Skip to main content
Annals of Pediatric Endocrinology & Metabolism logoLink to Annals of Pediatric Endocrinology & Metabolism
. 2026 Oct 31;31(5):319–326. doi: 10.6065/apem.2550394.197

Triptorelin-stimulated gonadotropin concentrations for precocious puberty

Eunjung Koh 1, Chan Jong Kim 1,✉
PMCID: PMC13745950  PMID: 43000826

Abstract

Purpose

Subcutaneous triptorelin stimulation testing is used to evaluate girls with early breast development when intravenous gonadotropin-releasing hormone is unavailable, but the optimal sampling times and minimum test duration remain uncertain.

Methods

We retrospectively analyzed 186 girls younger than 8 years who presented with breast development and underwent subcutaneous triptorelin testing (0.1 mg), with serum luteinizing hormone (LH) and follicle-stimulating hormone measured at 0, 30, 60, 90, and 120 minutes. Central precocious puberty (CPP) was defined as peak LH ≥5.0 IU/L; those below this threshold were classified as premature thelarche (PT). We performed receiver operating characteristic analyses for stimulated LH at 30, 60, 90, and 120 minutes and compared LH trajectories using mixed-effects modeling.

Results

CPP was identified in 88 girls (47.3%). Using the 5.0 IU/L threshold, cumulative detection reached 90.9% by 60 minutes, 95.5% by 90 minutes, and 100% only by 120 minutes. Peak timing differed between groups: LH peaked mainly at 30–60 minutes in the CPP group (33.0% at 60 minutes; 29.5% at 30 minutes) and at 120 minutes in the PT group (51.0%). Diagnostic discrimination was excellent at all time points (area under the curve, 0.993–0.999), and LH trajectories differed significantly between groups (group-by-time interaction, P<0.001).

Conclusions

Although stimulated LH showed excellent discrimination at each sampled time point, sampling through 120 minutes was required to maximize sensitivity and minimize false-negative classification using a peak LH threshold of 5.0 IU/L.

Keywords: Precocious puberty, Gonadotropin-releasing hormone stimulation test, Luteinizing hormone, Subcutaneous injection, Triptorelin

Highlights

· Using peak LH ≥5 IU/L within the 120-minute triptorelin test as the reference, stopping at 60 or 90 minutes missed 9.1% or 4.5% of central precocious puberty cases, respectively. LH peaked predominantly at 30–60 minutes in central precocious puberty, versus 120 minutes in premature thelarche. Marked bone-age advancement in premature thelarche overlapped with central precocious puberty despite distinct stimulated LH responses, highlighting the limitations of skeletal maturation alone.

Introduction

Premature development of secondary sexual characteristics in girls before age 8 years, often accompanied by accelerated growth velocity, advanced bone age, pubertal gonadotropin profiles, and radiologic enlargement of the uterus and ovaries, may raise suspicion of central precocious puberty (CPP) [1–3]. Clinical presentations such as premature thelarche (PT), exaggerated thelarche, and slowly or rapidly progressive CPP likely represent a spectrum of hypothalamic-pituitary-gonadal (HPG) axis activation; however, for diagnostic and therapeutic purposes, they are often classified dichotomously as CPP or PT [1,4-7].

During the study period, intravenous gonadotropin-releasing hormone (GnRH) was not readily available in our center, necessitating the use of subcutaneous triptorelin as a practical diagnostic alternative [8-10]. Initial protocols included sampling for up to 180 minutes after subcutaneous triptorelin administration and demonstrated high diagnostic accuracy (area under the curve [AUC], 0.973; luteinizing hormone [LH] cutoff, 6 IU/L) [8,9]. Subsequent studies suggested that a 120-minute sampling window might be sufficient (LH cutoff 4.52 IU/L; 100% sensitivity), indicating that a shorter test may be feasible without compromising diagnostic performance [11,12]. Nevertheless, the optimal sampling kinetics and minimum duration of a clinically efficient triptorelin stimulation test remain to be definitively established.

Therefore, we retrospectively analyzed a cohort of 186 girls presenting with premature breast development who underwent a 120-minute subcutaneous triptorelin stimulation test. We compared LH and follicle-stimulating hormone (FSH) levels at 0, 30, 60, 90, and 120 minutes to determine the optimal sampling time points and evaluate the reliability of a streamlined 120-minute protocol for diagnosing CPP in real-world pediatric endocrine practice.

Materials and methods

1. Subjects

This retrospective study reviewed the medical records of 194 girls with premature breast development before 8 years of age who were evaluated and followed at the Endocrine Clinic, Chonnam National University Hospital from January 2022 to February 2023. Breast development was assessed by a pediatric endocrinologist and staged according to the Marshall and Tanner classification. Height age was determined using the National Standard Growth Curve of Korea’s Ministry of Public Health. Bone age was assessed using the Greulich and Pyle method [13].

