At present, GnRH stimulation is the gold standard for the diagnosis of CPP, and bone age and ultrasound examination of uterus and ovary are auxiliary means. INHB and AMH are glycoproteins of transforming growth factor – β family, both of which are produced by ovarian granulosa cells. AMH gene is located in the short arm of chromosome 19 (19p13.3) and only expressed in gonad. However, AMH is not considered to be a gonadotropin and produced by preantral follicles in granulosa cells. The cells around the follicles are recruited from the primordial follicle pool, but some of them are not selected as dominant antral follicles . In women, it takes 36 weeks of gestation for fetal ovarian granulosa cells to begin to produce AMH. AMH inhibits the transformation of primordial follicles to primary follicles and maintains primordial follicles at rest . AMH is not expressed in primordial follicles, but can be slightly expressed in granulosa cells of primary follicles. The expression of AMH is the highest in preantral follicles, primary follicles and granulosa cells of antral follicles less than 4 mm in diameter, while the expression of larger antral follicles gradually decreases. About 60% of AMH in human serum is produced and secreted by granulosa cells of follicles with a diameter of 5-8mm . The expression of AMH is limited to primary follicles and preantral follicles, but it is not produced at the higher stage of follicular development . AMH not only inhibits the initial regeneration of follicles, but also inhibits FSH dependent follicular growth  . AMH in female ovaries has the functions of regulating initial recruitment and circulating recruitment, inhibiting excessive consumption of primordial follicle pool and formation of dominant follicles .
Mini puberty begins in the first few months of a girl's life. The level of AMH in infants increases significantly from birth to 3 months, suggesting that the hypothalamus pituitary ovary axis is temporarily activated . The increase of AMH level in puberty may be the response of ovary to FSH induced follicles. In our study, the elevated AMH levels in patients with PT and Tanner stage 2 CPP may be an ovarian response to prevent FSH induced follicle growth and premature activation. Former studies with small samples have found that there is a weak negative correlation between the levels of AMH and FSH in normal girls before puberty  , but no correlation was found in patients with precocious puberty in our study. There was no correlation between AMH and basal LH in healthy women aged 18–24 . AMH and basal LH showed no correlation with each other in CPP group and positively correlated in PT group in our study. Another explanation for AMH variation in healthy female infants may be related to birth weight , and there is a negative correlation between BMI and AMH level in adult obese women . There are few studies on the relationship between AMH, INHB and BMI in precocious puberty children. There is no correlation between BMI and AMH or between BMI and INHB in our study, which might be due to the relatively normal weight of patients. The serum AMH levels of healthy girls vary greatly among individuals . Hagen et al. have shown that AMH levels increase by 17% in the first three years before puberty, after the onset of puberty, AMH levels decreased by 30% in the first two years . Previous study showed that compared with the control group before puberty, the level of AMH in the PT group was significantly higher, and the level of AMH in the CPP group was similar to that in the PT group, but the sample was very small, and no statistical difference was found . Part of the former researches selected CPP girls with Tanner stage 2 , or CPP patients were composed of Tanner stage 2–4 mixture . Those researches could not explain the impact of Tanner stage on the AMH level of CPP patients. Our study had a larger sample and the Tanner stage of CPP patients was well classified. We demonstrated that the AMH level of PT group was significantly higher than that of prepubertal control group, and the AMH level of CPP group was also higher than that of control group, which was mainly due to Tanner stage 2 CPP patients rather than Tanner stage 3 CPP patients. Because the ovary of girls with Tanner stage 3 CPP was in the transition stage from early puberty to late puberty, which reduces the serum AMH level. The AMH level of patients with Tanner stage 3 CPP in our study is similar to that of fast progressive CPP patients with Tanner stage 2 in former studies, because the AMH level of patients with Tanner stage 2 fast progressive CPP is lower than that of patients with slow progressive CPP. The level of AMH decreased with the maturation of hypothalamic pituitary gonadal axis, which may be a gradual process of precocious puberty. The AMH of patients with Tanner stage 3 CPP was significantly lower than that of patients with Tanner stage 2 CPP, indicating that with the further maturity of puberty, the AMH increased from early adolescence, and the peak level of AMH might be in Tanner stage 2, then declined to a relatively stable state. The latest research also showed that AMH in CPP patients was the lowest in Tanner stage 3 , which was consistent with our research, and the change of AMH concentration was similar to a mathematical parabola.
INHB secretes from early antral follicles and dominant follicles by granules cells. The increase of serum INHB level in adolescence seems to be induced by FSH in early adolescence due to active follicular development. During puberty, the level of INHB increased from Tanner stage 1 to Tanner stage 3, increased sharply in Tanner stage 2, reached the peak level in the Tanner stage 3, and then decreased in the Tanner stage 4 of puberty . In our study, INHB was at the highest level in Tanner stage 2 CPP, and decreased significantly in Tanner stage 3 CPP, but it was still elevated compared with PT group and normal group. This suggested that the Tanner stage of peak level of INHB in CPP patients was earlier than that in girls with normal puberty. Previous studies chosed CPP girls in Tanner stage 2, without CPP patients in Tanner stage 3  , which easily ignored the influence of CPP in different Tanner stages on INHB level. INHB level might be particularly useful in early puberty. Due to elevated gonadotropin levels, INHB secretion appeared earlier and might be the first step in puberty before growth spurts and bone age. The INHB level of Tanner stage 2 CPP (74.06 pg/mL) was significantly higher than that of Tanner stage 3 CPP (49.13 pg/mL) and normal control group (37.41 pg/mL). Although there was a rise in the precocious girls, INHB is not suitable to distinguish PT group from CPP group, especially between PT group and Tanner stage 3 CPP group. Previous research also showed that INHB was inferior to basal LH in predicting the peak level of LH . Oestradiol and INHB are secreted by antral follicles stimulated by FSH, and the secretion of FSH is inhibited by the negative feedback of gonadal axis. Therefore, the consumption of follicular pool leads to the increase of FSH, the decrease of oestradiol and the decrease of INHB. These three indicators are interdependent rather than independent, and the changes occur relatively late. This suggested that the secretion of INHB was related to the activation of hypothalamus pituitary ovary axis, which might help to determine the early onset of CPP and observe the therapeutic effect of GnRHa. Previous studies showed that INHB level decreased significantly after GnRHa treatment which also reduced AMH level, but returned to the level before treatment after stopping treatment, while changes in AMH levels were not associated with changes in gonadotropin during or after treatment  . AMH and INHB in patients with precocious puberty were positively correlated in our study, which was consistent with previous study . But AMH or INHB showed no correlation with other indicators of ovarian function, including uterine volume, ovarian volume, E2 and FSH. The clinical significance of AMH and INHB in precocious puberty needs further verification.