Virtually all circulating insulin-like growth factors I and II (IGF-I and IGF-II) are bound to specific binding proteins (IGFBP), of which IGFBP-3 is the quantitatively most important. The mechanisms regulating the close coordination between serum levels of IGFs and IGFBP-3 is poorly understood. We therefore evaluated the temporal association of serum IGF-I, IGF-II, and IGFBP-3 measured by RIAs after well defined short-term GH exposure in GH-deficient patients. Six patients (mean +/- SE age: 20.5 +/- 1.1 yr) each underwent three GH study protocols in random order. Each study was preceded by 4 weeks without GH therapy. Two units of GH were administered iv as either: 1) two boluses, 2) eight boluses, or 3) a constant infusion. The duration of each study was 44 h including at least 16 h after termination of GH administration. Increments in serum IGF-I occurred 4-6 h after initiated GH exposure in all studies. In the two-bolus study the IGF-I increase was modest with mean +/- SE peak values of 12.4 +/- 2.1 nmol x L-1 after GH administration. In the eight bolus and constant infusion studies significantly higher IGF-I levels were generated: 17.0 +/- 2.2 nmol x L-1 (8 bolus) and 18.8 +/- 1.1 h nmol x L-1 (infusion). In contrast the time course change in serum IGF-II did not differ in the three studies, and it was characterised by a sluggish increase of approximately 30% evidenced after 16-20 h. The changes in IGFBP-3 were almost identical in the three studies. After a lag phase of approximately 18-20 h a gradual increase of approximately 40%, which had not ceased at the end of the study period, was observed. The molar ratio of serum IGF-I plus IGF-II:serum IGFBP-3 remained constant with values between 0.8-0.9 except in the constant infusion experiment, in which the ratio increased significantly with time reaching a mean peak value, which exceeded 1.0, after 24 h. Our data suggest that a pulsatile GH pattern is not superior to constant GH levels as regards generation of IGFs and IGFBP. The earlier increase in serum IGF-I compared to IGF-II and IGFBP-3 suggests that IGF-I may be the main regulator of IGFBP-3 production. Accordingly, the slow increase in serum IGF-II, which paralleled that of IGFBP-3, could indicate that serum IGF-II levels mainly depend on the concentration or binding site availability of IGFBP-3.