Insulin levels in prepubertal patients with isolated premature pubarche: The INPREPP study

Abstract.

We aimed to determine insulin levels and the homeostasis model assessment of insulin resistance (HOMA–IR) index in a cohort of children with isolated premature pubarche, to analyze the potential influence of adiposity, and to compare these parameters with those of a reference population. This study included 50 prepubertal patients with isolated premature pubarche between the ages of 5 and 8 yr (girls) and up to 9 yr (boys). Anthropometric parameters, abdominal circumference, body composition assessed by bioelectrical impedance analysis, and bone age were evaluated. Blood glucose, lipid profile, insulin, IGF-1, and androgen levels were measured. Standard deviation scores for insulin and the HOMA–IR index were calculated using age- and sex-specific reference values for our population. Among the participants, 76% were girls, and 20% had obesity. Insulin levels and the HOMA–IR index in normal-weight patients were similar to the population mean, whereas they were significantly increased in patients with excess adiposity. Higher levels were also observed in children who were formula-fed at birth compared with those who were breastfed, and in those with advanced bone age compared with those with bone age appropriate for chronological age; in both cases, this association was mediated by adiposity. The increase in insulin levels and the HOMA–IR index in patients with isolated premature pubarche appears to be influenced by adiposity.

Highlights

● Insulin resistance in isolated premature pubarche is associated with adiposity.

● Insulin levels were higher in formula-fed children than in breastfed children.

● Isolated premature pubarche does not result in a higher rate of obesity.

Introduction

Isolated premature pubarche is a condition characterized by the appearance of pubic and/or axillary hair in girls younger than 8 yr and boys younger than 9 yr, after other causes of hyperandrogenism, such as congenital adrenal hyperplasia, have been excluded. It is, therefore, a diagnosis of exclusion. Most cases are associated with premature adrenarche, defined as the early activation of adrenal androgen synthesis, with dehydroepiandrosterone sulfate (DHEA-S) levels above 40–50 mcg/dL (1). The prevalence remains unclear and varies according to ethnicity; however, there is a clear predominance in girls (2, 3).

Adrenocorticotropic hormone (ACTH) is necessary for adrenal steroidogenesis but is insufficient for the development and functional maturation of the zona reticularis; therefore, the mechanism underlying its early activation remains unknown (4). Several intrinsic and extrinsic factors have been proposed to influence this process, including being born small for gestational age (SGA), excessive weight gain, exposure to endocrine disruptors, and dietary factors. Obesity, in particular, has been associated with higher circulating concentrations of DHEA-S (5, 6).

Several authors have also proposed an association between isolated premature pubarche and hyperinsulinism; however, considerable controversy remains, as adiposity may act as a confounding factor, and it is unclear whether these patients have a higher prevalence of obesity and greater fat mass than the reference population (4, 7). In addition, there is substantial heterogeneity among studies regarding the definition of obesity and, particularly, adiposity, because most studies rely on body mass index (BMI) without considering body composition.

The aim of this study was to evaluate insulin and homeostasis model assessment of insulin resistance (HOMA–IR) levels in a cohort of children with isolated premature pubarche, to analyze the potential influence of adiposity, and to compare these levels with those of the reference population. In addition, we sought to assess the influence of familial and perinatal factors on clinical and metabolic parameters.

Materials and Methods

The study included 50 patients from the pediatric endocrinology clinic of a tertiary hospital who were diagnosed with isolated premature pubarche between November 2023 and November 2024. The minimum age at inclusion was 5 yr, and the maximum age was 8 yr for girls and 9 yr for boys. Premature pubarche was defined as the appearance of pubic or axillary hair before the age of 8 yr in girls and 9 yr in boys. Cases attributable to congenital adrenal hyperplasia, tumors, or syndromes associated with growth or pubertal disorders were excluded. Pubertal stage was determined according to Tanner staging; Tanner stage II was defined as the appearance of breast buds in girls and a testicular volume of 4 mL measured using a Prader orchidometer in boys. All patients were required to be in Tanner stage I at the time of inclusion. Perinatal data, type of feeding during the first months of life, and longitudinal weight and height data were obtained from electronic medical records. A patient was classified as SGA if birth weight and/or length was below −2 standard deviations (SD) for sex and gestational age, according to reference charts for our population.

