Study questions accuracy of blood pressure readings

A small-scale UNSW study suggests that many blood pressure readings may be questionable due to the way traditional cuffs are calibrated.

It is already known by experts that the accuracy of brachial cuffs is questionable, but they continue to be the dominant method for blood pressure measurements in hospitals and surgery practices across Australia and the world. 

Reliable measurements of blood pressure are critical for diagnostic decisions and treatment strategies. Hypertension affects more than 1.1 billion people worldwide and when not properly managed, increases the risk of cardiovascular diseases, including heart attacks and strokes.

Misdetection of what are known as Korotkoff sounds is a potential reason measurements using cuffs may be significantly unreliable, according to research published in the Journal of Hypertension by Professor Branko Celler, an Emeritus Professor at UNSW in the School of Electrical Engineering and Telecommunications.

Celler’s study, involving only 40 patients, indicates that the accurate detection of those Korotkoff sounds during calibration of the equipment to the accepted Universal Standard is problematic.

“We’ve actually known for 100 years or more that we systemically underestimate systolic blood pressure, but we’ve never had an answer to it,” Celler said.

“This is a fundamental and important problem that’s been identified. The implication of this study is really quite profound and I acknowledge the fact it’s hugely disruptive to current accepted practices.

“The disruption comes from the fact that, effectively, every non-invasive blood pressure device in the world is calibrated using Korotkoff sounds and may therefore be in error.

“It is not our intention to suggest that blood pressure readings should not be taken using these cuffs. The current methods have served very well in the battle against hypertension.

“What we are suggesting is that improvements need to be made to eliminate the sources of error we have identified.”

Korotkoff sounds

The process to calibrate a brachial cuff begins when it is inflated to the point where blood flow is temporarily stopped in the brachial artery, the main blood vessel in the upper arm.

Next, two trained operators listen for what are known as Korotkoff sounds using a stethoscope placed over the brachial artery at the elbow crease. These sounds, named after Russian physician Nikolai Korotkov, who discovered them in 1905, indicate blood flow changes as the pressure in the cuff is released.

The sounds are critical because they indicate when blood starts to flow back into the artery as the cuff pressure is slowly released. The first Korotkoff sound, a clear tapping noise as blood first begins to flow again, is deemed to match the systolic pressure — the maximum pressure in the arteries when the heart beats.

As the cuff continues to deflate, the sounds change and eventually disappear. The point at which the Korotkoff sounds completely stop indicates the diastolic pressure — the minimum pressure in the arteries when the heart is at rest between beats. This last audible sound is crucial for determining the lower number in a blood pressure reading.

Accurate calibration is required to ensure subsequent blood pressure measurements taken with the brachial cuff are precise and reflect true systolic and diastolic pressures.

Underestimation

However, the research led by Celler and involving the study of 40 people at the Concord Repatriation Hospital’s Catheterisation Laboratory in Sydney indicates that underestimation of systolic blood pressure, based on that first Korotkoff sound, can be as high as 30 mmHg.

Celler and his team analysed the data of tests on 40 people (35 men and 5 women, with an average age of 64) who had catheters inserted to record their blood pressure from directly within their artery.

Simultaneously, they also had their blood pressure measured using a non-invasive brachial cuff, and the Korotkoff sounds recorded as the pressure in the fully inflated cuff was released.

The results showed that for the group studied, the occurrence of the first Korotkoff sound did not represent an accurate systolic blood pressure reading due to a delayed reopening of the brachial artery as the cuff deflated.

In around one-third of the cases this discrepancy was 10 mmHg or less, but in the remaining two-thirds the error was more significant than that, with the maximum difference being 31 mmHg.

“What we showed is that when you deflate the cuff, the artery sometimes doesn’t open up — and the amount of time that goes on for is very variable,” Celler said.

“In some cases the artery does open immediately, but in other cases it takes a lot longer, when the pressure in the blood is up to 30 mm lower. In our study the average error was about 17 mm. The Universal Standard for the calibration of cuffs allows for an error of between 5 and 8 mm, so this is much above that.”

Although not part of the research team, Professor Alta Schutte, principal theme lead of Cardiac, Vascular and Metabolic Medicine in the Faculty of Medicine and Health at UNSW Sydney, said the results are intriguing.

It would take a lot of time and money to amend the current guidelines related to brachial cuffs, Schutte said.

“The findings from Professor Celler’s team are interesting as it shows that the blood pressure recorded with traditional non-invasive methods are not identical to the intra-arterial pressures,” Schutte said.

“Although profound, current medical practice and clinical guidelines are based on clinical trial evidence that showed clearly that cuff-derived blood pressures are strongly predictive of outcomes such as stroke and mortality.

“It is possible that more accurate blood pressure measurements that align better with intra-arterial pressure would have even stronger predictive value, but it would take major studies and millions of dollars to repeat such trials.

“I think it is thus highly unlikely that current medical practice and guidelines using cuff-based blood pressures would change anytime soon.”

Celler acknowledged that solving the known problem of underestimating blood pressure readings using cuffs will not be an easy task.

“I have presented this work at the European Hypertension Conference and some clinicians said these results were too disruptive and the consequences would be too hard to deal with given the widespread use of cuffs to record blood pressure,” he said.

“But others were keen to know more because they recognised that if a blood pressure reading from a cuff was 120 mmHg and the true BP was up to 30 mm higher, then their management of that patient would be very different.

“The problem is that the solution to this issue is not easy. I just hope that identifying the scale of this problem will now spark a raft of new research worldwide.”

Collaborative research required

In the paper, the researchers state clearly that more studies are needed to fully understand the reasons for the delayed reopening of the brachial artery as the cuff pressure is reduced.

Celler has developed mathematical models that allow the prediction of true systolic blood pressure (SBP), based on the data obtained during the research and a study of a number of variables including pulse pressure and peak cuff pressure, as well as the values recorded from the onset and cessation of the Korotkoff sounds.

However, he acknowledges those need to be validated in much larger trials. In the interim, the UNSW academic is looking into more direct instrument-based methods of estimating true SBP.

Via $75,000 of funding from UNSW’s Innovation Community, Celler has already developed a new experimental non-invasive blood pressure device that simultaneously displays different BP estimates from Korotkoff sounds, an oscillometric ratio method, an oscillometric gradient method and his own mathematical model estimate.

In terms of the errors in blood pressure readings when using the cuff, Celler had suspected arterial stiffness might be a cause, but the results did not back up his original hypothesis. Nor was any correlation found between those who had high, or indeed low, blood pressure and the level of discrepancy in their readings.

“This needs to be researched by many more people. I'm trying to build a collaborative team, and I’ve had very strong feedback around the world, including from people at Cambridge University and elsewhere,” he said.

“I don’t present this as if I have all the answers. I have identified the problem and I want to put this out there to see what other people think.

“Because this is so disruptive to traditional methods that have been in place for so long, I know it might take 10 years to find a solution and potentially amend the international standards.

Image credit: iStock.com/Jfanchin