A crossover randomized clinical trial revealed that arm position during blood pressure measurement may lead to substantial overestimation of readings, potentially resulting in misdiagnosis and overtreatment of hypertension.
In the ARMS (Arm Position and Blood Pressure Readings) study, published in JAMA Internal Medicine, researchers from Johns Hopkins University recruited 133 adult participants in Baltimore, Maryland. The study compared blood pressure (BP) readings obtained in three arm positions: supported on a desk with midcuff at heart level (reference), hand supported on lap, and arm unsupported at the side.
The study employed a rigorous methodology to account for intrinsic BP variability. The study participants underwent four sets of triplicate BP measurements in randomized order, including two sets with the arm in the reference position. The difference-in-differences approach was used to analyze the data, comparing the difference between nonstandard and reference positions with the difference between two reference measurements.
Study population characteristics included:
- Mean age: 57 years (standard deviation [SD] = 17)
- Female participants: 53%
- Black participants: 77%
- Body mass index ≥ 30: 41%
- SBP ≥ 130 mmHg: 36%
- Prescribed antihypertensive medications: 59%
The researchers used a validated oscillometric BP device (ProBP 2000 Digital Blood Pressure Device) and followed standardized measurement procedures. All measurements were conducted by trained research staff between 9 AM and 6 PM.
Among the key findings were:
- Lap position: SBP increased by 3.9 mmHg (95% confidence interval [CI] = 2.5–5.2) and DBP by 4.0 mmHg (95% CI = 3.1–4.9) compared with the reference position.
- Side position: SBP increased by 6.5 mm Hg (95% CI = 5.1–7.9) and DBP by 4.4 mmHg (95% CI = 3.4–5.4) compared with the reference position.
- Subgroup analysis: Results were largely consistent across subgroups, with notably larger differences observed in side SBPs among those with SBP ≥ 130 mmHg (8.5 mmHg, 95% CI = 5.7–11.4).
The researchers demonstrated that commonly used nonstandard arm positions elevated systolic BP (SBP) readings by up to 6.5 mmHg and diastolic BP (DBP) readings by up to 4.4 mmHg compared with the guideline-recommended position.
The researchers estimated that improper arm positioning could result in misclassification of hypertension in 16% of U.S. adults (40 million individuals) using an SBP cutoff of 140 mmHg, and 22% (54 million individuals) using an SBP cutoff of 130 mmHg.
The study noted several physiological mechanisms that may explain the higher BP readings in nonstandard arm positions, including increased hydrostatic pressure, decreased venous return, compensatory vasoconstriction, and muscle contraction in unsupported arms.
Average SBP/DBP was 126/74 mmHg for each of the desk 1 and desk 2 positions, 130/78 mmHg for the lap position, and 133/78 mmHg for the side position. The mean (SD) difference between desk 2 and desk 1 was −0.21 (7.26) mmHg for SBP and 0.09 (3.78) mmHg for DBP.
Subgroup analyses revealed a statistically significant larger difference in lap BPs among those who had not received medical care in the past year compared with those who had. Additionally, side SBPs showed a greater difference among those with an SBP of 130 mmHg or higher compared to those with lower SBP.
The study had limitations, including unequal random distribution of participants to each group because of the use of the RANDBETWEEN function. However, sensitivity analyses adjusting for participant characteristics and order demonstrated consistent results with the a priori analysis. Some subgroups included relatively small sample sizes, and the generalizability to other settings and BP devices was uncertain.
In conclusion, this crossover randomized clinical trial demonstrated that not adhering to the guideline-recommended arm position and support during BP measurement resulted in overestimation of BP by 4 to 10 mmHg. These findings highlighted the importance of proper arm positioning in clinical practice to ensure accurate BP measurements and appropriate diagnosis and management of hypertension.
Conflict of interest disclosures can be found in the study.