Introduction: Central systolic blood pressure (cSBP) has been shown to have a higher predictive value than brachial (cuff) pressure. Accurate cSBP, however, is difficult to obtain non-invasively and is often estimated from carotid or transformed peripheral pressures. In this study, the cSBP estimate from the Omron HEM 9000AI was compared to the SphygmoCor cSBP estimate and to carotid SBP. Whilst SphygmoCor uses a radial-to-aortic transfer function to calculate cSBP, the Omron HEM 9000 AI uses a regression equation which relies on the correlation between the second systolic peak of the radial pressure waveform and cSBP.

Methods: Radial applanation tonometry was performed in 251 rural black South Africans (aged 36–91 years) enrolled in the PURE study. Each subject was measured with an Omron HEM 9000AI and a SphygmoCor. Four different estimates of central pressure were calculated: (i) Omron device (cSBP-Omron); (ii) SphygmoCor, with calibration of the radial pressure by brachial SBP and DBP (cSBP-Sphygmo); (iii) SphygmoCor, with calibration of the radial pressure by brachial MAP and DBP obtained from brachial tonometry (cSBP-Sphygmo2, N=201) and (iv) carotid SBP obtained through carotid tonometry calibrated with brachial MAP and DBP (cSBP-carotid, N=143).

Results: All four cSBP estimates were highly correlated (R>0.97).

Conclusion: When using both devices as advocated by their manufacturers, the mean difference in cSBP is 18.6 (4.5) mmHg. Carotid SBP is in the middle between the Omron and SphygmoCor estimates, indicating that the ‘true’ central cSBP should be sought between the Omron and SphygmoCor estimates.