High Blood Pressure & Cardiovascular Prevention

, Volume 25, Issue 3, pp 303–307 | Cite as

Inter-arm Systolic Blood Pressure Difference in Physically Active, Adult Subjects

  • Daniel Neunhäuserer
  • Andrea Gasperetti
  • Sara Ortolan
  • Francesca Battista
  • Pietro Pettenella
  • Marco Zaccaria
  • Marco Bergamin
  • Andrea ErmolaoEmail author
Original Article
Part of the following topical collections:
  1. Epidemiology of Hypertension



Increased inter-arm systolic blood pressure difference (ΔPsys) has been associated with cardiovascular (CV) disease in elderly patients with CV risk factors. However, its significance in healthy subjects is unclear.


To determine the relationship between ΔPsys, the individual level of physical activity and the global CV risk in apparently healthy adults.


Systolic blood pressure was measured in both arms in 400 subjects aged 46.5 ± 12.2 years, using a simultaneous oscillometric device (WatchBP Office, Microlife, Widnau, Switzerland). In the subjects with ΔPsys ≥ 10 mmHg (Cases n = 20) and in a Control group (20 subjects without ΔPsys ≥ 10 mmHg), another simultaneous measurement was repeated during a second visit. The physical activity level was assessed via the International Physical Activity Questionnaire-Short Form (IPAQ-SF), the ankle brachial pressure index (ABPI) with a photoplethysmographic method (Angioflow-Microlab, Padova, Italy) and the CV risk via the Framingham Risk Score (FRS).


The prevalence of ΔPsys ≥ 10 mmHg in the whole population was 5% (95% CI 3.24–8.01%). Cases and Controls were comparable in gender, age, and BMI. ΔPsys ≥ 10 mmHg was only confirmed in 17.6% of the Cases. No statistically significant differences were found between groups for IPAQ-SF, ABPI, or FRS.


The prevalence of ΔPsys ≥ 10 mmHg in this population was only slightly lower than what observed in older, hypertensive or diabetic patients. Cases and Controls did not differ in physical activity level, ankle brachial pressure index and CV risk. However, low test–retest reliability might limit the use of ΔPsys as a reliable marker for CV screening in this population.


Cardiovascular risk Ankle brachial pressure index Framingham risk score IPAQ Inter-arm blood pressure difference 

1 Introduction

Prevalence of an increased inter-arm systolic blood pressure (BP) difference (ΔPsys ≥ 10 mmHg) has been estimated to be 7.5% in community settings [1], although new evidence suggests a prevalence of 11.2% in hypertensive patients, 7.4% in diabetic patients, and 3.6% in a population without hypertension or diabetes [2]. Moreover, elevated ΔPsys (≥ 15 mmHg) between the upper limbs might often be due to peripheral arterial disease or subclavian artery stenosis. However, recent studies on hypertensive patients have shown that also values of ΔPsys ≥ 10 mmHg might be related to peripheral arterial disease [3]. Indeed, data from the Framingham Heart Study cohort demonstrated that ΔPsys ≥ 10 mmHg was associated with a significant increased risk for future CV events, even when the difference is modest [4]. Therefore, it was suggested that an elevated ΔPsys could be a useful marker to unmask an increased risk of peripheral arterial disease and future CV events. This could support an expanded clinical use of this simple measurement [4].

Recent studies have investigated ΔPsys also in healthy subjects without CV disease [5, 6]. Results have shown that the prevalence of increased ΔPsys in healthy adults was lower, probably due to fewer CV risk factors.

In light of these evidences, the objective of this study was to estimate the prevalence of ΔPsys ≥ 10 mmHg between upper limbs in physically active adult subjects. In particular, the aim was to analyse how this parameter correlates with the individual level of physical activity, with other markers of peripheral arterial disease, and with the estimated risk of CV events. Therefore, the Ankle Brachial Pressure Index (ABPI) and the Framingham Risk Score (FRS) were determined respectively. The ABPI represents the ratio of systolic blood pressure in the ankle arteries to the systolic blood pressure in the brachial arteries, providing incremental prognostic utility over traditional risk factors. The ABPI is considered a reliable predictor of peripheral artery disease when it is < 0.9. Indeed, in population studies, a markedly increased CV disease risk has been found also in asymptomatic individuals with low ABPI [7]. Furthermore the FRS, a sex-specific multivariable risk factor algorithm, is commonly used to assess the risk of general CV diseases and individual CV events [8]. Both parameters are utilized in clinical routine and are helpful instruments to guide preventive care [7, 8].

Accordingly, the objective of this study was to evaluate the inter-arm systolic blood pressure difference, as easily determinable physical examination finding, within a general screening for the assessment of CV risk in adult, physically active subjects.

