Intended for healthcare professionals


Difference in blood pressure between arms might reflect peripheral vascular disease

BMJ 2001; 323 doi: (Published 18 August 2001) Cite this as: BMJ 2001;323:399
  1. C E Clark, general practitioner (Chris.Clark{at}
  1. School Surgery, Witheridge, Devon EX16 8AH

    EDITOR—I believe that McAlister and Straus underestimate the frequency and significance of a blood pressure difference between the arms.1 They quote 6% from the paper by Harrison et al, but this group reported a difference in 10/131 (that is, 7.6%) normotensive subjects for systolic or diastolic differences, 44/310 hypertensive patients (14%) for a systolic difference, and 31/310 hypertensive patients (10%) for a diastolic difference.2

    I have reviewed the English language literature and identified 11 studies with comparable data on at least 100 subjects. These studies (table) reported prevalences ranging from 12% to 18.4% for a systolic difference ≥20 mm Hg and 13% to 33.7% for a diastolic difference ≥10 mm Hg in selected populations. No publications were identified from primary care.

    Summary of previous studies searched

    View this table:

    I have been prospectively gathering pairs of readings from hypertensive patients. To date I have collected 435 pairs of recordings from 205 patients. The mean absolute systolic difference is ≥10 mm Hg in 64 (31%) patients and ≥20 mm Hg in eight (4%). The mean absolute diastolic difference is ≥10 mm Hg in 27 (13%) patients. These data suggest that identification of any difference in blood pressure between arms is a vital part of the assessment of hypertensive patients if their diagnosis is to be accurate and their response to treatment reliably monitored.

    The causes of a pressure difference between arms are unclear. Harrison et al found fewer differences with intra-arterial measurements than with indirect techniques, suggesting that variations in the measurement techniques or the soft tissues may play a part.2 For some patients the difference is vascular in origin. This may be due to characteristics of flow in a normal arterial tree,3 but I suggest that vascular disease may also cause a difference. One study found an increased prevalence of differences in patients with coronary heart disease or peripheral vascular disease,4 and another showed that 83% of vascular surgical patients with differences had angiographic evidence of innominate or subclavian artery stenosis on the side of the lower pressure.5

    The importance of a difference is already recognised between the arm and the leg, as measured by the ankle-brachial pressure index, which is reduced in the presence of asymptomatic peripheral vascular disease; a reduced index is associated with increased mortality. Why should the pathology, and prognostic implications, not be the same with differences between arms?

    Until more work is done, hypertensive patients with a reproducible difference in blood pressure between arms should be investigated and managed intensively, on the assumption that they have asymptomatic peripheral vascular disease.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    View Abstract