DEVICE-SPECIFIC EMPIRICAL CORRECTION EQUATIONS FOR ACCURATE NON-INVASIVE ESTIMATION OF CENTRAL SYSTOLIC BLOOD PRESSURE IN CHILDREN AND ADOLESCENTS. (April 2021)

• Record Type: Journal Article
• Authors: Mynard, Jonathan; Harrington, Hilary; Smolich, Joseph; Avolio, Alberto; Cheung, Michael
• Is Part Of: Journal of hypertension. Volume 39(2021)e-Supplement 1
• Publication Date: 2021-04
• Subjects: Hypertension -- Periodicals; Hypertension -- Periodicals; 616.132005
• DOI: 10.1097/01.hjh.0000749220.45798.c1
• Languages: English
• ISSNs: 1473-5598

Abstract

Abstract : Objective: Central systolic blood pressure (cSBP) represents the load experienced by the heart and other central organs, and differs from brachial blood pressure due to pulse pressure amplification (PPA). Two widely-used automated devices (SphygmoCor XCEL and Mobil-O-Graph) incorporate algorithms to estimate cSBP and PPA, but in the recent KidCoreBP study, neither passed validation against invasively-measured cSBP in children and adolescents (J Hypertens 38:821–828, 2020). We investigated whether empirical correction equations, using information provided by these automated devices and basic demographics, could provide cSBP and PPA estimates with acceptable accuracy (error mean < 5 mmHg) and precision (error standard deviation < 8 mmHg). Figure. No caption available. Design and method: Using data from the KidCoreBP study (n = 62, SphygmoCor; n = 52, Mobil-O-Graph), stepwise linear regression was used to identify empirical correction equations for predicting cSBP and PPA. Input variables included all blood pressures and pulse wave analysis indices provided by the device, as well as age, sex, height and weight. A leave-one-out cross-validation was then conducted to quantify accuracy and precision. Results: The empirical equations substantially improved estimation of cSBP (–0.1 ± 5.1 corrected vs 7.9 ± 6.8 mmHg uncorrected, SphygmoCor; and 0.03 ± 7.6 vs 5.7 ± 10.3 mmHg, Mobil-O-Graph) and PPA (0.0 ± 4.5 vs 4.2 ± 5.1 mmHg; and 0.0 ± 4.7 vs 6.8 ± 8.4 mmHg). Conclusions: This is the first study to demonstrate acceptable accuracy and precision of cSBP and PPA estimated using automated devices in children and adolescents, when compared to high fidelity invasive measurements. Device-specific empirical correction equations may overcome the present inaccuracy and/or imprecision of available devices.