2 Methods

This study included all regularly physically active adult subjects attending our Sports and Exercise Medicine Division for a pre-participation screening for competitive and non-competitive physical activities. The level of physical activity was determined by a previously validated questionnaire (International Physical Activity Questionnaire-Short Form—IPAQ-SF) [9]. Patients with a history of major CV events, myocardial ischemia, congenital heart disease, or arrhythmias were excluded.

Anthropometric data and the brachial systolic BP were determined in all participants. Moreover, the systolic BP in upper limbs was obtained by using an automatic simultaneous oscillometric device (WatchBP Office, Microlife, Widnau, Switzerland), in a seated position, and after a 5-min rest period. This evaluation was repeated three times, with 1-min intervals between measurements. ΔPsys was calculated as the difference between the means of both arms.

Subjects found with a ΔPsys ≥ 10 mmHg (“Cases”) were invited to attend a second visit, to re-evaluate ΔPsys and to perform the following further investigations. The ABPI was measured by using an automatic photoplethysmographic instrument (Angioflow-Microlab, Padova, Italy), and selecting the arm with higher absolute systolic BP [10]. Finally, also the overall 10-year CV risk was estimated, according to the algorithm of the Framingham Risk Score for general CV diseases [8].

These evaluations were also performed in a second group of systematically allocated subjects, which did not show a significant difference in ΔPsys (“Controls”). Specifically, every fifth participant with normal ΔPsys was selected for the Controls to be compared to the Cases.

Confidence intervals were calculated with the Clopper–Pearson method. Groups were compared with the Student’s t test for normally distributed, and the Wilcoxon-Rank-Sum test for not-normally distributed variables. The Chi square test was used for qualitative variables. A p value < 0.05 was considered statistically significant. Data of normally distributed variables were expressed as mean ± SD, whereas data of not-normally distributed variables were also expressed as median (IQR).

3 Results

A total of 400 adult subjects (82.5% males) were analysed during this study, with a mean age of 46.5 ± 12.2 years (range 20–74 years), a BMI of 24.49 ± 2.86 kg·m−2, and a high prevalence of moderately or very active life style (25 and 49%, respectively). Mean systolic BP was 133 ± 14 mmHg, mean ΔPsys 4 ± 3 mmHg.

Twenty subjects were identified with a ΔPsys ≥ 10 mmHg, indicating a prevalence of 5% (95% CI 3.24–8.01%). Cases and Controls were comparable in gender, age, and BMI; there was no statistically significant difference in terms of physical activity levels (IPAQ-SF: p = 0.353) and for the prevalence of other CV risk factors (smoking p = 0.222, hypercholesterolemia p = 1.0, hypertension p = 0.677; see Table 1a). Furthermore, both groups showed comparable levels of absolute blood pressure. Seventeen of 20 Cases and 13 of 20 Controls attended the second visit. Further statistical analyses did not show any significant differences on the ABPI (p = 0.378) or CV risk score (FRS: p = 0.392) (see Table 1b). However, only 17.6% of the Cases confirmed a ΔPsys ≥ 10 mmHg at the second visit. All these subjects were younger than 45 years, moderately or very active, none of them showing pathological ABPI and all of them with low 10-year FRS (< 10%).
Table 1

Cases vs. controls


Cases (n = 20)

Controls (n = 20)

p value

(a) Baseline characteristics of Cases and Controls



M: 19, F: 1

M: 18, F: 2

0.584 (Chi-square)

 BMI (kg·m−2)

25.81 ± 3.09

24.80 ± 2.85

0.289 (Student’s t)

 Age (years)

45.95 ± 15.36

49.80 ± 13.77

0.409 (Student’s t)

 Physical activity

  1. Low active

  2. Moderately active

  3. Very active







0.353 (Wilcoxon-RS)

 Cardiovascular risk factors

  1. Smoke

  2.  Hypercholesterolemia

  3. Hypertension

 4. Diabetes

2 (10%)

7 (35%)

3 (15%)


2 (10%)

7 (35%)

4 (20%)


0.222 (Chi square)

1.0 (Student’s t)

0.677 (Chi square)

 Psys left (mmHg)

135 ± 13

136 ± 12

0.929 (Student’s t)

 Psys right (mmHg)

130 ± 14

133 ± 12

0.538 (Student’s t)

 Individual ΔPsys (mmHg)

12 ± 2

11 (10–12)*

4 ± 3

4 (2–5)*

< 0.001 (Wilcoxon-RS)


Cases (n = 17)

Controls (n = 13)

p value

(b) Cases and Controls at follow-up evaluation

 ABP Index left ankle

1.13 ± 0.11

1.11 ± 0.09

0.745 (Student’s t)

 ABP Index right ankle

1.14 ± 0.06

1.11 ± 0.11

0.378 (Student’s t)

