Use of GnRH agonist and human chorionic gonadotrophin tests for differentiating constitutional delayed puberty from gonadotrophin deficiency in boys

OBJECTIVES: The differentiation of constitutional delayed puberty (CDP) from gonadotrophin deficiency (GD) in boys at referral poses a difficult challenge. The effectiveness of the GnRH agonist (GnRH-a) test in distinguishing between the two conditions was evaluated and compared with findings of the GnRH and hCG stimulation tests. PATIENTS, METHODS AND DESIGN: The study sample included 32 prepubertal boys aged 14 years or older. Thirteen entered spontaneous puberty within 1 year of referral (group A) and 19 remained prepubertal (group B). All underwent the GnRH test (Relefact, Hoechst AG, 0.1 mg/m² i.v. in one bolus), GnRH-a test (Decapeptyl, Ferring GmbH, 0.1 mg/m² s.c.) and hCG stimulation (Chorigon, Teva, 1500 units i.m. on three alternate days) at 1-week intervals. All tests were performed at referral at 0800 h. Blood samples were collected before testing and at 30 and 60 min (GnRH test) or 4 h (GnRH-a) for LH and FSH determination, and before testing and at 4 h (GnRH-a) or on the seventh day (hCG) after stimulation for serum testosterone measurement. RESULTS: The LH response to GnRH-a and the testosterone response to hCG stimulation were significantly higher in group A (LH, mean ± SD 20.4 ± 7.5 mlU/ml, range 10.8-32.6; testosterone, mean ± SD 18.0 ± 5.9 nmol/l, range 9.4-26, P < 0.0001) than in group B (LH, mean ± SD 2.3 ± 2.0 mlU/ml, range 0.7-6.9; testosterone, mean ± SD 1.0 ± 0.7 nmol/l, range 0.7-3.2), with no overlap between the groups. The cut-off for the LH response to GnRH-a was 8.0 mlU/ml, and for the testosterone response to hCG, 8 nmol/l. There were also significant differences between the groups in mean basal serum LH and FSH (LH, 1.1 ± 0.5 vs. 0.6 ± 0.2 mlU/ml, P < 0.05; FSH, 2.2 ± 2.0 vs. 0.4 ± 0.3 mlU/ml, P < 0.02) and their response to GnRH (LH, 11.4 ± 4.4 vs. 2.7 ± 1.1 mlU/ml, P < 0.0001; FSH, 5.1 ± 3.4 vs. 2.5 ± 2.4 mlU/ml, P < 0.0001), and mean serum testosterone level at 4 h after GnRH-a administration (1.9 ± 1.0 vs. 0.9 ± 0.4 nmol/l, P = 0.002), but all showed a great overlap in range. Mean age, testicular volume and basal serum testosterone levels were similar in the two groups at referral. One year later, the testicular volume of group A (5.0-12.0 ml) was significantly larger than that of group B (1.0-3.0 ml, P < 0.0001), which remained unchanged on reexamination 3.0 ± 0.5 years later. CONCLUSIONS: The GnRH-agonist test and the repeated-injection hCG test are reliable diagnostic tools for differentiating CDP from GD in boys.