Abstract
This is the first longitudinal study evaluating whether adiposity is associated with inter‐arm blood pressure difference. We evaluated 714 overweight/obese individuals aged 40‐65 years over a 3‐year follow‐up. Systolic and diastolic blood pressures were measured in both arms simultaneously using an automated machine. Linear regression assessed the associations of body mass index, fat %, waist, neck, thigh, and arm circumferences (cm), with absolute inter‐arm differences in systolic (IAS) and diastolic (IAD) blood pressure (mm Hg). Poisson regression was used for binary outcomes (IAS and IAD ≥ 10 mm Hg). All models were adjusted for age, gender, smoking, physical activity, and HOMA‐IR. Adiposity measures were associated with increased IAS and IAD (β range: 0.09‐0.20 and 0.09‐0.30). Neck circumference showed the strongest association with IAS (β = 0.20, 95% CI: 0.03, 0.37) and IAD (β = 0.30, 95% CI: 0.12, 0.47); arm circumference showed a similar association with IAS, but lower with IAD. Highest quartiles of BMI, thigh, and arm showed significant associations with IAS (IRR: 2.21, 2.46 and 2.70). Highest quartiles of BMI, waist, neck, and arm circumferences were significantly associated with IAD (IRR: 2.38, 2.68, 4.50 and 2.24). If the associations are corroborated in other populations, adiposity may be an important modifiable risk factor for inter‐arm blood pressure difference with a large potential public health impact.
Keywords: adiposity, cardiovascular morbidity and mortality, inter‐arm blood pressure difference
1. INTRODUCTION
The variation in blood pressure between arms, termed as inter‐arm blood pressure difference (IABPD), was described about a century ago.1 The left subclavian artery originates from the aorta, making an acute angle, in contrast to the right subclavian artery. Therefore, higher blood pressure (BP) is normally seen in the right arm compared with the left.2 An IABPD < 10 mm Hg has been established as normal.2 IABPD is higher among individuals with increased cardiovascular risk, such as people with hypertension, diabetes, chronic renal failure, or peripheral vascular disease (PVD). Prevalence of high inter‐arm systolic blood pressure difference (IAS) has been estimated to be 7.5% overall, 11.2% among hypertensive patients, 7.4% in diabetes patients, and 3.6% in people without hypertension or diabetes.3
High IABPD has been associated with subclavian stenosis, cardiovascular disease (CVD), chronic kidney disease, cardiovascular and overall mortality.4, 5, 6, 7, 8, 9 Each 5 mm Hg increase in IAS was related to a 12% higher risk of vascular events among participants without manifested vascular disease.10 IAD > 5 mm Hg has been associated with left ventricular mass index and with interventricular septal thickness and with posterior wall thickness at end‐diastole.11
Identifying modifiable risk factors for IABPD is important for reducing cardiovascular morbidity and mortality. Overall and central obesity have been associated with increased blood pressure.12, 13 However, only a few cross‐sectional studies have evaluated associations between adiposity and IABPD. In a representative sample of 484 Finnish adults, aged 25‐74 years, people with IAD > 5 (compared with ≤ 5 mm Hg) had higher body mass index (BMI) and arm circumference.11 People with IAS ≥ 10 (compared with < 10 mm Hg) showed higher BMI in the Framingham Heart Study,9 and among 806 participants aged 30‐64 years without history of major CVD.14 Obesity was also associated with IAS over 10 mm Hg among 1090 individuals with hypertension.15 In a sample of 335 women (238 HIV infected and 97 uninfected), participants with obesity had an increased risk of having IAS over 20 mm Hg.16 Significant associations between BMI and high IAS (multivariable odds ratios 1.11‐1.90) have been reported among Asians and Americans.14, 15, 17, 18
To the best of our knowledge, there are no longitudinal studies reporting on the association between adiposity and IABPD. This report evaluates the association between different adiposity measures and IABDP over a 3‐year follow‐up period.
2. METHODS
2.1. Study overview and sample
The analyses were conducted within a subsample of participants from the San Juan Overweight Adults Longitudinal Study (SOALS). The study was approved by the University of Puerto Rico Medical Sciences Campus Human Research Subjects Protection Office. All participants gave informed consent. SOALS data are available from the principal investigators upon reasonable request. SOALS is a 3‐year follow‐up longitudinal study conducted among civilian, non‐institutionalized adults recruited in Puerto Rico. The original purpose of the study was to assess the bidirectional relationship between periodontal disease and glucose abnormalities. Supplemental data were also collected in the study to evaluate additional hypotheses formulated before data collection. Recruitment efforts and retention have been previously described.19, 20 Eligibility criteria for this cohort study included the following: (a) age between 40 and 65 years, (b) overweight or obese (BMI ≥ 25.0 kg/m2), and (c) free of clinically diagnosed diabetes prior to the baseline examination. The baseline exclusion criteria were as follows: (a) physician‐diagnosed type 1 or type 2 diabetes or taking either insulin or oral anti‐hyperglycemic agents; (b) pregnancy; (c) reported physician‐diagnosed hypoglycemia, congenital heart murmurs, heart valve disease, congenital heart disease, endocarditis, rheumatic fever, and hemophilia or bleeding disorders; (d) active dialysis treatment; (e) having undergone procedures related to cardiovascular disease; (f) severe health conditions or psychological or physical disabilities that would interfere with participation in the study; (g) fewer than 4 teeth; wearing braces or orthodontic appliances which might affect periodontal assessment; and (h) plans on moving away in the next 3‐year period.
