Wearable Technology to Detect Stress-Induced Blood Pressure Changes: The Next Chapter in Ambulatory Blood Pressure Monitoring?

Without tradition, art is a flock of sheep without a shepherd. Without innovation, it is a corpse. Sir Winston Churchill Cuff-based blood pressure (BP) measurement is a century-old technique, dating back to 1896 when Riva-Rocci first used a cuff to measure the pressure required to occlude an artery.1 Oscillometric BP measurement, first described by Marey in 1876,2 has now surpassed auscultation as the preferred clinical measurement method because it is easier to perform and requires less user training. Given that both technologies have changed relatively little over the last 120 years, one cannot help but reflect that innovation in this field has been slow to evolve. However, an important aspect of seeking innovation is that it should not be pursued at the expense of accuracy. Self-monitoring of BP is gaining in popularity and strongly endorsed by clinical practice guidelines because it promotes self-engagement in care. The self-monitoring space is highly competitive, estimated to be worth $1.03 billion USD in 2017, with projected growth to 2.07 billion by 2025.3 Many wearable BP devices (WBPM) known as “cuffless BP” products that estimate BP using pulse transit time or pulse arrival time have been introduced into the market.4 Pulse transit time is inversely related to BP and is the time taken by a pressure wave to travel between 2 arteries as measured by photoplethysmography waveforms.4 Pulse arrival time is the measured time delay between 2 R-peaks of electrocardiogram and a characteristic point on the photoplethysmography waveform.4 To “measure” BP, A “calibration” measurement(s) is performed using the cuffless device and a conventional cuffed monitor to generate a personalized mathematical relationship between pulse transit time or pulse arrival time and BP and then this relationship is used to subsequently estimate (i.e., mathematically calculate) future BP.5 Manufacturers market these devices as accurate, comfortable, and convenient.6 However, in reality, they do not measure BP, they drift (lose accuracy of the original calibration), and are vulnerable to motion artifact.6 Lack of formal accuracy assessment (clinical validation) is another major limitation of most cuffless devices. Proper validation requires performance of an independent study done using a globally accepted BP measurement standard. Further eroding accuracy is the practice or performing initial calibration using an oscillometric device. While pragmatic, this practice introduces further error because each oscillometric device has its own margin of error relative to the true reference standard, blinded, 2 observer mercury-based auscultation. Notably, no current requirement exists mandating that manufacturers validate their device released into the market. An urgent call to action from the Lancet Commission on Hypertension group has been made for mandating independent validation of all BP devices (both cuffless and cuffed) according to the International Standards Organization (ISO) Standard and for the development of validation standards for new BP technologies.7 In this issue, Tomitani et al.8 describe a post hoc analysis of a study comparing the HeartGuide wearable watch-type device (HEM-6410T: Omron Healthcare, Kyoto, Japan) to the TM-2441 ambulatory blood pressure monitoring (ABPM) device (A&D, Tokyo, Japan). Both devices were applied to 50 outpatients who wore the ABPM device for 24 hours and the HeartGuide wearable watch-type device only during daytime hours. Every 30 minutes during daytime hours, the participants had an ABPM measurement and were instructed to stop their activities during these measurements. They were also instructed to self-measure a WBPM measurement after each ABPM measurement with the WBPM device held at heart level. The participants were provided a diary to document the location, physical activity, and emotional state during each measurement Emotional state was categorized into positive (happy/calm) and negative (anxious/tense), generating 575 positive emotional state reports and 67 negative emotional state instances. The study demonstrated a statistically significant BP difference between negative and positive emotional states (9.3 ± 2.1 (systolic blood pressure)/8.4 ± 1.4 (diastolic blood pressure) mm Hg, P < 0.001) with the WBPM. This BP difference between negative and positive emotional states was similar with ABPM (10.7 ± 2.1 (systolic blood pressure)/5.6 ± 1.4 (diastolic blood pressure) mm Hg, P <0.001). Some limitation of this study should be acknowledged up front. The analysis is post hoc and the groups and emotional categorizations were not specified a priori. Of 956 paired (ABPM/WBPM) readings, 100 were eliminated because they represented the initial readings for each participant, a further 139 were eliminated due to lack of accompanying emotional state at the time of the WBPM reading, and 18 readings were excluded due to lack of certainty of the associated emotional state. Overall, 27% of the readings were excluded. Further, the study population was predominately male, with normal body mass index and recorded a narrow range of BP measurements. Whether these results are widely generalizable and repeatable will require further study. Limitations aside, the study is highly innovative because it gives a window into future uses of oscillometric technology. The WBPM is much less obtrusive and has more streamlined ergonomics than a conventional ABPM device, yet it still produces valid BP measurements unlike typical cuffless devices.9 Notably, the accuracy of the WBPM has been confirmed by performing an independent study using to the rigorous ISO validation standard.10 A particularly compelling aspect of this study is that it shows how a WBPM can be used to conveniently study associations between BP variability and emotional state. Although the relationship between emotions and BP has been described for at least 90 years,11 measurement of BP during extremes of emotion for diagnostic, prognostic, and therapeutic purposes has not been routinely recommended. Importantly, foundational studies that have generated the prognostic and therapeutic evidence underpinning the BP thresholds and targets used for the diagnosis and management of hypertension have been performed using measurement protocols that remove emotional state or physical activity as influencers of the BP measurement. However, perhaps BP response to emotional stress is an important parameter to consider and use of this new technology thereby opens up new lines of investigation. One could envision that the WBPM technology, given its convenience factor, could be also be used to study broader relationships between BP and ambulatory activities in individuals and populations, such as ambulatory BP responses within different disease states, and across differing work environments or environmental states. Although this device is limited by the fact that the user has to stop activity, remain still, and use the recommended BP measurement procedures to obtain a measurement, this is still a substantial advance and we note that no technology yet exists that accurately measures BP during motion. If coupled with secure, real-time remote transmission of BP measurements to a cloud and big data analytic and artificial intelligence capabilities, future lines of inquiry could be pursued that assess and predict BP responses in individuals, communities, and populations. In conclusion, the study by Tomitani et al. while limited in size and scope, represents an initial step toward potential broadening use of oscillometric technology, enabling us to gain new insights into ambulatory BP assessment and its relationships to human health and disease. J.R. and R.P. are cofounders of mmHg, a university-based start-up company focused on innovations in BP measurement. R.P. is a member of the Canadian Standards Association and International Standards Organization sphygmomanometer committee.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.