Body composition analysis by dual energy x-ray absorptiometry and anthropometry in adults with childhood-onset growth hormone (GH) deficiency before and after six months of recombinant GH therapy

Abstract

Measurements of total body fat (BF) and fat free mass (FFM) obtained by anthropometry, using the Durnin and Womersley (DW) equations, and by total body dual energy x-ray absorptiometry (DXA) were compared in 8 adults with childhood-onset GH deficiency (GHD) and in 9 healthy subjects. The sensitivity of these two methods in detecting the changes in body composition produced by six months of GH therapy in patients with GHD was also compared. Anthropometric determination of percent BF was calculated from the sum of biceps, triceps, subscapular and suprailiac skinfolds, using the appropriate DW and Siri equations for body density and percent fat estimation. FFM was calculated by subtracting BF from body mass (BM). BF and FFM were also determined by DXA (QDR 1000/W, Hologic Inc). The data obtained from the GHD patients were compared with those recorded in a control group of healthy males, matched for sex, age and physical activity. Body composition obtained by anthropometry: before GH treatment, significant differences existed between patients and controls in terms of BM (mean +/- SD: 45.8 +/- 10.0 vs 71.7 +/- 6.6 kg), percent BF (21.0 +/- 3.2 vs 17.1 +/- 3.7%) and FFM (36.0 +/- 6.5 vs 59.3 +/- 3.7 kg), while body mass index (BMI, kg/m2) values were similar in the two groups. Six months of GH therapy did not change BM and BMI, but caused a significant reduction of percent BF (from 21.0 +/- 3.2 to 18.6 +/- 4.0%) and a rise of FFM (from 36.0 +/- 6.5 to 38.0 +/- 6.7 kg). After treatment, no significant differences were found between percent BF values of patients and controls. Body composition obtained by DXA: BF (22.0 +/- 3.9%) and FFM (37.2 +/- 8.0 kg) of patients significantly differed from those of controls (16.8 +/- 3.7% and 59.8 +/- 3.7 kg) before treatment; after GH treatment, percent BF values (17.7 +/- 4.9%) of patients were similar to those of controls. Anthropometry vs DXA: high correlation (p < 0.001-0.0001, R2 = 0.784-0.988) was found between the percent BF and FFM determined by anthropometry and by DXA for both patients, before and after treatment, and controls. It is noteworthy that, for both BF and FFM, most values were evenly distributed along the identity line, showing no systematic overestimation or underestimation by anthropometry. The relation between DXA and anthropometry was maintained even after GH treatment. These results indicate that body fat and FFM assessment by anthropometry are comparable to those by DXA. GH-induced changes in body composition in hypopituitary adults are detected with the same level of accuracy by the two techniques. The reliability, practicality and low cost of anthropometry favour its use for the assessment of body composition even in GHD patients.