This study was approved by the institutional review board (IRB) of Chonnam National University Hospital (IRB No. CNUH-2023-093). The requirements for informed consent and an opt-out procedure were waived due to the use of anonymized retrospective data.

2. Method

1) The GnRH agonist (triptorelin) stimulation test

An intravenous cannula was inserted into the left forearm for repeated blood sampling. Baseline blood was collected at 0 minutes for LH, FSH, and estradiol immediately before administration of the GnRH agonist triptorelin (0.1 mg; aqueous solution 0.1 mg/mL; Diphereline; Ipsen Pharma Biotech, France), which was injected subcutaneously into the left deltoid region. Subsequent blood samples for LH and FSH were obtained at 30, 60, 90, and 120 minutes after injection. The highest LH and FSH concentrations observed across poststimulation time points were defined as the peak values. Although some published subcutaneous triptorelin protocols include sampling up to 180 minutes, a 120-minute protocol is also commonly used and has demonstrated diagnostic utility [9,11,12]. Extending the test duration to 180 minutes generally increases visit duration and the number of blood draws, which may limit feasibility in routine outpatient practice.

2) Hormone assays

Serum LH, FSH, and estradiol concentrations were measured by direct chemiluminescence on an ADVIA Centaur Immunoassay System (Siemens Healthcare Diagnostics, USA). The manufacturer-reported limits of detection are 0.1 IU/L for LH, 0.05 IU/L for FSH, and 43.3 pmol/L for estradiol. However, during the study period, our laboratory’s validated lower reportable limits were 1.00 IU/L for LH and 2.00 IU/L for FSH, based on internal verification and quality-control criteria. Concentrations below these limits were reported in the laboratory information system as <1.00 IU/L for LH and <2.00 IU/L for FSH, without numerical values. For statistical analysis, LH and FSH values below their respective lower reportable limits were assigned values of 1.00 and 2.00 IU/L, respectively. Consequently, summary statistics and analyses involving these substituted values should be interpreted with caution. A peak stimulated LH cutoff of 5 IU/L is commonly used to support a diagnosis of CPP in clinical practice, although thresholds vary by assay and protocol [14-16]. Sellar magnetic resonance imaging (MRI) was performed in 70 of 88 patients with CPP (79.5 %) to exclude intracranial pathology. Patients diagnosed with idiopathic CPP were treated with monthly subcutaneous injections of leuprolide acetate depot (3.75 mg).

3. Statistical analyses

Statistical analyses were performed using IBM SPSS Statistics ver. 29.0 (IBM Corp., USA). Continuous variables are expressed as means and standard deviations. An independent-samples t-test was used to compare clinical data between the 2 groups, and 1-way analysis of variance (ANOVA) was used to compare data across 3 or 4 groups. When significant differences were detected by ANOVA, post hoc pairwise comparisons were conducted using Tukey honestly significant difference test to adjust for multiple comparisons. A 2-sided P value <0.05 was considered statistically significant.

Receiver operating characteristic (ROC) analyses were performed for stimulated LH at 30, 60, 90, and 120 minutes, using the final classification (CPP vs. PT) as the reference standard. At each sampling time, the AUC and its 95% confidence interval were calculated; sensitivity and specificity were estimated for the conventional LH cutoff of 5.0 IU/L and for the optimal cutoff determined by the Youden index.

To compare serial LH responses between groups, we fitted a linear mixed-effects model with fixed effects for group, time, and the group-by-time interaction and a subject-specific random intercept. A first-order autoregressive covariance structure was used to account for within-subject correlation across adjacent time points. Estimated marginal means were compared at each time point with Sidak adjustment for multiple comparisons.

Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared (kg/m2). BMI standard deviation scores (SDS) were derived from the 2017 Korean National Growth Charts.

Results

A total of 194 girls presenting with breast development before 8 years of age were assessed. Eight girls with conditions known to affect pubertal timing, including McCune-Albright syndrome and thyroid dysfunction, were excluded, leaving 186 girls for analysis. All 186 girls underwent a subcutaneous triptorelin stimulation test; 88 (47.3%) met the diagnostic criteria for CPP and 98 (52.7%) were classified as PT. Sellar MRI was recommended for all girls with CPP and performed in 70 of 88 (79.5%). Incidental findings included Rathke’s cleft cysts in 10 patients, anterior pituitary hyperplasia in 3, a pineal cyst in 1, and an arachnoid cyst in 1.

