Objective: Studies with GH substitution in GH-deficient (GHD) adults lasting more than 6 months have so far been uncontrolled. End-points such as physical fitness and body composition may be subject to a considerable placebo effect which weakens the validity of open studies. We therefore tested GH (2 IU/m2 per day) versus placebo treatment for 12 months.
Design: Twenty-nine patients (mean age 45.5 +/- 2.0 years) with adult-onset GHD were studied in a double-blind, parallel design. Measurements of body composition by means of conventional anthropometry, bioelectrical impedance (BIA), CT scan and DEXA scan, exercise capacity, and isometric muscle strength were performed at baseline and after 12 months treatment. For body composition measurements a control group of 39 healthy, age and sex-matched subjects was included.
Results: Sum of skinfolds (SKF) at 4 sites decreased significantly after GH treatment. Total body fat (TBF) as assessed by DEXA and BIA was elevated at baseline but normalized after GH. TBF assessed by SKF revealed significantly higher levels compared to DEXA and BIA, although all estimates intercorrelated closely. Visceral and subcutaneous abdominal fat decreased by 25 and 17%, respectively after GH (P < 0.01) to levels no longer different from the control group. CT of the mid thigh revealed a significant reduction in fat tissue and a significant increase in muscle volume after GH treatment, both of which resulted in a normalization of the muscle: fat ratio (%) (placebo: 58:42 (baseline) vs 58:42 (12 months); GH: 66:34 (baseline) vs 72:28 (12 months) (P = 0.002); normal subjects: 67:33 (P < 0.05 when compared to 12 months placebo data)). Total body resistance and resistance relative to muscle volume decreased significantly after GH treatment suggesting over-hydration as compared to normal subjects. Exercise capacity (kJ) increased significantly after GH treatment (placebo: 54.7 +/- 9.8 (baseline) vs 51.6 +/- 8.2 (12 months); GH: 64.9 +/- 13.3 (baseline) vs 73.5 +/- 13.6 (12 months) (P < 0.05)). Isometric quadriceps strength increased after GH but no treatment effect could be detected owing to a small increase in the placebo group. Serum IGF-I levels (microgram/l) were low baseline and increased markedly after GH treatment to a level exceeding that of normal subjects (270 +/- 31 (12 months GH) vs 156 +/- 8 (normal subjects (P < 0.01)). The levels of serum electrolytes and HbA1c remained unchanged. The number of adverse effects were higher in the GH group after 3 months, but not after 6 and 12 months.
Conclusions: (1) The reduction in excess visceral fat during GH substitution is pronounced and sustained; (2) beneficial effects on total body fat, muscle volume and physical fitness can be reproduced during prolonged placebo-controlled conditions; (3) uncontrolled data on muscle strength must be interpreted with caution; (4) a daily GH substitution dose of 2 IU/m2 seems too high in many adult patients.