Abstract
PURPOSE Doses of growth hormone in adults with growth hormone deficiency are now lower than previously. However, it is not clear they are as effective as higher doses. The objective of this meta-analysis was to assess efficacy of low to moderate dose (LD) GH replacement on standard endpoints of GH compared to higher doses. METHODS A meta-analysis was carried out using PubMed, Cochrane and Embase databases from 1960 to 9/23/12. Three reviewers identified randomized double-blind, placebo-controlled trials of 6 months duration. Of 173 publications, 28 representing 22 trials (591 GH-treated patients and 562 placebo) were included. Data were independently extracted by three reviewers. Endpoints were analyzed if ≥4 studies per dose group reported baseline and 6 month data. RESULTS Mean lean body mass (LBM) increased by 2.61 kg in GH-treated subjects versus 0.04 in the placebo group (P < 0.0001). Fat mass (FM) was reduced by -2.19 kg versus 0.31 (GH vs. placebo) (P = 0.0002). Changes in LBM and FM were dose-related (P = 0.02 and 0.007, respectively), high dose (HD) being more effective than low dose (LBM P = 0.03 and FM P = 0.04). In contrast, treatment with GH reduced total cholesterol -0.38 mmol/L versus. 0.01 (placebo) (P < 0.0001), and low density lipoprotein cholesterol (LDL-C) -0.42 mmol/L versus -0.1 (P = 0.0009), but there were no differences between LD and HD GH. CONCLUSIONS LDs of hGH improve total- and LDL-C, and body composition. Higher doses are more effective on body composition, but not lipids.
