Skip to main content
NCBI home page
Medicine logoLink to Medicine
. 2025 Aug 1;104(31):e41936. doi: 10.1097/MD.0000000000041936

Pharmacotherapy for children with central precocious puberty or early puberty: A systematic review and meta-analysis

Chunsong Yang a,b,c,d, Zheng Liu a,b,c,d, Linan Zeng a,b,c,d, Jin Wu d,e, Lingli Zhang a,b,c,d,f,g,*
PMCID: PMC12323909  PMID: 40760548

Abstract

Background:

The intervention of gonadotropin-releasing hormone agonist or in combination with recombinant human growth hormone (rhGH) is still controversial for children with central precocious puberty (CPP) or early puberty (EP). This study aimed to systematically review and compare the efficacy and safety of current therapeutic regimens for CPP or EP based on different timings of intervention.

Methods:

We searched 7 databases and 2 registers for available studies from inception until December 2023. Two independent reviewers screened literature, extracted data, and assessed the risk of bias. Eligible data from randomized controlled trials and non-randomized studies of interventions were synthesized with meta-analysis and further performed subgroup analysis.

Results:

A total of 55 studies were included in this review, including 11 randomized controlled trials and 44 non-randomized studies of interventions. The pharmacotherapy was beneficial to final height (FH) for the girls with CPP or EP (for combination therapy: mean difference [MD] 1.58, 95% confidence interval [CI] 0.66–2.51; for monotherapy: MD 2.51, 95% CI 1.26–3.76), and subgroup analysis showed that CPP with puberty onset aged before 8 years (for combination therapy: MD 3.69, 95% CI 1.59–5.79; for monotherapy: MD 4.83, 95% CI 3.29–6.17), early intervention before 8 years of age (for combination therapy: MD 3.50, 95% CI 1.75–5.25; for monotherapy: MD 6.50, 95% CI 4.99–8.01), and the course of combination with rhGH for more than 2 years (MD 3.31, 95% CI 2.14–4.47) improved higher FH. The pharmacologic intervention had less impact on the body mass index at FH (MD −0.06, 95% CI −0.50 to 0.38), but probably increase the risk of the incidence of polycystic ovary syndrome (risk ratio 1.93, 95% CI 1.26–2.98) in the long term.

Conclusion:

The gonadotropin-releasing hormone agonist improved the FH of children with CPP, and in combination with rhGH appeared to further increase the effectiveness of pharmacotherapy. The increasing risk of polycystic ovary syndrome should be considered before treatment and further confirmed through high-quality comparative studies in the future.

Keywords: central precocious puberty, early puberty, gonadotropin-releasing hormone agonist, recombinant human growth hormone, systematic review

1. Introduction

Central precocious puberty (CPP) traditionally refers to the early sexual development caused by the premature activation of the hypothalamic–pituitary–gonadal axis before the age of 8 years in girls and 9 years in boys, characterized by the appearance of secondary sexual characteristics, accelerated linear growth, advanced bone age (BA), and pubertal levels of luteinizing hormone and follicle-stimulating hormone.[1] It has an estimated prevalence of 200 to 410.6 cases per 100,000 girls and 10.9 to 50 cases per 100,000 boys in various regions, with an approximately 10-fold higher incidence in girls than that in boys.[2,3] The average age at thelarche has shifted towards younger in past decades, with a substantially increasing trend in the incidence of precocious puberty worldwide, which brings about the risk of short stature and severe psychosocial problems for children and adolescents if without appropriate intervention.[2,4,5]

As the gold-standard therapy, gonadotropin-releasing hormone agonist (GnRHa) is commonly administrated to treat CPP, such as leuprorelin, triptorelin, goserelin, and histrelin, which have been widely applied in clinical practice. It works by space-occupying the GnRH receptors to resist the physiologic activation from the hypothalamus and continuously stimulate the pituitary gonadotrophs, resulting in a desensitization of the gonadotroph cells, with subsequent gonadotropins suppression and sex steroids reduction that contributed to delay pubertal development and BA advance.[6] The decision of treatment mainly depends on the age and the rate of pubertal progression, for those whose puberty onset before the defined age with rapid progression of pubertal changes are considered to initiate GnRHa.[6,7] It seems unlikely to compromise the final height (FH) among slow progressive CPP, which is not suggested to offer pharmacologic intervention. In recent years, the application scope of GnRHa has gradually expanded to the population experiencing the early stage of normal puberty (generally between 8–10 years of age in girls and 9–11 years of age in boys) but with rapid progression in Tanner scale within 6 months, which was defined as early puberty (EP) in some studies.[8–10] EP has similar clinical presentations and potentially harmful effect with CPP, and is recommended to receive GnRHa treatment in some consensus.[6,7]

The effects on hypothalamic–pituitary–gonadal axis suppression and height gain of GnRHa have been well established, and it can maximally increase the final adult height when initiating treatment before 6 years old.[11] As for the effects on children with onset age of puberty of more than 6 years were still contradictory among current studies, particularly after the scale-up of application in those over 8 years of age. It was reported that the FH was similarly beneficial from GnRHa between the patients aged <8 years and more than 8 years,[12,13] while some published studies indicated that the GnRHa treatment may not improve the FH in patients those with puberty onset aged 6 to 8 years, let alone aged after 8 years, which was still controversial for EP.[14–16] Meanwhile, the long-term safety of GnRHa treatment was not yet confirmed and still controversial in the metabolic change and reproductive function. Several reports illustrated that GnRHa was associated with changes in the body mass index (BMI) during treatment,[17–19] but some authors considered it temporary and could be recovery after discontinuation of treatment.[20,21] Additionally, the Australasian Paediatric Endocrine Group indicated that GnRHa is not suitable for treatment in children outside of CPP due to the potential risk of polycystic ovary syndrome (PCOS),[22] which was also considered as an independent risk factor in the previous study,[23] but contradicted by a recent single-center cohort study showing no association.[24]

However, it appeared that the suppression from GnRHa might present over inhibiting effect on growth velocity (GV) in some patients, even resulting in impaired FH and failure in the achievement of target height (TH).[25] A perspective was proposed that recombinant human growth hormone (rhGH), as an adjunctive therapy, comminated with GnRHa might compensate for the loss of height, which takes effect by stimulating the hepatic insulin-like growth factor 1 (IGF-1) production to promote the cell growth. The rhGH is mainly used in patients with growth hormone deficiency, idiopathic short stature, small for gestational age, Noonan syndrome, etc, but belongs to off-label use for CPP. Nevertheless, due to the unsatisfactory effectiveness of GnRHa treatment, some parents still require to add rhGH for maximizing the final adult height of children. Present studies showed contradictory results on the final adult height between the GnRHa plus rhGH group and GnRHa alone groups,[26–28] and an international consensus published in 2009 issued it not routinely recommended to combine with rhGH in CPP due to the potential adverse effects and suggested larger studies to validation.[11]

It was regarded that the long-term outcomes were related to the onset age of puberty, the onset age of intervention, and the duration of treatment.[29] The degree of height gain was various and contradictory conclusions were generated from different studies, some of which were even with flaws in designs, such as single-arm studies without the control group, or only comparison with predicted adult height (PAH) that probably overestimated the adult height due to subjective difference in prediction methods.

Considering the dissimilar results, this study aims to investigate the treatment outcomes for patients with CPP and EP. CPP and EP are typically characterized by the premature appearance of secondary sexual characteristics, accelerated BA, and increased GV. Without timely intervention, patients may face various adverse outcomes, including reduced adult height, mental health issues (e.g., anxiety and depression), social difficulties, and abnormalities in reproductive system development. Furthermore, EP may lead to metabolic disorders and endocrine dysfunctions, potentially impacting the long-term health and quality of life of affected individuals. Given these potential risks, this study seeks to conduct a systematic review and meta-analysis to comprehensively evaluate the efficacy and safety of pharmacological interventions in the treatment of CPP and EP, providing high-quality evidence to support clinical decision-making.

2. Materials and methods

This systematic review was registered on PROSPERO (CRD42022377476) and followed the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) reporting guideline (Appendix 1, Supplemental Digital Content, https://links.lww.com/MD/P576).

2.1. Search strategy and selection criteria

We searched relevant literature in PubMed (1946–December 2023), Embase (Ovid) (1947–December 2023), CENTRAL (Ovid) (1993–December 2023), China National Knowledge Infrastructure (CNKI) (1979–December 2023), Chinese BioMedical Literature Database (CBM) (1978–December 2023), VIP Database for Chinese Technical Periodicals (VIP) (1989–December 2023), Knowledge service platform of WanFang Data (WanFang) (1998–December 2023), ClinicalTrals.gov (1997–December 2023), and the World Health Organization International Clinical Trials Registry Platform (ICTRP) (2007–December 2023) from inception until December 2023. The search strategy was developed based on the medical subject headings (MeSH) combined with free-text terms including “precocious puberty,” “early puberty,” “gonadotropin releasing hormone agonist*,” etc, without any language restrictions (Appendix 2, Supplemental Digital Content, https://links.lww.com/MD/P576). Additionally, the reference lists of included studies and relevant systematic reviews were screened for more eligible studies.

