Herbal Medicine for Idiopathic Central Precocious Puberty: A Systematic Review and Meta-Analysis

Abstract

Introduction:

Herbal medicine (HM) is widely used in East Asia to treat idiopathic central precocious puberty (ICPP). However, no systematic reviews of the efficacy and safety of HM for treatment of ICPP have been reported. Thus, this systematic review assessed the efficacy and safety of HM for ICPP. Most available clinical trials that investigated HM for ICPP were included in this review.

Methods and analysis:

Eleven databases, including Asian databases, were searched from the date of inception until June 2018. We included randomized controlled trials (RCTs) that assessed HM for ICPP.

Results:

Nine RCTs including 650 participants with ICPP, were included in this review. As a result of meta-analysis, response rate of HM group was 1.14 times higher compared with control group, with moderate level of evidence. Also, serum estradiol (E2) level was significantly lower in HM group compared to control group with moderate level of evidence. Notably, serum luteinizing hormone (LH) level (IU/L) and E2 level were significantly lower in HM group than triptorelin injection group with moderate level of evidence. Bone age index and uterine volume were also significantly lower in HM group than control group with very low level of evidence. The incidence of adverse events was lower in the HM group than control group, but it was not statistically significant.

Conclusions:

According to the results of the meta-analysis, it seems that HM treatment has a significant effect on the treatment of ICPP and can effectively reduce serum LH, E2 level, and bone age. In particular, as gonadotropin-releasing hormone analog therapy has a negative effect on the growth rate and final height if bone age ≥13 years, this study suggests that HM treatment may be helpful. Additional higher quality RCTs are needed to demonstrate the efficacy and safety of HM for treatment of ICPP.

Introduction

Central precocious puberty (CPP, also known as gonadotropin-releasing hormone [GnRH]-dependent precocious puberty) is defined as premature activation of the hypothalamic-pituitary-gonadal axis and appearance of secondary sex characteristics before the age of 8 years in girls and 9 years in boys.¹ CPP without an identifiable cause is known as idiopathic central precocious puberty (ICPP). The prevalence of CPP is much higher in girls, comprising 90% of all reported cases of ICPP²; the prevalence of CPP among Korean girls has increased markedly, from 3.3 per 100,000 girls in 2004 to 50.4 per 100,000 girls in 2010.³ Early secretion of estradiol (E2) accelerates growth and bone age, resulting in shortening of the growth period and reduction of adult height. CPP can cause psychological problems and early menarche in girls,⁴ and can increase the risk of metabolic and cardiovascular diseases in both boys and girls.^5,6

Currently, the most common treatment for CPP is GnRH analog therapy; such analogs inhibit the progression of puberty by blocking secretion of gonadotropin and the production of sex steroids.⁷ However, there have been reports of side effects. If the bone age is ≥13 years, GnRH analog therapy may negatively impact growth velocity and final height, and may lead to weight gain and reduced bone mineral density during treatment.⁸ Furthermore, adverse effects of GnRH analog therapy may include hypertriglyceridemia and increased low-density lipoprotein level,⁹ hyperprolactinemia,¹⁰ vaginal bleeding,¹¹ and skin reactions at the injection site.¹² In addition, many children with ICPP complain of fear and pain of injection treatment.¹³

Recent studies have suggested that GnRH analog therapy may not result in these side effects and is safe for long-term use,^7,14,15 but considerable controversy remains regarding its safety and further research is needed.⁹ In Korea, GnRH analog injection treatment, which is a typical treatment for ICPP, is not covered by health insurance in most cases, thus incurs a cost burden.¹⁶ Furthermore, according to one study conducted in Korea, it was found that 43.4% of children treated with GnRH analog therapy for ICPP also used growth hormone to compensate for the decrease in the growth rate due to GnRH analog therapy, causing increase in treatment cost, which is about 3,000,000 to 30,000,000 won per year.¹⁷ Because of concerns regarding safety and cost burden, the demand for complementary and alternative medicine approaches to treat ICPP has increased in East Asia.^18,19

Herbal medicine (HM) is widely used for the treatment of ICPP in Asian countries,^16,17 specifically to delay pubertal development and promote growth in affected patients.²⁰ In Traditional Chinese Medicine (TCM), ICPP is considered to be caused by the broken balance of yin and yang, and is treated by classifying it in to three main TCM patterns: hyperactivity of fire due to yin deficiency, liver-qi stagnation transforming into fire, and phlegm stagnation due to spleen deficiency.²¹ In previous studies, HM suppressed the hypothalamic-pituitary-gonadal axis and inhibited the synthesis and release of GnRH in an animal model,²² and has been reported to delay sexual development in animal studies.^23,24 In addition, several studies have reported that HM regulates the function of the hypothalamic-pituitary-ovarian axis and delays skeletal development and maturation in idiopathic precocious puberty patients.^25,26

Although several studies have been performed regarding the effects of HM for treatment of ICPP, there have been no systematic reviews of this therapeutic approach. Thus, this study aimed to assess the efficacy and safety of HM for the treatment of ICPP.

Methods

Study registration

The protocol for this systematic review has been registered in PROSPERO 2018 (No. CRD42018087988).

Data sources

The following databases were searched from their dates of inception until June 2018: MEDLINE, Embase, the Cochrane Central Register of Controlled Trials (CENTRAL), AMED, and CINAHL. We also searched six Korean medical databases (OASIS, the Korean Traditional Knowledge Portal, the Korean Studies Information Service System, KoreaMed, the Korean Medical Database, and DBPIA) and three Chinese databases: CNKI (containing data from the China Academic Journal, the China Doctoral Dissertations and Masters' Theses Full-text Database, the China Proceedings of Conference Full-Text Database, and the Century Journal Project), Wanfang, and VIP. In addition, we conducted nonelectronic searches of conference proceedings, our own article files, and nine traditional Korean medicine journals. Study selection was documented and summarized in a PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis)-compliant flow chart (Fig. 1). Additional search was conducted in March 2020, but no studies matching the criteria were found.

FIG. 1.

Flowchart depicting the results of the literature search.

Types of studies

Prospective randomized controlled trials (RCTs) and quasi-RCTs that evaluated the efficacy of HM for ICPP were included in this review. Both treatment with HM alone and concurrent treatment with HM and another therapy were considered acceptable if HM was administered solely to the intervention group and any other treatment was administered equally to both groups. Trials with any type of control intervention were included. There were no restrictions regarding publication language. Hard copies of all articles were obtained and read in full.

Types of participants

This study included patients who were diagnosed with ICPP. We excluded patients with CPP secondary to congenital hypothyroidism, germ cell tumors, various types of gonadal tumors, or organic diseases (e.g., congenital adrenal hyperplasia, adrenal tumors, various intracranial tumors, or central nervous system diseases). Due to the outcome determinations, only RCTs involving girls were included.

Types of interventions

Interventions involving any formulation (i.e., decoction, tablets, capsules, pills, powders, and extracts) of HM were included. This study included only previous studies using HM prescribed by traditional East Asian medicine doctors, and the compositions of all interventions were reviewed.

Data extraction and quality assessment

Hard copies of all articles were obtained and read in full. Two authors (J.A.L. and Y.B.L.) extracted the data and assessed the quality using a predefined data extraction form. The risk of bias was evaluated using the Cochrane risk of bias assessment tool, version 5.1.0, which considers random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, completeness of outcome data, selective reporting, and other sources of bias.²⁷ The results of such evaluations were presented by utilizing scores of “L,” “U,” and “H” to indicate a low risk of bias, an uncertain risk of bias, and a high risk of bias, respectively. Disagreements were resolved by discussion among all authors. When disagreements regarding selection could not be resolved through discussion, an arbiter (L.H.L.) made the final decision.

Data collection and synthesis

Outcome measures

Primary outcomes

Response rate

Hormone levels (follicle-stimulating hormone [FSH], luteinizing hormone [LH], and E2).

