BACKGROUND: A hypertensive response to exercise has prognostic significance. Patients with type 2 diabetes have vascular abnormalities which may predispose to exaggerated brachial and central blood pressure (BP) during exercise. This study aimed to test this hypothesis and to determine the clinical significance of high exercise BP by examining its relation to left ventricular (LV) mass. METHODS: Brachial and central BP were recorded at rest and in response to maximal exercise in 73 diabetic patients (aged 54 +/- 10 years) and 73 controls (aged 53 +/- 12 years). Brachial BP was recorded using mercury sphygmomanometry and LV mass using 2D-echocardiography. Central BP was estimated by radial tonometry using an exercise-validated generalized transfer function. RESULTS: At rest there were no significant (P > 0.05) differences between groups in brachial or central BP. The diabetic patients had significantly increased exercise brachial systolic BP (SBP: 199 +/- 25 mm Hg vs. 185 +/- 21 mm Hg; P = 0.002) and central SBP (158 +/- 17 mm Hg vs. 149 +/- 15 mm Hg; P = 0.002). There was a significantly higher prevalence of an exaggerated exercise BP response (> or =210/105 mm Hg; men and > or =190/105 mm Hg; women) in the diabetic patients (51% vs. 22%; P < 0.01). Compared with those with normal exercise BP, LV relative wall thickness (RWT) was significantly higher (0.41 +/- 0.09 vs. 0.36 +/- 0.08; P < 0.05) and LV hypertrophy was more prevalent (35% vs. 16%; P < 0.05) in those with a hypertensive response. After accounting for other confounding variables, exercise central SBP remained independently associated with LV RWT (beta = 0.22; P = 0.006). CONCLUSION: Diabetic patients are more likely to exhibit exaggerated exercise BP. Regardless of disease status, high exercise central SBP may contribute to cardiovascular risk via adverse cardiac remodeling.