The study consisted of multiracial individuals of Hispanic ethnicity, who reported their race as White (25%), Black (14%), and Mixed (61%). From the 1351 baseline eligible participants, 1028 (76%) came to the 3‐year follow‐up visit. We have shown in an earlier publication that baseline characteristics of participants retained and included in analyses are similar to the overall study sample at baseline.21 The simultaneous inter‐arm blood pressure could not be measured in all participants due to logistical reasons, therefore, 312 missing IABPD data were excluded. One individual with missing smoking information and one with missing physical activity information were additionally excluded. Thus, the analyses included 714 eligible individuals.
2.2. Exposures
Baseline adiposity measures include BMI, fat %, waist (WC), neck (NC), thigh (TC), and arm circumferences (AC). Anthropometric measurements were taken twice according to the NHANES III Anthropometry Procedures Manual. If the first two measurements differed by 0.5 cm, a third measurement was recorded. The average of all measures taken was computed. Height was measured in meters using a portable stadiometer (Seca Corporation). Body weight (in 0.2 kg weight graduation) and fat % were measured using a Tanita scale (Tanita Body Composition Analyzer‐TBF‐310A). BMI (kg/m2) was used to classify participants as overweight (25 ≤ BMI < 30), class I obesity (30 ≤ BMI < 35), class II obesity (35 ≤ BMI < 40), and class III obesity (BMI ≥ 40).22 Body circumferences were measured with a Gulick tape and recorded to the nearest 0.1 cm. WC was measured at the umbilicus. NC was measured below the laryngeal prominence and perpendicular to the long axis of the neck, and the minimal circumference was recorded. TC was measured around the mid‐thigh, and AC around the midpoint of the upper arm. We computed quartiles of fat % and waist, neck, thigh, and arm circumferences.
2.3. Outcomes
The outcomes were assessed at the 3‐year follow‐up examination. Participants were asked to sit down with both feet flat on the floor, legs uncrossed with right arm resting on a tabletop at the level of the heart. We measured the participant's arm length and circumference, and an appropriate‐sized cuff was selected for each participant for an accurate reading. After participants sat quietly for 5 minutes, resting systolic and diastolic blood pressure was measured on the arm 3 times with an interval of 1‐2 minutes of rest using the Microlife WatchBP Office ABI machine. Using the same machine and cuff size, measurements of BP on both arms were then simultaneously obtained. Absolute IAS and IAD were calculated and evaluated as separate continuous outcomes. We also computed binary IAS and IAD using a cutoff of ≥ 10 mm Hg.2, 6, 7, 18
2.4. Covariates
Trained interviewers administered questionnaires to gather information about age, gender, smoking (never, former, current),23 physical activity (based on WHO recommendations), 24 and health conditions, including sleep breathing disorders, hypertension status, and medication use. Fasting serum levels of glucose, glycated hemoglobin (HbA1C), total cholesterol, low‐density lipoprotein cholesterol (LDL‐C), and high‐sensitivity C‐reactive protein (hs‐CRP) were measured using standard procedures.19, 25 Diabetes status was defined based on the ADA cutoffs for fasting and 2‐hour post‐load glucose tolerance test and HbA1c.26 The homeostatic model assessment of insulin resistance (HOMA‐IR) was calculated as fasting insulin (mU/L) multiplied by fasting glucose (mmol/L) and divided by 22.5. We used standard cut‐offs from the literature as described below for classifying metabolic syndrome and its components. Participants were classified with elevated waist circumference (≥ 102 cm for men or ≥ 88 cm for women), triglycerides as levels (≥ 150 mg/dL or a history of drug treatment for hypertriglyceridemia), low HDL‐C (< 40 mg/dL in men and levels < 50 mg/dL in women or history of drug treatment for low HDL‐C), elevated blood pressure (systolic blood pressure ≥ 130 mm Hg or diastolic blood pressure ≥ 85 mm Hg or reported antihypertensive drug treatment), and elevated fasting blood glucose (≥ 100 mg/dL or a history of drug treatment for hyperglycemia).27 The presence of three of these five criteria suggests metabolic syndrome.
2.5. Statistics
Continuous variables are presented as mean ± standard deviation and categorical variables as percent. We assessed the associations between adiposity measures at baseline and IABPD at follow‐up using linear regression. We used Poisson regression models for the binary outcomes of high IABPD and obtained incidence rate ratios (IRR) with 95 percent confidence intervals (95% CI). Time between the baseline and follow‐up visits were included in the models as an offset, and we used robust standard errors for the parameter estimates.28 We controlled for baseline confounders. BMI overweight category was the references for classes I, II, and III obesity; and the first quartiles of WC, NC, TC, AC, and fat % were references for higher quartiles. All multivariate models were adjusted for age, sex, smoking, physical activity, and HOMA‐IR. We also evaluated additional potential confounders (described in results) by adding each covariate to the model, and keeping them in the model if they changed the estimate by 10% or more. Additionally, we evaluated the association between NC and IABPD among subgroups of sex, age, smoking, physical activity, HOMA‐IR, diabetes, metabolic syndrome, and sleep apnea.