Secondary outcomes

Ovarian volume and uterine volume

Adverse events

Bone age

Data synthesis

Differences between the intervention and control groups were assessed. The mean difference (MD) and 95% confidence interval (CI) were used to measure the effects of treatment for continuous data. We converted other forms of data into the MD. For outcome variables with different scales, we used the standard MD and 95% CI. For dichotomous data, we presented treatment effects as the relative risk (RR) and 95% CI; other binary data were converted into the RR.

All statistical analyses were conducted using the Cochrane Collaboration's software program Review Manager (RevMan) version 5.3 for Windows (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark). When missing information was encountered in a study, we contacted the corresponding author for clarification whenever possible. As appropriate, we pooled data across studies to conduct a meta-analysis using fixed- or random-effects models. We used GRADEpro software (McMaster University/Evidence Prime, Inc., Hamilton, Canada) to create a summary of findings table.

Results

Study search results

A total of 268 studies were selected through electronic search; all were screened, and there were no duplicates. The literature search identified 48 potentially relevant studies, 9 of which were included (Fig. 1). The main characteristics of the nine included studies are summarized in Table 1.^{28

–36} All included RCTs were performed in China and were published between 2001 and 2016.

Table 1.

Characteristics of Studies Included in the Meta-Analysis

First author (year); country	Disease	Sample size (I/C); Mean age ± SD (years)	Intervention group	Control group	Course of treatment	Follow-up	Main outcomes	Result	Adverse event (group: frequency)
HM group vs. Megestrol
Cai (2001)²⁸; China	IPP	86 (51/35); Tanner stage III; (A) 5.6 ± 1.6 (B) 5.3 ± 1.4 Tanner stage IV; (A) 6.5 ± 1.8 (B) 6.8 ± 1.7 Tanner stage V; (A) 7.4 ± 0.7 (B) 7.4 ± 0.6	(A) ZiYin Xiehuo decoction 30 mL^*2 times/day+(B)	(B) WM (Megestrol acetate tablets Tanner stage III; 6 mg/3 times/day Tanner stage IV; 7 mg/3 times/day Tanner stage V; 8 mg/3 times/day)	1–3.5 years	n.r	1. LH (IU/L) 2. Uterine volume (mL) 3. Ovary volume (mL) 4. BAI (ΔBA/ΔCA) 5. Breast nodule unit 6. Predicted final height (cm)	1. After LHRH stimulating test 15 min; (A) 10.4 ± 8.6 (B) 14.1 ± 13.6 p > 0.05 30 min; (A) 12.2 ± 9.3 (B) 15.7 ± 14.8 p > 0.05 60 min; (A) 10.1 ± 7.1 (B) 12.6 ± 8.3 p > 0.05 90 min; (A) 11.7 ± 8.6 (B) 13.4 ± 10.1 p > 0.05 2. Tanner stage III; (A) 2.6 ± 1.7 (B) 2.8 ± 1.4 p > 0.05 Tanner stage IV; (A) 9.5 ± 3.2 (B) 9.3 ± 2.7 p > 0.05 Tanner stage V; (A) 15 ± 6.2 (B) 15.2 ± 5.1 p > 0.05 3. Tanner stage III; (A) 1.2 ± 0.4 (B) 1.2 ± 0.3 p > 0.05	n.r
								Tanner stage IV;(A) 1.8 ± 0.3(B) 1.8 ± 0.3 p > 0.05Tanner stage V;(A) 2.3 ± 1(B) 2.4 ± 0.8 p > 0.054.(A) 0.65 ± 0.36(B) 1.43 ± 0.35 p < 0.00015.Tanner stage III;(A) 3.4 ± 2(B) 3.6 ± 2.2 p > 0.05Tanner stage IV;(A) 5.6 ± 2.4(B) 5.8 ± 2 p > 0.05Tanner stage V;(A) 10.3 ± 3.1(B) 10.6 ± 2.5 p > 0.056.(A) 158.5 ± 4.3(B) 151.5 ± 3.5 p < 0.00001
Yang (2015)²⁹; China	ICPP	62 (32/30)(A) 7.09 ± 0.76(B) 7.48 ± 0.93	(A)TCM syndrome differentiation treatment (decoction)+(B)	(B) WM (Megestrol acetate tabletsTanner stage B2; 6 mg/3 times/dayTanner stage B3; 7 mg/3 times/dayTanner stage B4; 8 mg/3 times/day)	6 months	6 months	1. Response rate (total/treated)2. LH (IU/L)3. FSH (IU/L)4. E2 (pg/mL)5. Uterine volume (mL)6. Ovary volume (mL)7. Ovary diameter (mm)8. BAI (ΔBA/ΔCA)	1.A vs. B, RR 1.28 (95% CI: 1.01–1.61)(A) 30/32(B) 22/30 p < 0.052.(A) 4.28 ± 1.41(B) 6.33 ± 2.53 p < 0.00013.(A) 5.43 ± 2.18(B) 7.21 ± 2.86 p < 0.054.(A) 18.47 ± 9.29(B) 23.22 ± 8.21 p < 0.055.(A) 2.18 ± 0.83(B) 2.25 ± 0.87 p > 0.056.(A) 1.47 ± 0.89(B) 1.77 ± 0.74 p > 0.057.(A) 3.57 ± 1.19(B) 4.06 ± 1.21 p > 0.058.(A) 0.69 ± 0.23(B) 1.29 ± 0.18 p < 0.00001	Mild diarrhea(A:2)Pain at the injection point(A:3)(B:4)
HM group vs. Triptorelin
Cui (2015)³⁰; China	ICPP	51 (27/24)6.9 ± 1.7	(A) Modified Zhibai Dihuang decoctionn.r^2 times/day +Acupoint sticking therapyn.r min^n.r time/day	(B) WM (triptorelin acetate subcutaneous injection60–100 μg/kg [maximum 3.75 mg]^*1 time/4 weeks)	6 months		1. Response rate (total/treated)2. LH (mIU/mL)3. FSH (mIU/mL)4. E2 (pg/mL)5. IGF-1 (ng/mL)6. Uterine volume (mL)7. Ovary volume (mL)	1.A vs. B, RR 0.98 (95% CI: 0.74–1.31)(A) 21/27(B) 19/24 p > 0.052.(A) 1.52 ± 0.82(B) 1.33 ± 1.06 p > 0.053.(A) 3.62 ± 2.19(B) 2.35 ± 1.77 p = 0.024.(A) 21.45 ± 9.73(B) 23.21 ± 11.24 p > 0.055.(A) 177 ± 72(B) 158 ± 61 p > 0.056.(A) 2.07 ± 0.66(B) 2.17 ± 0.82 p > 0.057.(A) 1.12 ± 0.43(B) 1.09 ± 0.52 p > 0.05	n.r
Huang (2015)³¹; China	ICPP	80 (40/40)(A) 7.06 ± 0.78(B) 7.13 ± 0.35	(A) Kangzao granules50 mL^*2 times/day	(B) WM (triptorelin intramuscular injection80–100 μg/kg^*1 time/4 weeks)	6 months		1. Response rate (total/treated)2. LH (mIU/mL)3. FSH (mIU/mL)4. E2 (ng/L)5. Uterine volume (mL)6. Ovary volume (mL)7. BAI (ΔBA/ΔCA)8. Breast nodule diameter (cm)9. Weight (kg)10. Height (cm)11. BMI (kg/m²)	1.A vs. B, RR 0.97 (95% CI: 0.73–1.28)(A) 28/40(B) 29/40 p > 0.052.(A) 2.79 ± 1.38(B) 3.42 ± 1.32 p = 0.043.(A) 3.42 ± 1.32(B) 3.76 ± 0.67 p > 0.054.(A) 18.2 ± 11.17(B) 19.3 ± 10.17 p > 0.055.(A) 2.51 ± 1.03(B) 2.56 ± 1.08 p > 0.056.(A) 2.03 ± 1.15(B) 1.85 ± 0.96 p > 0.057.(A) 1.12 ± 0.39(B) 1.13 ± 0.28 p > 0.058.(A) 1.28 ± 1.21(B) 1.14 ± 0.86 p > 0.059.(A) 29.14 ± 6.02(B) 31.4 ± 6.39 p > 0.05	None
