Context: Differential diagnosis between hypogonadotropic hypogonadism (HH) and constitutional delay of puberty in boys is challenging. Most tests use an acute GnRH stimulus, allowing only the release of previously synthesized gonadotropins. A constant GnRH infusion, inducing de novo gonadotropin synthesis, may allow a better discrimination.
Objective: We evaluated the diagnostic accuracy of basal and peak gonadotropins after GnRH infusion, measured by ultrasensitive assays, to confirm the diagnosis in boys with suspected HH.
Design and setting: We conducted a validation study following Standards for Reporting of Diagnostic Accuracy criteria at a tertiary public hospital.
Patients and methods: A GnRH i.v. infusion test was performed in 32 boys. LH and FSH were determined by immunofluorometric assay at 0-120 min. DIAGNOSIS ASCERTAINMENT: The following diagnoses were ascertained: complete HH (n = 19; testes < 4 ml at 18 yr), partial HH (n = 6; testes enlargement remained arrested for > or = 1 yr or did not reach 15 ml), and constitutional delay of puberty (n = 7; testes > or = 15 ml at 18 yr).
Main outcome measures: Sensitivity, specificity, positive and negative predictive values, and diagnostic efficiency were assessed.
Results: Basal FSH less than 1.2 IU/liter confirmed HH with specificity of 1.00 (95% confidence interval = 0.59-1.00), rendering GnRH infusion unnecessary. In patients with basal FSH of at least 1.2 IU/liter, the coexistence of peak FSH less than 4.6 IU/liter and peak LH less than 5.8 IU/liter after GnRH infusion had high specificity (1.00; 95% confidence interval = 0.59-1.00) and diagnostic efficiency (76.9%) for HH.
Conclusions: Basal FSH less than 1.2 IU/liter confirms HH, which precludes from further testing, reducing patient discomfort and healthcare system costs. In patients with basal FSH of at least 1.2 IU/liter, a GnRH infusion test has a high diagnostic efficiency.