Auxological and biochemical characteristics differed between CPP and PT (Table 1). Girls with CPP were older and had more advanced skeletal maturation than girls with PT, with higher chronological age, bone age, and bone age minus chronological age (BA–CA). Basal LH, peak stimulated LH, peak LH/FSH ratio, and estradiol concentrations were also higher in CPP than in PT (all P<0.001). Basal LH was below the laboratory reportable range (<1.0 IU/L) in 164 of 186 girls (88.2%), and these values were not released numerically. This included all girls with PT (98 of 98, 100%) and 66 of 88 girls with CPP (75.0%). Accordingly, formal diagnostic performance analyses of basal LH at low thresholds were not feasible in this cohort.

Table 1.

Auxological and biochemical parameters of patients with CPP and PT

Characteristic CPP group (n=88) PT group (n=98) P-value
CA (yr) 7.77±0.73 7.21±1.01 <0.001*
BA (yr) 9.85±1.05 8.95±1.26 <0.001*
BA–CA (yr) 2.04±0.89 1.72±0.93 0.020*
Height age (yr) 8.80±1.24 8.17±1.15 <0.001*
Height SDS 1.33±1.17 1.32±1.13 0.935
BMI SDS 0.60±1.08 0.71±1.16 0.524
Breast budding onset (yr) 7.14±1.01 6.84±1.38 0.101
Basal LH (IU/L) 1.42±1.17 <1.00 <0.001*
Basal FSH (IU/L) 4.16±1.92 2.91±0.90 <0.001*
Basal LH/FSH 0.36±0.17 0.37±0.10 0.380
Basal E2 (pmol/L) 26.44±15.54 18.72±7.33 <0.001*
Peak LH (IU/L) 17.87±20.72 2.74±1.00 <0.001*
Peak FSH (IU/L) 20.67±9.01 18.53±6.07 0.056
Peak LH/FSH 1.13±0.88 0.38±0.11 <0.001*

Values are presented as mean ± standard deviation unless otherwise indicated.

LH and FSH values below their respective lower reportable limits were assigned values of 1.00 and 2.00 IU/L, respectively, for statistical analysis. In the PT group, all basal LH values were below the lower reportable limit and are therefore presented as <1.00 IU/L. CPP, central precocious puberty; PT, premature thelarche; CA, chronological age; BA, bone age; SDS, standard deviation score; BMI, body mass index; LH, luteinizing hormone; FSH, follicle-stimulating hormone; E2, estradiol.

*

P <0.05 indicates statistical significance based on an independent-samples t-test.

Using a peak LH cutoff of 5.0 IU/L to define CPP, most girls with CPP exceeded the cutoff early, but additional cases were captured with later sampling (Table 2). At 30 minutes, 73 of 88 patients (83.0%) had LH ≥5 IU/L. At both 90 and 120 minutes, 82 of 88 (93.2%) met the cutoff at each individual time point. When the test was interpreted cumulatively (i.e., positivity at any time up to a given sampling point), 80 of 88 (90.9%) met the cutoff by 60 minutes, 84 of 88 (95.5%) by 90 minutes, and 88 of 88 (100%) by 120 minutes.

Table 2.

Proportion of CPP patients with LH levels ≥5.0 IU/L at each sampling time point after triptorelin stimulation

Variable Basal (0 min) 30 Min 60 Min 90 Min 120 Min
LH ≥5.0 IU/L at each time point 2 (2.3) 73 (83.0) 77 (87.5) 82 (93.2) 82 (93.2)
LH ≥5.0 IU/L up to each time point 2 (2.3) 73 (83.0) 80 (90.9) 84 (95.5) 88 (100)

Values are presented as number (%).

CPP, central precocious puberty; LH, luteinizing hormone.

The distribution of time to peak LH (the sampling time with the highest observed LH during the 0- to 120- minute test) differed between diagnostic groups (Fig. 1). In the CPP group, peak LH most commonly occurred at 60 minutes (29 of 88 [33.0%]) or 30 minutes (26 of 88 [29.5%]). In the PT group, peak LH most commonly occurred at 120 minutes (50 of 98 [51.0%]). In 4 girls with PT, LH did not rise after triptorelin; therefore, the maximal value occurred at baseline (0 minute).

Fig. 1.

Fig. 1.

Distribution of time to peak LH after subcutaneous triptorelin stimulation in CPP and PT groups. Peak LH was defined as the maximal observed LH concentration during the 120-minute sampling period. Values are shown as the number of patients within each group whose peak LH occurred at each time point. LH, luteinizing hormone; CPP, central precocious puberty; PT, premature thelarche.

