The validity of estimating total body fat and fat-free mass from skinfold thickness in adults with growth hormone deficiency.

The regression equations of Durnin and Womersley for estimating total body fat (TBF) and fat-free mass (FFM) from skinfold thickness were validated for adult GH-deficient (GHD) patients by comparing the values of TBF and FFM from the prediction equations with the directly measured values from dual energy x-ray absorptiometry. Twenty-seven male and 24 female patients (aged 21-61 yr) were studied. GHD was isolated in 5 cases and was part of a spectrum of hypopituitarism due to a variety of causes in 46 cases. The mean period of GHD was 6.9 +/- 4.6 yr. All patients were receiving stable replacement therapy. The validation statistics showed no significant differences (P > 0.05) between measured and predicted values of TBF and FFM in either males (24.5 vs. 24.9 and 65.5 vs. 65.8 kg) or females (24.6 vs. 26.3 and 44.7 vs. 43.9 kg). Mean differences were smaller in males (0.4 and 0.2 kg) than females (1.7 and -0.8 kg); they were less than 1% in males and less than 2% in females. Therefore, the Durnin and Womersley equations are suitable for general use with GHD patients. Using TBF (kilograms) from dual energy x-ray absorptiometry as the dependent variable and the log of the sum of skinfold thickness as the independent variable, linear regression equations were formulated to predict TBF in GHD patients. The lowest SE of estimate was 3.8 kg in males and 4.6 kg in females. To determine their general applicability, these equations will need to be cross-validated.