Validation of Two Pulse Transit Time‐Based Blood Pressure Measurement Devices

Abstract

BackgroundBlood pressure (BP) has been measured with a cuff for over a 100 years. Recently, ‘tricorders’ and smartwatches that measure BP without a cuff using pulse transit time (PTT) have become available. These BP measurements are based on the inverse relationship between BP and PTT. PTT can be measured as the timing delay in a QRS complex on an EKG and the onset of a photoplethysmography wave, for example measured from a finger. Since these measurements are relatively more user‐friendly than conventional cuff‐based measurements they may aid in more frequent BP monitoring. Using a guidelines‐based protocol, we investigated the accuracy and precision of two popular PTT‐based BP measuring devices: the Everlast TR10 fitness watch (Everlast, New York City, NY) and the BodiMetrics tricorder (BodiMetrics, Manhattan Beach, CA).MethodsWe enrolled 127 patients (>18 y) from the Thomas Jefferson University Hospital Preadmission Testing Center. For the Everlast smartwatch and Bodimetrics tricorder measurements, research staff were trained to measure BP according to manufacturer guidelines; Bodimetrics measurements required entry of date of birth, sex, height, weight, and an initial calibration using a cuff. The Everlast smartwatch provides measurements of systolic and diastolic BP, whereas the Bodimetrics tricorder only provides systolic BP. For reference BP measurements, research staff were trained to follow a standard protocol using a hospital‐grade, validated, automated sphygmomanometer (Cardiocap 5, Datex‐Ohmeda, Madison, WI). After 5 minutes of quiet sitting, four standard and three investigational BP measurements were taken with sixty seconds in between each measurement. The reference BP value was calculated by determining the average of the 2 standard BP measurements that bounded the investigational measurements. Using this method, we determined 3 comparison pairs for each investigational device in each subject. We calculated the mean (SD) of the absolute difference between the respective investigational devices and the reference for systolic and diastolic BP. After excluding data from 42 individuals because of excessive variation in sequential standard BP measurements per prespecified dropping rules, data from 85 subjects were used for final analysis.ResultsOf 85 participants, 36 (42%) were women. The mean (SD) age was 53 (21) years, body mass index was 28 (7) kg/m^2; 32 (38%) self‐reported hypertension, and 97% of these (n=31) reported taking antihypertensive medications. The average differences between the Everlast watch and reference were 22.7 (27.4) mm Hg for systolic BP and 6.9 (6.2) mm Hg for diastolic BP (Figure). The average difference between the Bodimetrics tricorder and the reference was 5.3 (4.7.) mm Hg for systolic BP.ConclusionsThe fitness watch and the tricorder we tested are not accurate enough to be used as BP measurement devices. Important sources of physiological bias in PTT‐based BP measurement may be changes in vascular tone and left ventricular isovolumetric contraction time that are independent of the static relationship between BP and PTT that is based on the mechanical properties of the vessel wall.Support or Funding InformationThis work received no specific grant from any funding agency.This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.