Limitations of body mass index for counseling individuals with unilateral lower extremity amputation

Background: Body composition is important for health screening, but appropriate methods for unilateral lower extremity amputees have not been validated. Objectives: To compare body mass index adjusted using Amputee Coalition equations (body mass index-Amputee Coalition) to dual-energy X-ray absorptiometry in unilateral lower limb amputees. Study design: Cross-sectional, experimental. Methods: Thirty-eight men and women with lower limb amputations (transfemoral, transtibial, hip disarticulation, Symes) participated. Body mass index (mass/height(2)) was compared to body mass index corrected for limb loss (body mass index-Amputee Coalition). Accuracy of classification and extrapolation of percent body fat with body mass index was compared to dual-energy X-ray absorptiometry. Results: Body mass index-Amputee Coalition increased body mass index (by similar to 1.1 kg/m(2)) but underestimated and misclassified 60% of obese and overestimated 100% of lean individuals according to dual-energy X-ray absorptiometry. Estimated mean percent body fat (95% confidence interval) from body mass index-Amputee Coalition (28.3% (24.9%, 31.7%)) was similar to dual-energy X-ray absorptiometry percent body fat (29.5% (25.2%, 33.7%)) but both were significantly higher (p < 0.05) than percent body fat estimated from uncorrected body mass index (23.6% (20.4%, 26.8%)). However, total errors for body mass index and body mass index-Amputee Coalition converted to percent body fat were unacceptably large (standard error of the estimate = 6.8%, 6.2% body fat) and the discrepancy between both methods and dual-energy X-ray absorptiometry was inversely related (r = -0.59 and r = -0.66, p < 0.05) to the individual's level of body fatness. Conclusions: Body mass index (despite correction) underestimates health risk for obese patients and overestimates lean, muscular individuals with lower limb amputation.