Objective: Previous studies evaluating the lipolytic effect of GH have in general been performed in subjects on chronic GH therapy. In this study we assessed the lipolytic effect of GH in previously untreated patients and examined whether the negative effect of enhanced lipolysis on glucose metabolism could be counteracted by acute antilipolysis achieved with acipimox. Methods: Ten GH-deficient (GHD) adults participated in four experiments each, during which they received in a double-blind manner: placebo (A); GH (0.88^0.13 mg) (B); GH þ acipimox 250 mg b.i.d. (C); and acipimox b.i.d. (no GH) (D), where GH was given the night before a 2 h euglycemic, hyperinsulinemic clamp combined with infusion of [3-3 H]glucose and indirect calorimetry. Results: GH increased basal free fatty acid (FFA) levels by 74% (P ¼ 0.0051) and insulin levels by 93% (P ¼ 0.0051). This resulted in a non-significant decrease in insulin-stimulated glucose uptakes (16.61^8.03 vs 12.74^5.50 mmol/kg per min (S.D.), P ¼ 0.07 for A vs B). The rates of insulin-stimulated glucose uptake correlated negatively with the FFA concentrations (r ¼ 2 0.638, P , 0.0001). However, acipimox caused a significant improvement in insulin-stimulated glucose uptake in the GH-treated patients (17.35^5.65 vs 12.74^5.50 mmol/kg per min, P ¼ 0.012 for C vs B). The acipimox-induced enhancement of insulin-stimulated glucose uptake was mainly due to an enhanced rate of glucose oxidation (8.32^3.00 vs 5.88^2.39 mmol/kg per min, P ¼ 0.07 for C vs B). The enhanced rates of glucose oxidation induced by acipimox correlated negatively with the rate of lipid oxidation in GH-treated subjects both in basal (r ¼ 2 0.867, P ¼ 0.0093) and during insulinstimulated (r ¼ 2 0.927, P ¼ 0.0054) conditions. GH did not significantly impair non-oxidative glucose metabolism (6.86^5.22 vs 8.67^6.65 mmol/kg per min, P ¼ NS for B vs A). The fasting rate of endogenous glucose production was unaffected by GH and acipimox administration (10.99^1.98 vs 11.73^2.38 mmol/kg per min, P ¼ NS for B vs A and 11.55^2.7 vs 10.99^1.98 mmol/kg per min, P ¼ NS for C vs B). On the other hand, acipimox alone improved glucose uptake in the untreated GHD patients (24.14^8.74 vs 16.61^8.03 mmol/kg per min, P ¼ 0.0077 for D vs A) and this was again due to enhanced fasting (7.90^2.68 vs 5.16^2.28 mmol/kg per min, P ¼ 0.01 for D vs A) and insulin-stimulated (9.78^3.68 vs 7.95^2.64 mmol/kg per min, P ¼ 0.07 for D vs A) glucose oxidation. Conclusion: The study of acute administration of GH to previously untreated GHD patients provides compelling evidence that (i) GH-induced insulin resistance is mainly due to induction of lipolysis by GH; and (ii) inhibition of lipolysis can prevent the deterioration of insulin sensitivity. The question remains whether GH replacement therapy should, at least at the beginning of therapy, be combined with means to prevent an excessive stimulation of lipolysis by GH.