								10.(A) 132.96 ± 8.11(B) 132.91 ± 8.61 p > 0.0511.(A) 16.06 ± 2.06(B) 17.63 ± 2.27 p = 0.001
Gan (2016)³²; China	ICPP	83 (45/38)(A) 8.11 ± 1.97(B) 8.02 ± 2.01	(A)Dabuyin Pills6 g^*2–3 times/day+(B)	(B) WM (triptorelin acetate intramuscular injectionFirst; 3.75 mg;Later;80–100 μg/kg^*1 time/4 weeks)	1 year	1 year	1. Response rate (total/treated)2. LH (IU/L)3. FSH (IU/L)4. E2 (pmol/L)5. Uterine volume (mL)6. Ovary volume (mL)7. BAI (ΔBA/ΔCA)8. Growth rate (cm/year)	1.A vs. B, RR 1.25 (95% CI: 1.00–1.57)(A) 40/45(B) 27/38 p = 0.052.(A) 2.5 ± 0.3(B) 3.7 ± 0.5 p < 0.000013.(A) 3.7 ± 0.5(B) 4.9 ± 0.6 p < 0.000014.(A) 85.1 ± 8.9(B) 101.3 ± 9.7 p < 0.000015.(A) 2.3 ± 1.1(B) 3.4 ± 1.4 p < 0.00016.(A) 1.4 ± 0.4(B) 1.1 ± 0.3 p < 0.00017.(A) 0.9 ± 0.1(B) 1.1 ± 0.3 p < 0.0001	None
								8.(A) 4.96 ± 1.34(B) 7.15 ± 1.75 p < 0.00001
Qi (2016)³³; China	ICPP	72 (36/36)(A) 6.80 ± 0.32(B) 7.10 ± 0.45	(A) modified Dabuyin Wan decoction150 mL^2 times/day for 2 months+triptorelin injection80–100 μg/kg^1 time/4 weeks for 4 months80–100 μg/kg^1 time/4–5 weeks for 6 months80–100 μg/kg^1 time/6 weeks for 4 months+Dabuyin Wan addition and subtraction decoction150 mL^*2 times/day for 2 months	(B) WM (triptorelin injection80–100 μg/kg^1 time/4 weeks when gonad is controlled;50–80 μg/kg^1 time/4 weeks	1.5 years	1.5 year	1. Response rate (total/treated)2. LH (IU/L)3. FSH (IU/L)4. E2 (lU/L)5. Uterine volume (mL)6. Ovary volume (mL)7. Ovary diameter (mm)8. BAI (ΔBA/ΔCA)9. Treatment cost (yuan)	1. A vs. B, RR 1.04(95% CI: 0.82–1.31)(A) 29/36(B) 28/36 p > 0.052.(A) 6.07 ± 2.36(B) 5.96 ± 4.45 p > 0.053.(A) 6.1 ± 3.43(B) 5.85 ± 2.77 p > 0.054.(A) 19.75 ± 11.21(B) 22.31 ± 10.72 p > 0.055.(A) 2.22 ± 0.92(B) 2.09 ± 1.07 p > 0.056.(A) 1.54 ± 1(B) 1.75 ± 0.92 p > 0.057.(A) 4.17 ± 2.29(B) 3.57 ± 2.33 p > 0.058.(A) 0.72 ± 0.21(B) 0.67 ± 0.27 p > 0.05	Transient vaginal bleeding(B:5)
								9.(A) 22340 ± 4795.6(B) 34794.67 ± 4227.15 p < 0.00001
Zhang (2016)³⁴; China	IPP	70 (35/35)(A) 6.79 ± 0.11(B) 6.78 ± 0.12	(A) Dabuyin pill6 g^*2–3 times/day	(B) WM (triptorelin acetate intramuscular injectionFirst; 3.75 mgLater;80–100 μg/kg^*1 time/4 weeks)	6 months	6 months	1. Response rate (total/treated)2. LH (IU/L)3. FSH (IU/L)4. E2 (pmol/L)5. Uterine volume (mL)6. Ovary volume (mL)7. BAI (ΔBA/ΔCA)	1. A vs. B, RR 1.24 (95% CI: 0.97–1.58)(A) 31/35(B) 25/35 p > 0.052.(A) 2.4 ± 0.2(B) 3.7 ± 0.6 p < 0.000013.(A) 3.2 ± 0.3(B) 4.9 ± 0.5 p < 0.000014.(A) 85.1 ± 8.3(B) 101.2 ± 10.8 p < 0.000015.(A) 2.2 ± 1(B) 3.3 ± 1.4 p = 0.00026.(A) 1.1 ± 0.1(B) 1.3 ± 0.3 p = 0.00027.(A) 3.57 ± 1.19(B) 4.06 ± 1.21 p > 0.058.(A) 0.8 ± 0.1(B) 1.1 ± 0.2 p < 0.00001	None
HM group vs. Leuprolide
Yang (2013)³⁵; China	ICPP	46 (25/21)(A) 7.35 ± 1.36(B) 7.60 ± 1.73	(A) 栀早颗粒 granules4–7 years; 1/3ch^2 times/day7–10 years; 1/2ch^2 times/day	(B) WM (leuprolide acetate intramuscular injectionFirst; 100 μg/kg (maximum 3.75 mg)Later;50–100 μg/kg^*1 time/4 weeks)	6 months		1. E2 (pmol/L)2. BGP (pmol/L)3. BAI (ΔBA/ΔCA)	1.(A) 70.73 ± 80(B) 76.35 ± 84.6 p > 0.052.(A) 13.12 ± 1.93(B) 13.17 ± 2.14 p > 0.053.(A) 1.14 ± 0.1(B) 1.14 ± 0.12 p > 0.05	n.r
HM group vs. No treatment
Cai (2001)²⁸; China	IPP	71 (51/20);Tanner stage III;(A) 5.6 ± 1.6(B) 5.5 ± 1.9Tanner stage IV;(A) 6.5 ± 1.8(B) 6.7 ± 1.3Tanner stage V;(A) 7.4 ± 0.7(B) 6.7 ± 1.93	(A)Zi Yin Xiehuo decoction 30 mL^*2 times/day+WM (Megestrol acetate tabletsTanner stage III; 6 mg/3 times/dayTanner stage IV; 7 mg/3 times/dayTanner stage V; 8 mg/3 times/day)	(B)No treatment	1–3.5 years	n.r	1. Uterine volume (mL)2. Ovary volume (mL)3. BAI (ΔBA/ΔCA)4. Breast nodule unit5. Predicted final height (cm)	1.Tanner stage III;(A) 2.6 ± 1.7(B) 5.3 ± 1.2 p < 0.0001Tanner stage IV;(A) 9.5 ± 3.2(B) 18.4 ± 5.7 p = 0.0002Tanner stage V;(A) 15 ± 6.2(B) 31.2 ± 5.4 p < 0.000012.Tanner stage III;(A) 1.2 ± 0.4(B) 2.5 ± 0.7 p < 0.0001Tanner stage IV;(A) 1.8 ± 0.3(B) 3.2 ± 0.6 p < 0.00001	n.r
								Tanner stage V;(A) 2.3 ± 1(B) 5.1 ± 1.5 p < 0.000013.(A) 0.65 ± 0.36(B) 1.54 ± 0.27 p < 0.000014.Tanner stage III;(A) 3.4 ± 2(B) 14.2 ± 1.5 p < 0.00001Tanner stage IV;(A) 5.6 ± 2.4(B) 20.5 ± 2.8 p < 0.00001Tanner stage V;(A) 10.3 ± 3.1(B) 27.2 ± 2.4 p < 0.000015.(A) 158.5 ± 4.3(B) 149.3 ± 4.5 p < 0.00001
Wang (2012)³⁶; China	IPP	80 (40/40)(A) 6.35 ± 0.46(B) 6.78 ± 0.24	(A) Dabuyin pill 10 g^*3 times/day	(B)No treatment	1 year		1. LH (IU/L)2. FSH (IU/L)3. E2 (pmol/L)4. Uterine volume (mL)5. Ovary volume (mL)6. BAI	1.(A) 2.72 ± 0.55(B) 12.35 ± 0.48 p < 0.000012.(A) 3.46 ± 0.42(B) 9.48 ± 0.71 p < 0.000013.(A) 81.67 ± 8.94(B) 191.45 ± 18.65 p < 0.00001	None
								4.(A) 2.4 ± 1.6(B) 5.2 ± 1.3 p < 0.000015.(A) 1.1 ± 0.4(B) 2.7 ± 0.6 p < 0.000016.(A) 0.73 ± 0.06(B) 1.68 ± 0.31 p < 0.00001

BA, bone age; BAI, bone age index (ΔBA/ΔCA); BGP, osteocalcin; BMI, body mass index; CA, chronological age; CI, confidence interval; E2, estradiol; FSH, follicle-stimulating hormone; HM, herbal medicine; LH, luteinizing hormone; ICPP, idiopathic central precocious puberty; IPP, idiopathic precocious puberty; LHRH, luteinizing hormone releasing hormone; RR, risk ratio; TCM, Traditional Chinese Medicine; WM, Western medicine.

