Cuffless Blood Pressure Measurement: Great Potential, Uncertain Accuracy

An Interview With George Stergiou, MD

Tricia Ward


March 18, 2022

The promise and pitfalls of cuffless blood pressure measurement were addressed during the 2022 Cardiovascular Health Tech virtual conference, sponsored by the Institute of Electrical and Electronics Engineers Engineering in Medicine and Biology Society technical committee on Cardiopulmonary Systems and Physiology-Based Engineering.

George Stergiou, MD

Medscape spoke with presenter George Stergiou, MD, Professor of Medicine & Hypertension at the Hypertension Center, University of Athens, Greece.

This interview has been edited for length and clarity.

Is the condensed version of your presentation that cuffless blood pressure devices cannot be relied on because of their uncertain accuracy?

Yes. In order to recommend something for clinical use, it has to be able to provide reliable information on which to base patient decisions. For something new to be useful in medicine, it need only do something a little bit better by certain criteria — maybe patients prefer it, maybe it improves outcome — but we need it to show some benefit on top of what we have.

We don't just add new things because they provide information. We have to prove that the information is reliable and can improve how we provide care.

The only FDA-approved cuffless blood pressure is the Biobeat smartwatch and patch. What other products are in development?

There are a lot of cuffless devices close to being approved, but none are recommended by the medical community for use in patients for clinical decision-making.

Approval is a validation that focuses on safety and less so on accuracy. So a cuffless device may be approved, but the other question is whether it is recommended.

Consumers like wearables, so they may use them even if the medical profession doesn't recommend or like them.

The medical profession very much likes cuffless blood pressure devices in terms of the concept. For almost a century, we have been dependent on the cuff for blood pressure readings. But the cuff is problematic. It's difficult to fit for people with small arms or very big arms, and patients don't like it. Cuff readings provide very little information.

For example, my blood pressure sitting and relaxed is different from when I walk. We would like to know the dynamic behavior of the blood pressure of each individual without the annoyance of the cuff. With ambulatory blood pressure monitoring, the cuff inflates every 20 minutes — some people are annoyed by that and won't use it.

So we like the concept of cuffless devices and the idea of getting another level of information for blood pressure behavior.

Is your concern that they are being approved on the basis of standards for cuff devices?

Yes, this is a major concern. The standards we have been using for many years that work well and have been helpful in evaluating cuffed devices are not sufficient for cuffless devices.

For cuffless measurement, we are looking at very different technologies, and they have different intended uses. For example, I could have a wrist-worn cuffless blood pressure monitor; I wear it for weeks and it takes lots of measurements. Or I could use a mobile phone app; I have to press the button with my finger to take a measurement. This is a different kind of measurement because I decide when to take a reading. Another type is facial video technology that can pick up signals from the face and measure blood pressure. Yet another measures beat to beat for patients under anesthesia or in the intensive care unit (ICU).

I just described four very different scenarios not only in terms of technology, but also in terms of intended use. Their validation might be different. We can't have a standard for cuffless blood pressure measurement without knowing which type of cuffless measurement.

Let's focus on the wrist-worn smartwatch type of device. Most of these have to be calibrated with a cuff measurement. Is the concern that if the calibrated reading is, say, 140/80 mm Hg and then the device detects a reading that deviates from that, the algorithm will assume that reading is wrong and display the calibrated reading instead?

You've picked up a key problem. The devices that currently require calibration do not measure blood pressure. They measure blood pressure changes.

We worked for several years to develop a standard to validate the cuff devices. We succeeded a few years ago and got everyone to agree on a single protocol. This universal protocol involves sequential blood pressure measurements using the test device alternated with the reference device which is manual auscultatory.

When you calibrate a cuffless device, you essentially tell the device your blood pressure measurement, and it uses this as a reference. The cuffless device will pass the universal validation protocol because it repeatedly shows the calibrated blood pressure.. This protocol provides zero information on cuffless devices — zero. It's really a joke.

You and others propose a validation method that includes measurements taken after the blood pressure changes, such as when the patient takes an antihypertensive.

Yes; we need to know whether these devices are able to track blood pressure changes. If I'm sitting down, my systolic blood pressure might be 120 mm Hg; then, when I walk, it's 160 mm Hg; and when I go to bed, it drops to 100 mm Hg. A cuffless device must prove that it can track such blood pressure changes using an established protocol — this is the most important as well as the trickiest part of evaluating cuffless blood pressure devices.

The devices that are dependent on calibration are unable to track these blood pressure changes. I'm not an engineer, so don't expect me to fully understand these technologies, but I know what I need to care for my patients.

The engineers realized the weaknesses of these technologies several years ago.

Do the engineers think they're close to getting there?

They are not happy as yet with the cuffless devices. We haven't seen consistent and convincing data.

There are two types of tests we need to assess any kind of technology: lab tests and clinical field tests. You may have a device that is accurate under sterilized research conditions, but I also need a clinical field test to see whether it works for my needs.

At this stage, we need to be very careful before using this new technology in patients. Don't misunderstand me — I'm very excited about cuffless devices. If the questions I have are properly addressed, I may forget all the other tools I have.

As you said, you're not an engineer. but can you explain the technologies used? Is it mainly pulse transit time or pulse wave amplitude?

The main technology is pulse wave velocity; it's the speed with which the blood travels across the arterial tree, based on different calculations. It's been used for several years in medicine to evaluate the stiffness of the arteries, for example, so it's not entirely new.

But whether you can extract data from this to make a good estimate of blood pressure — that is something we are still debating.

Then there's PPG — the green lights on the back of a Fitbit or Apple Watch.