We included randomized controlled trials (RCTs) and non-randomized studies of interventions (NRSIs, including non-randomized controlled trials and cohort studies) that assessed the efficacy and safety of pharmacotherapy for children with CPP or EP, comparing GnRHa plus rhGH to GnRHa alone, or GnRHa alone to no treatment, which was administrated at least 1 year; and reporting the outcome of height, adverse effects and other relevant indicators. Studies were considered to exclude if they focused on those populations caused by secondary diseases or combined with any other conditions that may affect height, such as growth hormone deficiency, idiopathic short stature, small for gestational age, or familial inherited disease, etc. We also excluded trials that duplicated publications or were not available in the full text.

2.2. Data collection and quality assessment

After duplicate records removal by Endnote X9 (Clarivate Analytics, UK), 2 independent reviewers (Z.L. and C.S.Y.) screened titles and abstracts and reviewed full texts for potentially eligible studies according to the inclusion criteria. A standardized form was thereafter used to extract data on study identifiers, study design, setting and sample, participant demographics, intervention characteristics, and outcome measures. Important but missing information was obtained, if necessary, through contact with authors by e-mail. Two reviewers cross-checked in each stage and discussed to consensus, or if necessary, resolved by the third reviewer (J.W.).

The primarily interesting outcomes in this meta-analysis were (i) FH, defined as the height at the time when BA was over 15 years and/or GV was <1 cm per year and/or at least 2 years after menarche, and (ii) height gain, defined as the difference between FH and initial PAH (FH–initial PAH). The secondary outcomes were (i) FH standard deviation score (SDS), calculated as (FH − mean height)/standard deviation of corresponding age, (ii) FH SDS − PAH SDS (the difference between FH SDS and initial PAH SDS), (iii) FH − TH (the difference between FH and TH), and (iv) ΔPAH (the difference of PAH between cessative and initial treatment). With regard to safety, we captured the short-term and long-term adverse events simultaneously, including endocrine and metabolism adverse events, reproductive function adverse events, and total general adverse events, if reported enough.

The risk of bias was assessed independently by 2 reviewers (Z.L. and C.S.Y.) based on the Cochrane Collaboration Risk of Bias tool (ROB 2.0, 2019 version) for RCTs, the Risk of Bias in Non-randomized Studies of Interventions tool (ROBINS-I) for NRSIs, respectively. Disagreements were resolved by discussion among themselves or by third-party adjudication (J.W.).

2.3. Statistical analysis

The meta-analysis for girls and boys was separately performed considering the gender difference in growth and puberty development in children, and the pooled effect size of continuous variables and dichotomous variables were calculated as mean difference (MD) and risk ratio with 95% confidence interval (CI) using the inverse variance statistical method and the Mantel–Haenszel method, which were estimated the efficacy and safety outcomes, respectively. The heterogeneity among included studies was assessed with the Q test and I² statistic, among which I² < 50% and P > .1 was considered as mild-to-moderate heterogeneity and the fixed-effects model was adopted to synthesize effects, while I² ≥ 50% or P < .1 indicated high heterogeneity and the random-effects model was preferred. Univariate meta-regression was performed to explore the source of heterogeneity for primary outcomes based on region, study design, sample size, risk of bias, and population difference. Subgroup analyses were further conducted to estimate the efficacy difference under different conditions stratified by onset age of puberty (<6 years, 6–8 years, or ≥8 years), onset age of intervention (<6 years, 6–8 years, or ≥8 years), and course (<2 years or ≥2 years), but omitting the studies brought about the high heterogeneity between the 2 groups. Publication bias was assessed with Egger tests and funnel plots if reported studies were 10 or more, and Duval and Tweedy trim-and-fill analysis was carried out as a sensitivity analysis when there was a publication bias, otherwise by sequentially eliminating each study to verify the robustness of the results. The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach was used to rate the quality of evidence and summarized the evidence profiles. The statistical analysis was performed using Revman (version 5.3, Cochrane Collaboration, UK) and Stata (version 15.1, Stata Corp., USA), and P < .05 was considered statistically significant.

3. Results

3.1. Search results

We identified 5360 citations from 7 databases and 2 registers. After removing 2236 duplicate citations, 3124 articles were preliminary screened based on titles and abstracts, 141 of which were selected for full-text review and assessed for eligibility, with 55 studies ultimately included in this review, including 11 RCTs and 44 NRSIs (all were cohort studies) (Fig. 1, Appendix 3, Supplemental Digital Content, https://links.lww.com/MD/P576).

Figure 1.

Figure 1.

Open in a new tab

PRISMA flow diagram of study selection.

3.2. Characteristics and risk of bias of included studies

The 55 included studies had a total sample size of 4633 pediatric patients, involving 4516 girls (97.47%) in 55 studies and 117 boys (2.53%) in 4 studies reported respectively, which were recruited from 13 different countries, covering China (19 studies), Italy (9 studies), France (6 studies), Israel (4 studies), South Korea (4 studies), Netherlands (2 studies), Thailand (3 studies), Turkey (2 studies), the USA (2 studies), Greece (1 studies), Jordan (1 studies), Sweden (1 studies), and the UK (1 studies). Additionally, 32 studies (58.18%) were conducted to compare efficacy and safety between GnRHa alone and no treatment, 20 studies (36.36%) compared GnRHa plus rhGH to GnRHa alone, and 3 studies (5.45%) simultaneously compared GnRHa plus rhGH, GnRHa alone, and no treatment. The baseline characteristics of the included studies are reported in Table 1.

Table 1.

Baseline characteristics of the included studies.