3. RESULTS
Tables 1A and B describe baseline adiposity measures and characteristics by high and low IAS and IAD (≥ 10 vs < 10 mm Hg). The group with high IAS (compared to low IAS) had more adiposity, more current smokers, and higher hs‐CRP. The group with high IAD had more adiposity, more women, more past smokers, higher hs‐CRP levels, and less physical activity compared with the group with low IAD.
Table 1.
A, Baseline adiposity measures by systolic and diastolic inter‐arm blood pressure difference. B, Baseline characteristics by systolic and diastolic inter‐arm blood pressure difference
| Table 1A | IASa < 10 mm Hg (n = 574) | IAS ≥ 10 (n = 140) | IADb < 10 (n = 646) | IAD ≥ 10 (n = 68) |
|---|---|---|---|---|
| BMI (kg/m2) | 32.8 ± 6.0 | 34.8 ± 7.1 | 32.7 ± 5.8 | 37.5 ± 8.4 |
| Overweight | 39.7 | 30.0 | 39.2 | 25.0 |
| Class I obesity: 30 ≤ BMI < 35 | 31.9 | 31.4 | 33.6 | 14.7 |
| Class II Obesity: 35 ≤ BMI < 40 | 16.6 | 17.1 | 15.0 | 32.4 |
| Class III obesity: BMI ≥ 40 | 11.9 | 21.4 | 12.2 | 27.9 |
| Waist circumference (cm) | 105.0 ± 12.8 | 109.4 ± 18.1 | 105.1 ± 13.58 | 113.0 ± 17.4 |
| 1st quartile | 25.6 | 22.9 | 26.5 | 11.8 |
| 2nd quartile | 26.3 | 20.0 | 25.5 | 20.6 |
| 3rd quartile | 25.1 | 24.3 | 24.5 | 29.4 |
| 4th quartile | 23.0 | 32.9 | 23.5 | 38.2 |
| Neck circumference (cm) | 37.4 ± 3.8 | 38.1 ± 4.0 | 37.4 ± 3.8 | 38.5 ± 4.2 |
| 1st quartile | 26.5 | 21.4 | 27.1 | 10.3 |
| 2nd quartile | 24.7 | 23.6 | 23.8 | 30.9 |
| 3rd quartile | 24.9 | 27.1 | 25.2 | 26.5 |
| 4th quartile | 23.9 | 27.9 | 23.8 | 32.4 |
| Thigh circumference (cm) | 55.3 ± 7.0 | 59.5 ± 7.5 | 55.4 ± 6.7 | 63.7 ± 11.0 |
| 1st quartile | 26.7 | 18.4 | 26.4 | 12.5 |
| 2nd quartile | 27.1 | 7.9 | 25.3 | 12.5 |
| 3rd quartile | 23.9 | 31.6 | 25.6 | 12.5 |
| 4th quartile | 22.3 | 42.1 | 22.7 | 62.5 |
| Arm circumference (cm) | 34.6 ± 4.1 | 36.2 ± 4.4 | 34.6 ± 4.0 | 37.5 ± 5.4 |
| 1st quartile | 32.2 | 19.6 | 31.0 | 20.4 |
| 2nd quartile | 23.2 | 23.4 | 24.2 | 13.0 |
| 3rd quartile | 25.6 | 23.4 | 25.1 | 25.9 |
| 4th quartile | 18.9 | 33.6 | 19.6 | 40.7 |
| Fat % | 38.9 ± 8.0 | 39.9 ± 9.1 | 38.7 ± 8.2 | 42.3 ± 8.7 |
| 1st quartile | 25.2 | 26.8 | 26.3 | 17.9 |
| 2nd quartile | 27.2 | 16.7 | 26.2 | 14.9 |
| 3rd quartile | 24.7 | 29.0 | 24.8 | 32.8 |
| 4th quartile | 22.9 | 27.5 | 22.7 | 34.3 |
| Table 1B | IASa < 10 mm Hg (n = 574) | IAS ≥ 10 (n = 140) | IADb < 10 (n = 646) | IAD ≥ 10 (n = 68) |
|---|---|---|---|---|
| Age (y) | 50.7 ± 6.6 | 50.2 ± 6.9 | 50.7 ± 6.7 | 49.9 ± 6.7 |
| Sex (% women) | 73.0 | 71.4 | 71.8 | 80.9 |
| Smoking % | ||||
| Past | 14.1 | 13.6 | 13.3 | 20.6 |
| Current | 16.6 | 24.3 | 18.1 | 17.7 |
| Physical activityc | 47.6 | 50.7 | 49.1 | 39.7 |
| hs‐CRP (nmol/L) | 54.0 ± 57.6 | 69.8 ± 73.2 | 54.8 ± 59.8 | 79.5 ± 69.9 |
| HbA1c% | 5.8 ± 0.6 | 5.8 ± 0.7 | 5.8 ± 0.6 | 5.8 ± 0.6 |
| LDL (mmol/L) | 3.2 ± 0.8 | 3.1 ± 0.9 | 3.2 ± 0.9 | 3.2 ± 0.7 |
| Pre‐hypertension% | 30.1 | 35.0 | 31.6 | 26.5 |
| Hypertension% | 47.9 | 47.1 | 47.7 | 48.5 |
| Pre‐diabetes | 53.1 | 45.0 | 52.2 | 45.6 |
| Diabetes | 9.6 | 8.6 | 9.3 | 10.3 |
Inter‐arm systolic blood pressure difference.