Risk of bias

Only two trials reported the use of a random sequence generation method.^29,33 Both trials used a randomized digital table method and were considered to exhibit a “low risk” of selection bias. The randomization procedure was not clearly stated in the other seven trials. No trials described allocation concealment or blinding of outcome assessment. In all trials, blinding of participants and personnel was incomplete. Because the control group did not receive placebo treatment, participants and researchers were able to identify the assigned intervention. There were no dropouts in any of the studies; thus, the risk of bias for incomplete outcome data was considered to be “low risk.” The risk of bias for selective reporting was “unclear” because none of the trials had published protocols. Three trials did not state whether baseline characteristics were comparable, so those studies were considered to have an “unclear” risk of other sources of bias (Supplementary Figure S1).

Effects of interventions

Images of meta-analysis results not shown in the text are shown in Supplementary Figure S2.

HM versus Megestrol

Two studies tested four different HM interventions compared with megestrol.^28,29 Both studies compared HM plus megestrol with megestrol alone.

Response rate

One study compared response rate between HM plus megestrol and megestrol alone, and reported response rate of HM plus megestrol was significantly higher (p < 0.05).²⁹ This study evaluated clinical efficacy in four stages; cured, significantly improved, improved, and ineffective. It was considered effective if breasts reduced to stage B1 or reduced after treatment, clinical symptoms disappeared or decreased, and laboratory indicators returned to normal or improved. Total response rate is defined as the proportion of participants who were cured, significantly improved, or improved after treatment.

Hormone levels (FSH, LH, and E2)

In one study, serum LH, FSH, and E2 level of HM plus megestrol and megestrol alone were compared, and all were significantly lower in HM plus megestrol group (p < 0.05).²⁹ In other study, only serum LH level (IU/L) after LH releasing hormone stimulating test was compared, but there was no significant difference between two groups (p > 0.05).²⁸

Uterine and ovarian volume

Both studies compared uterine and ovarian volume between HM plus megestrol and megestrol alone,^28,29 and one study compared according to Tanner stage.²⁸ Both studies showed no significant differences between two groups (p > 0.05).^19,20

Bone age index

Both studies compared bone age index (ΔBA/ΔCA) between HM plus megestrol and megestrol alone^28,29; meta-analysis revealed that bone age index (ΔBA/ΔCA) was significantly lower in HM plus megestrol group (n = 148, MD: −0.68, 95% CI: −0.86 to −0.51, p < 0.00001, I ² = 73%). Bone age index is defined as bone age divided by chronological age.

Other

In one study, predicted final height of HM plus megestrol group was significantly higher than control group (p < 0.05).²⁸

HM versus triptorelin (GnRH analog) injection

Five studies tested four different HM interventions compared with triptorelin.^30

–34 Two studies compared HM plus triptorelin injection with triptorelin injection alone^32,33; another two studies compared HM alone with triptorelin injection alone.^31,34 One study compared HM plus acupoint sticking with triptorelin injection alone.³⁰

Response rate

Five studies compared response rate between HM group and triptorelin injection group^30

–34; meta-analysis revealed that the response rate of HM group was higher than triptorelin injection-treated group, but this finding was not statistically significant (n = 356, RR: 1.11, 95% CI: 0.99–1.24, p = 0.08, I ² = 0%). Four studies evaluated clinical efficacy in three stages: cured or significantly improved, improved, and ineffective.^30,32
–34 And one study evaluated clinical efficacy in four stages: cured, significantly improved, improved, and ineffective.³¹ Evaluation criteria varied among studies, but were generally considered to be effective when breast reduces, clinical symptoms disappeared or decreased, and laboratory indicators returned to normal or improved. Total response rate is defined as the proportion of participants who were cured, significantly improved, or improved after treatment.

Hormone levels (LH, FSH, and E2)

All of five studies compared serum LH level between HM group and triptorelin injection group^30

–34; meta-analysis revealed that the serum LH level (IU/L) of HM group was significantly lower than triptorelin injection group (n = 356, MD: −0.74, 95% CI: −1.22 to −0.27, p = 0.002, I ² = 88%, Fig. 2).

FIG. 2.

Forest plot comparing LH level (IU/L) between HM group and triptorelin-treated group. HM, herbal medicine; LH, luteinizing hormone.

Five studies compared serum E2 level (pg/mL) between HM group and triptorelin injection group^30

–34; meta-analysis showed that serum E2 level was also significantly lower in HM group (n = 356, MD: −4.21, 95% CI: −5.00 to −3.43, p < 0.00001, I ² = 0%, Fig. 3).

FIG. 3.

Forest plot comparing E2 level (pg/mL) between HM group and triptorelin-treated group. E2, estradiol; HM, herbal medicine.

Five studies reported serum FSH level (IU/L)^30

–34; meta-analysis revealed that serum FSH level of HM group was lower than triptorelin injection group, but this finding was not statistically significant (n = 356, MD: −0.57, 95% CI: −1.26 to 0.12, p = 0.11, I ² = 93%).

Uterine and ovarian volume

Five studies compared uterine volume (mL) between HM group and triptorelin injection group^30

–34; meta-analysis revealed that the uterine volume (mL) of HM group was lower than triptorelin injection group, but not statistically significant (n = 356, MD: −0.42, 95% CI: −0.92 to 0.08, p = 0.10, I ² = 81%).

Five studies compared ovarian volume (mL) of HM group and triptorelin injection group^30

–34; meta-analysis showed that ovarian volume (mL) of HM group was higher than triptorelin injection group, but this finding was not statistically significant (n = 356, MD: 0.02, 95% CI: −0.25 to 0.29, p = 0.87, I ² = 87%).

Bone age index

Four studies compared bone age index (ΔBA/ΔCA) between HM group and triptorelin injection group^31

–34; meta-analysis revealed that the bone age index (ΔBA/ΔCA) of HM group was lower than triptorelin injection group, but this finding was not statistically significant (n = 305, MD: −0.09, 95% CI: −0.25 to 0.08, p = 0.30, I ² = 78%). Bone age index is defined as bone age divided by chronological age.

Other

In one study, body max index (BMI) was compared between HM and triptorelin injection, and BMI (kg/m³) of HM group was significantly lower than triptorelin injection group (p = 0.001).³¹ Another study compared growth rate between HM plus triptorelin injection and triptorelin alone, and it was significantly lower in HM plus triptorelin injection group (p < 0.00001).³² One study compared treatment cost between HM plus triptorelin injection and triptorelin injection alone, and it was significantly lower in HM plus triptorelin injection group (p < 0.00001).³³

HM versus leuprolide (GnRH analog) injection

Only one study compared HM with leuprolide injection.³⁵ This study compared serum E2 level (pg/mL), osteocalcin level (pmol/L), and bone age index (ΔBA/ΔCA) between HM and leuprolide injection, and there were no significant differences between two groups (p > 0.05).

HM versus no treatment

Two studies compared tested two different HM interventions compared with no treatment.^28,36 One study compared HM plus megestrol with no treatment,²⁸ and the other study compared HM alone with no treatment.³⁶

Response rate

Both studies did not report response rate.