Study ID Country Study design Sample size Intervention and Comparator Duration of treatment Onset age of puberty Onset age of intervention H BMI BA PAH TH GV
Girls
Adan (2002) France NRSI (cohort study) 43 GnRHa 2 6.4 ± 0.2 7.9 ± 0.2 132.8 ± 1.2 NR 10.3 ± 0.2 156 ± 1.2 161.2 ± 0.7 NR
29 Control – 6.9 ± 0.2 8.0 ± 0.1 NR NR NR 164.1 ± 1.2 161.6 ± 0.7 NR
Allali (2011) France NRSI (cohort study) 70 GnRHa 2.8 ± 1.3 6.5 ± 1.5 7.4 ± 1.5 128.9 ± 9.6 NR 9.1 ± 1.8 159.5 ± 7.2 161.0 ± 5.0 2.3 ± 2.1
52 Control – 7.1 ± 0.9 8.0 ± 1.0 134.2 ± 7.6 NR 9.1 ± 1.6 164.5 ± 7.2 161.4 ± 4.1 2.3 ± 2.2
Antoniazzi (1994) Italy NRSI (cohort study) 15 GnRHa (a) 2.67 ± 1.0 NR 7.7 ± 0.9 131.9 ± 5.0 NR 10.2 ± 1.1 152.9 ± 6.6 155.5 ± 5.3 7.8 ± 1.2
15 GnRHa (b) 3.25 ± 0.83 NR 7.6 ± 0.5 133.2 ± 7.6 NR 9.8 ± 1.0 154.1 ± 4.9 157.6 ± 5.9 7.9 ± 1.4
10 Control – NR 7.2 ± 0.9 130.2 ± 8.6 NR 9.6 ± 2.2 153.3 ± 8.0 156.4 ± 1.3 7.7 ± 1.3
Bertelloni (1998) Italy NRSI (cohort study) 14 GnRHa 6.2 NR 6.2 ± 1.8 127.8 ± 11.5 NR 9.6 ± 1.6 153.5 ± 7.2 163.3 ± 6.2 NR
9 Control – NR 6.5 ± 1.0 125.4 ± 7.2 NR 7.3 ± 1.1 163.1 ± 6.2 161.0 ± 5.9 NR
Bouvattier (1999) France RCT 20 GnRHa 2 9.3 ± 0.5 NR 135.2 ± 4.2 17.2 ± 0.3 10.9 ± 0.5 154.1 ± 3.9 157.6 ± 4.2 8.3 ± 1.8
10 Control – 9.4 ± 0.3 NR 136.1 ± 4.2 17.8 ± 0.8 10.9 ± 0.3 155.2 ± 3.7 157.8 ± 4.7 8.3 ± 0.4
Brauner (1994) France NRSI (cohort study) 19 GnRHa 3.1 ± 0.3 6.2 ± 0.4 7.5 ± 0.4 131 ± 2 NR 10.5 ± 0.3 152.1 ± 1.4 160.2 ± 5.3 NR
15 Control – 7.2 ± 0.1 7.9 ± 0.2 NR NR NR 162.5 ± 1.4 NR NR
Bridges (1995) UK NRSI (cohort study) 11 GnRHa + rhGH NR NR 10.32 NR NR NR NR NR NR
18 GnRHa NR NR 8.22 NR NR NR NR NR NR
25 Control – NR 6.2 NR NR NR NR NR NR
Carel (1999) France NRSI (cohort study) 58 GnRHa 3.7 ± 1.5 6.3 ± 1.5 7.5 ± 1.3 NR NR 10.1 ± 1.5 156.4 ± 6.3 160.1 ± 4.4 8.4 ± 2.2
86 Control – 5.3 ± 1.9 NR NR NR NR NR NR NR
Cassio (1999) Italy RCT 23 GnRHa 2.08 NR 8.5 ± 0.6 134.9 ± 6.8 NR 10.6 ± 0.8 157.8 ± 9.1 158.0 ± 7.6 10.5 ± 7.2
21 Control – NR 8.4 ± 0.5 135.5 ± 5.9 NR 10.3 ± 0.6 159.3 ± 5.4 158.8 ± 3.9 9.9 ± 2.4
Cassio (2006) Italy RCT 22 GnRHa 2.08 NR NR NR NR NR NR NR NR
18 Control – NR NR NR NR NR NR NR NR
Chiavaroli (2010) Italy NRSI (cohort study) 25 GnRHa 1.9 ± 0.5 8.7 ± 0.2 8.9 ± 0.2 136.3 ± 3.8 20.7 ± 2.1 11 ± 0.6 157.4 ± 3.6 160.2 ± 4 NR
55 Control – 8.8 ± 0.1 9 ± 0.2 138.1 ± 3.8 20.7 ± 2.8 11.1 ± 0.5 160.1 ± 3.8 161.7 ± 3.6 NR
Cho (2023) Korea NRSI (cohort study) 22 GnRHa + rhGH 3.00 ± 0.58 NR 8.34 ± 0.44 128.52 ± 3.84 17.25 ± 1.73 10.51 ± 0.61 150.41 ± 5.32 158.93 ± 4.01 NR
188 GnRHa 3.21 ± 0.73 NR 8.20 ± 0.62 132.24 ± 5.17 17.64 ± 2.50 10.30 ± 0.77 156.35 ± 6.34 159.01 ± 3.69 NR
Couto-Silva (2002) France NRSI (cohort study) 9 GnRHa 2.4 ± 0.2 8.6 ± 0.2 9.3 ± 0.1 129.5 ± 1.6 NR 9.8 ± 0.4 151.9 ± 1.7 157.2 ± 1.6 NR
31 Control – 8.7 ± 0.1 9.7 ± 0.2 134.6 ± 1.2 NR 10.2 ± 0.3 156.7 ± 1.0 157.6 ± 1.0 NR
Dai (2019) China RCT 55 GnRHa NR NR 6.31 ± 1.18 NR NR NR NR NR NR
55 Control – NR 6.72 ± 1.25 NR NR NR NR NR NR
Faienza (2017) Italy NRSI (cohort study) 56 GnRHa 3.4 ± 0.8 NR 7.0 ± 0.6 135.5 ± 2.8 19.4 ± 2.4 10.1 ± 1.6 158.4 ± 3.6 160.8 ± 4.7 8.1 ± 1.5
38 Control – NR 7.0 ± 0.8 135.3 ± 2.5 20 ± 2.8 10.2 ± 1.4 159.1 ± 3.8 161.2 ± 4.2 8.2 ± 1.7
Fang (2004) China RCT 5 GnRHa + rhGH 1.42 ± 0.39 7.25 10.17 139.92 ± 4.25 NR 11.72 ± 0.94 148.88 ± 2.22 155.80 ± 1.75 NR
5 GnRHa 1.1 ± 0.26 7.50 9 135.32 ± 6.43 NR 10.80 ± 0.57 149.86 ± 7.10 156.30 ± 3.29 NR
Feng (2014) China NRSI (cohort study) 62 GnRHa 1.77 ± 0.54 NR 10.2 ± 0.8 NR 16.87 ± 1.31 NR 151.6 ± 5.2 NR NR
58 Control – NR NR NR NR NR NR NR NR
Fu (2020) China NRSI (cohort study) 95 GnRHa + rhGH 2.5 ± 0.08 NR 8.7 ± 0.1 131.3 ± 0.7 17.14 ± 0.22 11 ± 0.14 150.81 ± 0.66 156 ± 0.42 NR
244 GnRHa 2 ± 0.05 NR 8.4 ± 0.06 133.75 ± 0.4 16.66 ± 0.14 10.6 ± 0.07 152 ± 0.42 157 ± 0.24 NR
54 Control – NR 8.4 ± 0.12 134.4 ± 0.79 16.98 ± 0.3 11 ± 0.16 151 ± 0.94 157.5 ± 0.57 NR
Jaruratanasirikul (2011) Thailand NRSI (cohort study) 32 GnRHa 2.9 ± 1.5 NR 6.8 ± 1.9 126.7 ± 14.2 NR 10.7 ± 2.2 147.2 ± 4.9 152.7 ± 4.5 NR
20 Control – NR 7.7 ± 1.9 137.9 ± 9.6 NR 13.4 ± 0.5 144.1 ± 5.8 151.1 ± 5.5 NR
Jung (2014) Korea NRSI (cohort study) 23 GnRHa + rhGH 1.9 ± 0.99 NR 8.8 ± 0.59 133.8 ± 5.47 17.3 ± 1.55 10.5 ± 0.86 154.6 ± 2.55 158.1 ± 3.31 NR
59 GnRHa 2~ NR 8.7 ± 0.78 136.9 ± 7.40 18.2 ± 2.36 10.5 ± 0.86 156.6 ± 3.96 159.9 ± 3.52 NR
Karavani (2021) Israel NRSI (cohort study) 27 GnRHa 2.03 ± 1.07 NR 7.3 ± 0.6 130.6 ± 10.4 80.5 ± 19.3 9.6 ± 1.7 NR NR NR
24 Control – NR 7.6 ± 0.6 135.6 ± 14.2 75.7 ± 24.9 10.7 ± 1.7 NR NR NR
Kauli (1997) Israel NRSI (cohort study) 48 GnRHa 3.2 ± 1.3 6.2 ± 1.5 8.3 ± 1.5 136.7 ± 7.9 NR 10.8 ± 2.0 152.3 ± 6.0 157.7 ± 5.7 NR
28 Control – 6.6 ± 1.0 7.8 ± 1.0 134.8 ± 9.7 NR 10.2 ± 1.3 155.8 ± 8.3 159.3 ± 6.1 NR
Kim (2019) Korea NRSI (cohort study) 31 GnRHa + rhGH 4.97 ± 0.95 NR 7.81 ± 0.97 NR 0.06 ± 1.15 9.25 ± 1.10 146.44 ± 3.56 156.5 ± 3.4 NR
135 GnRHa 3.69 ± 0.63 NR 7.91 ± 0.77 NR 0.39 ± 0.91 9.77 ± 0.84 150.76 ± 3.69 159.1 ± 3.6 NR
Korkmaz (2019) Turkey NRSI (cohort study) 25 GnRHa 2.1 ± 0.9 7.1 ± 1.0 8.0 ± 1.6 138.7 ± 7.5 NR 9.7 ± 2.3 165.4 ± 7.9 158.4 ± 5.1 NR
9 Control – 7.3 ± 0.5 8.4 ± 1.4 136.4 ± 9.6 NR 10.3 ± 2.1 163.5 ± 8.1 159.8 ± 5.8 NR
Lanes (2004) USA NRSI (cohort study) 16 GnRHa 2.7 ± 1.0 NR 8.8 ± 1.4 NR NR 10.8 ± 1.3 153.7 ± 1.2 157.7 ± 4.2 8.7 ± 1.1
21 Control – NR 8.5 ± 1.0 NR NR 9.7 ± 1.4 164.1 ± 4.1 162.2 ± 8.4 8.5 ± 1.4
Lazar (2002) Israel NRSI (cohort study) 63 GnRHa 2–4 8.34 ± 0.32 9.06 ± 0.44 137.1 ± 5.45 18.04 ± 1.54 11.11 ± 0.58 151.8 ± 5.8 157.67 ± 4.88 8.35 ± 2.45
63 Control – 8.5 ± 0.4 9.25 ± 0.47 135.8 ± 6.50 18.29 ± 2.38 11.27 ± 0.76 151.8 ± 5.8 157.96 ± 5.29 7.97 ± 2.46
Lazar (2015) Israel NRSI (cohort study) 100 GnRHa 2.9 ± 1.0 7.3 ± 1.2 8.3 ± 0.9 NR NR NR NR NR NR
42 Control – 7.5 ± 0.6 7.5 ± 0.6 NR NR NR NR NR NR
Lee (2007) Korea NRSI (cohort study) 17 GnRHa + rhGH 1 NR 9.75 ± 1.98 NR NR 11.29 ± 1.86 147.6 ± 5.2 155.0 ± 4.2 NR
18 GnRHa 1 NR 7.93 ± 1.85 NR NR 10.44 ± 1.62 149 ± 6.6 157.7 ± 3.6 NR
Li (2005) China NRSI (cohort study) 10 GnRHa + rhGH 1.55 ± 0.64 NR 9.7 ± 1.1 137.8 ± 8.0 NR 12.2 ± 0.5 149.5 ± 4.1 154.2 ± 4.1 NR