Inter‐arm diastolic blood pressure difference.
Complying with WHO Physical activity recommendations.
The results of linear regression are shown in Table 2. The assumption of normality of residuals between adiposity measures and IABPD measures was slightly violated. However, when the sample size is sufficiently large (>200), the central limit theorem ensures that the distribution of disturbance term will approximate normality. The continuous associations were significant for all measures except for the association between fat % and IAD (P = .09); 8 out of the 12 P‐values for the associations were < .01, suggesting strong highly significant results. The estimates (β) ranged from 0.09 to 0.20 for IAS, and from 0.09 to 0.30 for IAD, with NC showing the highest association with both IAS (β = 0.20, 95% CI: 0.03, 0.37) and IAD (β = 0.30, 95% CI: 0.12, 0.47).
Table 2.
Adiposity measures and inter‐arm blood pressure difference multivariate linear regressionsa
| IASb (mm Hg) β (95%CI) | P‐value | IADc (mm Hg) β (95%CI) | P‐value | |
|---|---|---|---|---|
| BMI (kg/m2) n = 714 | 0.18 (0.07, 0.30) | < .01 | 0.28 (0.11, 0.44) | <.01 |
| Waist (cm)b n = 714 | 0.09 (0.04, 0.14) | <.01 | 0.10 (0.03, 0.17) | <.01 |
| Neck (cm) n = 714 | 0.20 (0.03, 0.37) | .02 | 0.30 (0.12, 0.47) | <.01 |
| Thigh (cm) n = 289 | 0.11 (0.02, 0.20) | .01 | 0.14 (0.05, 0.22) | <.01 |
| Arm (cm) n = 599 | 0.20 (0.08, 0.32) | <.01 | 0.23 (0.10, 0.37) | <.01 |
| Fat (%) n = 709 | 0.11 (0.01, 0.20) | .03 | 0.09 (−0.01, 0.20) | .09 |
Adjusted for age, sex, smoking and physical activity, and HOMA‐IR. Time between visits included as an offset.
Inter‐arm systolic blood pressure difference.
Inter‐arm diastolic blood pressure difference.
Table 3 presents results for categories of the anthropometric measures, adjusting for age, sex, smoking, physical activity, and HOMA‐IR. BMI (class III obesity compared with overweight IRR = 2.21), and TC (highest vs lowest quartiles of IRR = 2.46), and AC (IRR = 2.70) were significantly associated with high IAS. BMI (IRR = 2.38), WC (IRR = 2.68), NC (IRR = 4.50), and AC (IRR = 2.24) were significantly associated with high IAD, comparing extreme categories. BMI class II obesity (IRR = 2.34) and the 3rd quartile of NC (IRR = 2.70) had significantly higher IAD compared with the lowest quartile. Fat % in quartiles was not significantly associated with IAS or IAD. For IAD, tests for trend were significant for all the adiposity measurements. For IAS, tests for trend were significant for BMI, WC, TC, and AC.
Table 3.
Incidence rate ratios and trend associating adiposity and inter‐arm blood pressure Differencea
| IASb ≥ 10 mm Hg IRR (95%CI) | IADc ≥ 10 mm Hg IRR (95%CI) | |
|---|---|---|
| BMI (kg/m2), n = 714: reference overweight 25 ≤ BMI < 30 | ||
| Class I obesity: 30 ≤ BMI < 35 | 1.34 (0.91‐1.98) | 0.63 (0.29‐1.37) |
| Class II obesity: 35 ≤ BMI < 40 | 1.44 (0.90‐2.29) | 2.34 (1.24‐4.45)* |
| Class III obesity: BMI ≥ 40 | 2.21 (1.40‐3.48)* | 2.38 (1.24‐4.57)* |
| P‐value for trend | <.01 | <.01 |
| Waist circumference, n = 714: reference 1st quartile | ||
| 2nd quartile | 0.88 (0.54‐1.43) | 1.85 (0.80‐4.30) |
| 3rd quartile | 1.13 (0.71‐1.80) | 2.26 (1.00‐5.10) |
| 4th quartile | 1.54 (0.97‐2.43) | 2.68 (1.20‐6.00)* |
| P‐value for trend | .04 | <.01 |
| Neck circumference, n = 714: reference 1st quartile | ||
| 2nd quartile | 1.16 (0.74‐1.82) | 2.96 (1.30‐6.78)* |
| 3rd quartile | 1.35 (0.85‐2.15) | 2.70 (1.13‐6.48)* |
| 4th quartile | 1.47 (0.85‐2.55) | 4.50 (1.85‐11.00)* |
| P‐value for trend | .15 | .01 |
| Thigh circumference, n = 289: reference 1st quartile | ||
| 2nd quartile | 0.46 (0.12‐1.76) | 0.79 (0.09‐7.12) |
| 3rd quartile | 1.74 (0.72‐4.20) | 0.96 (0.13‐7.36) |
| 4th quartile | 2.46 (1.04‐5.82)* | 3.80 (0.69‐20.84) |
| P‐value for trend | <.01 | <.01 |
| Arm circumference, n = 599: reference 1st quartile | ||
| 2nd quartile | 1.63 (0.94‐2.82) | 0.79 (0.31‐2.18) |
| 3rd quartile | 1.54 (0.89‐2.66) | 1.50 (0.69‐3.27) |
| 4th quartile | 2.70 (1.59‐4.59)* | 2.24 (1.05‐4.76)* |
| P‐value for trend of odds | <.01 | <.01 |
| Fat %, n = 709: reference 1st quartile | ||
| 2nd quartile | 0.69 (0.39‐1.22) | 0.67 (0.26‐1.73) |
| 3rd quartile | 1.29 (0.76‐2.18) | 1.37 (0.54‐3.50) |
| 4th quartile | 1.31 (0.75‐2.28) | 1.28 (0.47‐3.47) |
| P‐value for trend | .26 | <.01 |
Poisson regressions adjusted for age, sex, smoking, physical activity and HOMA‐IR. Time between visits included as an offset.