Hormone levels (FSH, LH, and E2)

In one study, serum LH, FSH, and E2 level of HM alone and no treatment were compared, and all were significantly lower in HM group (p < 0.05).³⁶

Uterine and ovarian volume

Both studies compared uterine and ovarian volume between HM group and no treatment group,^28,36 and one study compared according to Tanner stage.²⁸ Both studies showed significantly smaller uterine and ovarian volume in HM group (p < 0.05).^28,36

Bone age index

Both studies compared bone age index (ΔBA/ΔCA) between HM alone and no treatment^28,36; meta-analysis revealed that bone age index (ΔBA/ΔCA) was significantly lower in HM group (n = 151, MD: −0.93, 95% CI: −1.02 to −0.85, p < 0.00001, I ² = 0%, Fig. 4). Bone age index is defined as bone age divided by chronological age.

FIG. 4.

Forest plot comparing bone age index (ΔBA/ΔCA) between HM group and No treatment group. HM, herbal medicine.

Other

In one study, predicted final height of HM plus megestrol group was significantly higher than control group (p < 0.05).²⁸

HM versus control group

Overall, nine studies tested nine different HM interventions compared with control group.^{28

–36}

Response rate

Six studies reported response rate^29

–34; meta-analysis of these six studies showed that the response rate was significantly higher in the HM group (n = 418, RR: 1.14, 95% CI: 1.03–1.26, p = 0.01, I ² = 2%, Fig. 5).

FIG. 5.

Forest plot comparing response rate between HM group and control group. HM, herbal medicine.

Hormone levels (LH, FSH, and E2)

Seven studies reported serum LH level (IU/L).^{29

–34,36}; meta-analysis showed that the LH level (IU/L) was lower in the HM group, but this result was not statistically significant (n = 498, MD: −2.09, 95% CI: −5.30 to 1.13, p = 0.20, I ² = 100%).

Seven studies compared serum FSH level (IU/L) between HM group and control group^{29

–34,36}; meta-analysis showed that the serum FSH level was lower in the HM group, but this finding was not statistically significant (n = 498, MD: −1.39, 95% CI: −3.21 to 0.42, p = 0.13, I ² = 99%).

Seven studies compared serum E2 level (pg/mL) between HM group and control group^29

–35; meta-analysis showed that the serum E2 level was significantly lower in the HM group (n = 464, MD: −4.22, 95% CI: −5.00 to −3.45, p < 0.00001, I ² = 0%).

Uterine and ovarian volume

Seven studies reported uterine volume (mL) after treatment^{29

–34,36}; meta-analysis showed that the uterine volume was significantly smaller in the HM group (n = 498, MD: −0.70, 95% CI: −1.37 to −0.04, p = 0.04, I ² = 92%).

Seven studies compared ovarian volume (mL) after treatment^{29

–34,36} between HM group and control group; meta-analysis showed that the ovarian volume (mL) was smaller in HM group, but this finding was not statistically significant (n = 498, MD: −0.26, 95% CI: −0.74 to 0.22, p = 0.29, I ² = 97%).

Bone age index (ΔBA/ΔCA)

Eight studies reported bone age index (ΔBA/ΔCA)^{28,29,31

–36}; meta-analysis showed that the bone age index (ΔBA/ΔCA) was significantly lower in the HM group (n = 579, MD: −0.35, 95% CI: −0.60 to −0.09, p = 0.007, I ² = 98%). Bone age index is defined as bone age divided by chronological age.

Adverse events

Six studies reported adverse events.^{29,31

–34,36} In one study comparing HM plus megestrol with megestrol alone, two cases of mild diarrhea and three cases of pain at the injection site were reported in the group receiving HM with oral megestrol acetate tablets; four cases of pain at the injection site were reported in the group taking oral megestrol acetate tablets alone.²⁹ The difference between groups was not statistically significant (n = 62, RR: 1.17, 95% CI: 0.35–3.96, p = 0.80, one trial). Among five studies that reported adverse events comparing HM group with triptorelin injection, four of these revealed that there were no side effects in both groups^31,32,34,36; and in one study, five cases of mild vaginal bleeding were reported in the group treated with triptorelin (GnRH analog) injection.³³ The difference between groups was not statistically significant (n = 72, RR: 0.09, 95% CI: 0.01–1.59, p = 0.10, one trial).

Details of HM approaches and formulations

Details regarding the HM approaches and formulations in the included studies are presented in Table 2. Only two studies reported TCM patterns.^29,35 In one study, participants were divided into three TCM patterns—hyperactivity of fire due to yin deficiency, liver-qi stagnation transforming into fire, and phlegm stagnation due to spleen deficiency—and were administered different HM interventions.²⁹ In another study, only people with liver depression and phlegm stagnation were recruited.³⁵ Regarding the preparation form, four studies used decoction,^28
–30,33 two studies used granules,^31,35 and three studies used pills.^32,34,36 In seven studies, a detailed composition of the HM was provided^{28

–31,33,35}; all seven studies reported different compositions. The Dabuyin pill was used in all three studies involving pill usage.^32,34,36 There were six studies describing the dose of HM,^{28,31,33

–36} and one study applied different doses according to age.³⁵ In five studies, participants took HM twice a day,^{28,30,31,33,35} in one study, participants took HM two to three times a day,³⁴ and in one study, participants took HM three times a day.³⁶ Duration of taking HM was provided in all nine studies, and it varied from 4 months to 3.5 years.^{28

–36}

Table 2.

