Repeatability of individual calibration for cuffless estimation of blood pressure from arterial pulse wave velocity

Objective: The determination of ambulatory blood pressure (BP) holds strong ground both clinically and commercially. Conventional, cuff-based methods are discontinuous and inconvenient, requiring compressive inflation of a cuff on the arm. Estimation of BP using arterial pulse wave velocity (PWV) is one technique potentially eliminating these drawbacks. Accuracy depends on precise prediction of BP from PWV (calibration). This study investigates the repeatability of the BP/PWV relationship using a postural-induced change in BP . Design and Method: Nineteen participants (age 29±13 years, 11 female) underwent an oscillometric, cuff-based brachial BP assessment in a seated position, followed by measuring carotid-femoral PWV using a tonometer (carotid pulse) and a femoral cuff (SphygmoCor XCEL, AtCor Medical, Australia). Measurements were repeated in the supine position, inducing a hydrostatic change in BP, and therefore PWV, resulting in a BP/PWV relationship. Measurements were repeated 1 to 5 days later to quantify repeatability as analysed by Bland-Altman plots and paired t-tests. Results: The BP/PWV relationship in session one was 19±2 mmHg/m/s and in session 2 20±2 mmHg/m/s (difference 1±11 mmHg/m/s, relative error 6%, p=0.65). Variability was driven through PWV measurement variability (difference 0.2±0.8 m/s, relative error 8%, p=0.41) more than BP measurement (difference 0.6±5.0 mmHg, relative error 1%, p=0.61). Conclusion: On average the postural change produced a reproducible calibration curve though there was variability across the cohort driven by the apparent change in PWV, as reflected by the relative error. Such individual calibration techniques have merit but require proper quantification and investigation of repeatability before deployment in devices.