Inter‐arm systolic blood pressure difference.
Inter‐arm diastolic blood pressure difference.
P < .05.
Additionally controlling for baseline systolic and diastolic blood pressures, hs‐CRP, HbA1c, fasting blood glucose, diabetes status, total cholesterol, LDL cholesterol, HOMA‐IR, hypertension status, medications (lipid lowering, hypertension and cardiovascular), coronary heart disease, angina, enlarged heart, and lower extremity arterial diseases did not change the main effect estimates, therefore, we did not include these covariates in the models.
Table 4 shows the results of linear regression between NC and IABPD within subgroups based on potential effect modifiers. All the coefficients were positive, suggesting an increase in IABPD with higher NC consistently across all subgroups. Subgroups with age ≥ 55 years, ever smokers, people meeting recommended levels of physical activity, HOMA‐IR ≥ 2.5, having diabetes (detected from baseline examination), having metabolic syndrome, and sleep apnea, showed higher estimates (β) for the associations between NC and IAS compared with their counterpart subgroups. For IAD, subgroups with females, never smokers, not meeting recommended levels of physical activity, HOMA‐IR < 2.5, diabetes and sleep apnea showed higher associations with NC.
Table 4.
Neck circumference and inter‐arm blood pressure difference linear regressions by subgroups of baseline characteristicsa
| Subgroups | n | IASb | IADc | ||
|---|---|---|---|---|---|
| β | 95% CI | β | 95% CI | ||
| Sex | |||||
| Male | 195 | 0.18 | −0.11, 0.46 | 0.13 | −0.00, 0.26 |
| Female | 519 | 0.21 | 0.00, 0.43 | 0.38* | 0.13, 0.64 |
| Age | |||||
| <55 | 510 | 0.18 | −0.02, 0.37 | 0.29* | 0.11, 0.48 |
| ≥55 | 204 | 0.26 | −0.11, 0.62 | 0.30 | −0.12, 0.71 |
| Smoking | |||||
| Never | 485 | 0.16 | −0.03, 0.34 | 0.33* | 0.11, 0.55 |
| Ever | 229 | 0.31 | −0.07, 0.69 | 0.23 | −0.04, 0.49 |
| Physical activityd | |||||
| Yes | 344 | 0.27* | 0.04, 0.51 | 0.19* | 0.03, 0.36 |
| No | 370 | 0.15 | −0.10, 0.40 | 0.43* | 0.12, 0.74 |
| HOMA‐IR | |||||
| <2.5 | 434 | 0.11 | −0.16, 0.38 | 0.40* | 0.08, 0.72 |
| ≥2.5 | 280 | *0.26 | 0.06, 0.47 | 0.30* | 0.12, 0.49 |
| Diabetes | |||||
| Normal | 279 | 0.05 | −0.21, 0.32 | 0.19* | 0.03, 0.35 |
| Pre‐diabetes | 368 | 0.26c | 0.00, 0.52 | 0.40* | 0.08, 0.72 |
| Diabetes | 67 | 0.49 | −0.04, 1.02 | 0.46 | −0.03, 0.96 |
| Metabolic syndrome | |||||
| No | 347 | 0.07 | −0.21, 0.34 | 0.35 | −0.01, 0.71 |
| Yes | 367 | 0.30* | 0.09, 0.51 | 0.30* | 0.15, 0.44 |
| Sleep apnea | |||||
| No | 600 | 0.16 | −0.02, 0.33 | 0.20* | 0.09, 0.32 |
| Yes | 114 | 0.39 | −0.10, 0.88 | 0.43 | −0.16, 1.01 |
Adjusting for age, gender, smoking status, physical activity, and HOMA‐IR; except for the stratifying variable for the smoking, physical activity and HOMA‐IR subgroups. Time between visits was included in the models as an offset.
Inter‐arm systolic blood pressure difference.
Inter‐arm diastolic blood pressure difference.