Details of Herbal Medicine in the Included Studies

First author (year)	TCM pattern	HM	Composition	Preparation form	Dose and treatment period
HM group vs. Megestrol
Cai (2001)²⁸	n.r	ZiYin Xiehuo	Rehmanniae Radix 15 g	Decoction	30 mL^*2 times/day for 1–3.5 years
			Chinemydis Carapax 12 g
			Phellodendri Cortex 12 g
			Anemarrhenae Rhizoma 12 g
			Scrophulariae Radix 12 g
			Moutan Cortex 9 g
			Gentianae Radix 6 g
			Crataegii Fructus 9 g
Yang (2015)²⁹	Hyperactivity of fire due to yin deficiency	n.r	Rehmanniae Radix 12 g	Decoction	n.r for 6 months
			Ecliptae Herba 12 g
			Prunellae Spica 12 g
			Phellodendri Cortex 12 g
			Anemarrhenae Rhizoma 8 g
			Ligustri Lucidi Fructus 8 g
			Lycii Radicis Cortex 8 g
			Artemisiae Apiaceae Herba 8 g
			Moutan Cortex 6 g
			Glycyrrhizae Radix 6 g
	Liver-qi stagnation transforming into fire	n.r	Bupleuri Radix 15 g	Decoction
			Curcumae Radix 15 g
			Curcumae Radix 12 g
			Angelicae Gigantis Radix 12 g
			Cyperi Rhizoma 8 g
			Paeoniae Radix Alba 8 g
			Rehmanniae Radix 8 g
			Hordei Fructus Germiniatus 8 g
			Glycyrrhizae Radix 6 g
	Phlegm stagnation due to spleen deficiency	n.r	Rehmanniae Radix 15 g	Decoction
			Pseudostellariae Radix 15 g
			Poriae cocos 12 g
			Anemarrhenae Rhizoma 8 g
			Corni Fructus 8 g
			Angelicae Gigantis Radix 8 g
			Phellodendri Cortex 8 g
			Paeoniae Radix Alba 8 g
			Cyperi Rhizoma 8 g
			Trionycis Carapax 8 g
			Cyathulae Radix 8 g
			Glycyrrhizae Radix 6 g
HM group vs. Triptorelin
Cui (2015)³⁰	n.r	Modified Zhibai Dihuang	Anemarrhenae Rhizoma 10 g	Decoction	n.r^*2 times/day for 6 months
			Pulvis Cornus Bubali Concentratus 6 g
			Prunellae Spica 10 g
			Phellodendri Cortex 6 g
			Arctii Fructus 10 g
			Rehmanniae Radix 5 g
			Sargassum 10 g
			Moutan Cortex 10 g
			Laminariae Thallus 10 g
			Corni Fructus 10 g
Huang (2015)³¹	n.r	Kangzao granules	Fraxini Cortex 6 g	Granule	50 mL^*2 times/day for 6 months
			Pinelliae Rhizoma 6 g
			Poriae cocos 10 g
			Rehmanniae Radix 10 g
			Phellodendri Cortex 6 g
			Anemarrhenae Rhizoma 10 g
			Moutan Cortex 10 g
			Crataegii Fructus 10 g
			Aurantii Fructus Pericarpium 10 g
			Prunellae Spica 10 g
Gan (2016)³²	n.r	Dabuyin	n.r	Pill	n.r for 1 year
Qi (2016)³³	n.r	Modified Dabuyin Wan	Rehmanniae Radix Preparat 12 g	Decoction	150 mL^*2 times/day for 2 months and discontinued.After 14 months of triptorelin injection treatment, HM was taken again for 2 months
			Anemarrhenae Rhizoma 8 g
			Trionycis Carapax 8 g
			Phellodendri Cortex 6 g
			Dioscoreae Rhizoma 9 g
			Corni Fructus 6 g
			Poriae cocos 8 g
			Alismatis Rhizoma 8 g
			Scrophulariae Radix 8 g
			Fleshfingered Citron 6 g
			Citrus medica L. 6 g
			Crataegii Fructus 9 g
			Atractylodis Rhizoma Alba 9 g
			Hordei Fructus Germiniatus 6 g
Zhang (2016)³⁴	n.r	Dabuyin	n.r	Pill	6 g^*2–3 times/day for 6 months
HM group vs. Leuprolide
Yang (2013)³⁵	Liver depression and phlegm stagnation	Zhizao granules	Bupleuri Radix	Granule	4–7 years; 1/3ch^2 times/day7–10 years; 1/2ch^2 times/day for 6 months
			Gardeniae Fructus
			Paeoniae Radix Alba
			Trichosanthis Fructus
			Prunellae Spica
			Fritillariae Thunbergii Bulbus
			Curcumae Radix
			Ostreae Concha
			Litchi Semen
HM group vs. No treatment
Cai (2001)²⁸	n.r	ZiYin Xiehuo	Rehmanniae Radix 15 g	Decoction	30 mL^*2 times/day for 1–3.5 years
			Chinemydis Carapax 12 g
			Phellodendri Cortex 12 g
			Anemarrhenae Rhizoma 12 g
			Scrophulariae Radix 12 g
			Moutan Cortex 9 g
			Gentianae Radix 6 g
			Crataegii Fructus 9 g
Wang (2012)³⁶	n.r	Dabuyin	n.r	Pill	10 g^*3 times/day for 1 year

HM, herbal medicine; TCM, Traditional Chinese Medicine.

Level of evidence

The level of evidence for each outcome is presented in Tables 3 –6. When comparing HM group with megestrol, the evidence level for bone age index was “very low.” When comparing HM group with triptorelin injection, the evidence level for serum LH level (IU/L) and E2 level (pg/mL) was “moderate,” and evidence level for uterine volume, ovarian volume, and bone age index was “very low.” When comparing HM group with no treatment, there was a “moderate” level of evidence for bone age index. Overall, there were “moderate” level of evidences for response rate and serum E2 level in HM group compared to control group, and “very low” level of evidences for serum LH level, serum FSH level, uterine volume, ovarian volume, and bone age index in HM group compared to control group. The low level of evidence was largely due to a high risk of bias (especially due to incomplete blinding of participants and personnel), high heterogeneity, and overlap in the 95% CIs, leading to a lack of significant effects.

Table 3.

Summary of Findings: Herbal Medicine Group Compared to Megestrol for Idiopathic Central Precocious Puberty

Patient or population: ICPPIntervention: HM groupComparison: Megestrol
Outcomes	Anticipated absolute effects^* (95% CI)		Relative effect (95% CI)	No. of participants (studies)	Certainty of the evidence (GRADE)	Comments
Outcomes	Risk with Megestrol	Risk with HM group	Relative effect (95% CI)	No. of participants (studies)	Certainty of the evidence (GRADE)	Comments
BAI	The mean bone age index was 0	MD 0.68 lower (0.86 lower to 0.51 lower)	—	148 (2 RCTs)	⨁◯◯◯VERY LOW^a,b	Bone age index is bone age/chronological age, with higher results (>1) indicating accelerated bone growth.

GRADE Working Group grades of evidence:

High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.

Moderate certainty: we are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

In all trials, blinding of participants and personnel was incomplete.

Heterogeneity is very high (I ² = 73%).

BAI, bone age index; CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation; HM, herbal medicine; ICPP, idiopathic central precocious puberty; MD, mean difference; RCT, randomized controlled trial.

Table 4.

Summary of Findings: Herbal Medicine Group Compared to Triptorelin Injection for Idiopathic Central Precocious Puberty

Patient or population: ICPPIntervention: HM groupComparison: Triptorelin injection
Outcomes	Anticipated absolute effects^* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
Outcomes	Risk with triptorelin injection	Risk with HM group	Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
Response rate	740 per 1000	821 per 1000 (732 to 917)	RR 1.11 (0.99–1.24)	356 (5 RCTs)	⨁◯◯◯VERY LOW^a,b	Response rate was calculated as cured participants+improved participants/total
LH (IU/L)	The mean LH (IU/L) was 0	MD 0.74 lower (1.22 lower to 0.27 lower)	—	356 (5 RCTs)	⨁⨁⨁◯MODERATE^a	Normal range of LH level in girls: before puberty; <0.5 IU/LTanner stage II; 1.0–7.0 IU/LTanner stage IV; 2.0–8.0 IU/L
FSH (IU/L)	The mean FSH (IU/L) was 0	MD 0.57 lower (1.26 lower to 0.12 higher)	—	356 (5 RCTs)	⨁◯◯◯VERY LOW^a,c,d	Normal range of FSH level in girls: before puberty; 0.4–3.0 IU/LTanner stage II; 1.6–7.0 IU/LTanner stage III; 4.0–7.0 IU/LTanner stage IV; 3.0–8.0 IU/L
E2 (pg/mL)	The mean E2 (pg/mL) was 0	MD 4.21 lower (5 lower to 3.43 lower)	—	356 (5 RCTs)	⨁⨁⨁◯MODERATE^a	Normal range of E2 level in girls: before puberty; <16 pg/mL
Uterine volume (mL)	The mean uterine volume (mL) was 0	MD 0.42 lower (0.92 lower to 0.08 higher)	—	356 (5 RCTs)	⨁◯◯◯VERY LOW^a,d,e	During puberty, uterine volume increases significantly, but cutoff values of pubertal uterus volume vary among studies.
Ovarian volume (mL)	The mean ovarian volume (mL) was 0	MD 0.02 higher (0.25 lower to 0.29 higher)	—	356 (5 RCTs)	⨁◯◯◯VERY LOW^a,d,f	During puberty, ovarian volume increases significantly, but cutoff values of pubertal uterus volume vary among studies.
Bone age index	The mean bone age was 0	MD 0.09 lower (0.25 lower to 0.08 higher)	—	305 (4 RCTs)	⨁◯◯◯VERY LOW^a,d,g	BAI is bone age/chronological age, with higher results indicating accelerated bone growth.

GRADE Working Group grades of evidence:

High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.

Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

Very low certainty: we have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect.

The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

In all trials, blinding of participants and personnel was incomplete.

Heterogeneity is very high (I ² = 88%).

Heterogeneity is very high (I ² = 93%).

p > 0.05.

Heterogeneity is very high (I ² = 81%).

Heterogeneity is very high (I ² = 87%).

Heterogeneity is very high (I ² = 78%).

BAI, bone age index; CI, confidence interval; E2, estradiol; FSH, follicle-stimulating hormone; GRADE, Grading of Recommendations Assessment, Development and Evaluation; HM, herbal medicine; LH, luteinizing hormone; ICPP, idiopathic central precocious puberty; MD, mean difference; RCT, randomized controlled trial; RR, relative risk.

