To determine the usefulness of a GnRH agonist analog as a diagnostic test to distinguish between constitutional delay of growth (CGD) in boys with Tanner stage I of sexual development and patients with hypogonadotropic hypogonadism (HH), we evaluated six boys (mean age 15 yr 4 m) and five HH patients (mean age 20 yr 4 m). In addition, 20 normal healthy men aged 21 yr to 50 yr received either nafarelin or GnRH followed two weeks later by the other test in order to compare the efficacy of each of these tests and to evaluate the optimal sampling times for the nafarelin test. All subjects were healthy, and had not received hormonal replacement for at least 2 months prior to enrollment in the study. Each man had four baseline blood samples before and at timed intervals following the administration of either GnRH or nafarelin. Each of the patients had blood withdrawn every 15 min during 12 h overnight followed by a single s.c. injection of nafarelin (1 microgram(s)/kg up to 100 microgram(s)), except two HH patients who did not have an overnight study. Blood samples were obtained at timed intervals for 24 h. LH, FSH, T and E2 were measured by RIA. Baseline concentrations of plasma LH, FSH and T were similar before the administration of either GnRH or nafarelin in the group of normal men. Peak stimulation of plasma LH, FSH and T released by nafarelin was significantly higher, and it took a longer time to reach the peak maximum, than after GnRH (p < 0.001). Mean nocturnal LH was 5.5 +/- 0.9 IU/I for the CGD group, and 2.7 +/- 0.7 IU/I for HH (p < 0.02). Mean nocturnal FSH was 5.1 +/- 1.0 and 2.5 +/- 0.2 IU/I whereas mean nocturnal T concentrations were 4.2 +/- 0.8 and 0.7 +/- 0.2 nmol/I (CGD vs HH, respectively, p < 0.02). Peak LH responses to nafarelin were 36.9 +/- 8.9 IU/I for the CGD group, and 7.0 +/- 2.0 IU/I for the HH group (p < 0.001). Peak FSH released by nafarelin was 14.2 +/- 2.4 IU/I for the CGD group and 4.8 +/- 2.0 IU/I for the HH group (p < 0.02). Peak T was reached 24 h following nafarelin injection and was 5.7 +/- 1.7 nmol/I for the CGD group and 0.3 +/- 0.2 nmol/I for the HH group (p < 0.001). The results obtained indicate that in early stages of puberty (before detectable changes of sexual maturation) the nafarelin test, with measurements of LH, FSH and T in blood or in urine, is superior to and more practical than overnight hormonal estimates to clearly distinguish CGD from HH.