BMI was calculated using the formula BMI (kg/m²) = weight/height², according to the age- and sex-specific reference charts from the 2010 Spanish Growth Study (8). Results were expressed as absolute values and SD scores. A BMI between 1.5 and 2.0 SD was classified as overweight, and ≥ 2.0 SD as obesity. Abdominal circumference (AC) was measured using a non-stretchable tape measure with an accuracy of 1 mm, with the participant standing upright. The tape was placed at the level of the umbilicus, at the midpoint between the lowest rib and the iliac crest. Reference values for the Spanish population were obtained from the study by Arribas Muñoz et al. (9). Body composition was assessed using bioelectrical impedance analysis (Tanita TBF-300A, Tokyo, Japan) after a fasting period of at least 2 h. Blood pressure was measured using a Dynamap XL device, and reference values published in Pediatrics in 2004 by the National High Blood Pressure Education Program Working Group were applied.

All blood analyses were performed in the hospital laboratory after a 12-h fasting period, between 8:00 and 9:00 AM. Serum DHEA-S (mcg/dL), androstenedione (ng/mL), 17-hydroxyprogesterone (17-OHP; ng/mL), and total testosterone (ng/mL) were measured by chemiluminescence. An ACTH stimulation test for 17-OHP was performed when the baseline value was > 1 ng/mL. The test was conducted using 250 mcg of intravenous ACTH (Synacthen), and 17-OHP levels were measured at baseline and at 60 min. Genetic testing was requested when stimulated 17-OHP values were > 7 ng/mL. Biochemical parameters were analyzed using a Cobas c701 analyzer (Roche). Insulin levels were determined by electrochemiluminescence immunoassay (Abbott). The homeostasis model assessment (HOMA) index was calculated as fasting insulin (mcU/mL) × fasting glucose (mmol/L) / 22.5. Insulin levels and the HOMA–IR index were compared with reference values published for our population by García-Cuartero et al., according to sex and age, and were also expressed as standard deviation (SD) scores (10). IGF-I levels (ng/mL) were measured using chemiluminescence. Bone age was assessed by the same observer using the Greulich and Pyle atlas; it was considered advanced or delayed when it differed from chronological age by 1.5 yr.

Statistical analysis was performed using SPSS version 21.0 (IBM Corp., Armonk, NY, USA). Descriptive results were expressed as mean and SD. Parametric tests were applied to quantitative variables with a normal distribution (Student’s t test for two groups and analysis of variance for more than two groups), whereas nonparametric tests were used when normality was not assumed (Mann–Whitney U test for two independent continuous variables and Kruskal–Wallis test for more than two independent continuous variables). Pearson’s correlation coefficient was used to assess linear associations between quantitative variables. Analysis of covariance was performed to assess the contribution of adiposity to differences in insulin levels between patient groups. A p value < 0.05 was considered statistically significant.

This study was approved by the hospital’s Research Ethics Committee (A1481; June 28, 2023). Written informed consent was obtained at the first consultation, and parents signed the consent form in accordance with the ethical principles outlined in the Declaration of Helsinki.

Results

Fifty prepubertal children diagnosed with isolated premature pubarche participated in the study; 76% were girls (n = 38) and 24% were boys (n = 12). Regarding perinatal history, in vitro fertilization was reported in 4 cases (8%). The mean gestational age was 39.1 ± 2.0 wk; 3 children were born preterm, and 2 were SGA. The mean birth weight was 3,100 ± 600 g, and the mean birth length was 49.2 ± 3.0 cm. A significant negative correlation was observed between birth weight and DHEA-S levels (r = −0.45; p = 0.03), as well as total testosterone levels (r = −0.37; p = 0.02). In this cohort, 72% were Caucasian, 12% Hispanic, 6% Arab, and 2% African. Only 8% were born outside Spain.

The mean age at first consultation was 7.75 ± 0.7 yr for boys and 7.23 ± 1.0 yr for girls. Mean weight was 0.76 ± 1.4 SD, height 0.62 ± 1.1 SD, BMI 0.55 ± 1.2 SD, and AC 0.79 ± 1.8 SD. No statistically significant differences were observed by sex, although there was a trend toward higher values in boys. Nine children had obesity (4 boys and 5 girls), and one girl was classified as overweight.

