Only two previous studies have assessed the effects of long-term GH replacement therapy on bone mineral density (BMD) in patients with adult onset GH deficiency. To date no study has looked at the long-term impact on BMD after a short course (6-12 months) of GH replacement. In two groups of patients with adult onset GH deficiency we have studied BMD either (a) after 3 years of continuous GH replacement or (b) 2 years after completion of a short course of GH.DESIGN
An open GH therapeutic study in which patients were recruited from a previous double-blind placebo-controlled study. The BMD status of all patients was unknown to the physician and patient at the time of recruitment.PATIENTS
Group A (n = 7, three females) all received GH replacement continuously for 3 years. Group B (n = 8, five females) included six patients who received GH replacement for 6 months and two who received GH replacement for 12 months with BMD being measured at 6-monthly intervals.METHODS
Single photon absorptiometry (SPA) and later single X-ray absorptiometry (SXA) were used to measure forearm cortical BMD. Dual-energy X-ray absorptiometry (DXA) was used to measure lumbar spine, trochanteric, femoral neck and Ward's area BMD.RESULTS
In group A lumbar spine and trochanter BMD had increased significantly from baseline by 3.7% (DXA: median change = 0.045 g/cm2; P = 0.028) and 4.0% (DXA: median change = 0.031 g/cm2; P = 0.046), respectively. There were non-significant decreases in femoral neck (1.9%) (DXA: median change = −0.02 g/cm2; P = 0.39), Ward's area (6.5%) (DXA: median change = −0.06 g/cm2; P = 0.09) and forearm (2.6%) (SPA/SXA: median change = −0.013 g/cm2; P = 0.18). In group B, compared with baseline, only trochanter BMD changed significantly, increasing by 5.9% (DXA: median change = 0.0485 g/cm2; P = 0.049). Lumbar spine (DXA: median change = −0.001 g/cm2) Ward's area (DXA: median change = 0.0135 g/cm2), femoral neck (DXA: median change = −0.005 g/cm2) and forearm cortical (SPA/SXA; median change = −0.01 g/cm2) BMD did not change significantly (P = 0.67, P = 0.57, P = 0.86 and P = 0.31, respectively). Median percentage changes compared with baseline were −0.1%, 1.8%, −0.5% and −2.1%, respectively. From the time of completion of GH therapy however, BMD increased significantly at lumbar spine, (median change=0.023 g/cm2), Ward's area (median change = 0.03 g/cm2) and trochanter (median change = 0.056 g/cm2) (P = 0.036, P = 0.049 and P = 0.012, respectively) but not at the femoral neck (median change = 0.017 g/cm2; P = 0.31) or forearm (median change = 0 g/cm2; P = 0.75).CONCLUSION
Long-term GH replacement therapy for three years appears to have beneficial effects on bone in patients with adult onset GH deficiency particularly at the lumbar spine and trochanter; the effects on femoral neck and forearm cortical BMD, however, are less impressive. A short course (6-12 months) of GH replacement therapy results in an increase in trochanter BMD several years later, and after an initial decline in BMD whilst on GH replacement, lumbar spine and Ward's area BMD return towards their baseline values. These results emphasize that not all types of bone and skeletal sites respond to GH therapy identically. Furthermore a short course of GH replacement over 6-12 months may result in significant changes in BMD several years later.