11 GnRHa 1.41 ± 0.54 NR 9.3 ± 0.8 138.2 ± 6.0 NR 11.8 ± 1.1 150.3 ± 3.0 158.6 ± 4.1 NR
Li (2013) China NRSI (cohort study) 21 GnRHa + rhGH 2.49 ± 1.10 7.31 ± 0.60 10.83 ± 1.07 NR NR 12.22 ± 0.83 152.61 ± 3.92 154.39 ± 4.72 2.80 ± 0.50
22 GnRHa 1.86 ± 0.69 6.96 ± 1.05 10.63 ± 1.26 NR NR 12.41 ± 0.89 152.54 ± 5.86 155.60 ± 4.52 3.92 ± 1.10
Liang (2015) China NRSI (cohort study) 23 GnRHa + rhGH >2 6.18 ± 0.39 7.63 ± 0.28 126.57 ± 1.83 15.96 ± 0.31 8.36 ± 0.30 159.99 ± 0.99 156.15 ± 0.80 7.25 ± 0.29
17 GnRHa >2 6.77 ± 0.32 8.13 ± 0.20 132.79 ± 1.57 17.85 ± 0.53 9.18 ± 0.29 161.56 ± 0.91 158.29 ± 0.91 7.57 ± 0.36
Lu (2018) China NRSI (cohort study) 253 GnRHa NR NR NR NR NR NR 149.3 ± 3.5 159.7 ± 3.1 NR
50 Control – NR NR NR NR NR 149.7 ± 5.2 158.4 ± 4.3 NR
Luo (2015) China RCT 30 GnRHa + rhGH 1.63 ± 0.41 7.05 ± 1.76 8.46 ± 2.12 127.99 ± 32.01 16.93 ± 4.24 10.53 ± 2.64 151.36 ± 37.84 158.64 ± 39.66 NR
30 GnRHa 1.59 ± 0.41 7.13 ± 1.78 8.52 ± 2.13 128.37 ± 32.13 17.26 ± 4.32 10.69 ± 2.68 151.28 ± 37.82 158.45 ± 39.62 NR
Ma (2020) China NRSI (cohort study) 32 GnRHa + rhGH 2.00–5.75 NR 8.75 ± 1.50 NR NR 10.89 ± 1.12 151.10 ± 6.25 NR NR
48 GnRHa 2.00–5.75 NR NR NR 154.08 ± 6.27 NR NR
Magiakou (2010) Greece NRSI (cohort study) 33 GnRHa 2.75 (Median) 6.75 (Median) 7.92 (Median) NR NR 10 (Median) 151.53 (Median) NR NR
14 Control – 6.625 (Median) 7.96 (Median) NR NR 10.75 (Median) 154.265 (Median) NR NR
Mul (2005) Netherlands RCT 14 GnRHa + rhGH 3 NR 9.6 ± 0.9 135.1 ± 5.7 NR 11.6 ± 0.8 146.8 ± 4.8 NR NR
12 GnRHa 3 NR 9.6 ± 0.9 133.8 ± 8.7 NR 10.7 ± 1.1 149.8 ± 5.6 NR NR
Mul (2001) Netherlands RCT 14 GnRHa + rhGH 3 NR 9.6 ± 0.9 135.1 ± 5.7 NR 11.6 ± 0.8 146.8 ± 4.8 NR 6.7 ± 1.59
13 GnRHa 3 NR 9.7 ± 0.9 133.6 ± 8.4 NR 10.8 ± 1.1 149.3 ± 5.7 NR 5.4 ± 1.15
Pasquino (1999) Italy NRSI (cohort study) 10 GnRHa + rhGH 3.07 ± 1.33 6.3 ± 0.4 7.9 ± 0.6 NR NR 10.6 ± 0.4 152.7 ± 1.7 155.6 ± 2.0 8.3 ± 0.8
10 GnRHa 2–3 5.7 ± 0.6 7.6 ± 0.2 NR NR 10.4 ± 0.3 155.5 ± 2.0 155.5 ± 2.1 NR
Pasquino (2008) Italy NRSI (cohort study) 87 GnRHa 4.2 ± 1.6 6.5 ± 1.5 8.4 ± 1.5 134.8 ± 9.3 18.5 ± 2.4 11.1 ± 1.6 150.0 ± 5.1 NR 8.2 ± 1.8
32 Control – 6.8 ± 1.6 8.3 ± 1.2 136.0 ± 8.9 NR 11.2 ± 1.4 151.0 ± 3.9 NR NR
Paul (1995) USA NRSI (cohort study) 20 GnRHa 5.7 ± 2.3 3.9 ± 2.1 NR NR NR NR NR NR NR
93 Control – 4.9 ± 2.3 NR NR NR NR NR NR NR
Peng (2016) China NRSI (cohort study) 224 GnRHa 1.5 7.24 ± 2.16 8.5 120.44 ± 7.82 16.67 ± 3.04 NR NR NR NR
64 Control – 7.16 ± 1.97 NR 118.25 ± 6.93 16.52 ± 3.76 NR NR NR NR
Poomthavorn (2011) Thailand NRSI (cohort study) 47 GnRHa 3.4 ± 1.5 7.2 ± 1.0 8.3 ± 1.0 134.9 ± 7.5 19.2 ± 3.0 11.1 ± 1.7 150.8 ± 5.5 155.8 ± 4.1 9
11 Control – 7.2 ± 0.8 8.6 ± 1.4 135.4 ± 8.1 19.2 ± 3.7 11.2 ± 2.4 152.8 ± 7.2 154.5 ± 4.2 NR
Pucarelli (2003) Italy NRSI (cohort study) 17 GnRHa + rhGH 2–4 6.6 ± 1.0 8.3 ± 1.6 NR NR 11.0 ± 1.4 153.2 ± 5. 0 157.4 ± 4.8 NR
18 GnRHa 2–4 6.0 ± 2.0 7.9 ± 0.8 NR NR 10.7 ± 1.2 153.9 ± 3.8 157.2 ± 6.0 NR
Satitpatanapan (2020) Thailand NRSI (cohort study) 41 GnRHa 2.5 ± 1.5 NR 7.0 ± 1.7 NR NR 10.5 ± 2.0 149.0 ± 5.1 151.6 ± 3.2 NR
26 Control – NR 8.4 ± 1.2 NR NR 13.3 ± 1.2 147.1 ± 7.3 151.2 ± 4.2 NR
Savas-Erdeve (2016) Turkey NRSI (cohort study) 41 GnRHa 2 NR 8.6 ± 0.8 NR NR 10.7 ± 1.4 158.9 ± 6.3 158.1 ± 5.2 NR
8 Control – NR 8.83 ± 1.04 NR NR 9.98 ± 2.05 161.6 ± 4.06 158.9 ± 3.65 NR
Swaiss (2017) Jordan NRSI (cohort study) 39 GnRHa 3.0 ± 1.4 NR 7.11 ± 0.7 131.3 ± 9.2 NR 10.1 ± 1.6 158.5 ± 10.8 160.4 ± 5.6 NR
11 Control – NR 7.3 ± 0.6 130.7 ± 12.3 NR 10.0 ± 1.9 153.8 ± 8.7 160.4 ± 3.6 NR
Tuvemo (2004) Sweden RCT 24 GnRHa + rhGH 2.8 ± 0.66 NR 8.4 ± 0.78 132.3 ± 5.55 NR 10.8 ± 0.9 163.1 NR 8.2
22 GnRHa 3.4 ± 0.74 NR 8.2 ± 0.83 130.0 ± 7.39 NR 10.2 ± 1.2 163.6 NR 6
Wang (2014) China NRSI (cohort study) 31 GnRHa + rhGH 2.08 ± 0.58 7.7 ± 0.5 9.2 ± 0.7 134 ± 7 17.1 ± 2.1 11.2 ± 0.53 153 ± 7 157 ± 4 NR
49 GnRHa 2.17 ± 0.75 7.6 ± 0.8 8.9 ± 0.6 134 ± 6 17.8 ± 2.7 11.0 ± 0.50 153 ± 5 157 ± 3 NR
Wu (2014) China NRSI (cohort study) 15 GnRHa + rhGH 1.30 ± 0.51 NR 8.20 ± 0.90 132.37 ± 3.65 NR 11.05 ± 0.63 149.55 ± 3.62 NR NR
28 GnRHa 1.52 ± 0.34 NR 7.85 ± 0.61 131.56 ± 3.25 NR 10.35 ± 0.83 151.87 ± 4.05 NR NR
41 Control – NR 7.92 ± 0.60 129.92 ± 3.88 NR 10.66 ± 0.65 151.92 ± 3.35 NR NR
Wu (2023) China NRSI (cohort study) 101 GnRHa 2.48 ± 0.52 NR 8.15 ± 0.73 130.65 ± 4.84 16.27 ± 1.83 10.29 ± 0.76 150.93 ± 3.31 157.20 ± 3.40 NR
41 Control – NR 7.61 ± 1.06 129.16 ± 10.01 16.14 ± 1.83 9.18 ± 1.83 154.98 ± 4.74 156.90 ± 3.79 NR
Yan (2007) China RCT 6 GnRHa + rhGH 1 ± 0.25 NR 9.5 ± 1.0 135.0 ± 3.0 NR 11.5 ± 0.8 149.0 ± 2.0 NR NR
6 GnRHa 1.25 ± 0.21 NR 9.4 ± 0.8 133.2 ± 3.5 NR 11.3 ± 0.5 150.0 ± 1.5 NR NR
Yuan (2011) China NRSI (cohort study) 57 GnRHa 1.69 ± 0.43 7.03 ± 0.94 7.92 ± 0.93 130.63 ± 6.45 16.30 ± 1.46 10.12 ± 1.08 155.64 ± 6.40 158.66 ± 3.11 NR
77 Control – 7.26 ± 0.91 8.16 ± 0.76 132.49 ± 6.35 16.44 ± 1.79 9.78 ± 1.24 158.38 ± 7.46 158.29 ± 3.81 NR
Zheng (2011) China NRSI (cohort study) 26 GnRHa + rhGH 2.57 NR 9.41 ± 1.4 134.0 ± 7.8 NR 11.3 ± 1.4 148.1 ± 4.6 154.4 ± 4.6 3.2 ± 1.0
23 GnRHa 1.85 NR 9.5 ± 1.4 139.0 ± 6.1 NR 11.9 ± 0.9 150.3 ± 6.0 155.6 ± 4.3 NR
Zhu (2009) China RCT 8 GnRHa + rhGH 1.72 ± 0.42 NR 9.96 ± 1.3 NR NR NR NR NR NR
8 GnRHa 2.05 ± 0.23 NR 8.95 ± 0.8 NR NR NR NR NR NR
Zhu (2017) China NRSI (cohort study) 23 GnRHa + rhGH NR NR 10.9 ± 1.4 133.2 ± 7.6 NR 11.4 ± 1.3 147.9 ± 4.7 NR 3.1 ± 1.0
23 GnRHa NR NR 11.2 ± 1.3 138.8 ± 6.2 NR 12.0 ± 0.9 150.4 ± 6.1 NR NR
Boys
Carel (1999) France NRSI (cohort study) 8 GnRHa 4.7 ± 1.8 7.1 ± 3 9.1 ± 1.7 NR NR 11.6 ± 1.7 174.2 ± 6.6 171.8 ± 3.7 8.9 ± 1.9
26 Control – 4.8 ± 2.9 NR NR NR NR NR NR NR
Couto-Silva (2002) France NRSI (cohort study) 9 GnRHa 2 ± 0.1 10.2 ± 0.2 11.3 ± 0.2 140.3 ± 2.1 NR 11.2 ± 0.7 173.2 ± 3.1 170.4 ± 1.2 NR
31 Control – 10.3 ± 0.2 12.0 ± 0.6 147.0 ± 3.7 NR 12.5 ± 0.8 170.8 ± 2.7 170.2 ± 1.2 NR
Paul (1995) USA NRSI (cohort study) 6 GnRHa 5.7 ± 2.3 3.9 ± 2.1 NR NR NR NR NR NR NR
23 Control – 4.9 ± 2.3 NR NR NR NR NR NR NR
Ma (2020) China NRSI (cohort study) 7 GnRHa + rhGH NR NR NR NR NR NR NR NR NR
7 GnRHa NR NR NR NR NR NR NR NR NR
Open in a new tab