Table 5.

Summary of Findings: Herbal Medicine Group Compared to No Treatment for Idiopathic Central Precocious Puberty

Patient or population: ICPPIntervention: HM groupComparison: No treatment
Outcomes	Anticipated absolute effects^* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
Outcomes	Risk with No treatment	Risk with HM group	Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
BAI	The mean bone age was 0	MD 0.93 lower (1.02 lower to 0.85 lower)	—	151 (2 RCTs)	⨁⨁⨁◯MODERATE^a	Bone age index is bone age/chronological age, with higher results indicating accelerated bone growth.

GRADE Working Group grades of evidence:

High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.

Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

In all trials, blinding of participants and personnel was incomplete.

Table 6.

Summary of Findings: Herbal Medicine Group Compared to Control Group for Idiopathic Central Precocious Puberty

Patient or population: ICPPIntervention: HM groupComparison: Control group
Outcomes	Anticipated absolute effects^* (95% CI)		Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
Outcomes	Risk with control group	Risk with HM group	Relative effect (95% CI)	No of participants (studies)	Certainty of the evidence (GRADE)	Comments
Response rate	739 per 1000	842 per 1000 (761 to 931)	RR 1.14 (1.03–1.26)	418 (6 RCTs)	⨁⨁⨁◯MODERATE^a	Response rate was calculated as cured participants+improved participants/total
LH (IU/L)	The mean LH (IU/L) was 0	MD 2.09 lower (5.3 lower to 1.13 higher)	—	498 (7 RCTs)	⨁◯◯◯VERY LOW^a,b,c	Normal range of LH level in girls: before puberty; <0.5 IU/LTanner stage II; 1.0–7.0 IU/LTanner stage IV; 2.0–8.0 IU/L
FSH (IU/L)	The mean FSH (IU/L) was 0	MD 1.39 lower (3.22 lower to 0.43 higher)	—	495 (7 RCTs)	⨁◯◯◯VERY LOW^a,c,d	Normal range of FSH level in girls: before puberty; 0.4–3.0 IU/LTanner stage II; 1.6–7.0 IU/LTanner stage III; 4.0–7.0 IU/LTanner stage IV; 3.0–8.0 IU/L
E2 (pg/mL)	The mean E2 (pg/mL) was 0	MD 4.22 lower (5 lower to 3.45 lower)	—	464 (7 RCTs)	⨁⨁⨁◯MODERATE^a	Normal range of E2 level in girls: before puberty; <16 pg/mL
Uterine volume (mL)	The mean uterine volume (mL) was 0	MD 0.7 lower (1.37 lower to 0.04 lower)	—	498 (7 RCTs)	⨁◯◯◯VERY LOW^a,e	During puberty, uterine volume increases significantly, but cutoff values of pubertal uterus volume vary among studies.
Ovarian volume (mL)	The mean ovarian volume (mL) was 0	MD 0.26 lower (0.74 lower to 0.22 higher)	—	498 (7 RCTs)	⨁◯◯◯VERY LOW^a,c,f	During puberty, ovarian volume increases significantly, but cutoff values of pubertal uterus volume vary among studies.
BAI	The mean bone age index was 0	MD 0.35 lower (0.6 lower to 0.09 lower)	—	579 (8 RCTs)	⨁◯◯◯VERY LOW^a,g	Bone age index is bone age/chronological age, with higher results indicating accelerated bone growth.

GRADE Working Group grades of evidence:

High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.

Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).

In all trials, blinding of participants and personnel was incomplete.

Heterogeneity is very high (I ² = 100%).

p > 0.05.

Heterogeneity is very high (I ² = 99%).

Heterogeneity is very high (I ² = 92%).

Heterogeneity is very high (I ² = 97%).

Heterogeneity is very high (I ² = 98%).

Discussion

Summary of evidence

The results of this systematic review and meta-analysis showed that, for treatment of ICPP, HM is more effective than control group. Response rate of HM group was 1.14 times higher compared with control group with moderate level of evidence. Also, serum E2 level was significantly lower in HM group compared to control group with moderate level of evidence. Bone age index and uterine volume were also significantly lower in HM group than control group with very low level of evidence. According to the above results, it is considered that HM treatment is effective for ICPP. In particular, serum LH level (IU/L) and E2 level were significantly lower in HM group than triptorelin injection group with moderate level of evidence. Triptorelin is a GnRH analog, which is widely used as a treatment for ICPP. Since GnRH analog may have a negative effect on the final height if bone age is over 13 years,⁸ combination with HM or HM alone might be effective for treating ICPP without affecting final height.

The incidence of adverse events was lower in the HM group than control group, but it was not statistically significant. Five cases of mild vaginal bleeding were reported in the group treated with triptorelin injection.³³ Two cases of mild diarrhea and three cases of pain at the injection site were reported in the group treated with HM combined with oral megestrol acetate tablets; four cases of pain at the injection site were reported in the group treated with oral megestrol acetate tablets alone.²⁹ Side effects of GnRH analog therapy, currently used as a standard treatment for ICPP, may include hypertriglyceridemia and increased low-density lipoprotein level,⁹ hyperprolactinemia,¹⁰ vaginal bleeding,¹¹ and skin reactions at the injection site.¹² In one study included in this review, vaginal bleeding was observed in triptorelin injection group, while no adverse events occurred in HM plus triptorelin injection group.³³ Therefore, it is thought that combination use of HM with GnRH analog may reduce side effects. Overall, since only mild adverse events were reported in HM group, and no adverse events were reported in the group treated with HM alone, it is considered relatively safe to use HM compared to triptorelin injection and megestrol acetate tablets, but there is a limit to judging the clinical significance.

The major TCM patterns of ICPP are known as hyperactivity of fire due to yin deficiency, liver-qi stagnation transforming into fire, and phlegm stagnation due to spleen deficiency.²² In this systematic review, only two studies reported TCM patterns.^29,35 In one study, patients were divided into hyperactivity of fire due to yin deficiency, liver-qi stagnation transforming into fire, and phlegm stagnation due to spleen deficiency²⁹; in the other study, only patients with liver depression and phlegm stagnation types were recruited.³⁵

Research on the treatment cost was conducted in only one study, and it was reported that treatment cost of HM plus triptorelin injection group was significantly lower than triptorelin injection alone group (p < 0.00001).³³ Since many patients are feeling the burden of the cost treating ICPP, research on the treatment cost will be helpful for the further use of HM for treating ICPP.

Limitations

There are several limitations to this review. The first limitation of this review is that the studies included in the meta-analysis exhibited high heterogeneity. High heterogeneity among these studies was likely caused by various clinical factors (population, study duration, and herbal ingredients), as well as research design and the interventions used in treatment or control groups.

The second limitation of this review is that the included studies were of low quality. When assessing the risk of bias, all studies were considered to be “high risk” in more than one aspect; in particular, performance bias was considered to be “high risk” in all studies. Because there were no control groups using placebo, both participants and researchers were aware of the assigned interventions and blinding could not be performed.

The third limitation of this review is that the evaluation criteria are not standardized for assessing treatment outcomes of ICPP. Regarding the response rate, it was difficult to determine morbidity because the criteria were subjective and differed among studies; thus, the response rate reliability was poor.

The fourth limitation of this review is that all of the included studies were conducted in China; thus, the results cannot be readily generalized.

Conclusions

According to the results of the meta-analysis, it seems that HM treatment has a significant effect for treating ICPP and can effectively reduce serum LH, E2 level, and bone age. In particular, since serum LH, E2 level, and bone age index were significantly lower in HM group than triptorelin injection group, it seems that HM can supplement the limitations of the GnRH analog treatment, which is widely used for ICPP. In conclusion, HM treatment for treating ICPP is effective and safe. However, because of the limitations described above, more large-scale high-quality studies are needed to clarify the safety and efficacy of HM for treatment of ICPP. In addition, more accurate analysis, achieved through application of standardized and objective outcome variables, is needed. Because treatment of ICPP is performed in children and side effects have been reported when HM are used, further studies of these side effects are needed.