The mean values of blood and metabolic parameters are presented in Table 1, with no statistically significant differences according to sex. In 82% of patients, DHEA-S levels were above 40 mcg/dL, consistent with premature adrenarche. No patients met the criteria for metabolic syndrome; however, 5 had high-density lipoprotein cholesterol levels < 40 mg/dL, another 5 had low-density lipoprotein cholesterol levels > 130 mg/dL, and 1 had systolic blood pressure > 2 SD. Twelve patients required an ACTH stimulation test for 17-OHP; in 2 cases, the results were compatible with heterozygous carrier status for a mild variant in the 21-hydroxylase gene, which was confirmed by sequencing. Regarding bone age, 54% had bone maturation consistent with chronological age, 44% had advanced bone maturation (mean advancement of 19.6 mo ± 6.0), and only one patient had delayed bone age.

Table 1. Anthropometric, biochemical, and body composition parameters at the first consultation in patients with isolated premature pubarche according to sex (no statistically significant differences unless otherwise indicated).

With respect to insulin levels and the HOMA–IR index, no significant differences were observed between sexes. However, significantly higher levels were found in overweight or obese patients compared with normal-weight patients, whereas the latter had values similar to those of the reference population (Table 2). A statistically significant positive correlation was observed between insulin and HOMA–IR values and fat mass (r = 0.58, p = 0.001; r = 0.56, p = 0.002, respectively), BMI (SD) (r = 0.38, p = 0.009; r = 0.36, p = 0.014), and AC (SD) (r = 0.38, p = 0.013; r = 0.36, p = 0.015). These associations remained significant both in the overall sample and when restricted to normal-weight patients. No association was observed with androgen levels. In addition, overweight or obese patients had higher IGF-I levels than normal-weight patients (1.28 ± 0.8 SD vs. 0.73 ± 0.5 SD; p = 0.046), as well as greater AC (3.04 ± 1.7 SD vs. 0.18 ± 1.4 SD; p < 0.001) and a higher fat mass percentage (32.96% ± 6.0% vs. 19.38% ± 8.0%; p < 0.001).

Table 2. Insulin levels and HOMA–IR index according to weight status (BMI < 1.5 SD vs. BMI ≥ 1.5 SD).

Regarding type of feeding, 58% of patients were breastfed from birth, and 44% of the total cohort were exclusively breastfed until 6 mo of age. The mean duration of breastfeeding was 10.82 ± 9.6 mo. In contrast, 28% were formula-fed from birth, and an additional 14% received mixed feeding (breast milk and formula). No differences in birth weight were observed according to type of feeding. However, significant differences in glucose metabolism parameters were observed between patients who were exclusively breastfed and those who were formula-fed from birth, with higher values in the latter group (Table 3); these differences did not persist after adjustment for BMI and fat mass. When analyzing weight gain up to 12 mo of age, patients who were exclusively breastfed until 6 mo had lower weight gain (in SD) between 2 and 6 mo compared with those who had initiated formula feeding before 6 mo (−0.60 ± 0.55 SD vs. −0.17 ± 0.6 SD; p = 0.025). Moreover, greater weight gain during this period was positively associated with higher BMI (SD) and AC (SD) at the time of consultation (r = 0.53, p = 0.005; r = 0.48, p = 0.014, respectively).

Table 3. Analysis of metabolic parameters and body composition according to neonatal feeding type.

Regarding family history, the mean maternal age at menarche was 12.06 ± 1.6 yr. Sixteen percent of mothers had been diagnosed with polycystic ovary syndrome (PCOS), and another 16% had isolated hirsutism. A trend toward higher BMI (SD) and larger AC was observed among patients whose mothers had PCOS; however, no differences in metabolic profile were identified.

Patients with advanced bone age showed differences in body composition and laboratory parameters compared with those with bone age consistent with chronological age (Table 4).

Table 4. Metabolic parameters and body composition according to bone maturation status.

Discussion

In this cohort of children with isolated premature pubarche and normal weight, insulin and HOMA–IR index values, both absolute and expressed as SD scores, were similar to the mean values of the reference population, whereas they were increased in overweight or obese patients. A significant positive correlation was also observed with fat mass and AC, which are parameters that more accurately reflect adiposity than BMI.