Complying with WHO Physical activity recommendations.
P < .05.
4. DISCUSSION
4.1. Main findings
Our results show that baseline adiposity measures are associated with IABPD at follow‐up after controlling for several covariates. This is the first longitudinal study to report that adiposity is associated with higher inter‐arm blood pressure difference. Significant results for both continuous and categorical measurements of adiposity and IAS and IAD suggest strong robust associations. For significant associations, depending on the adiposity measure, being in the highest category doubled, tripled, or quadrupled the risk of high IABPD compared to the lowest category. The direction of the associations was consistent across all subgroups for both continuous and categorical measures with some variation in the magnitude of the effect estimates, indicating robust findings.
4.2. Explanation and comparison with other studies
Our results from this longitudinal study corroborate earlier findings from cross‐sectional reports, but our findings (IRR range: 1.31‐4.50) are much stronger than previous studies (OR ranging from 1.11 to 1.90).14, 15, 17, 18 The association between thigh circumference and IABPD in our study, along with the fact that IABPD is associated with atherosclerosis,2 is in concordance with previous studies reporting that TC is associated with carotid intima‐media thickness, an established marker for subclinical atherosclerosis.29 Prior studies on IABPD have not evaluated thigh circumference. Studies have reported a positive association between NC and several cardiovascular risk factors, including obesity and metabolic syndrome in adults,30, 31 change in blood pressure,32 BMI, body weight, waist and hip circumference,33, 34 total cholesterol, and LDL cholesterol in both women and men, and triglycerides only in men.35 NC reflects upper‐body fat in humans, which makes it a valuable predictor of metabolic syndrome. However, the relationship between NC and hypertension is not fully understood; some studies suggest that upper‐body subcutaneous fat might increase fatty acid release, induce insulin resistance and cause endothelial damage. On the contrary, some studies suggest that overall adiposity may be more important than central adiposity for blood pressure36; suggesting that the predictors for blood pressure may be somewhat different from those for IABPD. Also, NC is highly correlated with obstructive sleep apnea syndrome, a well‐known cardiovascular risk factor.37 NC showed the strongest association with IAS and IAD. This is in concordance with our previous study reporting that NC might be a better predictor of cardio‐metabolic risk.38 Fat % has not been evaluated in previous studies and we found a significant association with continuous IAS, a significant trend for IAD, but no significant associations were seen between extreme quartiles of fat % and high IABPD. This might be related to our use of bioelectric impedance analysis, which is a simple and quick surrogate measure compared with densitometry or imaging techniques with better accuracy, which may be considered for future studies.
4.3. Strengths and limitations
Our study has several strengths. It is the first longitudinal study evaluating these relationships. Also, we used standardized procedures to assess outcomes and covariates. Moreover, we were able to evaluate the associations independent of major potential confounders. Our study also has some limitations. The sample consisted of overweight/obese Hispanic participants, 40‐65 years of age; hence, results cannot be directly generalized to other age groups, non‐Hispanics, and normal‐weight individuals.
4.4. Conclusions and implications for practice and future research
Our study suggests that adiposity is strongly associated with inter‐arm blood pressure difference. IABPD can be easily evaluated clinically, and routine evaluation of IABPD may be potentially helpful in early detection and prevention of cardiovascular disease, especially among people with high adiposity. The European Society of Hypertension has included IABPD assessment as part of the recommended physical examination of patients with arterial hypertension, indicating that high IABPD findings should trigger further investigation of vascular abnormalities.39 However, the more recent “Guidelines for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults”, from the American College of Cardiology and the American Heart Association, do not mention inter‐arm blood pressure differences.40
Further studies replicating our findings in other populations, as well as relating obesity and clinical measurements of hypertension‐related organ damage, would be important to corroborate our findings, expand the current knowledge, and inform clinical practice. Adiposity measurements may be particularly relevant to identify high‐risk overweight patients, who could benefit from screening for inter‐arm blood pressure differences in the primary care setting to prevent diseases. For example, IABPD may be associated with PVD in the legs: hence, IABPD may be an indicator to predict PVD which may be otherwise occult.7, 41, 42
Adiposity may be an important predictor and modifiable risk factor for inter‐arm blood pressure differences, with a large potential impact in public health. It may be worth assessing IABPD in healthy individuals with adiposity, since the prevalence of IASBPD ≥ 10 mm Hg in a young, healthy, and physically active population of adults was only slightly lower than that observed in older, hypertensive, or diabetic patients.3 Also, results from another study suggested that the clinical impact of IASBPD was more prominent in patients without underlying diseases like hypertension, diabetes, or obesity, than in patients with these diseases.8 If the results of our study are corroborated in additional populations, simultaneous blood pressure measurement in both arms could be recommended routinely in clinical practice to help predict the development of future cardiovascular complications for high‐risk individuals.
CONFLICT OF INTEREST
We have no conflict of interest to declare.
AUTHOR CONTRIBUTIONS
FM contributed to conceptualization of the manuscript and data collection, conducted the data management and analyses, wrote the first draft of the manuscript, and contributed to the interpretation. OMA and JIR contributed to analyses and interpretation and drafting of the manuscript. KJ is the PI of the parent study, conceptualized the hypothesis and contributed to data collection, analysis, interpretation, and writing. All authors reviewed and approved the manuscript.