To further characterize heterogeneity within the PT group, patients were stratified by skeletal advancement at presentation (BA–CA ≥2 years vs. <2 years) (Table 3). Girls in the PT subgroup with BA–CA ≥2 years had a bone age similar to that of girls with CPP (mean, 9.82 years vs. 9.85 years), whereas girls in the PT subgroup with BA–CA <2 years had a lower bone age (mean, 8.58 years; P<0.001). The PT subgroup with BA–CA ≥2 years was younger than the CPP group (mean chronological age, 7.03 vs. 7.77 years) and therefore had a larger BA–CA difference (2.79 years vs. 2.04 years; P<0.001). Height SDS was highest in the PT subgroup with BA–CA ≥2 years (2.12), intermediate in the CPP group (1.33), and lowest in the PT subgroup with BA–CA <2 years (0.98; P<0.001), whereas BMI SDS and age at breast budding did not differ significantly. Basal LH, basal FSH, peak LH, and the peak LH/FSH ratio were higher in CPP than in either PT subgroup (peak LH, 17.87 IU/L vs. 2.64 and 2.78 IU/L; P<0.001). Thus, girls in the PT subgroup with BA–CA ≥2 years resembled girls with CPP in terms of absolute bone age but not in their biochemical response to stimulation.

Table 3.

Clinical characteristics of PT subgroups (BA–CA ≥2 years vs. <2 years) and CPP

Characteristic CPP (n=88) BA–CA ≥2 yr in PT (n=29) BA–CA <2 yr in PT (n=69) P-value
CA (yr) 7.77±0.73†,§ 7.03±0.86 7.28±1.06 <0.001*
BA (yr) 9.85±1.05§ 9.82±1.22‡ 8.58±1.09 <0.001*
BA–CA (yr) 2.04±0.89†,§ 2.79±0.73‡ 1.28±0.57 <0.001*
Height age (yr) 8.80±1.24§ 8.36±1.19 8.10±1.14 0.001*
Height SDS 1.33±1.17† 2.12±0.84‡ 0.98±1.06 <0.001*
BMI SDS 0.60±1.08 1.03±1.26 0.57±1.10 0.141
Breast budding onset (yr) 7.14±1.01 6.76±1.13 6.87±1.47 0.242
Basal LH (IU/L) 1.42±1.17†,§ 1.00±0.00 1.00±0.00 0.002*
Basal FSH (IU/L) 4.16±1.92†,§ 2.67±0.93 3.02±1.60 <0.001*
Basal LH/FSH 0.36±0.17 0.41±0.10 0.36±0.10 0.204
Basal E2 (pmol/L) 26.44±15.54†,§ 19.15±9.06 18.53±6.53 <0.001*
Peak LH (IU/L) 17.87±20.72†,§ 2.64±1.13 2.78±0.96 <0.001*
Peak FSH (IU/L) 20.67±9.01 16.98±6.37 19.18±5.87 0.070
Peak LH/FSH 1.13±0.88†,§ 0.41±0.11 0.37±0.10 <0.001*

Values are presented as mean±standard deviation.

PT, premature thelarche; BA, bone age; CA, chronological age; CPP, central precocious puberty; SDS, standard deviation score; BMI, body mass index; LH, luteinizing hormone; FSH, follicle-stimulating hormone; E2, estradiol.

*

P <0.05 indicates statistical significance based on 1-way analysis of variance. Post hoc comparisons were conducted using Tukey honestly significant difference test.

†

P <0.05 for CPP versus BA–CA ≥2 years in PT.

‡

P <0.05 for BA–CA ≥2 years in PT versus BA–CA <2 years in PT.

§

P <0.05 for CPP versus BA–CA <2 years in PT.

Among girls with CPP, clinical and hormonal features were broadly similar across peak-LH timing groups (30, 60, 90, and 120 minutes) (Table 4). Bone age and BMI SDS did not differ significantly by peak timing. By contrast, the basal LH/FSH ratio and peak FSH differed across groups (overall P<0.05): the basal LH/FSH ratio was higher in the The basal LH/FSH ratio was higher in the 60-minute peak group than in the 120-minute peak group (P=0.022), and peak FSH concentrations were significantly higher in the 60-minute peak group than in the 120-minute peak group (P=0.025).

Table 4.

