Continuous infusion versus daily injections of growth hormone (GH) for 4 weeks in GH-deficient patients.

Abstract

Endogenous GH secretion is pulsatile. Animal studies indicate that GH administered in a pulsatile manner induces growth and insulin-like growth factor I (IGF-I) generation more effectively than continuous administration. Short term human studies, however, have reported similar metabolic effects with constant and pulsatile GH delivery. This study was carried out to compare the metabolic effects of longer term continuous infusion vs. daily injections of GH. Thirteen GH-deficient patients were studied in a cross-over design. The patients were randomized to receive GH as a continuous sc infusion by means of a portable pump for 1 month and as daily sc injections (at 1900 h) for another month. An average daily GH dosage (+/- SEM) of 3.15 +/- 0.27 IU was administered during both periods. Steady state 24-h profiles of GH, IGF-I, IGF-binding proteins (IGFBPs), insulin, glucose, lipid intermediates, and other metabolites were monitored after each treatment period. At the end of each study period (at 0800 h), an oral glucose tolerance test was performed. The mean (+/- SEM) integrated levels of serum GH (micrograms per L) were higher after GH injection [2.51 +/- 0.54 (injection) vs. 1.77 +/- 0.35 (infusion); P < 0.02]. Continuous infusion induced higher nighttime than daytime GH levels (P = 0.01), indicating a diurnal variation in the absorption or clearance of GH. Serum IGF-I levels (micrograms per L) were slightly higher (P < 0.05, by analysis of variance) after continuous GH infusion [312.5 +/- 50.2 (injection) and 334.6 +/- 46.6 (infusion)]. Similarly, constant GH delivery induced higher IGFBP-3 levels (P < 0.05, by analysis of variance). Serum IGFBP-1 levels were similar on the two occasions. Daily GH injections increased daytime insulin levels (P < 0.05), whereas 24-h levels were similar (P = 0.14). The trend toward increased insulin levels after GH injections was also found during the oral glucose tolerance test (P = 0.07). Blood glucose levels were identical on the two occasions. Nocturnal levels of nonesterified fatty acids were higher (P < 0.05) after GH injection. We conclude that continuous sc infusion of GH induced serum IGF-I and IGFBP-3 levels more effectively than daily sc injections. The constant appearance of GH in the circulation did not impair glucose tolerance, but resulted in a less physiological diurnal pattern of nonesterified fatty acids. Our data do not support the concept that a pulsatile GH pattern is of critical physiological significance.