BA = bone age, BA–CA = bone age–chronological age, BMI = body mass index, GnRHa = gonadotrophin releasing hormone analogue, GV = growth velocity, H = height, NR = no report, NRSI = non-randomized studies of interventions, PAH = predicted adult height, RCT = randomized controlled trials, rhGH = recombinant human growth hormone, TH = target height.

Among the 11 RCTs, 2 studies (18.18%) were assessed as low risk of bias, 2 study (18.18%) was some concerns, and 7 studies (63.64%) were high risk of bias. While of the 44 NRSIs, there were 13 studies (29.55%) of moderate risk of bias and 31 studies (70.45%) of serious risk of bias. The results of the quality assessment of the included studies are provided in Appendix 4 and Appendix 5, Supplemental Digital Content, https://links.lww.com/MD/P576.

3.3. GnRHa alone versus no treatment

3.3.1. FH

A total of 32 studies (including 2 RCTs and 30 NRSIs) involving 2995 girls showed that a significantly higher FH was observed in the treated group than the untreated group (MD 2.51, 95% CI 1.26 to 3.76, P < .001; I² = 97%), among which the pooled effect from 2 RCTs showed no significant difference (MD 0.54, 95% CI −2.15–3.23, P = .69; I² = 0%) and from 12 NRSIs showed significant difference (MD 2.63, 95% CI 1.34–3.92, P < .001; I² = 97%) (Fig. 2, Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S1, Supplemental Digital Content, https://links.lww.com/MD/P578).

Figure 2.

Figure 2.

Open in a new tab

Forest plots of efficacy outcomes of the included studies. CI = confidence interval, FH = final height, MD = mean difference, NA = not available, NRSI = non-randomized studies of interventions, PAH = predicted adult height, RCT = randomized controlled trials, SDS = standard deviation score, TH = target height.

3.3.2. FH-PAH

Eight studies (including 1 RCT and 7 NRSIs) involving 897 girls showed that the treated group obtained a greater height gain than the untreated group (MD 3.25, 95% CI 1.96–4.55, P < .001; I² = 89%), among which the pooled effect respectively from 1 RCT and 7 NRSIs showed a similar result (RCT: MD 2.50, 95% CI 1.71–3.29, P < .001; NRSIs: MD 3.47, 95% CI 1.68–5.26, P < .001, I² = 91%) (Fig. 2, Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S2, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.3.3. FH SDS

Eleven studies (all NRSIs) involving 918 girls showed that no significant difference was observed in FH SDS between the treated group and the untreated group (MD 0.09, 95% CI −0.36 to 0.53, P = .69; I² = 98%) (Fig. 2, Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S3, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.3.4. FH–TH

Ten studies (all NRSIs) involving 914 girls showed that significant difference was observed in the difference between FH and TH between the treated group and the untreated group (MD 2.90, 95% CI 0.82–4.98, P = .006; I² = 96%) (Fig. 2, Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S4, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.3.5. Safety

Concerning the endocrine and metabolism adverse events compared GnRHa alone to no treatment, 12 studies (including 1 RCT and 11 NRSIs) involving 1238 girls reported BMI at FH and the meta-analysis showed that no significant difference was observed in BMI when the treated and untreated girls achieved to FH (MD −0.06, 95% CI −0.50 to 0.38, P = .78; I² = 48%), among which the BMI of only 1 NRSI was over 25.00 with no significant difference (Lazer et al, 2015) (Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S5, Supplemental Digital Content, https://links.lww.com/MD/P578).

Additionally, Pasquino et al evaluated the bone mineral density (BMD) at discontinuation of treatment that was significantly lower than in the control group (treated vs untreated: 0.82 ± 0.01 vs 1.001 ± 0.11, P < .001), while Magiakou et al found that there was no significant difference between the 2 groups on BMD at the FH of patients with CPP (treated vs untreated: 1.045 [median] vs 1.078 [median], P = .713). Regarding the reproductive function adverse events, 8 studies (including 1 RCT and 7 NRSIs) involving 850 girls showed that the incidence of PCOS significantly occurred in the treated group than in the untreated group (risk ratio 1.93, 95% CI 1.26–2.98, P = .003; I² = 0%) (Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S6, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.4. GnRHa plus rhGH versus GnRHa alone

3.4.1. FH

In total, 16 studies (including 4 RCTs and 12 NRSIs) involving 1181 girls indicated a significantly higher FH in receiving GnRHa plus rhGH group than those in receiving GnRHa alone group (MD 1.58, 95% CI 0.66–2.51, P < .001; I² = 84%), among which the pooled effect respectively from 4 RCTs and 12 NRSIs showed a similar result (RCTs: MD 4.30, 95% CI 0.60–8.00, P = .02, I² = 76%; NRSIs: MD 0.96, 95% CI 0.10–1.83, P = .03, I² = 81%) (Fig. 2, Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S7, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.4.2. FH-PAH

Thirteen studies (including 2 RCTs and 11 NRSIs) involving 1073 girls reported that the combination therapy group obtained a greater height gain than the monotherapy group (MD 3.43, 95% CI 2.24–4.63, P < .001; I² = 91%), among which the pooled effect respectively from 2 RCTs and 11 NRSIs showed a similar result (RCTs: MD 3.38, 95% CI 1.17–5.59, P = .003, I² = 0%; NRSIs: MD 3.43, 95% CI 2.14 to 4.72, P < .001, I² = 92%) (Fig. 2, Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S8, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.4.3. FH SDS

Six studies (including 2 RCTs and 4 NRSIs) involving 550 girls reported FH SDS, with no significant difference between the combination therapy group and the monotherapy group (MD −0.01, 95% CI −0.41 to 0.38, P = .94; I² = 83%), among which the pooled effect respectively from 2 RCTs and 4 NRSIs showed a similar result (RCTs: MD 0.29, 95% CI −0.19 to 0.78, P = .23, I² = 17%; NRSIs: MD −0.12, 95% CI −0.61 to 0.37, P = .62, I² = 89%) (Fig. 2, Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S9, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.4.4. FH SDS–PAH SDS