ACKNOWLEDGMENTS
We would like to acknowledge the SOALS team (Mr José L. Vergara‐Arroyo, Ms Yadiris Santaella, Ms Tania Ginebra, Mr Jeanpaul Fernández, Ms Elaine Rodríguez, and Ms Rosalyn Roman) for their help with the study. We would also like to acknowledge the PRCTRC nurses and laboratory personnel (Nilda Gonzalez and Aracelis Arroyo) for their help during the blood processing and analysis. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Muñoz‐Torres FJ, Andriankaja OM, Ruiz JI, Joshipura KJ. Longitudinal association between adiposity and inter‐arm blood pressure difference. J Clin Hypertens. 2019;21:1519–1526. 10.1111/jch.13678
Funding information
Research reported in this publication was supported by the National Institute of Dental and Craniofacial Research Grants R01DE020111 and K23DE025313‐03 and the National Institute on Minority Health and Health Disparities Grants U54MD007600, 2U54MD007587 and S21MD001830 of the National Institutes of Health.
REFERENCES
- 1. Cyriax EF. Unilateral alterations in blood‐pressure caused by unilateral pathological conditions: the differential blood pressure sign. QJM. 1920;os‐13:148‐164. [Google Scholar]
- 2. Park S‐J, Son J‐W, Park S‐M, Choi H‐H, Hong K‐S. Relationship between inter‐arm blood pressure difference and severity of coronary atherosclerosis. Atherosclerosis. 2017;171‐176. [DOI] [PubMed] [Google Scholar]
- 3. Neunhäuserer D, Gasperetti A, Ortolan S, et al. Inter‐arm systolic blood pressure difference in physically active, adult subjects. High Blood Press Cardiovasc Prev. 2018;25:303‐307. [DOI] [PubMed] [Google Scholar]
- 4. Clark CE, Taylor RS, Butcher I, et al. Inter‐arm blood pressure difference and mortality: a cohort study in an asymptomatic primary care population at elevated cardiovascular risk. Br J Gen Pract. 2016;66:e297‐308. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Agarwal R, Bunaye Z, Bekele DM. Prognostic significance of between‐arm blood pressure differences. Hypertension. 2008;51:657‐662. [DOI] [PubMed] [Google Scholar]
- 6. Clark CE, Taylor RS, Shore AC, Campbell JL. The difference in blood pressure readings between arms and survival: primary care cohort study. BMJ. 2012;e1327‐e1340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Clark CE, Taylor RS, Shore AC, Ukoumunne OC, Campbell JL. Association of a difference in systolic blood pressure between arms with vascular disease and mortality: a systematic review and meta‐analysis. Lancet. 2012;379:905‐914. [DOI] [PubMed] [Google Scholar]
- 8. Kim GS, Byun YS, Lee HY, Jung IH, Rhee KJ, Kim BO. Difference in SBP between arms is a predictor of chronic kidney disease development in the general Korean population. J Hypertens. 2019;37:790‐794. [DOI] [PubMed] [Google Scholar]
- 9. Weinberg I, Gona P, O'Donnell CJ, Jaff MR, Murabito JM. The systolic blood pressure difference between arms and cardiovascular disease in the Framingham heart study. Am J Med. 2014;127:209‐215. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kranenburg G, Spiering W, de Jong PA, et al. Inter‐arm systolic blood pressure differences, relations with future vascular events and mortality in patients with and without manifest vascular disease. Int J Cardiol. 2017;244:271‐276. [DOI] [PubMed] [Google Scholar]
- 11. Johansson JK, Puukka PJ, Jula AM. Interarm blood pressure difference and target organ damage in the general population. J Hypertens. 2014;32:260‐266. [DOI] [PubMed] [Google Scholar]
- 12. Zhao Y, Zhang M, Luo X, et al. Association of 6‐year waist circumference gain and incident hypertension. Heart. 2017;103:1347‐1352. [DOI] [PubMed] [Google Scholar]
- 13. Zhao Y, Zhang M, Luo X, et al. Association of obesity categories and high blood pressure in a rural adult Chinese population. J Hum Hypertens. 2016;30:613‐618. [DOI] [PubMed] [Google Scholar]
- 14. Song BM, Kim HC, Shim JS, Lee MH, Choi DP. Inter‐arm difference in brachial blood pressure in the general population of Koreans. Korean Circ J. 2016;46:374‐383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Kimura A, Hashimoto J, Watabe D, et al. Patient characteristics and factors associated with inter‐arm difference of blood pressure measurements in a general population in Ohasama. Japan. J Hypertens. 2004;22:2277‐2283. [DOI] [PubMed] [Google Scholar]
- 16. Lazar J, Holman S, Minkoff HL, Dehovitz JA, Sharma A. Interarm blood pressure differences in the women's interagency HIV study. AIDS Res Hum Retroviruses. 2008;24:695‐700. [DOI] [PubMed] [Google Scholar]
