Differences in blood pressure between arms

BMJ 2012; 344 doi: http://dx.doi.org/10.1136/bmj.e2033 (Published 20 March 2012) Cite this as: BMJ 2012;344:e2033
  1. Dae Hyun Kim, instructor in medicine
  1. 1Division of Gerontology, Department of Medicine, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA, 02215, USA
  1. dkim2{at}bidmc.harvard.edu

May be diagnostically useful but needs further evaluation as a prognostic marker

A difference in blood pressure readings between arms can be seen in congenital heart disease, aortic dissection, peripheral vascular disease, and unilateral neuromuscular abnormalities. In the absence of these conditions, any discrepancy is small (mean difference: 5 mm Hg and 4 mm Hg for systolic and diastolic blood pressure, respectively).1 About 20% of patients in primary care or secondary care have a between arm blood pressure difference of 10 mm Hg or more and 4% have a difference of 20 mm Hg or more.2 Although such a difference in blood pressure is thought to be a marker of atherosclerosis,3 its clinical significance is not fully understood. In a linked research study (doi:10.1136/bmj.e1327),4 and in a recent meta-analysis,5 Clark and colleagues provide additional evidence on the diagnostic and prognostic relevance of this phenomenon.

In their meta-analysis of cross sectional studies,5 Clark and colleagues reported that a between arm blood pressure difference of 15 mm Hg or more was associated with peripheral vascular disease (sensitivity 15% and specificity 96%) and with cerebrovascular disease (sensitivity 8% and specificity 93%), but not with coronary artery disease. A difference of 10 mm Hg or more was associated only with peripheral vascular disease (sensitivity 32% and specificity 91%).5 The overall low sensitivity and high specificity suggest that the measurement of blood pressure in both arms is not a good screening test for asymptomatic peripheral vascular disease.

On the basis of these prevalence data, it might seem sensible to examine whether, as a marker of asymptomatic peripheral vascular disease of the upper extremities, a difference in blood pressure between arms predicts future cardiovascular events and mortality. Although extensive data show an association between peripheral vascular disease of the lower extremities and both all cause mortality and mortality from cardiovascular disease,6 little evidence exists for the prognostic value of between arm blood pressure differences.4 7 The authors of a prospective study of 1872 community dwelling adults and patients evaluated for peripheral vascular disease of the lower extremities in the United States found that a between arm systolic blood pressure difference of 15 mm Hg or more might be associated with a modest increase in all cause mortality and mortality from cardiovascular disease, but the findings were of borderline statistical significance.7 The linked study, however, reports that a difference of 15 mm Hg or more was significantly associated with all cause mortality and mortality from cardiovascular disease in 230 adults with hypertension in a primary care setting in the United Kingdom over a median follow-up of 9.8 years.4 In the authors’ meta-analysis, the pooled hazard ratios were 1.6 (95% confidence interval 1.1 to 2.3) for all cause mortality and 1.7 (1.1 to 2.5) for mortality from cardiovascular disease.5

The strengths of the study include its prospective design, generalisability, and long term follow-up. However, there are some limitations to consider when interpreting the findings. Because true within person variations in blood pressure exist and measurement errors occur,8 repeated simultaneous blood pressure measurements are needed for accurate measurement of differences between arms.1 Clark and colleagues used the average of a single set of sequential readings taken over three visits for their estimate of the difference between arms, so that people with highly variable blood pressure could have been misclassified as having a large difference.4 Evidence suggests that people with such variability have an increased risk of cardiovascular events.9 The Framingham risk score was computed using crude categories for risk factors (for example, current smoking instead of pack years), and this may have allowed residual confounding.10 In addition, unmeasured confounding as a result of the use of antiplatelet agents, antihypertensives, and lipid lowering drugs might have led to overestimation of the association. Furthermore, given the modest sample size, some changes in the associations after adjusting for pre-existing cardiovascular disease may reflect unstable estimation from sparse strata, rather than true adjustment for confounding (for example, only one of 47 patients with pre-existing cardiovascular disease had a between arm difference of 15 mm Hg or more).4 These limitations mean that the study cannot be considered to be definitive, and the prognostic value of blood pressure differences between arms remains to be replicated in future studies.

How should clinicians use measurements of bilateral arm blood pressure in their practice? As recommended by current hypertension guidelines, including the new update from the National Institute for Health and Clinical Excellence,11 bilateral blood pressure measurements should be done routinely to avoid delayed diagnosis or undertreatment of hypertension in those with a 10 mm Hg difference between arms (20% of primary and secondary care patients). What about measurement technique? On the basis of available evidence and practicality,1 12 a sequential measurement, followed by confirmation with at least two simultaneous measurements using two automatic devices, seems to be a reasonable approach. The optimal number of repeated measurements and monitoring intervals are unknown.

If the difference is 10 mm Hg or more on repeated simultaneous measurements, the positive predictive value for peripheral vascular disease of the upper extremities is high and further diagnostic evaluation is warranted, especially in people with risk factors for cardiovascular disease. Subsequent blood pressure monitoring should be performed in the arm with the higher readings. In the absence of obstructive peripheral vascular disease, there is insufficient evidence for evaluating prognosis or guiding treatment with antiplatelet drugs or statins on the basis of between arm differences in blood pressure. More work is needed to examine how well between arm difference in blood pressure compares with other markers of subclinical atherosclerosis, biomarkers, and clinical predictors as a predictor of mortality from cardiovascular disease.


Cite this as: BMJ 2012;344:e2033


  • Research, doi:10.1136/bmj.e1327
  • Competing interests: The author has completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declares: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Commissioned; not externally peer reviewed.