Two studies (all NRSIs) involving 248 girls reported the difference between FH SDS and PAH SDS compared the combination therapy group to the monotherapy group, and the results of the meta-analysis showed that there was a significant difference between the 2 groups (MD 0.33, 95% CI 0.04–0.62, P = .02; I² = 0%) (Fig. 2, Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S10, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.4.5. FH–TH

Ten studies (including 1 RCT and 9 NRSIs) involving 994 girls reported the difference between FH and TH, with significant difference between the combination therapy group and the monotherapy group (MD −2.52, 95% CI 1.63 to 3.40, P < .001; I² = 87%), among which the pooled effect respectively from 1 RCT and 9 NRSIs showed a similar result (RCT: MD 4.00, 95% CI 1.93–6.07, P < .001; NRSIs: MD 2.39, 95% CI 1.47–3.31, P < .001, I² = 88%) (Fig. 2, Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S11, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.4.6. ΔPAH

Eleven studies (including 4 RCTs and 7 NRSIs) involving 894 girls reported the difference of PAH between cessative and initial treatment, with significant difference between the combination therapy group and the monotherapy group (MD 3.36, 95% CI 3.23–3.49, P < .001; I² = 0%), among which the pooled effect respectively from 4 RCTs and 7 NRSIs showed a similar result (RCTs: MD 4.10, 95% CI 3.41–4.79, P < .001, I² = 0%; NRSIs: MD 3.33, 95% CI 3.20–3.46, P < .001, I² = 0%) (Fig. 2, Table S1, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S12, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.4.7. Safety

Concerning the endocrine and metabolism adverse events compared GnRHa plus rhGH group to GnRHa alone group, 1 NRSI involving 82 girls reported BMI at FH indicating that no significant difference was observed on BMI when the patients receiving GnRHa with or without rhGH achieved to FH (GnRHa + rhGH: 18 ± 4.27, GnRHa alone: 19.6 ± 1.86, P = .08). Besides, Li et al and Zheng et al reported that 4 of 21 and 5 of 26 patients were higher in IGF-1 and 3 of 21 and 4 of 26 patients were above the normal range in fasting insulin during receiving treatment of GnRHa plus rhGH, respectively. Regarding the reproductive function adverse events, only 1 NRSI involving 29 girls showed that the incidence of PCOS was not significantly different between the combination therapy group and the monotherapy group (GnRHa + rhGH: 54.55%, GnRHa alone: 22.22%, P = .08).

3.5. Effects in boys

A total of 3 studies (all NRSIs) involving 80 boys reported the FH compared receiving GnRHa group alone to no treatment group, and the results of the meta-analysis showed that no significant difference was observed between the 2 groups (MD 7.06, 95% CI −9.07 to 23.18, P = .39; I² = 97%). Two studies (all NRSIs) involving 46 boys that reported FH SDS showed that no significant difference was observed between the treated group and the untreated group (MD 0.21, 95% CI −0.07 to 0.50, P = .14; I² = 0%). Only 1 NRSI involving 14 boys reported height gain between the combination therapy group and the monotherapy group but with no significant difference (GnRHa + rhGH: 8.78 ± 5.2, GnRHa alone: 7.99 ± 4.82, P = .77) (Fig. 2, Table S2, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S13–S14, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.6. Meta-regression and subgroup analysis

The results of univariate meta-regression showed that the pooled effects of the primary outcomes between the treated group and the untreated group and between the combination therapy group and the monotherapy group were all not associated with the region (former groups: P = .256; latter groups: P = .399), study design (former groups: P = .508; latter groups: P = .059), sample size(former groups: P = .393; latter groups: P = .060), and risk of bias (former groups: P = .756; latter groups: P = .751), but were significantly affected by the population difference (former groups: P < .001; latter groups: P = .019), indicating that the population difference may be the source of high heterogeneity.

After the removal of the studies presenting population differences from the subgroup analysis, that is Adan (2002), Allali (2011), Bertelloni (1998), Brauner (1994), Lanes (2004), Yuan (2011) for GnRHa alone versus no treatment, and Cho (2023), Fu (2020), Kim (2019), Liang (2015), Wang (2014), Wu (2014), Zheng (2011) for GnRHa plus rhGH versus GnRHa alone, the heterogeneity decreased to below 90% but were still high in some subgroups. The FH and height gain in girls with onset puberty aged 6 to 8 years were significantly improved in the treated group than untreated group (FH: MD 4.83, 95% Cl 3.29–6.17; FH-PAH: MD 4.69, 95% Cl 3.16–6.23). A significant difference was shown in the height gain (MD: 3.16, 95% Cl 1.80–4.53) but no significant difference was observed in the FH (MD 0.28, 95% Cl −1.51 to 2.07) between the 2 groups among those patients whose puberty onset aged after 8 years. The pharmacologic intervention that began before 8 years of age was probably more effective than that after 8 years of age. Similar results were presented in the combination therapy group compared to the monotherapy group. In addition, there was no significant difference observed between the combination therapy group and the monotherapy group when the course of treatment was <2 years (Fig. 3, Table S3, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S15–S26, Supplemental Digital Content, https://links.lww.com/MD/P578).

Figure 3.

Figure 3.

Open in a new tab

Forest plots of subgroup analysis of primary outcomes omitting population difference in girls. CI = confidence interval, FH = final height, MD = mean difference, NA = not available, PAH = predicted adult height; P interaction = P value for heterogeneity between groups.

3.7. Publication bias and sensitivity analysis

Publication bias was suggested in the outcome of the height gain (FH-PAH) compared GnRHa plus rhGH to GnRHa alone in girls, but the Duval and Tweedy trim-and-fill analysis showed a similar result to the main findings. The other outcomes were not observed the publication bias, and the sensitivity analysis sequentially eliminating each study did not substantially differ from the primary results, other than the outcome of FH–TH compared GnRHa alone and no treatment in girls, which the result was reversed (MD 3.20, 95% CI −0.08 to 6.47, P = .06; I² = 96%) after omitting the study reported by Fu et al (Table S4, Supplemental Digital Content, https://links.lww.com/MD/P577, Figure S27–S35, Supplemental Digital Content, https://links.lww.com/MD/P578).

3.8. Certainty of evidence

For outcomes from RCTs, the quality of evidence was from moderate to low, while for outcomes from NRSIs (all were cohort studies) was from low to very low, which mainly result from the risk of bias and inconsistency. The detailed quality assessment of evidence is outlined in Appendix 6, Supplemental Digital Content, https://links.lww.com/MD/P576.

4. Discussion

The specific onset age of puberty has various definitions across different races and nations, while the appearance of secondary sexual characteristics after 10 years of age in girls and 11 years of age in boys is generally accepted as the normal initial age of puberty. The onset age before 8 years of age is defined as CPP and that between 8 and 10 years of age is considered EP in girls. In this meta-analysis of 55 studies, we assessed the comparative efficacy and safety among GnRHa plus rhGH, GnRHa alone, and no treatment for pediatric patients with CPP or EP, and estimated the efficacy difference under different age of puberty onset, age of intervention onset, and treatment course. The findings of our study showed that the GnRHa plus rhGH presented a better effectiveness in FH and height gain for girls than GnRHa alone, as well as GnRHa alone better than that of no treatment, with less impact on BMI at FH but probably increase in the risk of PCOS, while no significant difference was found among outcomes of height in boys due to the less available studies.

We included more published studies in the final analysis than previous systematic reviews, but meanwhile, high heterogeneity was also observed in most outcomes. The meta-regression showed that the population difference among the included studies might be the main reason accounting for the high heterogeneity. We noticed inconsistent baselines between the 2 groups in some trials, such as the comparison of fast progressive CPP in the treated group versus slow progressive CPP in the untreated groups, or that of suppressed GV (<5 cm per year) in the combination group versus normal GV in the monotherapy group, which were critical factors affecting the outcomes and contributing to the selection bias and population difference with no comparability before the intervention, and thus underestimating the actual effect size.

