BACKGROUND: Brachial pulse pressure (BPP) is a predictor of outcome in epidemiologic studies, but brachial and aortic pulse pressure (AoPP) may not correspond and both are influenced by multiple parameters including arterial properties and cardiac output. We sought to what extent pulse pressure (PP) measurements reflected direct measurement of arterial properties, assessed as total arterial compliance (TAC). METHODS: We studied 123 patients (76 men; age 55 +/- 11); 31 normal controls, 46 patients with coronary artery disease and 46 patients with hypertension. PP was determined from brachial cuff pressure and TAC was measured by simultaneous radial applanation tonometry and pulsed wave Doppler of the LV outflow. AoPP was calculated using a transfer function from the radial waveform. RESULTS: There was a significant difference between BPP and AoPP (57 +/- 16 vs. 45 +/- 14; P < 0.0001), although TAC correlated with BPP (r = -0.72; P = 0.01) and AoPP (r = -0.66; P = 0.01). In a multiple linear regression, the difference between BPP and AoPP was predicted by cardiac output (CO) (P = 0.002) and gender (P = 0.03). Bland-Altman analysis showed the best correspondence between BPP and AoPP in the middle tertile (CO 4.7 to 5.7 L min(-1)) with less correlation in the low and high tertiles. The same analysis by gender showed a higher difference in women than men (14 +/- 6 vs. 10 +/- 5; P < 0.0001). The difference between BPP and AoPP showed the best correlation in the control group and the worst in the hypertension group. CONCLUSION: BPP correlates with TAC in men with normal cardiac function. However, in women, in patients at the low and high extremes of function, and in patients with preclinical and overt cardiovascular disease, there appears to be incremental value in measuring TAC.