Accuracy of anthropometric measurements in estimating fat mass in individuals with 21-hydroxylase deficiency

Objective The use of anthropometric measurements to estimate the percentage of body fat (%BF) is easy and inexpensive. However, the accuracy of these methods in patients with 21-hydroxylase deficiency (21OHD) has not been explored. The objective of this study was to evaluate the accuracy of skinfold-based models, body mass index (BMI), and waist circumference (WC) in estimations of %BF using dual-energy X-ray absorptiometry (DXA) as the reference method in individuals with 21OHD. Methods Fifty-four 21OHD patients (32 women and 22 men), aged 7 to 20 y, were recruited for the study. DXA was used to determine %BF; four predictive skinfold equations, BMI, and WC were assessed for accuracy in determining %BF. Results All predictive skinfold equations were highly associated ( R , range: 0.82-0.89) with DXA %BF values. In women, BMI and WC showed moderate correlations ( R = 0.69 for both BMI and WC) with DXA values. In contrast, among men there was a low explanatory power for BMI (13%) and WC (4%) and high errors (BMI, 6.9%; WC, 7.4%). All predictive equations significantly underestimated %BF (range of differences, −4.1 to −8.9) compared with DXA (women, 31.3 ± 6.1; men, 24.4 ± 7.3), and large limits of agreement were observed (range, −15.3 to 1.7 and −15.5 to 4.2 for women and men, respectively). Conclusion In children and adolescents with 21OHD, %BF as estimated by skinfold measurements was associated more strongly with DXA-assessed %BF than both BMI and WC. However, still, the skinfold-based assessment underestimated DXA %BF and showed moderate agreement.