 FRS (%)

8.8 ± 0.1

7.3 (1.8–11.0)*

13.1 ± 0.1

7.0 (5.8–16.4)*

0.392 (Wilcoxon-RS)

 Individual ΔPsys (mmHg)

4 ± 5

3 (1–7)*

3 ± 4

2 (1–6)*

0.460 (Wilcoxon-RS)

(a) Compares subjects with increased individual ΔPsys (= difference of systolic blood pressure between arms in each subject) ≥ 10 mmHg (Cases) with systematically allocated subjects not showing a significant difference in ΔPsys (Controls). Both groups were comparable in sex, BMI, age, physical activity levels and prevalence of other CV risk factors. (b) Compares Cases and Controls at the second examination during follow-up, where 10 patients dropped out. Data are expressed as mean ± SD for normally distributed variables; as mean ± SD and as median (IQR) for non-normally distributed variables

n number of subjects, Psys arm systolic blood pressure, ABP Index Ankle Brachial Pressure Index, FRS cardiovascular risk at 10 years calculated using the Framingham Risk Score for General Cardiovascular Disease, IQR Interquartile Range (first quartile–third quartile)

*Median (IQR)

4 Discussion

The main outcomes of our study are threefold: (1) An increased ΔPsys is not a rare finding in physically active, adult subjects. (2) A single recording of ΔPsys ≥ 10 mmHg does not seem to be a reliable measurement. In this sample, it was confirmed only in about one-fifth of the monitored Cases. (3) ΔPsys does not appear to be an accurate marker for cardiovascular screening in this population.

Compared to preceding clinical trials, our study was conducted in a younger population, with some participants having < 30 years of age, and all engaged in regular physical activity. Moreover, among study participants a significantly lower than estimated prevalence for smoking, hypertension, and diabetes was found, compared to the most recent Italian epidemiological data [11]. Even though current evidences support the use of global risk scores in individuals ≥ 40 years of age, the FRS was also recommended to assess the CV risk even in adults as young as age 20 [12, 13]. Although subjects of this study had a lower CV risk, the prevalence of ΔPsys ≥ 10 mmHg seemed only slightly lower than what reported for populations with increased CV risk, where its prevalence was estimated between 7.4 and 11.2% for diabetes and hypertension, respectively [1, 2]. However, it must be noted that previous studies reported heterogeneous prevalence rates, partly due to methodological issues [6, 14]. In particular, it was demonstrated that subsequent BP measurements overestimated ΔPsys compared to simultaneous measurements [4, 6, 14].

In our study, no significant difference was shown between Cases and Controls for physical activity levels, smoking, hypercholesterolemia, hypertension, and diabetes. Although inter-arm and leg BP differences were identified as strong predictors of peripheral arterial disease in an older population, our data showed comparable ABPI between both groups [4, 15]. These results do not contradict previous studies’ outcomes, because our sample was different and represents a low risk population. Indeed, it could be considered that Cases and Controls had a comparable cardiovascular risk (FRS median values: Cases 7.3%, Controls 7.0%). Moreover, in our study, not even the BMI, previously associated with increased ΔPsys, differed between Cases and Controls [1]. Thus, study results did not show any association between ΔPsys and CV risk. Previous studies also have failed to find any correlation between ΔPsys ≥ 10 mmHg and coronary artery disease, cerebrovascular disease and CV/all-cause mortality [15]. However, Clark et al. concluded that a difference of 15 mmHg or more could be a useful risk factor for vascular diseases and death [3].

In addition, the test–retest reliability of ΔPsys-measurements in our subjects’ follow-up evaluation was low, thus suggesting high intra-individual variability. Indeed, the presence of an increased ΔPsys was confirmed at the second visit only in 17.6% of the subjects (3 of 17 Cases), which raises some concern regarding its prognostic and clinical significance. It is possible that the younger age, higher/unequal muscle mass, as well as the low probability of obstructive arterial disease, could explain low test reproducibility. However, this inconsistency over time suggests the need for a second evaluation before drawing clinical conclusions. Moreover, it seems to be more accurate not only using an average of ≥ 2 readings obtained on ≥ 2 occasions to estimate the individuals’ level of blood pressure [16], but also obtaining them in both arms simultaneously to get a real ΔPsys and to better define patients at risk.

The present study was conducted in younger, physically active patients and may thus add a different perspective to the current understanding of ΔPsys. Although this simple measurement was proposed as method for risk prediction in cohorts of elderly subjects or patients with increased cardiovascular risk [2, 4, 15], our findings showed that an increased ΔPsys in adult, physically active subjects does not suffice to assume a higher risk of peripheral arterial disease or CV events. However, if a significant ΔPsys is confirmed by a second exam, it might be cautious to perform further analyses. In order to avoid delay in the diagnosis of a potential new hypertensive condition, it should be recommended to measure blood pressure in the arm with the higher recorded value.

