Context: Clinical use of single serum gonadotropin measurements in children is limited by thepulsatile secretion of follicle-stimulating hormone (FSH) and luteinizing hormone (LH). However,first morning voided (FMV) urine may integrate the fluctuating gonadotropin serum levels.
Objective: We aimed to evaluate urinary and serum gonadotropin levels according to age, sex, andpubertal stage in healthy children and to assess the clinical use of FMV urinary gonadotropins inchildren with disordered puberty.
Design: Cross-sectional part of the COPENHAGEN Puberty Study and longitudinal study of patients.
Setting: Population-based and outpatient clinic.
Patients or Other Participants: Eight hundred forty-three healthy children from the COPENHAGENPuberty Study and 25 girls evaluated for central precocious puberty (CPP).
Main Outcome Measures: Clinical pubertal staging, including serum and urinary gonadotropin levels.
Results: Urinary gonadotropins increased with advancing age and pubertal development and weredetectable in FMV urine before physical signs of puberty. FMV urinary LH correlated strongly withbasal (r = 0.871, P , 0.001) and gonadotropin-releasing hormone (GnRH)-stimulated serum LH(r = 0.82, P , 0.001). Urinary LH was superior to urinary FSH in differentiating the pubertal stage.Receiver operating curve analysis revealed that a cut-off standard deviation (SD) score of 2 forurinary LH (IU/L) gave a sensitivity of 75% and a specificity of 92% in predicting a positive GnRHstimulation test (LHmax . 5 IU/L). Urinary concentrations of LH decreased after 3 months of GnRHtreatment to levels below +2 SDs.
Conclusions: Urinary gonadotropin levels increased before the onset of puberty and were elevatedin girls with CPP. We suggest urinary LH as an alternative noninvasive method to improve diagnosingand therapeutic management of children with disordered puberty.