Background:  Patients with HIV-associated lipodystrophy have increased prevalence of relative growth hormone (GH) deficiency. GH exerts significant effects on body composition, glucose homeostasis, and cholesterol metabolism. A long-term study investigating the effects of low-dose physiologic GH replacement in HIV patients with lipodystrophy and relative GH deficiency has not previously been performed.

Methods: We enrolled 56 HIV-infected men and women with lipodystrophy and relative reduction in GH secretion (peak GH <7.5 ng/mL by GHRH-arginine stimulation) enrolled in an 18-month prospective, randomized, double-blind, placebo-controlled trial. Low-dose GH (average dose 4.1 µg/kg/d) or identical placebo was administered daily subcutaneously to achieve target IGF-1 level in the upper quartile of normal range. Using intent-to-treat analysis, the treatment efficacy was examined using repeated measures mixed effects ANCOVA adjusting for potential confounders including age, gender, race, study drug dosage, testosterone use, duration of PI and NRTI use for all outcome measures. Additional adjustment for blood pressure, smoking, anti-hypertensive and lipid-lowering medications were included for carotid intima-media thickness (IMT), anti-hypertensive medication for BP, and lipid-lowering medications for lipids.

Results: In the treatment group compared to placebo, visceral adipose tissue area (–22 ± 6 vs –4 ± 4 cm²; p = 0.049), trunk-to-extremity fat ratio (–0.4 ± 0.1 vs –0.007 ± 0.1; p = 0.0002), and trunk fat (–0.5 ± 0.2 vs 0.2 ± 0.2 kg; p = 0.04) decreased, and IGF-1 increased (109 ± 11 vs –25 ± 8 ng/mL; p <0.0001). Diastolic blood pressure (–3 ± 1 vs 4 ± 1 mm Hg; p = 0.006) and triglycerides improved (–7, IQR –43–41 vs 0, –56–51 mg/dL), but carotid IMT did not change with GH compared to placebo (0.003 ± 0.008 vs –0.003 ± 0.009 mm, p = 0.78); 2-hour post-oral glucose challenge glucose increased (16 ± 5 vs –4 ± 4 mg/dL, p = 0.009). Frequency of GH-related adverse effects were not increased compared to placebo (22% in GH group vs 28% in placebo group; p = 0.64) and viral load and CD4 were unchanged.

Conclusions: Physiologic GH replacement significantly reduced visceral fat and truncal obesity, triglycerides, and diastolic blood pressure, and was well-tolerated. However, GH increased 2-hour glucose and had no effect on carotid IMT in patients with HIV lipodystrophy and relative GH deficiency over a long period of 18 months.