5 Conclusions

Larger studies providing hard long-term outcomes are needed to investigate the clinical significance of increased ΔPsys in this population. Further studies might be useful to identify the correct diagnostic procedure to follow.



The study was not supported by any funding or grant.

Compliance with Ethical Standards

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The protocol has been approved by the Internal Review Board.

Informed consent

Written informed consent was obtained from each patient.


  1. 1.
    Singh S, Sethi A, Singh M, et al. Prevalence of simultaneously measured interarm systolic blood pressure difference and its clinical and demographic predictors: a systemic review and meta-analysis. Blood Press Monit. 2015;20:178–85.CrossRefPubMedGoogle Scholar
  2. 2.
    Clark CE, Taylor RS, Campbell JL, et al. Prevalence of systolic inter-arm differences in blood pressure for different primary care populations: systematic review and meta-analysis. Br J Gen Pract. 2016;66:e838–47.CrossRefPubMedPubMedCentralGoogle Scholar
  3. 3.
    Clark CE, Taylor RS, Shore AC, et al. Association of a difference in systolic blood pressure between arms with vascular disease and mortality: a systematic review and meta-analysis. Lancet (London, England). 2012;379:905–14.CrossRefGoogle Scholar
  4. 4.
    Weinberg I, Gona P, O’Donnell CJ, et al. The systolic blood pressure difference between arms and cardiovascular disease in the Framingham Heart Study. Am J Med. 2014;127:209–15.CrossRefPubMedGoogle Scholar
  5. 5.
    Kim KB, Oh MK, Kim HG, et al. Inter-arm differences in simultaneous blood pressure measurements in ambulatory patients without cardiovascular diseases. Korean J Fam Med. 2013;34:98–106.CrossRefPubMedPubMedCentralGoogle Scholar
  6. 6.
    Grossman A, Prokupetz A, Gordon B, et al. Inter-arm blood pressure differences in young, healthy patients. J Clin Hypertens. 2013;15:575–8.CrossRefGoogle Scholar
  7. 7.
    Greenland P, Smith SC, Grundy SM. Improving coronary heart disease risk assessment in asymptomatic people: role of traditional risk factors and noninvasive cardiovascular tests. Circulation. 2001;104:1863–7.CrossRefPubMedGoogle Scholar
  8. 8.
    D’Agostino RB, Vasan RS, Pencina MJ, et al. General cardiovascular risk profile for use in primary care: the Framingham heart study. Circulation. 2008;117:743–53.CrossRefPubMedGoogle Scholar
  9. 9.
    Craig CL, Marshall AL, Sjöström M, et al. International physical activity questionnaire: 12-Country reliability and validity. Med Sci Sports Exerc. 2003;35:1381–95.CrossRefGoogle Scholar
  10. 10.
    Jönsson B, Laurent C, Eneling M, et al. Automatic ankle pressure measurements using PPG in ankle-brachial pressure index determination. Eur J Vasc Endovasc Surg. 2005;30:395–401.CrossRefPubMedGoogle Scholar
  11. 11.
    Donfrancesco C, Palmieri L, Vannucchi S, et al. Salute cardiovascolare degli italiani: i dati preliminari dell’indagine Oec/Hes 2008–12. Accessed 06 July 2018.
  12. 12.
    Greenland P, Alpert JS, Beller GA, et al. 2010 ACCF/AHA guideline for assessment of cardiovascular risk in asymptomatic adults: a report of the american college of cardiology foundation/american heart association task force on practice guidelines. Circulation. 2010;122:584–637.CrossRefGoogle Scholar
  13. 13.
    Armstrong AC, Jacobs DR, Gidding SS, et al. Framingham score and LV mass predict events in young adults: CARDIA study. Int J Cardiol. 2014;172(2):350–5.CrossRefPubMedPubMedCentralGoogle Scholar
  14. 14.
    Verberk WJ, Kessels AGH, Thien T. Blood pressure measurement method and inter-arm differences: a meta-analysis. Am J Hypertens. 2011;24:1201–8.CrossRefPubMedGoogle Scholar
  15. 15.
    Singh S, Sethi A, Singh M, et al. Simultaneously measured inter-arm and inter-leg systolic blood pressure differences and cardiovascular risk stratification: a systemic review and meta-analysis. J Am Soc Hypertens. 2015;9:640–50.CrossRefPubMedGoogle Scholar
  16. 16.
    Pk W, Whelton PK, Carey RM, et al. Guideline for the prevention, detection, evaluation, and management of high blood pressure in adults a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2017;2017:283.Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Sport and Exercise Medicine Division, Department of MedicineUniversity of PadovaPaduaItaly

Personalised recommendations