Whether patients with premature pubarche exhibit hyperinsulinism remains unclear because of heterogeneity among studies and small sample sizes, particularly in boys. Because insulin sensitivity varies across pubertal stages, the ideal approach would be to analyze these parameters in Tanner stage I patients, thereby minimizing confounding factors such as obesity or a history of SGA. Studies such as that by Ibáñez et al. reported higher basal insulin levels in 20 prepubertal girls compared with controls (11), and Denburg et al. observed similar findings in 11 boys compared with 8 controls (12). In contrast, in this cohort of normal-weight patients, both insulin levels and the HOMA–IR index corresponded to the mean values of Tanner stage I reference populations and were significantly increased only in overweight patients. A recent study conducted in a Turkish population also concluded that hyperinsulinism was not present in 35 prepubertal girls with normal weight compared with controls (13). Although both insulin and insulin-like growth factors appear to modulate steroidogenesis, it remains unclear whether this effect is mediated solely by adiposity.

Perinatal factors and early childhood nutrition are recognized as important determinants of metabolic programming. The present results indicate that insulin levels and the HOMA–IR index were higher in patients who were formula-fed from birth compared with those who were exclusively breastfed; however, this association was influenced by adiposity, suggesting that breast milk may exert a protective effect against obesity during the prepubertal period. These findings support the hypothesis that differences in weight gain during infancy may vary according to feeding type, with greater weight gain observed in children who receive formula. Previous studies have linked greater weight gain in the first months of life with higher DHEA-S levels in childhood and earlier onset of adrenarche (14,15,16); however, no such association was identified in this cohort.

Similarly, there is ongoing controversy in the literature regarding the prevalence of obesity in prepubertal patients with isolated premature pubarche. Some studies have reported a higher prevalence compared with control groups, including differences in fat mass percentage, abdominal fat, and waist circumference (11, 17). However, in this study, the prevalence of excess weight was 33.3% in boys and 15.7% in girls, which are lower figures than those reported in the Spanish reference population, particularly among girls (37.5% and 34.4%, respectively, according to the ALADINO 2023 study) (18). It should be noted that the socioeconomic status of this population is predominantly medium to low. In this cohort, patients with a BMI > 1.5 SD had a significantly higher fat mass percentage and larger AC than those with an adequate BMI, demonstrating a correlation between excess weight and adiposity. This finding is relevant because few studies have included body composition assessment. No association with excess weight was identified in other studies conducted in the Spanish population, such as that by Sancho et al., which included 71 prepubertal boys and girls with premature pubarche and/or adrenarche (19), or that by Mejorado et al., which included only girls (20). Ethnicity may contribute to variability in findings related to premature pubarche, obesity, and insulin resistance. In a study conducted in an Arab population, Atay et al. reported that premature pubarche in a Turkish population was not associated with higher BMI (SD) values (21). Despite these discrepancies, accumulating evidence supports an association between obesity and premature thelarche, as well as earlier pubertal onset (21,22,23).

Patients with isolated premature pubarche may exhibit accelerated bone maturation, with reported frequencies varying across studies; in this cohort, 44% had bone age advanced relative to chronological age. The precise mechanism underlying the timing and progression of this acceleration remains unclear. A synergistic effect between premature adrenarche and adiposity has been suggested; however, this phenomenon has also been observed in normal-weight patients. Elevated DHEA-S and IGF-I levels, as well as increased aromatization of androgens within adipose tissue, may contribute to this process (24,25,26,27). In this cohort, boys and girls with advanced bone age had a higher fat mass percentage, which may explain the higher insulin and 17-OHP levels observed in this group.

Traditionally, isolated premature pubarche has been considered a benign condition; however, Ibáñez et al. demonstrated that a subgroup of girls may later develop a clinical profile compatible with PCOS (28, 29). Identifying these patients during the prepubertal stage remains challenging; nevertheless, existing studies suggest that these girls tend to have higher BMI and waist circumference without overt hyperandrogenism, consistent with the present findings (30).

The limitations of this study include the sample size, particularly the small number of boys. In addition, the insulin data used for comparison were derived from a local reference population. Although the chemiluminescence immunoassay was used in both groups, different analytical systems were employed, which may have introduced measurement bias; ideally, a control group assessed under identical laboratory conditions should have been included. Regarding androgen levels, SHBG, an important parameter in the evaluation of hyperandrogenism, was not measured. Finally, increasing evidence highlights the potential role of endocrine disruptors, which should be explored in future research.

Conclusion

Insulin levels and the HOMA–IR index in children with isolated premature pubarche appear to be associated with adiposity. Type of feeding and excessive weight gain may contribute to this association, highlighting the importance of providing appropriate dietary guidance throughout childhood. Further multicenter, prospective studies are needed to clarify the role of each factor in the development of premature pubarche and its long-term implications.

Conflicts of interests

The authors declare that they have no conflicts of interest.