Abstract: Background：The 2019 international consensus on central precocious puberty (CPP) proposed that random serum luteinizing hormone (LH) has important reference value for the confirmation or exclusion of CPP, but there is no corresponding diagnostic criteria in China. Objective：By verifying the diagnostic value of the cutoff value of serum LH basal value proposed in the international consensus in 2019 for CPP, it is expected to reduce unnecessary GnRH stimulation tests and provide guidance for clinical practice. Design：Diagnostic accuracy study. Methods：According to the Chinese Expert Consensus on the Diagnosis and Treatment of Central Precocious Puberty (2022 edition) (referred to as the Chinese consensus), the recommended value of LH (chemiluminescence method) (peak LH ≥5 IU·L-1and peak LH /FSH ≥0.6 combined with clinical) was used as the gold standard for the diagnosis of CPP. According to the two cutoff values (0.83 and 0.20 IU·L-1) recommended by the international consensus, girls aged 49 years who met the diagnostic criteria of precocity, did not have menarche, had breast development Tanner stage Ⅲ or below, and underwent GnRH stimulation test were enrolled. Peripheral precocity was excluded. The clinical data of age, height, weight, body mass index (BMI), Tanner stage, gonadal axis hormones and sex hormones, bone age (TW3), pituitary MR and uterine and ovarian ultrasound were collected. Main outcome measures：Sensitivity and specificity of different basic luteinizing hormone cutoff values for diagnosing CPP. Results：A total of 352 girls with precocious puberty were included in the analysis. Among them, 203 cases were diagnosed as CPP and 149 cases were diagnosed as PT according to the gold standard. The average age of precocious puberty was (7.3±1.0) years, and the average height was (130.6±10.3) cm. The CPP girls had significantly higher mean age, height and bone age, and baseline serum LH, FSH, LH/FSH, and E2 levels than the PT girls. The differences were statistically significant. The specificity of baseline serum LH value ≥0.83 IU·L-1 (100%, 95%CI:97%100%) was better than that of baseline serum LH value <0.20 IU·L-1(45%, 95%CI:37%53%) and from 0.20 to 0.82 IU·L-1 (56%, 95%CI:47%63%). The CPP girls with a baseline LH level of ≥0.83 IU·L-1 had significantly higher breast Tanner stage, bone age, and bone agechronological age difference than those with a baseline LH level of <0.83 IU·L-1 (P<0.05), and a significantly lower pituitary height than those with a baseline LH level of < 0.83 IU·L-1 (P<0.05). Conclusions：When chemiluminescence method is used to detect gonadal hormone in children with precocious puberty, CPP can be diagnosed with serum LH base value ≥ 0.83 IU·L-1, and GnRH excitation test is not required LH <0.20 IU·L-1 cannot be used as the exclusion criteria for CPP, and sexual GnRH excitation test should be selected to assist diagnosis based on clinical characteristics.

Key words: Central precocious puberty, Luteinizing hormone, Gonadotropin-releasing hormone stimulation test, Precocious puberty