- 17. Su H‐M, Lin T‐H, Hsu P‐C, et al. Association of interarm systolic blood pressure difference with atherosclerosis and left ventricular hypertrophy. PLoS ONE. 2012;7:e41173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Canepa M, Milaneschi Y, Ameri P, et al. Relationship between inter‐arm difference in systolic blood pressure and arterial stiffness in community‐dwelling older adults. J Clin Hypertens. 2013;15:880‐887. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Andriankaja OM, Jimenez JJ, Munoz‐Torres FJ, Perez CM, Vergara JL, Joshipura KJ. Lipid‐lowering agents use and systemic and oral inflammation in overweight or obese adult Puerto Ricans: the San Juan overweight adults longitudinal study (SOALS). J Clin Periodontol. 2015;42:1090‐1096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Pérez CM, Muñoz F, Andriankaja OM, et al. Cross‐sectional associations of impaired glucose metabolism measures with bleeding on probing and periodontitis. J Clin Periodontol. 2017;44:142‐149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Joshipura KJ, Munoz‐Torres FJ, Morou‐Bermudez E, Patel RP. Over‐the‐counter mouthwash use and risk of pre‐diabetes/diabetes. Nitric Oxide. 2017;71:14‐20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser. 2000;894:i‐xii, 1‐253. [PubMed] [Google Scholar]
- 23. Centers for Disease Control and Prevention (CDC) . Cigarette smoking among adults–United States, 2002. MMWR Morb Mortal Wkly Rep. 2004;53:427‐431. [PubMed] [Google Scholar]
- 24. World Health Organization . Global Recommendations on Physical Activity for Health. Geneva, Switzerland: WHO;2010:24‐27. [PubMed] [Google Scholar]
- 25. Joshipura KJ, Andriankaja MO, Hu FB, Ritchie CS. Relative utility of 1‐h oral glucose tolerance test as a measure of abnormal glucose homeostasis. Diabetes Res Clin Pract. 2011;93:268‐275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. American Diabetes Association . Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33(Suppl 1):S62‐S69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis and management of the metabolic syndrome: an American heart association/national heart, lung, and blood institute scientific statement. Curr Opin Cardiol. 2006;21:1‐6. [DOI] [PubMed] [Google Scholar]
- 28. Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159:702‐706. [DOI] [PubMed] [Google Scholar]
- 29. Ge W, Wu F, Islam T, et al. Association between anthropometric measures of obesity and subclinical atherosclerosis in Bangladesh. Atherosclerosis. 2013;232:234‐241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Ben‐Noun L, Sohar E, Laor A. Neck circumference as a simple screening measure for identifying overweight and obese patients. Obes Res. 2001;9:470‐477. [DOI] [PubMed] [Google Scholar]
- 31. Cui T, Yan B‐H, Liu Z, Yang H, Gyan M, Ma Y‐X. Neck circumference: a valuable anthropometric measurement to detect metabolic syndrome among different age groups in China. Diabetes Metab Res Rev. 2018;34(3):e2966. [DOI] [PubMed] [Google Scholar]
- 32. Ben‐Noun LL, Laor A. Relationship between changes in neck circumference and changes in blood pressure. Am J Hypertens. 2004;17:409‐414. [DOI] [PubMed] [Google Scholar]
- 33. Yan Q, Sun D, Li XU, et al. Neck circumference is a valuable tool for identifying metabolic syndrome and obesity in Chinese elder subjects: a community‐based study. Diabetes Metab Res Rev. 2014;30:69‐76. [DOI] [PubMed] [Google Scholar]
- 34. Saka M, Türker P, Ercan A, Kızıltan G, Baş M. Is neck circumference measurement an indicator for abdominal obesity? A pilot study on Turkish Adults. Afr Health Sci. 2014;14:570‐575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Ben‐Noun LL, Laor A. Relationship between changes in neck circumference and cardiovascular risk factors. Exp Clin Cardiol. 2006;11:14‐20. [PMC free article] [PubMed] [Google Scholar]
- 36. Gnatiuc L, Alegre‐Diaz J, Halsey J, et al. Adiposity and blood pressure in 110 000 mexican adults. Hypertension. 2017;69:608‐614. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Fan S, Yang B, Zhi X, et al. Neck circumference associated with arterial blood pressures and hypertension: a cross‐sectional community‐based study in northern Han Chinese. Sci Rep. 2017;7:2620. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Joshipura K, Munoz‐Torres F, Vergara J, Palacios C, Perez CM. Neck circumference may be a better alternative to standard anthropometric measures. J Diabetes Res. 2016;2016:6058916. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Mancia G, Fagard R, Narkiewicz K, et al. 2013 ESH/ESC guidelines for the management of arterial hypertension: the task force for the management of arterial hypertension of the European society of hypertension (ESH) and of the European society of cardiology (ESC). Eur Heart J. 2013;34:2159‐2219. [DOI] [PubMed] [Google Scholar]
- 40. Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA 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. 2018;71:e127‐e248. [DOI] [PubMed] [Google Scholar]
- 41. Mehlsen J, Wiinberg N. Interarm difference in blood pressure: reproducibility and association with peripheral vascular disease. Int J Vasc Med. 2014;2014:841542. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Clark CE. Difference in blood pressure between arms might reflect peripheral vascular disease. BMJ. 2001;323:399‐400. [DOI] [PMC free article] [PubMed] [Google Scholar]