Recent automated applanation tonometry can measure radial pulse wave-derived central blood pressure (CBP), which has shown a prognostic value independently of peripheral blood pressure. However, CBP's clinical significance has not been fully established. We examined the associations between CBP and cardiac structure and function by comparing them with those of arterial stiffness assessed by cardio-ankle vascular index (CAVI) in treated hypertensive patients. Enrolled in the study were 102 patients (71+/-7 years) with treated hypertension. The transmitral early-to-atrial velocity ratio (E/A), peak systolic (S'), early diastolic (E') mitral annular velocities and the Tei index were measured as indexes of cardiac function derived from conventional and tissue Doppler echocardiography. Left ventricular mass index (LVMI) was measured as an index of LV hypertrophy. CBP and CAVI were measured just after echocardiographic examination. CBP, but not CAVI, correlated with LVMI (r=0.306, P<0.01). Although CBP correlated only with the Tei index (r=0.201, P<0.05), CAVI correlated with E/A (r=-0.387, P<0.001), S' (r=-0.270, P<0.01), E' (r=-0.362, P<0.01) and the Tei index (r=0.339, P<0.01). Stepwise regression analysis revealed that neither CBP nor CAVI was independently associated with E/A, S' or E'. However, CAVI, but not CBP, was independently associated with the Tei index (beta coefficient=0.311, P<0.001), reflecting both LV systolic and diastolic function. In conclusion, CBP may be suitable for detecting LV hypertrophy. In contrast, CAVI may be suitable for detecting LV dysfunction. This difference, suggesting the clinical value of each parameter, should be kept in mind when we use CBP and CAVI for assessing arteriosclerosis in treated hypertension.