Why is this research needed?
Stroke is a leading cause of death and disability in the UK. High blood pressure (BP) is the biggest cause of stroke. However, if it is identified and managed correctly stroke can be prevented.
BP is usually measured on the upper arm using an inflatable cuff. However, sometimes this isn’t possible – this could be due to arm problems caused by stroke or missing limbs.
BP can be measured in the leg instead. However, we currently don’t know how leg and arm readings are related. This is a problem because current clinical guidelines for treating high BP are based on readings taken in the arm. This means that people who can’t have their BP taken in their arm are at risk of not being properly diagnosed with, and treated for, high BP, which could mean they are at increased risk of having a stroke.
What is this research aiming to do?
This research project will investigate the relationship between BP in the arm and leg, and the link to cardiovascular health conditions including stroke. The researchers will create a model that can be used to predict arm BP readings based on leg BP and patient characteristics, such as age, medical history and body mass index.
How did the researchers do it?
This study will use data from ‘The Inter-arm blood pressure difference individual patient data collaborations’ (INTERPRESS-IPD) dataset. This collates information from a number of research studies to form one large set of data. In total, the INTERPRESS-IPD dataset contains information from 34,000 people from around the world.
The research team will analyse the data in the INTERPRESS-IPD dataset to answer the following questions:
1. What is the relationship between arm and leg BP?
2. Can a person’s leg BP and other characteristics (such as age) predict their arm BP?
3. How do leg BP readings (compared with arm BP) predict strokes and death?
4. Can someone’s leg BP, and the relationship between their arm and leg BP, predict their risk of stroke?
Using data that has been collected in previous studies means that the research team will be able to answer these questions much more quickly, and for a lot less money, than if they had to collect the data of another 34,000 people.
What did the researchers achieve?
The researchers analysed the data to understand the characteristics of the over 33,000 patient participants. They found the mean age of the cohort was 58 years, 45% were female, 20% were smokers, 60% had high blood pressure, 15% had diabetes, 6% had a previous stroke, and the majority were of White ethnicity (77.5%). Prior to further analysis, some patient participants were removed to avoid introducing bias to the models, for example, if blood pressure values were extremely high/low.
On average, leg blood pressure was 12.00 mmHg higher than arm blood pressure. The researchers identified groups of patients with certain characteristics where this difference was smaller or greater.
They used data to make a model that can predict the highest arm blood pressure from leg blood pressure very accurately. They found that the highest leg blood pressure was the better predictor of highest arm blood pressure than lowest leg blood pressure.
Another model was made to predict mortality and cardiovascular events over a 10 year period using leg blood pressure measurements. It was found that the lowest leg blood pressure was a predictor of death from all causes and cardiovascular events, but not death from cardiovascular events. They compared the accuracy of the new model to existing risk scores and found they performed similarly.
What difference will the research make?
The team will publish their results in research journals, and are making a calculator based on the new model that will be available online to clinicians and patients. This can make a big difference in detecting and managing hypertension more effectively for people whose arm blood pressure measurements may not be possible to collect, or accurate.