Comparative study of two ambulatory blood pressure monitors that use differing methods of calibration to determine central aortic blood pressure from the brachial waveform

Objective: Central aortic blood pressure (aBP), a potentially superior indicator of cardiovascular risk, can be measured non-invasively using ambulatory blood pressure monitors (ABPM). Current devices utilize differing methods of calibration to determine aBP from the brachial pulse waveform. Calibration methods affect accuracy of measurements and the ability for interchangeable measurements between devices. This study study aims to compare simultaneous measurements from two ABPM devices to determine differences in peripheral pressure and derived aBP. Design and method: Eleven participants (3 male, aged 33?15 years) simultaneously wore the Mobil-O-Graph (M-o-G; I.E.M,Germany) and BPLab (Petr Telegin, Russia) ABPM with arm assignment randomized. The devices utilize differing calibration methods for the brachial pulse wave (MoG: mean/diastolic; BPLab: systolic/diastolic). Brachial and aBP readings were taken simultaneously (3 minute range) every 15 minutes over a period of 7 hours during the day. Parameter means produced by each device for the 7-hour period were analysed for significant differences using paired t-tests, linear regression and Blant-Altman plots. Results: Despite small differences between overall means, the Mobil-O-Graph produced significantly higher estimations of diastolic blood pressure (DBP), mean arterial pressure (MAP) and aortic diastolic blood pressure (aDBP), while the BPLab produced a significantly higher estimation of peripheral pulse pressure (pPP) (Table). There was a very large difference in aortic augmentation index (aAIx, %), with the Mobil-O-Graph consistently producing a higher value. Linear regression indicated strong correlation coefficient for all parameters except pressure pulse amplification (pAmp). Conclusions: The two devices with differing calibration methods produced essentially comparable results for calibrated parameters of blood pressure. However, large differences were observed in aAIx, an uncalibrated, waveform-dependent parameter. Further studies on larger sample sizes and additionally taking into account potential inter-arm differences are advised.