Evaluation of limb measurements for predicting stature in men ≥ 60 years of age.

Abstract

Stature is often difficult to determine among elderly subjects due to postural changes, contractures, osteoporosis, etc. Formulae have been developed to estimate stature from knee height. However, knee height measurements are difficult to perform in some patients, such as those with paralysis, amputation, or the acutely ill. Subjects for this pilot study included 100 male inpatients (85 white, 15 black) at a large VA medical center. Mean age was 70.7 ± 6.5 (SD) years. Mean values for stature, knee height, upper arm length (UAL), lower arm length (LAL), and total arm length (TAL) were 170.1 ± 6.6, 53.5 ± 2.9, 35.6 ± 2.3, 47.8 ± 2.5 and 83.4 ± 4.2cm, respectively. Knee height was the strongest correlate of stature (r = 0.82, p <0.0001 ). Lower arm length (olecranon to third finger) was the best upper limb predictor (r = 0.74, p < 0.0001). Upper arm length (r = 0.49) and total arm length (r = 0.70) were also significantly related to height (p < 0.0001). The regression model for estimating stature using lower arm length was calculated to be: stature = 1.93 (LAL) + 77.75. We conclude that lower arm length may be a simple and more practical means for estimating stature in patients for whom standing height and knee height are impractical to measure.