CPP patients stratified by time to peak LH (30, 60, 90, and 120 minutes)

Characteristic 30 Min (n=26) 60 Min (n=29) 90 Min (n=9) 120 Min (n=24) P-value
CA (yr) 7.63±1.09 7.80±0.51 7.72±0.69 7.87±0.42 0.676
BA (yr) 9.63±1.09 9.83±1.14 9.93±1.25 10.07±0.78 0.514
BA–CA (yr) 2.00±0.83 1.88±1.00 2.20±0.81 2.20±0.88 0.570
Height age (yr) 8.84±1.47 8.64±1.14 8.81±1.21 8.92±1.14 0.878
Height SDS 1.53±1.23 1.26±1.10 1.35±1.09 1.18±1.25 0.746
BMI SDS 0.72±1.14 0.42±1.10 0.98±1.00 0.54±1.00 0.512
Breast budding onset (yr) 6.91±1.33 7.23±0.76 6.82±1.09 7.39±0.75 0.266
Basal LH (IU/L) 1.38±0.84 1.67±1.69 1.43±1.31 1.14±0.29 0.442
Basal FSH (IU/L) 4.55±1.81 3.75±1.98 3.88±1.74 4.33±2.03 0.437
Basal E2 (pmol/L) 27.71±14.71 21.10±13.30 24.25±17.52 31.78±16.39 0.078
Peak LH (IU/L) 18.96±14.16 23.24±26.62 19.54±34.09 9.57±5.28 0.113
Peak FSH (IU/L) 21.21±9.99 22.99±8.40 23.89±12.73 16.08±4.82 0.022*
Basal LH/FSH 0.31±0.11 0.43±0.22 0.36±0.16 0.30±0.10 0.019*
Peak LH/FSH 1.17±0.68 1.39±1.15 0.96±1.10 0.81±0.44 0.111

Values are presented as mean±standard deviation.

CPP, central precocious puberty; LH, luteinizing hormone; CA, chronological age; BA, bone age; SDS, standard deviation score; BMI, body mass index; FSH, follicle-stimulating hormone; E2, estradiol.

*

P <0.05 indicates statistical significance based on 1-way analysis of variance. Post hoc comparisons were conducted using Tukey honestly significant difference test.

P <0.05 for 60 minutes versus 120 minutes.

In ROC analyses using the full 0- to 120-minute test (peak LH ≥5 IU/L) as the reference standard, LH values at each sampled time point discriminated CPP from PT with high accuracy (Table 5; Fig. 2). The AUCs were 0.993 at 30 minutes, 0.998 at 60 minutes, 0.999 at 90 minutes, and 0.996 at 120 minutes. The optimal cutoff values at each time point were determined using the Youden index, and the corresponding sensitivity and specificity are summarized in Table 5.

Table 5.

Receiver operating characteristic performance of LH at 30, 60, 90, and 120 minutes during the triptorelin stimulation test for classification of CPP, using the 0- to 120-minute test (peak LH ≥5.0 IU/L). as the reference standard

Sampling time (min) AUC (95% CI) Cutoff value (IU/L) Sensitivity (%) Specificity (%) Youden J
30 0.993 (0.987–1.000) 4.01 93.2 98.0 0.911
60 0.998 (0.995–1.000) 4.14 100.0 95.9 0.959
90 0.999 (0.998–1.000) 4.56 97.7 100.0 0.977
120 0.996 (0.991–1.000) 4.16 98.9 94.9 0.938

LH, luteinizing hormone; GnRH, gonadotropin-releasing hormone; CPP, central precocious puberty; AUC, area under the curve; CI, confidence interval.

Fig. 2.

Fig. 2.

ROC curves for LH at each sampling time point during triptorelin stimulation. ROC curves for LH measured at 30, 60, 90, and 120 minutes were used to discriminate CPP from PT, with CPP defined by the full 0- to 120-minute test (peak LH ≥5 IU/L). The areas under the curve were 0.993, 0.998, 0.999, and 0.996 at 30, 60, 90, and 120 minutes, respectively. The diagonal line indicates chance discrimination. Sensitivity and specificity at the Youden-indexderived optimal cutoff values for each time point are provided in Table 5. LH, luteinizing hormone; ROC, receiver operating characteristic; CPP, central precocious puberty; PT, premature thelarche.

To compare poststimulation LH dynamics between CPP and PT, we fitted a linear mixed-effects model with group, time (30, 60, 90, and 120 minutes), and a group-by-time interaction. The interaction was significant (F [3,736]=8.30, P<0.001), indicating that LH trajectories differed between groups over time. The main effects of group (F [1,736]=57.53, P<0.001) and time (F [3,736]=9.01, P<0.001) were also significant. In Sidak-adjusted comparisons at each time point, LH concentrations were higher in CPP than in PT at 30 minutes (mean difference, 13.45 IU/L; P<0.001), 60 minutes (14.26 IU/L, P<0.001), 90 minutes (13.10 IU/L, P<0.001), and 120 minutes (12.02 IU/L, P<0.001).

