Arm Based on LEg blood pressures (ABLE-BP): Can leg blood pressure measurements predict brachial blood pressure? An individual participant data meta-analysis from the INTERPRESS-IPD Collaboration

Talk Code: 
Sinead McDonagh
James Sheppard, Fiona Warren, Kate Boddy, Leon Farmer, Philip Lewis, Rachel Baumber, Una Martin, Christopher E Clark
Author institutions: 
University of Exeter Medical School, University of Oxford, Volunteer patient and public advisor, Stepping Hill Hospital Stockport NHS Foundation Trust, Royal National Orthopaedic Hospital NHS Trust, University of Birmingham


Hypertension, a key modifiable risk factor for the prevention of stroke, is diagnosed and managed using blood pressures (BP) measured on the upper arm. Amputations, altered muscle tone after stroke or limb deformities can prevent accurate measurement of brachial BP and cardiovascular risk estimation, thus placing a cohort with known vascular risks at a disadvantage. Leg BP measurement is often used as a practical alternative to brachial measurement, but limited data exist to guide clinicians’ interpretation of leg BP values in terms of hypertension diagnosis and treatment.Findings from our study-level systematic review and meta-analysis suggested that, on average, systolic BP was 15 mmHg higher in the leg than the arm. However, substantial heterogeneity between contributing studies existed, due to factors such as BP measurement method and patient characteristics, limiting applicability of this finding to individuals.Using arm and leg BP data from the Inter-arm BP difference individual participant data (INTERPRESS-IPD) Collaboration, which holds 34,543 individual records from 15 international studies, we now aim to:1) Examine the relationship between arm and leg BP.2) Develop and validate a multivariable model predicting arm BP from leg BP.3) Investigate the prognostic role of leg BP in cardiovascular event and mortality risk prediction.


In an observational cohort design, IPD meta-analyses will be undertaken to explore the cross-sectional relationships between arm and leg BP in one- and two-stage multivariable models. Using hierarchical linear regression models with participants nested by study, we will investigate the association between arm and leg BP and participant characteristics. Planned predictor variables will include age, sex, body mass index, cardiovascular disease risk (defined by various cardiovascular risk scores), past medical history and use of antihypertensive medication. Prognostic models will also be derived for all-cause and cardiovascular mortality and cardiovascular events. Heterogeneity will be assessed using I2 and tau2. Study quality will be assessed using a modified version of the Quality In Prognostic Studies tool.


Arm and leg BP records exist for 34,543 individuals (mean age: 61.7 years, mean arm systolic/diastolic BP at baseline: 138/79 mmHg, 52% female). A total of 20,576 (59.6%) have hypertension, 5,433 (15.8%) have diabetes and 7,565 (22.3%) have cardiovascular disease (including 2,233 (6.5%) with stroke or transient ischaemic attacks and 1,145 (3.6%) with peripheral artery disease). The median follow-up period is 8.5 years, with 3,870 (11.6%) participants experiencing cardiovascular events and 3,096 (9.0%) dying within 10 years. Further analyses are underway and arm-leg BP models will be presented at the conference.


We will provide the first evidence-based method for estimating individual brachial systolic BP, and cardiovascular risk, from leg BP measurements. Our findings will support clinicians and patients in detecting and managing hypertension more effectively, where leg measurements are required.

Submitted by: 
Sinead McDonagh
Funding acknowledgement: 
Funding: Stroke Association and The Thalidomide Trust. The views expressed are those of the authors and not necessarily those of the funding bodies. The INTERPRESS-IPD Collaboration was funded by National Institute for Health Research (NIHR) Research for Patient Benefit programme (PB-PG-0215-36009). The views expressed are those of the authors and not necessarily those of the NIHR, the NHS or the Department of Health.’