pressure [SBP], diastolic pressure [DBP], pulse pressure [PP], mean arterial pressure [MAP], and mid-blood pressure [Mid-BP, the average of SBP and DBP]) and the 25-year risk of coronary artery disease (CAD) in the Pittsburgh Epidemiology of Diabetes Complications (EDC) study of childhood onset T1D (N=605, without baseline CAD). CAD was defined as fatal coronary artery event, non-fatal myocardial infarction, angina, revascularization, or ischemic ECG. Baseline mean age and diabetes duration were 27 and 19 years, respectively. In Cox Proportional Hazards modes, the baseline age-adjusted hazard ratios (HR) associated with one increment in SD for the risk of CAD events (n=219) were 1.49 (95% CI: 1.31, 1.70) for SBP; 1.46 (1.28, 1.67) DBP; 1.26 (1.09, 1.46) PP; 1.50 (1.32, 1.71) MAP and 1.51 (1.32, 1.71) for Mid-BP. There was an interaction between PP and HbA1c that the age-adjusted HR of PP was 1.20 (0.99, 1.45) and 1.38 (1.07, 1.77) in those with HbA1c ≤ 9 and > 9%, respectively. Though area under the receiver operating characteristic curve (AUC) was significantly lower for PP compared with the other four BP indices in overall subjects, the predictive utility of PP for CAD becomes similar to that of other indices in those with higher HbA1c levels (>9 %). Using the most recent data and adjusting for age, the predictive value of SBP was superior to DBP (∆ AUC=0.037, p=.006), MAP (∆ AUC=0.021, p=.018), and Mid-BP (∆ AUC=0.014, p=.031), but not PP (∆ AUC=0.007, p=.552). In sum, SBP appears to be the strongest determinant of CAD risk in T1D patients overall but PP becomes a determinant as strong as SBP in older age and in those with worse glycemic control, consistent with PP reflecting vascular stiffness.


J. Guo: None. T.J. Orchard: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at