Discussion

Over the past several years, increasing referrals for early pubertal development have heightened the need for stimulation protocols that preserve diagnostic accuracy while minimizing the burden on patients and clinics. In this retrospective cohort of 186 girls evaluated for breast development before 8 years of age, 88 (47.3%) were classified as having CPP and 98 (52.7%) as having PT using a peak LH threshold of 5 IU/L within a 0- to 120-minute subcutaneous triptorelin stimulation test. Two findings are most directly applicable to clinical practice: LH response kinetics differed between girls classified as CPP and those classified as PT, and when CPP was defined as peak LH ≥5 IU/L within the 0- to 120- minute reference test, sampling through 120 minutes was required to maximize sensitivity in this cohort.

Although we operationally classified participants as CPP or PT using a biochemical threshold, these entities are not strictly mutually exclusive in biological terms [7]. Contemporary clinical experience and the literature support a continuum of HPG-axis activation, with substantial overlap at first presentation [1,4-7]. A stimulation test therefore captures endocrine responsiveness during a single assessment and may not fully predict the subsequent pace of pubertal progression. When initial findings are equivocal, our data support a longitudinal approach incorporating serial pubertal staging, growth-velocity monitoring, and interval bone age assessment, with hormonal results interpreted within the overall clinical trajectory [1].

The subgroup of PT patients with BA–CA ≥2 years warrants particular attention. Despite biochemical classification as PT, these girls showed skeletal maturation patterns that overlapped with CPP: their absolute bone age was similar to that of girls classified as CPP, but they were younger chronologically, resulting in a larger BA–CA difference (Table 3). Their hormonal profile, however, remained distinct from CPP, with substantially lower peak LH and peak LH/FSH ratios (Table 3), consistent with the possibility that marked bone-age advancement can occur without commensurate biochemical activation [17]. Clinically, BA–CA ≥2 years should be viewed as a marker for closer surveillance rather than a stand-alone surrogate for CPP. Among patients with follow-up data, progression to CPP occurred in 3 of 29 patients (10.3%) with BA–CA ≥2 years and 6 of 69 (8.7%) with BA–CA <2 years, indicating only a modest difference in progression risk and supporting management guided by the pace of progression over time rather than a single bone-age measure.

We observed distinct LH kinetic patterns after subcutaneous triptorelin: most girls classified as CPP reached peak LH concentrations at 30–60 minutes, whereas those with PT more commonly peaked at 120 minutes (Fig. 1). Delayed peaks in a substantial proportion of PT patients and in a subset of CPP patients may reflect interindividual variability in pituitary responsiveness and the specific pharmacokinetics of subcutaneous administration [8,9,11,12]. Within the CPP group, the absence of an association between peak timing and bone age or BMI SDS (Table 4) suggests that peak timing reflects pituitary response dynamics rather than somatic maturation alone.

Although ROC analyses demonstrated exceptionally high discriminative performance at individual time points, with AUC values reaching 0.999 at 90 minutes (Table 5), these metrics require careful interpretation. Because the reference standard for CPP in this study was defined by the peak LH achieved within the same 0- to 120-minute window, the high AUC values primarily reflect internal concordance within our cohort rather than independent external validation.

From an operational standpoint, the key issue is the trade-off between shorter protocols and the risk of false-negative classification. Although 90.9% of CPP cases exceeded the diagnostic threshold by 60 minutes, cumulative sensitivity reached 100% only when sampling was extended to 120 minutes (Table 2). Thus, although single-time-point measurements, such as those at 60 or 90 minutes, provide substantial information, they do not eliminate the practical risk of missing delayed responders. Given that assay results are typically unavailable in real time, completing the full 120-minute protocol remains essential to ensure diagnostic sensitivity and avoid false negatives in real-world pediatric practice.

Basal LH is a useful adjunct in the evaluation of early pubertal development, and a clearly elevated basal LH can support HPG-axis activation [1,5,14,18]. Basal LH is particularly useful in girls presenting at more advanced pubertal stages (Tanner stage ≥3), in whom basal LH in the pubertal range may already support HPG-axis activation and reduce reliance on stimulation testing in appropriate clinical contexts [14]. However, a nonelevated basal LH should not be used as a stand-alone rule-out test in early presentations because of pulsatile secretion and the limitations of a single random sample [1,18]. In addition, interpretation of low basal LH concentrations is assay-dependent, and thresholds may vary by platform and reporting practices [19]. Accordingly, basal LH should be viewed as complementary: potentially informative when elevated but insufficient by itself to exclude CPP in early Tanner stages [1,18].

