The effect of recombinant human growth hormone on regulation of growth hormone secretion and blood glucose in insulin-dependent diabetes.

Abstract

GH hypersecretion in insulin-dependent diabetes (IDDM) is well documented. Although it has recently been shown that residual insulin secretion determines the magnitude of this GH hypersecretion, the underlying mechanisms of the disorder have not yet been clarified. The 24-h GH and blood glucose profiles, insulin-like growth factor I (IGF-I) concentrations and GH responses to GRF were analyzed in 21 insulin-dependent diabetics and 4 healthy subjects before and after 7 days treatment with recombinant human GH (rhGH) (4 IU given sc at 0800 h). According to C-peptide response to glucagon IDDM patients were subdivided into C-peptide negative (CpN, n = 12) patients without endogenous pancreatic beta-cell activity and C-peptide positive (CpP, (n = 9) patients with endogenous insulin secretion. No significant difference could be observed between the mean 24-h blood glucose profile before and after rhGH treatment in any treated group. Before and on rhGH treatment the highest 24-h GH values were observed in CpN patients when compared to CpP and controls. The rhGH treatment induced a similar increase in the mean 24-h GH concentrations in all groups studied which was statistically significant only in CpP diabetics. Mean pretreatment serum IGF-I concentrations were not significantly different between CpN, CpP patients and controls. The net increase in IGF-I concentrations after rhGH treatment was however, significantly lower in CpN patients than in CpP and control subjects. GRF-induced GH response before and after rhGH treatment was significantly greater in diabetics than in controls. The response of GH to GRF in CpN diabetics was however, almost unchanged after treatment whereas it became lower in CpP diabetics and controls. The dose of 4 IU of rhGH increased significantly GH levels in diabetics with preserved beta-cell function with consequent increase in IGF-I levels and attenuation of GRF induced GH response. In contrast, the same dose of rhGH failed to induce significant increase in GH levels in diabetics without residual beta-cell activity, most probably due to already high pretreatment levels. In addition, neither increase in IGF-I levels nor suppression of GH response to GRF on rhGH treatment was observed in CpN diabetics. The results are in keeping with an important role of portal insulin in GH-induced hepatic IGF-I secretion.