Puberty is a transitional period between childhood and adulthood during which individuals experience a number of physical and psychological changes that eventually lead to their final height and full development of sexual capacity.1 The onset of normal puberty occurs between 9 and 13 years of age for girls1 and between 10 and 14 years for boys.1 The process is initiated through the secretion of gonadotropin-releasing hormone (GnRH) in the hypothalamus, leading to the stimulation of the pituitary gland. The pituitary gland subsequently secretes luteinizing hormone (LH) and follicle-stimulating hormone (FSH), which in turn stimulate the secretion of gonadal steroids.2 The initial clinical sign of puberty is typically thelarche (breast bud development, Tanner stage B2) in girls1 and testicular enlargement to volumes ≥4mL (Tanner stage G2) in boys.1 A recognized variant of normal pubertal development, termed early puberty (PP), is characterized by pubertal onset between 8 and 9 years in girls1 and 9 and 10 years in boys.1 While this pattern typically does not compromise final adult height, a subset of children may develop a more rapid progression of pubertal signs, potentially requiring clinical intervention.1

PP is well-defined as the onset of puberty before 8 years of age in girls and 9 years of age in boys.1 The incidence rate of PP has been reported to range from 0.10% to 1.42% new cases per 1000 children per year,1,3 with a female-to-male ratio of approximately 10:1.1,3 The classification of PP is typically divided into two categories: true PP, which can have a central (e.g., hypothalamic or pituitary disorders) or peripheral (e.g., gonadal or adrenal disorders, or effect of exogenous steroids) etiology, and normal variants of puberty (NVP).1,2 Central precocious puberty (CPP) is the most common type of true PP.2 CPP begins, as in normal puberty, with the release of GnRH in the hypothalamus.2 CPP is idiopathic in 75–90% of girls and 25–60% of boys.2 In contrast, peripheral precocious puberty (PPP) is characterized by the onset of pubertal changes regardless of the hypothalamic activity.1,2 On the other hand, NVP include the isolated appearance of breast bud development (premature thelarche), pubic hair (premature pubarche), or vaginal bleeding (premature menarche) without any other signs of puberty before the age of 8 in girls and 9 in boys.1 Unlike PP, NVP are isolated, non-progressive, and sometimes reversible conditions.4 A cutoff point of 6 months without the development of other pubertal signs (e.g., progressive pubarche or breast development, uterine growth) is typically used by clinicians to distinguish between PP and NVP.4

Recent, large-scale studies have indicated a rise in the incidence rate of PP over the past few decades.3 Moreover, additional studies have shown a progressive decline in the mean age of pubertal onset in both girls and boys, with estimated decreases of 3 months5 and 1.2 months6 per decade, respectively. The causes of the increase in the incidence rate of PP remain to be fully elucidated. Although genetics are considered the primary determinant of pubertal timing,7 emerging evidence suggests that modifiable environmental factors may also influence its onset.7 These include the increasing prevalence of childhood obesity8 (with particularly strong associations in girls),9 psychological stress,7 reduced melatonin secretion due to the use of electronic devices,10 and exposure to endocrine-disrupting chemicals (EDCs)11 or air pollutants.12–14 However, conflicting studies report no significant associations between PP and EDCs15 or air pollution.16 Given the uncertain role of air pollutants in pubertal development, this study aims to investigate their potential effects on the timing of puberty. To our knowledge, this is the first investigation of this association in a Southern European population.

Of the 96 cases, 56 (58.3%) were diagnosed with PP – 53 (55.2%) with CPP and 3 (3.1%) with PPP – while 40 (41.6%) were classified as NVP, including 23 (23.9%) with premature thelarche and 17 (17.7%) with premature pubarche. The characteristics of patients with CPP and NVP are presented in Table 1. Patients diagnosed with CPP were taller than those diagnosed with NVP. However, when adjusted for standard deviations (SD), no significant differences in height were observed across the groups (Table 1). As expected, significant differences in Tanner stage, radiographic bone age, and peak LH levels following GnRH stimulation test were found among the different diagnostic categories. No statistically significant differences were observed between patients with CPP and NVP for the other variables analyzed, including exposure to air pollutants (Table 1).

The population of children aged 0–9 years, along with the CPP and NVP rates for each geographical zone, is shown in Supplementary Table 1. The incidence rates of PP in 2021, 2022, 2023, and 2024 ranged from 0.42 to 0.77 cases per 1000 children per year (0.42–0.76 cases per 1000 children per year for CPP). The inciden

ce rates of NVP ranged from 0.30 to 0.65 cases per 1000 children per year (Supplementary Table 1). Differences in air pollutant concentrations were observed among the zones within the health area, as well as a trend toward different CPP rates across these zone-years (Supplementary Table 1).