Although the present study focused on optimizing biochemical classification and sampling duration, stimulation test results should not be construed as an automatic indication for GnRH agonist therapy. Treatment decisions should integrate the tempo of pubertal progression, growth velocity, bone age progression, predicted adult height, and psychosocial context [1,3,4,20,21], rather than relying on a single stimulation test result.

Several prior studies of subcutaneous triptorelin stimulation have used longer sampling windows (e.g., extending to 180 minutes) and have proposed varying diagnostic thresholds. Our findings complement these reports by defining, within a pragmatic 0- to 120-minute protocol, the trade-off between procedural burden and sensitivity under a commonly used diagnostic definition (peak LH ≥5 IU/L within the reference test). In this context, our data support sampling through 120 minutes as a practical minimum when the clinical objective is to minimize false-negative classification.

This study has several limitations. First, its retrospective, single-center design may limit generalizability and introduce potential referral bias typical of tertiary-care populations. Second, sampling ended at 120 minutes; therefore, we could not determine whether extending sampling to 180 minutes would identify additional delayed responders, as suggested in prior reports [8,9,11]. Third, follow-up was not collected uniformly or prospectively for all participants, which limited our ability to characterize the natural history of PT subgroups, identify robust predictors of progression, or assess longer-term outcomes such as adult height. Fourth, diagnostic performance depends not only on assay characteristics but also on institutional reporting practices; in our laboratory, values below the reportable range were not released as numerical results, which precluded evaluation of basal LH at low concentrations and limited direct comparisons with studies using different reporting conventions. Finally, our operational classification of CPP versus PT based on a stimulated LH threshold necessarily simplifies a biological continuum and may misclassify slowly progressive or transitional cases.

In this cohort of girls evaluated for breast development before 8 years of age, subcutaneous triptorelin testing demonstrated distinct LH kinetics between those classified as CPP and those classified as PT and showed excellent discrimination at individual sampling time points. However, when CPP was defined by a peak LH ≥5.0 IU/L observed within a 0- to 120-minute reference test, sampling through 120 minutes was required to maximize diagnostic sensitivity. In similar clinical settings and under comparable diagnostic definitions, a minimum test duration of 120 minutes may help minimize false-negative classification; prospective studies are needed to confirm generalizability, evaluate whether shorter algorithms can be implemented without clinically meaningful loss of sensitivity, and clarify how kinetic patterns relate to longitudinal pubertal progression and treatment-relevant outcomes.

Footnotes

Conflicts of interest

No potential conflict of interest relevant to this article was reported.

Funding

This study received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Data availability

The data that support the findings of this study can be provided by the corresponding author upon reasonable request.

Author contribution

Conceptualization: CJK; Data curation: EK; Formal analysis: EK; Methodology: EK; Project administration: CJK; Visualization: EK; Writing - original draft: EK; Writing - review & editing: CJK