For more reliable results, we excluded the studies with inconsistent baselines before subgroup analysis. In terms of the final heigh, subgroup analysis found that pharmacotherapy was not beneficial to pediatric patients with puberty onset after 8 years old, neither GnRHa plus rhGH nor GnRHa alone. This effect might be attributed to the compensation mechanism that children who entered puberty after the age of 8 years still had remaining genetic potential and developed their height during a longer duration of growth,[30] even without treatment, so that they could achieve similar FH to those who puberty onset at 12 or 13 years old, which suggested pharmacotherapy did not largely affect the outcome on these population. That would also explain why no significant difference was presented in the pooled effects of FH from 2 RCTs (Bouvattier 1999 & Cassio 1999) compared GnRHa alone to no treatment. A similar finding was reported by Bertelloni et al[31] who published a meta-analysis indicating GnRHa had little effect on final adult height for EP with onset at age of 7 to 10 years. As for height gain, however, a significant difference was observed in the subgroup of puberty onset after 8 years that differed from the outcome of FH, probably because PAH was more damaged in the group who tended to be treated than to be untreated. While for CPP whose puberty onset before 8 years old, prematurely lost genetic potential may be associated with their shorter FH. We considered the clinical significance of FH was more important than that of height gain, which directly expressed the demand and satisfaction of treatment for pediatric patients and their families. In our analysis, pharmacotherapy that aiming to restrain maturation presented a significant effect than no treatment, among which GnRHa plus rhGH performed a higher FH than GnRHa alone, indicating rhGH had a certain active function on growth for CPP and was also in accordance with the previous review.[32]

We also found that receiving treatment began either before or after 8 years old can obtain better height than those no treatment, of which the 6 to 8 years group seemed to earn more height than after 8 years group, suggesting the early timing of intervention may have an important impact on the long-term outcome for patients. These results were consistent with Park findings[10] that showed the mean difference in adult height was higher in the girls who began treatment before 8 years old than after 8 years old. Furthermore, the pharmacologic intervention was still positive for precocious patients regardless of whether the course was sufficient for 2 years. However, compared to GnRHa alone, it appeared that GnRHa plus rhGH was no improvement in height if the course of treatment was <2 years, indicating the positive effects of rhGH may be also related to the duration of combination therapy. In contrast to the earlier findings from Liu et al,[32] the significant effectiveness was presented in the subgroup of 1 to 2 years treatment period of combination, however, we found that the subgroup in Liu et al analysis was incorrect in classification and indeed more than 2 years instead for some studies, leading to the different results with our study.

Additionally, we discussed the SDS of primary outcomes between 2 groups for the first time, which eliminated the differences among races, regions, and ages, and provided a comparatively synthesized efficacy of the included studies from different countries. While our findings showed that there was no significant difference in FH SDS either between GnRHa plus rhGH and GnRHa alone or between GnRHa alone and no treatment. The increase in FH SDS was limited even though taking medication, which might be because the genetics, nutrition, exercise, or sleep quality were potentially associated with height changes during puberty development of children in different income countries.[33] However, those factors were not quantitatively estimated in each trial and not reflected in our analysis due to the lack of corresponding measuring tools. Moreover, the pooled effect of (FH–TH) and ΔPAH also showed that pharmacotherapy can make the FH exceed the genetic height, and combination therapy improved more PAH than monotherapy after treatment.

The safety of the intervention with GnRHa alone or combined with rhGH was likewise of concern during treatment, however, there were little available data about safety reported in present studies, especially for combination with rhGH. With regard to the endocrine and metabolism adverse events, our review showed that the BMI at FH was not statistically different between the treated and untreated groups, and most evaluations had not exceeded the threshold of overweight and obesity, which was similar to the studies reported by Luo et al and Franzini et al,[34,35] indicating that the pharmacotherapy have fewer impacts on BMI for patients. Some trials reported that the BMI was changed during the intervention but recovered after treatment, which may be associated with the obesity status of the individual patient before pharmacotherapy.[36] The change of BMD was also pointed out in some studies but not synthesized in our analysis due to the different timing of follow-up, and it similarly appeared unaffected by the intervention with GnRHa as showed in previous studies.[36] On the other hand, the fluctuation of IGF-1 and fasting insulin potentially caused by rhGH was getting more attention as adverse events than BMI or BMD among the combination therapy group. Besides, other adverse events, such as hypothyroidism, scoliosis, and malignancy, were reported in some cases when rhGH was used for different diseases,[7,37] but not presented in the included studies, thus the confirmed association between the above adverse events and GnRHa combined with rhGH therapy for CPP was not yet confirmed by current evidence.

Our study pooling current evidence suggested that the incidence of PCOS was significantly higher in the GnRHa group than the untreated group, which was contrary to the results of the previous review[34] that included studies less than this meta-analysis. The long-term risk of the development of PCOS was a controversial issue for GnRHa intervention, and precocious puberty was also considered as one of the risk factors for PCOS due to the premature adrenarche presented at puberty onset in some patients,[38] thus it was still unclear whether the increased risk of PCOS associated with the GnRHa treatment or with the disease itself.[2] Considering the evidence mainly from NRSIs with low certainty, the interpretation should be cautious. It is recommended that large cohort studies are conducted to confirm the causality between GnRHa treatment and PCOS.

This systematic review also has some limitations. Current studies were most low quality and the pooling results had high heterogeneity, which we found mainly generated from the selection bias and different eligible criteria among original trials through meta-regression. Although subgroup analysis improved the reliability of our results to some extent, the positive results should still be interpreted with caution due to the evidence with low certainty. Besides, the targeted outcomes were not totally covered and reported by the included studies, and the available data about boys were insufficient due to their lower prevalence than girls, rendering that a valid conclusion was not drawn for these groups as planned. Despite the positive impact of rhGH plus GnRHa on height for CPP presented in our analysis, its safety was not yet confirmed based on fewer data. Given the expensive expenditure of rhGH and the stress associated with periodic injections, the necessity of combination therapy is insufficient, and the balance between benefit and risk should be taken into consideration, especially for poor families.

5. Conclusions

In conclusion, the pharmacologic intervention of GnRHa can improve the FH of girls with CPP and probably obtains more height for those whose initial treatment age is before 8 years. A better efficacy is presented in the GnRHa combined with rhGH than GnRHa alone, while the positive effects do not appear in the course of combination therapy for <2 years. The girls with EP whose puberty onset after 8 years of age seem to not be beneficial from the pharmacologic intervention. Additionally, pharmacotherapy has less impact on the BMI of those receiving treatment, but the risk of the development of PCOS likely increases in the long term. However, the information about the effects of pharmacotherapy on boys is still lacking and should be likewise taken seriously in clinical practice. Considering the weak evidence, prospective comparative studies with high quality are necessary to further confirmed our results.

Acknowledgments

The authors would like to thank the affiliated health facilities and organizations, the Group of People with Evidence-based Pharmacy Committee of Chinese Pharmaceutical Association for providing methodologies and valuable comments.

Author contributions

Data curation: Chunsong Yang.

Formal analysis: Chunsong Yang.

Investigation: Lingli Zhang.

Methodology: Zheng Liu, Linan Zeng.

Supervision: Zheng Liu, Jin Wu.

Validation: Zheng Liu, Linan Zeng.

Visualization: Linan Zeng.

Writing – original draft: Linan Zeng, Jin Wu.

Writing – review & editing: Lingli Zhang.

Supplementary Material

medi-104-e41936-s002.pdf (409.8KB, pdf)

Abbreviations:

BA
advanced bone age
BMD
bone mineral density
BMI
body mass index
CI
confidence interval
CPP
central precocious puberty
EP
early puberty
FH
final height
GnRHa
gonadotropin-releasing hormone agonist
GV
growth velocity
IGF-1
insulin-like growth factor 1
MD
mean difference
NRSI
non-randomized studies of intervention
PAH
predicted adult height
PCOS
polycystic ovary syndrome
RCT
randomized controlled trial
rhGH
recombinant human growth hormone
SDS
standard deviation score
TH
target height

This study was supported by the Fundamental Research Funds for the University (SCU2022D006), the Science and Technology Project of Health Commission of Sichuan Province (No. 23LCYJ030), the Cross-Straits Medicine Exchange Association (No. 2024-05), and Sichuan Provincial Natural Science Foundation (2025ZNSFSC0736).

The authors have no conflicts of interest to disclose.

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Supplemental Digital Content is available for this article.

How to cite this article: Yang C, Liu Z, Zeng L, Wu J, Zhang L. Pharmacotherapy for children with central precocious puberty or early puberty: A systematic review and meta-analysis. Medicine 2025;104:31(e41936).

Contributor Information

Chunsong Yang, Email: yangchunsong_123@126.com.

Zheng Liu, Email: 312235407@qq.com.

Linan Zeng, Email: egg15@aliyun.com.

Jin Wu, Email: wangdo620@163.com.

