Aim: To determine the role of different blood pressure (BP) components as predictors of cardiovascular disease (CVD) in diabetic patients. Methods: We followed 828 diabetic patients, aged >= 30 years, without baseline CVD and not taking antihypertensive drugs. The hazard ratios (HRs) for CVD were calculated for a 1 standard deviation (SD) change in each BP measure using Cox proportional regression analysis. Results: During median 8.4 year-follow up, 134 CVD events occurred. Systolic BP (SBP), pulse pressure (PP) and mean arterial pressure (MAP) were independent predictors of CVD with multivariate adjusted hazard ratios (HRs) of 1.35, 1.36 and 1.23 respectively (all p < 0.05) and similar discriminatory power (C statistics approximate to 75%). The multivariate models containing MAP plus one other BP component highlighted the significant improvement in the goodness of fit to predict incident CVD (likelihood ratio test=0.01); however this was not so for SBP and PP. In the elderly subjects, only PP and SBP were significantly associated with CVD. Conclusion: The SBP and PP are significant independent predictors of CVD in all diabetic populations (including younger and elder groups) and other BP components did not add to these measures in risk prediction.