References

  • 1.Carel JC, Léger J. Clinical practice. Precocious puberty. N Engl J Med. 2008;358:2366–77. doi: 10.1056/NEJMcp0800459. [DOI] [PubMed] [Google Scholar]
  • 2.Parent AS, Teilmann G, Juul A, Skakkebaek NE, Toppari J, Bourguignon JP. The timing of normal puberty and the age limits of sexual precocity: variations around the world, secular trends, and changes after migration. Endocr Rev. 2003;24:668–93. doi: 10.1210/er.2002-0019. [DOI] [PubMed] [Google Scholar]
  • 3.Carel JC, Eugster EA, Rogol A, Ghizzoni L, Palmert MR, ESPE-LWPES GnRH Analogs Consensus Conference Group et al. Consensus statement on the use of gonadotropin-releasing hormone analogs in children. Pediatrics. 2009;123:e752–62. doi: 10.1542/peds.2008-1783. [DOI] [PubMed] [Google Scholar]
  • 4.Fuqua JS. Treatment and outcomes of precocious puberty: an update. J Clin Endocrinol Metab. 2013;98:2198–207. doi: 10.1210/jc.2013-1024. [DOI] [PubMed] [Google Scholar]
  • 5.Brito VN, Spinola-Castro AM, Kochi C, Kopacek C, Silva PC, Guerra-Júnior G. Central precocious puberty: revisiting the diagnosis and therapeutic management. Arch Endocrinol Metab. 2016;60:163–72. doi: 10.1590/2359-3997000000144. [DOI] [PubMed] [Google Scholar]
  • 6.Lee PA. Central precocious puberty. An overview of diagnosis, treatment, and outcome. Endocrinol Metab Clin North Am. 1999;28:901–18. doi: 10.1016/s0889-8529(05)70108-0. , xi. [DOI] [PubMed] [Google Scholar]
  • 7.Stanhope R, Brook CC. Thelarche variant: a new syndrome of precocious sexual maturation? Acta Endocrinol (Copenh) 1990;123:481–6. doi: 10.1530/acta.0.1230481. [DOI] [PubMed] [Google Scholar]
  • 8.Freire AV, Escobar ME, Gryngarten MG, Arcari AJ, Ballerini MG, Bergadá I, et al. High diagnostic accuracy of subcutaneous triptorelin test compared with GnRH test for diagnosing central precocious puberty in girls. Clin Endocrinol (Oxf) 2013;78:398–404. doi: 10.1111/j.1365-2265.2012.04517.x. [DOI] [PubMed] [Google Scholar]
  • 9.Poomthavorn P, Khlairit P, Mahachoklertwattana P. Subcutaneous gonadotropin-releasing hormone agonist (triptorelin) test for diagnosing precocious puberty. Horm Res. 2009;72:114–9. doi: 10.1159/000232164. [DOI] [PubMed] [Google Scholar]
  • 10.Harrington J, Palmert MR, Hamilton J. Use of local data to enhance uptake of published recommendations: an example from the diagnostic evaluation of precocious puberty. Arch Dis Child. 2014;99:15–20. doi: 10.1136/archdischild-2013-304414. [DOI] [PubMed] [Google Scholar]
  • 11.Vukovic R, Milenkovic T, Soldatovic I, Pekic S, Mitrovic K, Todorovic S. Triptorelin stimulated luteinizing hormone concentrations for diagnosing central precocious puberty: study of diagnostic accuracy. Endocrine. 2022;75:934–41. doi: 10.1007/s12020-021-02947-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Ahn J, Lee Y, Gim S, Jeong H. Use of the subcutaneous triptorelin stimulation test for diagnosis of central precocious puberty. Children (Basel) 2023;10:1830. doi: 10.3390/children10111830. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Greulich WW, Pyle SI. Radiographic atlas of skeletal development of the hand and wrist. 2nd ed. Stanford (CA): Stanford University Press, 1959. [Google Scholar]
  • 14.Yeh SN, Ting WH, Huang CY, Huang SK, Lee YC, Chua WK, et al. Diagnostic evaluation of central precocious puberty in girls. Pediatr Neonatol. 2021;62:187–94. doi: 10.1016/j.pedneo.2020.12.001. [DOI] [PubMed] [Google Scholar]
  • 15.Kim HK, Kee SJ, Seo JY, Yang EM, Chae HJ, Kim CJ. Gonadotropin-releasing hormone stimulation test for precocious puberty. Korean J Lab Med. 2011;31:244–9. doi: 10.3343/kjlm.2011.31.4.244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Cavarzere P, Sandri M, Arrigoni M, Guardo C, Gaudino R, Antoniazzi F. Revising LH cut-off for the diagnosis of central precocious puberty via triptorelin stimulation assay. Endocrine. 2025;87:842–9. doi: 10.1007/s12020-024-04055-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Su H, Su Z, Pan L, Wang L, Xu Z, Peng G, et al. Factors affecting bone maturation in Chinese girls aged 4–8 years with isolated premature thelarche. BMC Pediatr. 2020;20:356. doi: 10.1186/s12887-020-02256-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Houk CP, Kunselman AR, Lee PA. Adequacy of a single unstimulated luteinizing hormone level to diagnose central precocious puberty in girls. Pediatrics. 2009;123:e1059–63. doi: 10.1542/peds.2008-1180. [DOI] [PubMed] [Google Scholar]
  • 19.Neely EK, Hintz RL, Wilson DM, Lee PA, Gautier T, Argente J, et al. Normal ranges for immunochemiluminometric gonadotropin assays. J Pediatr. 1995;127:40–6. doi: 10.1016/s0022-3476(95)70254-7. [DOI] [PubMed] [Google Scholar]
  • 20.Lazar L, Kauli R, Pertzelan A, Phillip M. Gonadotropin-suppressive therapy in girls with early and fast puberty affects the pace of puberty but not total pubertal growth or final height. J Clin Endocrinol Metab. 2002;87:2090–4. doi: 10.1210/jcem.87.5.8481. [DOI] [PubMed] [Google Scholar]
  • 21.Adan L, Chemaitilly W, Trivin C, Brauner R. Factors predicting adult height in girls with idiopathic central precocious puberty: implications for treatment. Clin Endocrinol (Oxf) 2002;56:297–302. doi: 10.1046/j.1365-2265.2002.01488.x. [DOI] [PubMed] [Google Scholar]

Articles from Annals of Pediatric Endocrinology & Metabolism are provided here courtesy of Korean Society of Pediatric Endocrinology

RESOURCES