ESPE Abstracts

Background and Aims: Central Precocious Puberty (CPP) caused by the early activation of the hypothalamic-pituitary-gonadal axis is a rare affection that occurs in 0.2% in girls. Clinical signs are suggestive and LHRH-test could be sometimes difficult to use and not always contributive. Gonadotropin assay in twenty-four hours-urinary collection could be an interesting option. The present work intended to evaluate the diagnostic interest of 24-hour urinary gonadotropin assay (FSHU, LHU) in the early pubertal onset in girls.

Methods: Urinary samples of >4 year old girls were addressed in Hormonology laboratory of Lyon University Hospital from October 2012 to July 2015 with puberty stage data for urinary FSH and LH dosages (setting cohort). We excluded incomplete urine collection, uncertain puberty stages, and interfering endocrine diseases further diagnosed (congenital adrenal hyperplasia, peripheral puberty onset, syndromic pathologies …). Distribution of values regarding Tanner stages were analyzed to propose thresholds to confirm puberty onset. We tested thus established cut-offs values in a second monocentric cohort of girls (validation cohort) consulting in Lyon University Hospital, from January 2021 to December 2022, who met the following criteria: (1) addressed for suspicion of precocious puberty; (2) ≤8 year old; (3) had pelvic ultrasonography (US) when stage of breast development was S2 or more. Pelvic US classified patients with estrogenic signs of uterine impregnation (S2+US+) or without (S2+US-). Other exclusion criteria were identical than in the setting cohort. FSH and LH were always measured with Abbott Laboratories® reagents (FSH: ref. 7K75. LH: ref. 2P40) on Architect i2000.

Results and Perspectives: Final numbers of included urine samples were 632 (455 girls) and 43 (33 girls) in setting and validation cohorts respectively. FSHU and LHU were significantly different in the setting cohort regarding Tanner stage (P<0.0001). Areas under ROC curves were 0,709 (0,662-0,755), 0,767 (0,725-0,809) and 0,753 (0,710-0,796) for FSHU, LHU and LHU/FSHU ratio respectively. Best thresholds were chosen using Youden’s Index and minimum sensitivity of 90% for FSHU and minimum specificity of 90% for LHU. In the validation cohort, combined positive FSHU (>1.1U/24h) and LHU (>0.08U/24h) detected S2+US+ girls with 83% prognostic value. Urines under one or the other threshold (FSHU<1.1U/24h or LHU<0.08U/24h) detected S1 and S2+US- girls with 93.5% prognostic value. These results deserves large-scale reproduction in greater cohorts, but show the usefulness of urinary gonadotropins for girls presenting with CPP suspicion as it is simple, non-invasive and overcomes the limits of circadian variations.