Abstract Body: Background: Validation guidelines require the width of auscultatory blood pressure (BP) cuffs to cover 37.5%-50% of the mid-arm circumference but this can be problematic for short and wide arms, which are common among obese adults. Till date, there has not been a way to quantify the extent of implausible cuff requirements for obese adults due to the absence of inner-arm length measurements in large cohorts. Equations estimating inner-arm length based on standard anthropometric measures are critical to quantify the extent of this cuff mismatch and inform scientific priorities around cuff dimensions for adults with obesity.
Objective: To estimate inner arm length from common body measurements using screening data from two community trials to inform best practices in BP cuff selection.
Methods: During screening visits for GoFresh and GoFreshRx, we obtained the following measurements from 544 participants: outer arm length (acromion to elbow), inner arm length (anterior axillary fold to antecubital fossa), upper arm circumference (around arm directly under the axillary fold), mid arm circumference, lower arm circumference (above the antecubital fossa), and upper forearm circumference (below the antecubital fossa). Height and weight was obtained from a subset of 186 participants. We used univariable and multivariable linear regression models of these arm measures, height, and weight to estimate inner-arm length. Model performance was assessed using the adjusted R-squared and compared via likelihood ratio tests. We determined accuracy within ±1 cm and ±2 cm of the actual inner-arm length.
Results: Mean age was 47 years among the 544 participants and 53 years in the subset of 186; most were female (73% vs 71%) (Table 1). Outer arm length and height showed strongest association with inner arm length (Table 2). Likelihood ratio tests indicated that the addition of height and weight significantly improved model fit (P < 0.05) (Table 3). In the subset, height and weight, in combination with outer arm length and all arm circumference measures, produced the best overall performance; although it was still low (adjusted R² = 0.18).
Conclusion: Outer arm length and height are most strongly associated with inner arm length and could be used to estimate inner arm length in large population databases. However, despite their associations, model performance was poor and likely should not be a substitute for direct measurement of inner arm length among adults with obesity.