Footnotes

Authors' Contributions

Y.B.L. and L.H.L. conceived the study, developed the criteria, searched the literature, performed the data analysis, and wrote the protocol. Y.B.L. and J.A.L. conducted the preliminary search. J.A.L. and L.H.L. assisted in searching the Chinese literature and extracting the data. All authors have read and approved the final article.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This research was supported by the Daejeon University Research Grants (2019).

Supplementary Material

Supplementary Figure S1

Supplementary Figure S2

References

Chen

, Eugster

. Central precocious puberty: Update on diagnosis and treatment. Paediatr Drugs, 2015; 17:273–281.

W-J

, Gong

C-X

, Guo

M-J

, et al. Efficacy and safety of domestic leuprorelin in girls with idiopathic central precocious puberty: A multicenter, randomized, parallel, controlled trial. Chin Med J (Engl), 2015; 128:1314.

Kim

, Huh

, Won

, et al. A significant increase in the incidence of central precocious puberty among Korean girls from 2004 to 2010. PLoS One, 2015; 10:e0141844.

Wojniusz

, Callens

, Sutterlin

, et al. Cognitive, emotional, and psychosocial functioning of girls treated with pharmacological puberty blockage for idiopathic central precocious puberty. Front Psychol, 2016; 7:1053.

Lakshman

, Forouhi

, Sharp

, et al. Early age at menarche associated with cardiovascular disease and mortality. J Clin Endocrinol Metab, 2009; 94:4953–4960.

Prentice

, Viner

. Pubertal timing and adult obesity and cardiometabolic risk in women and men: A systematic review and meta-analysis. Int J Obes (Lond), 2013; 37:1036–1043.

Arcari

, Gryngarten

, Freire

, et al. Body mass index in girls with idiopathic central precocious puberty during and after treatment with GnRH analogues. Int J Pediatr Endocrinol, 2016; 2016:15.

Tonini

, Lazzerini

. Side effects of GnRH analogue treatment in childhood. J Pediatr Endocrinol Metab, 2000; 13(Suppl 1):795–804.

Karamizadeh

, Tabebordbar

, Saki

, et al. The side effects of gonadotropin releasing hormone analog (diphereline) in treatment of idiopathic central precocious puberty. Acta Med Iran, 2013; 51:41–46.

10.

Kendirci

HNP

, Ağladıoğlu

, Önder

, et al. Effects of GnRH analogue treatment on anterior pituitary hormones in children with central precocious puberty. J Pediatr Endocrinol Metab, 2015; 28:1145–1151.

11.

Partsch

C-J

, Peter

, Sippell

, Deutsche

L-SJMK

. Treatment of progressive central precocious puberty with leuprorelin depot. Monatsschr Kinderheilkd, 1999; 147:638–647.

12.

Carel

J-C

, Lahlou

, Guazzarotti

, et al. Treatment of central precocious puberty with depot leuprorelin. Eur J Endocrinol, 1995; 132:699–704.

13.

, Young

. Experiences of precocious puberty in primary school girls with hormone therapeutics. Korean Acad Soc Nurs Educ, 2019; 25:459–470.

14.

Partsch

C-J

, Sippell

. Treatment of central precocious puberty. Best Pract Res Clin Endocrinol Metab, 2002; 16:165–189.

15.

Ying

, Tang

, Chen

, et al. GnRH agonist treatment for idiopathic central precocious puberty can improve final adult height in Chinese girls. Oncotarget, 2017; 8:109061–109067.

16.

Hee

, Chan

. An increase of patients diagnosed as precocious puberty among Korean children from 2010 to 2015. J Pediatr Korean Med, 2016; 30:60–65.

17.

Hoon

, Gil

, Hyun

. Estimated factors and clinical characteristics of children diagnosed with idiopathic precocious puberty. J Pediatr of Korean Med, 2011; 25:111–120.

18.

Lin

, Chang

, Chen

, et al. Characteristics of traditional Chinese medicine usage in children with precocious puberty: A nationwide population-based study. J Ethnopharmacol, 2017; 205:231–239.

19.

Trinh

, Park

, Oh

, et al. Preventive effect and safety of a follicle stimulating hormone inhibitory formulation containing a mixture of Coicis Semen and Artemisia capillaris for precocious puberty: A preliminary experimental study using female rats. Evid Based Complement Alternat Med, 2017; 2017:2906014.

20.

Wang

, Zhu

, Yu

, et al. Effect of nourishing “Yin”-removing “Fire” Chinese herbal mixture on hypothalamic NKB/NK3R expression in female precocious rats. Evid Based Complement Alternat Med, 2014; 2014:217424.

21.

Lee

M-J

, Chang

G-T

, Han

Y-J

. The study for precocious puberty in recent journals of traditional Chinese medicine. J Pediatr Korean Med, 2008; 22:163–187.

22.

Cheng

, Ye

, Miao

, et al. Experimental study on therapeutic effect of Dabuyin Wan on true precocious puberty in female rats [In Chinese]. Zhongguo Zhong Yao Za Zhi, 2013; 38:386–390.

23.

Tian

, Zhao

, Sun

, et al. Evaluation of the true precocious puberty rats induced by neonatal administration of Danazol: Therapeutic effects of nourishing “Yin”-removing “Fire” Chinese herb mixture. Reprod Biol Endocrinol, 2005; 3:38.

24.

Sun

, Perry

, Yu

, et al. Effect of nourishing “Yin”-removing “Fire” Chinese herbal mixture on hypothalamic kisspeptin expression in female precocious rats. J Ethnopharmacol, 2010; 127:274–279.

25.

Cai

, Ji

, Shi

. Clinical study on treatment of female idiopathic precocious puberty with combined therapy of Chinese medicine and megestrol acetate [In Chinese]. Zhongguo Zhong Xi Yi Jie He Za Zhi, 2001; 21:732–735.

26.

, Cai

, Huang

. Pattern of abnormal skeletal development in girls with idiopathic precocious puberty and therapeutic effect of Chinese herbs for nourishing yin and purging fire [In Chinese]. Zhongguo Zhong Xi Yi Jie He Za Zhi, 2004; 24:979–982.

27.

Higgins

JPT

, Altman

, Sterne

JAC

. Chapter 8: Assessing risk of bias in included studies. In: Higgins

JPT

, Green

, eds. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). London: The Cochrane Collaboration, 2011.

28.

Cai

, Ji

, Shi

29.

Yang

, Ru

. Research on the effectiveness of TCM combined with Western medicine treatment of girls idiopathic precocious puberty. Liaoning J Tradit Chin Med, 2015; 42:1278–1280.

30.

Cui

, Wang

, et al. WDe. Study on the relationship between serum IGF-1 and therapeutic effect in girls with idiopathic precocious puberty treated with traditional Chinese medicine. J Qiqihar Med Univ, 2015; 36:1772–1773.

31.

Huang

, Ye

, Kong

, Li

. Clinical observation on 40 cases of girl with idiopathic central precocious puberty treated by Kangzao granules. Guiding J Tradit Chin Med Pharm, 2015; 21:49–50.

32.

Gan

. Clinical study of Dabuyin Pills combined with Gonadotropin-releasing hormone antagonist in treating female central sexual precocious puberty. J Pediatr TCM, 2016; 12:30–33.

33.

, Zhang

, Xiang

. Clinical analysis of 36 cases of idiopathic central precocious puberty treated by traditional Chinese medicine and Western medicine. J Tianjin Univ Tradit Chin Med, 2016; 35:88–91.

34.

Zhang

. Clinical study on treatment of female idiopathic precocious puberty with Dabuyin Pill. J New Chin Med, 2016; 48:127–129.

35.

Yang

, Zhang

, Hou

, et al. HXe. Effect of Zhizao Granule on serum osteocalcin in girls with idiopathic central precocious puberty. Jilin J Chin Med, 2013; 162–164.

36.

Wang

, Liu

, Mou

, et al. Clinical study on treatment of female idiopathic precocious puberty with Chinese medicine Dabuyin pill. Natl Med Front China, 2012; 7:19.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.25 MB

0.53 MB