References

  • [1].Cheuiche AV, da Silveira LG, de Paula LCP, Lucena IRS, Silveiro SP. Diagnosis and management of precocious sexual maturation: an updated review. Eur J Pediatr. 2021;180:3073–87. [DOI] [PubMed] [Google Scholar]
  • [2].Kim YJ, Kwon A, Jung MK, et al. Incidence and prevalence of central precocious puberty in Korea: an epidemiologic study based on a national database. J Pediatr. 2019;208:221–8. [DOI] [PubMed] [Google Scholar]
  • [3].Teilmann G, Pedersen CB, Jensen TK, Skakkebaek NE, Juul A. Prevalence and incidence of precocious pubertal development in Denmark: an epidemiologic study based on national registries. Pediatrics. 2005;116:1323–8. [DOI] [PubMed] [Google Scholar]
  • [4].Eckert-Lind C, Busch AS, Petersen JH, et al. Worldwide secular trends in age at pubertal onset assessed by breast development among girls: a systematic review and meta-analysis. JAMA Pediatrics. 2020;174:e195881. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Bräuner EV, Busch AS, Eckert-Lind C, Koch T, Hickey M, Juul A. Trends in the incidence of central precocious puberty and normal variant puberty among children in Denmark, 1998 to 2017. JAMA Network Open. 2020;3:e2015665. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [6].Bangalore Krishna K, Fuqua JS, Rogol AD, et al. Use of gonadotropin-releasing hormone analogs in children: update by an international consortium. Horm Res Paediatr. 2019;91:357–72. [DOI] [PubMed] [Google Scholar]
  • [7].Subspecialty Group of Endocrinologic, Hereditary and Metabolic Diseases, the Society of Pediatrics, Chinese Medical Association; Editorial Board, Chinese Journal of Pediatrics. Expert consensus on the diagnosis and treatment of central precocious puberty (2022). Chin J Pediatr. 2023;61:16–22. [DOI] [PubMed] [Google Scholar]
  • [8].Mul D, Oostdijk W, Drop SL. Early puberty in girls. Best Pract Res Clin Endocrinol Metab. 2002;16:153–63. [DOI] [PubMed] [Google Scholar]
  • [9].Carel JC, Lahlou N, Roger M, Chaussain JL. Precocious puberty and statural growth. Hum Reprod Update. 2004;10:135–47. [DOI] [PubMed] [Google Scholar]
  • [10].Park HK, Choo MS, Shim YS. Adult height after gonadotropin-releasing hormone agonist treatment in girls with early puberty: a meta-analysis. Clin Endocrinol (Oxf). 202093:135–45. [DOI] [PubMed] [Google Scholar]
  • [11].Carel JC, Eugster EA, Rogol A, et al. Consensus statement on the use of gonadotropin-releasing hormone analogs in children. Pediatrics. 2009;123:e752–762. [DOI] [PubMed] [Google Scholar]
  • [12].Lin YC, Lin C-Y, Chee S-Y, et al. Improved final predicted height with the injection of leuprolide in children with earlier puberty: a retrospective cohort study. PLoS One. 2017;12:e0185080. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Onat P, Erdeve Ş S, Çetinkaya S, Aycan Z. Effect of gonadotropin-releasing hormone analog treatment on final height in girls aged 6–10 years with central precocious and early puberty. Turk Pediatri Arsivi. 2020;55:361–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Bouvattier C, Coste J, Rodrigue D, et al. Lack of effect of GnRH agonists on final height in girls with advanced puberty: a randomized long-term pilot study. J Clin Endocrinol Metab. 1999;84:3575–8. [DOI] [PubMed] [Google Scholar]
  • [15].Korkmaz O, Sari G, Mecidov I, Ozen S, Goksen D, Darcan S. The gonadotropin-releasing hormone analogue therapy may not impact final height in precocious puberty of girls with onset of puberty aged 6 - 8 years. J Clin Med Res. 2019;11:133–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [16].Savaş-Erdeve S, Şiklar Z, Hacihamdioğlu B, et al. Gonadotropin-releasing hormone analogue treatment in females with moderately early puberty: no effect on final height. J Clin Res Pediatr Endocrinol. 2016;8:211–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Anik A, Çatli G, Abaci A, Böber E. Effect of gonadotropin-releasing hormone agonist therapy on body mass index and growth in girls with idiopathic central precocious puberty. Indian J Endocrinol Metab. 2015;19:267–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Arcari AJ, Freire AV, Escobar ME, et al. One-year treatment with gonadotropin-releasing hormone analogues does not affect body mass index, insulin sensitivity or lipid profile in girls with central precocious puberty. J Pediatr Endocrinol Metab. 2019;32:181–6. [DOI] [PubMed] [Google Scholar]
  • [19].Park J, Kim JH. Change in body mass index and insulin resistance after 1-year treatment with gonadotropin-releasing hormone agonists in girls with central precocious puberty. Ann Pediatr Endocrinol Metab. 2017;22:27–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Leite AL, Galo E, Antunes A, et al. Do GnRH agonists really increase body weight gain? Evaluation of a multicentric Portuguese cohort of patients with central precocious puberty. Front Pediatr. 2022;10:816635. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Sinthuprasith P, Dejkhamron P, Wejaphikul K, Unachak K. Near final adult height, and body mass index in overweight/obese and normal-weight children with idiopathic central precocious puberty and treated with gonadotropin-releasing hormone analogs. J Pediatr Endocrinol Metab. 2019;32:1369–75. [DOI] [PubMed] [Google Scholar]
  • [22].Australasian Paediatric Endocrine Group. Recommendations from the Australasian paediatric endocrine group on measuring circadian hormones, short stature and delaying puberty. 2017. https://www.choosingwisely.org.au/recommendations/apeg4. Accessed May 2017.
  • [23].Chiavaroli V, Liberati M, D'Antonio F, et al. GNRH analog therapy in girls with early puberty is associated with the achievement of predicted final height but also with increased risk of polycystic ovary syndrome. Eur J Endocrinol. 2010;163:55–62. [DOI] [PubMed] [Google Scholar]
  • [24].Karavani G, Chill HH, Schachter-Safrai N, Lomnitz G, Gillis D, Bauman D. Gonadotropin releasing hormone analogue treatment of central precocious puberty is not associated with altered prevalence of polycystic ovary syndrome: a single center cohort study. Clin Diabetes Endocrinol. 2021;7:14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Pasquino AM, Municchi G, Pucarelli I, Segni M, Mancini MA, Troiani S. Combined treatment with gonadotropin-releasing hormone analog and growth hormone in central precocious puberty. J Clin Endocrinol Metab. 1996;81:948–51. [DOI] [PubMed] [Google Scholar]
  • [26].Cho AY, Shim YS, Lee HS, Hwang JS. Effect of gonadotropin-releasing hormone agonist monotherapy and combination therapy with growth hormone on final adult height in girls with central precocious puberty. Sci Rep. 2023;13:1264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].Jung MK, Song KC, Kwon AR, Chae HW, Kim DH, Kim H-S. Adult height in girls with central precocious puberty treated with gonadotropin-releasing hormone agonist with or without growth hormone. Ann Pediatr Endocrinol Metab. 2014;19:214–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [28].Liang Y, Wei H, Li J, et al. Effect of GnRHa 3.75 mg subcutaneously every 6 weeks on adult height in girls with idiopathic central precocious puberty. J Pediatr Endocrinol Metab. 2015;28:839–46. [DOI] [PubMed] [Google Scholar]
  • [29].Kim EY. Long-term effects of gonadotropin-releasing hormone analogs in girls with central precocious puberty. Korean J Pediatr. 2015;58:1–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Llop-Viñolas D, Vizmanos B, Closa Monasterolo R, Escribano Subías J, Fernández-Ballart JD, Martí-Henneberg C. Onset of puberty at eight years of age in girls determines a specific tempo of puberty but does not affect adult height. Acta Paediatr. 2004;93:874–9. [DOI] [PubMed] [Google Scholar]
  • [31].Bertelloni S, Massart F, Miccoli M, Baroncelli GI. Adult height after spontaneous pubertal growth or GnRH analog treatment in girls with early puberty: a meta-analysis. Eur J Pediatr. 2017;176:697–704. [DOI] [PubMed] [Google Scholar]
  • [32].Liu S, Liu Q, Cheng X, Luo Y, Wen Y. Effects and safety of combination therapy with gonadotropin-releasing hormone analogue and growth hormone in girls with idiopathic central precocious puberty: a meta-analysis. J Endocrinol Invest. 2016;39:1167–78. [DOI] [PubMed] [Google Scholar]
  • [33].World Health Organization. WHO child Growth Standards: Training Course on Child Growth Assessment. World Health Organization; 2008. [Google Scholar]
  • [34].Luo X, Liang Y, Hou L, Wu W, Ying Y, Ye F. Long-term efficacy and safety of gonadotropin-releasing hormone analog treatment in children with idiopathic central precocious puberty: a systematic review and meta-analysis. Clin Endocrinol (Oxf). 2021;94:786–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Franzini IA, Yamamoto FM, Bolfi F, Antonini SR, Nunes-Nogueira VS. GnRH analog is ineffective in increasing adult height in girls with puberty onset after 7 years of age: a systematic review and meta-analysis. Eur J Endocrinol. 2018;179:381–90. [DOI] [PubMed] [Google Scholar]
  • [36].van der Sluis IM, Boot AM, Krenning EP, Drop SL, de Muinck Keizer-Schrama SM. Longitudinal follow-up of bone density and body composition in children with precocious or early puberty before, during and after cessation of GnRH agonist therapy. J Clin Endocrinol Metab. 2002;87:506–12. [DOI] [PubMed] [Google Scholar]
  • [37].Bell J, Parker KL, Swinford RD, Hoffman AR, Maneatis T, Lippe B. Long-term safety of recombinant human growth hormone in children. J Clin Endocrinol Metab. 2010;95:167–77. [DOI] [PubMed] [Google Scholar]
  • [38].Rosenfield RL. Clinical review: identifying children at risk for polycystic ovary syndrome. J Clin Endocrinol Metab. 2007;92:787–96. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

medi-104-e41936-s001.pdf (745KB, pdf)
medi-104-e41936-s002.pdf (409.8KB, pdf)
medi-104-e41936-s003.pdf (2.7MB, pdf)

Articles from Medicine are provided here courtesy of Wolters Kluwer Health

RESOURCES