Secondary Logo

Institutional members access full text with Ovid®

Comparison of central pressure estimates obtained from SphygmoCor, Omron HEM-9000AI and carotid applanation tonometry

Kips, Jan Ga,b; Schutte, Aletta Ec; Vermeersch, Sebastian Ja,b; Huisman, Hugo Wc; Van Rooyen, Johannes Mc; Glyn, Matthew Cc; Fourie, Catharina Mc; Malan, Leonec; Schutte, Rudolphc; Van Bortel, Luc Ma; Segers, Patrickb

doi: 10.1097/HJH.0b013e328346a3bc
Original papers: Blood pressure measurement

Background The Omron HEM-9000AI is the first automated tonometer to provide an estimate of central SBP (cSBP), which is considered to be more predictive of cardiovascular events than brachial pressure. However, considerable differences between the cSBP estimate of Omron and that of SphygmoCor have been reported, but not explained. This study assesses the sources of differences between both cSBP estimates and provides a handle on which estimate is closest to reality.

Method For this purpose, aortic cSBP derived from calibrated carotid SBP was used as device- and algorithm-independent reference. Radial, brachial and carotid applanation tonometry were performed in 143 black South Africans, aged 39–91 years. Each individual was measured with an Omron HEM-9000AI and a SphygmoCor.

Results When using both devices as advocated by their manufacturers, the corresponding cSBP estimates correlated strongly (r = 0.99, P < 0.001), but the Omron estimate was 18.8 (4.3) mmHg higher than the SphygmoCor estimate. Aortic SBP was in between both estimates: 11.7 (5.5) mmHg lower than cSBP-Omron and 7.1 (5.0) mmHg higher than cSBP-SphygmoCor. Alternative calibration of the radial SphygmoCor-curves with radial instead of brachial pressures yielded a cSBP that was 3.0 (4.2) mmHg lower than aortic SBP. The shape of the recorded pressure waves was similar in both devices: less than 5% of the observed cSBP difference was caused by differences in wave shape.

Conclusion The results from this study demonstrate that the considerable difference between the central pressure estimates of Omron HEM-9000AI and SphygmoCor is due to algorithm differences, and suggest that the overestimation by Omron HEM-9000AI is larger than the underestimation by SphygmoCor.

aHeymans Institute of Pharmacology, Belgium

bIBiTech-bioMMeda, Gent University, Gent, Belgium

cHypertension in Africa Research Team (HART), North-West University, Potchefstroom, South Africa

Received 28 October, 2010

Revised 28 February, 2011

Accepted 10 March, 2011

Correspondence to Jan Kips, Clinical Pharmacology, Ghent University, Block B, De Pintelaan 185, 9000 Gent, Belgium Tel: +32 9 332 0042; e-mail:

© 2011 Lippincott Williams & Wilkins, Inc.