Also not something new. It has been successfully used to give us information about the cardiovascular system. But the estimation of blood pressure is quite tricky.

The problem is that we have a lot of clinical evidence on hypertension that is based on the classic cuff measurement. We can't throw that out the window. Whatever we use must accommodate the information we have.

Products like the Omron and Huawei D watch are wrist-worn, but they have a little cuff that you inflate, and then you hold the device across your heart to take a reading. Are they categorized as cuff devices?

These are very sophisticated devices. They are wrist-worn cuff devices. They use interesting algorithms, but they are not cuffless, so whatever I said about cuffless devices does not apply to these devices.

Would they be equivalent to a home blood pressure monitor or an ambulatory blood pressure monitor?

They don't provide ambulatory readings. These wrist-worn cuff devices are more like home blood pressure monitors. You have to sit with your wrist at this level (across the heart), so this means that you decide when to take a reading.

A quite different use for cuffless blood pressure monitoring is in the ICU patient. In those cases, you're not looking to measure blood pressure with an accuracy of, say, ± 5 mm Hg; instead, you're looking to detect a significant drop telling you to rush emergently to this patient. I assume that would have a different standard.

This is an important point. Blood pressure is two things. One is a risk factor for cardiovascular disease —say, 140 mm Hg, 160 mm Hg. The other is a vital sign — I'm alive, I'm dead, or I'm close to dead. For this, it's zero, 50, 150 mm Hg. In a patient with hypertension, an increase of 5 mm Hg makes a big difference; I may add another drug. But in the ICU, it tells me nothing.

In the ICU, we want beat-to-beat blood pressure monitoring to detect sudden changes. We are talking about a different technology. The way to validate it is different. In this case, your reference standard is the real intra-arterial measurement, which is different from millimeters of mercury.

Now you see how complex it is. Because it took a long time to get a universal standard for cuffed devices, and now we realize we can't use it for cuffless devices and that we will need several standards for the different cuffless technologies.

Looking at traditional cuff blood pressure measurement, an engineer will wonder why on earth you're using such an indirect measurement. But your counter as a clinician is, "I have a body of evidence that tells me what this means clinically."

You are right that occluding the artery with a cuff and using a stethoscope is not a measurement; it is an estimation. To measure the blood pressure in the real sense of the word, you would put a catheter directly in the artery.

A cuff measurement with auscultation is not the same as an intra-arterial reading. But it works very well in diagnosing hypertension and in preventing the complications.

Hypertension is one of the most well-studied conditions in medicine, if you look at the amount of evidence and the number of studies we have, and the thousands of people in each study who were randomized to this and that and followed for 5 years. All of this is based on auscultatory measurement.

We can't have another tool that gives us different information because we wouldn't know how to use it.

You said that it might take 20 years of research to know what a new measurement meant in terms of clinical outcomes. Do you think that there might be a clever algorithm that could correlate that evidence base to a more direct measurement, so you could get there quicker vs having to do all the work again?

In theory, yes. We had been working with the auscultatory method for many years, when oscillometry appeared. The developers realized that in order to be successful, they had to replicate the numbers that auscultation gives

This is the way to go for anything new, because hypertension affects 40% of the adult population. It's not me, the hypertension expert, who is treating all the hypertensive patients — it falls to the general practitioners.

If we change what hypertension is, that would take longer than 20 years. We have to build on top of what we have. We can't measure something different.

So millimeters of mercury aren't going away anytime soon.

Medicine is famous for resistance to change and for being conservative. Pressure is measured in kilopascals everywhere except for blood pressure. We can't go to doctors and tell them if you want to measure blood pressure, from now on, we're measuring kilopascals.

We need to be able to translate it to usual care because at the end of the day, it's about improving patient care.

There are unapproved devices on sale that claim to provide cuffless blood pressure measurement. Do you think consumers know how inaccurate these are?

Consumers do not know and doctors do not know. As I said, I'm very excited about this technology, and because I'm involved in research on blood pressure monitoring, I know the problems. But what about other doctors, patients, and the public? Some companies mislead people in what they claim their technologies provide.

I'm chairman of the European Society of Hypertension (ESH) Working Group on Blood Pressure Monitoring. I'm 100% sure it's the strongest blood pressure monitoring group in the world — we involved people from Australia, Japan, Germany, the UK, and the United States. We're working on a statement about cuffless blood pressure monitoring that will probably be published in the summer.

I think it is important to distribute this information because doctors need to know more about these devices.

I'm also involved in the website STRIDE BP, where we list validated cuff devices for home, ambulatory, office, or in-hospital use and for use in pregnant women and children. We are supported by ESH, the International Society of Hypertension, and the World Hypertension League.

We considered including cuffless devices a few years ago, but for now, we do not recommend cuffless devices.

Is there anything else you want to add?

I've been working in blood pressure monitoring for 30 years, covering the methodology and technology. What's different about cuffless blood pressure monitoring manufacturers is that we have many small start-ups and we have huge companies.

I won't name them, but if you think of the three top technology companies in the world, they are interested in cuffless blood pressure monitoring. Investment is not an issue for these huge companies, and yet so far they have failed. What they have said is that they are not ready.

If these people are not ready and then you see a small start-up throwing devices on the market, you should be a bit skeptical.

George Stergiou, MD, discloses that he has worked in an advisory capacity with several blood pressure monitoring companies and has received consulting fees and research grants from several pharmaceutical companies in the field of hypertension and cardiovascular medicine.

Tricia Ward is an executive editor at Medscape who primarily covers cardiology. She is based in New York City and you can follow her on Twitter @_triciaward

Follow Tricia Ward on Twitter

Follow | Medscape Cardiology on Twitter

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.