Table 2 shows the association between zones of high or low exposure to environmental pollutants and the rates of CPP and NVP. Although there were no statistically significant differences, the results revealed higher CPP rates in zones with high exposure to all the measured pollutants (e.g., PM10, PM2.5, SO2, NO2, CO, O3, NO, NOx). This association was particularly notable for NO2 levels (Table 2). Similarly, no significant differences were found between higher and lower exposure to air pollutants in relation to NVP rates (Table 2).

The correlation between the levels of air pollutants and the rates of CPP and NVP is presented in Table 3. Our results revealed a trend toward a weak and positive correlation between higher CPP rates and higher ambient CO concentrations (Table 3). In addition, a positive but non-significant correlation was observed between CPP rates and the concentrations of all other measured pollutants, including PM10, PM2.5, SO2, NO2, O3, NO and NOx (Table 3). No significant correlations were found between NVP rates and air pollutant levels across the different zones. Additionally, a supplementary analysis was performed including all 96 cases of PP (e.g., the 93 cases of CPP and the 3 cases of PPP). In this analysis, a statistically significant correlation was found between higher rates of overall PP and increased environmental levels of CO (rho=0.479, p=0.024) and NOx (rho=0.391, p=0.033).

The aim of this study was to evaluate the association between air pollution and pubertal development. To the best of our knowledge, this is the first study of its kind carried out in a Spanish population. According to our results, the PP rates in our health area (0.42–0.77 cases per 1000 children per year) were comparable to rates reported in former studies.1,3 Our study found a statistically significant correlation between higher overall PP rates (including both CPP and PPP) and increased ambient levels of CO and NOx. However, this association was attenuated when analyzing exclusively the relationship between environmental pollution and CPP rates. Although CPP rates were higher in high-exposure areas for all eight analyzed air pollutants, these differences did not reach statistical significance.

The role of environmental contaminants in pubertal development has not been clearly established. A recent systematic review found that most studies suggested exposure to environmental pollution could accelerate puberty.19 However, other studies were inconclusive or indicated that such exposure might delay pubertal onset.19 When analyzing the role of each environmental contaminant separately, the studies showed similar discrepancies. The effect of exposure to NOx on the development of PP has been a subject of research. Former studies have associated exposure to higher levels of NO and NO2 with an increased risk of PP,13 while others have not.12,16 For PM2.5, most studies demonstrate a clear association between higher ambient concentrations and an increased incidence rate of PP.12–14 Despite that, some studies did not find any association.16 The literature to date is also inconclusive about the role of PM10, SO2 and O3 in pubertal development. While some studies found a positive association between their concentrations and PP,12 particularly in the case of PM10,12–14 others found a negative16 or null association.13,16 Despite Although our study did not reveal other statistically significant correlations between CPP incidence and air pollutants, we observed a non-significant, weak, positive correlation between CPP rates and the concentrations of air pollutants, particularly CO. As far as we are aware, this is the first study to analyze the relationship between ambient CO levels and the incidence rate of PP.

The mechanisms by which air pollution disrupts normal pubertal development are not fully understood. On the one hand, air pollutants have been shown to contain EDCs.20 The occurrence of PP and premature thelarche has been associated with the exposure to EDCs.15 These EDCs may act not only through a direct effect on the hypothalamic–pituitary–gonadal axis, but also through their action on the peripheral tissue (e.g., the adrenal gland or the breast tissue), obesogenic mechanisms, or epigenetic effects.15 Despite these particulate-bound EDCs may have estrogenic,20,21 anti-estrogenic,21 androgenic21 or anti-androgenic21 effects, most EDCs have demonstrated the combination of estrogenic and antiandrogenic effects.15 This could potentially promote not only the development of CPP but also that of PPP. For this reason, in a supplementary analysis, we examined the association between the overall PP incidence rate (including both CPP and PPP) and air pollution levels and found a significant association between overall PP incidence rate and ambient CO and NOx concentrations. On the other hand, air contaminants may have indirect and direct effects. These particles may increase serum cortisol levels,22 induce oxidative stress23 and systemic inflammation,23 and activate particular pathways, all of which could play a role in the pubertal disturbances observed in these patients. Specifically, air pollutants NO and CO may be stimulators of GnRH secretion and both stimulators and inhibitors of CRH secretion.24 Former studies have shown that NO can increase GnRH in an indirect way by increasing GABA and glutamate release, and in a direct way by altering GnRH neuronal excitability by accelerating recovery from prior exposure to kisspeptin-10, which is a GnRH stimulating molecule.25 In addition, it has been postulated that CO may also stimulate the secretion of GnRH by having a direct effect on the GnRH neurons.26 Further research is needed to characterize the physiopathological mechanisms by which air pollutants may disrupt normal pubertal development.

A major limitation of our study is its observational design, which has the inherent limitations of retrospective studies.27 Considering that the pediatric check-up program in Spain included approximately three annual visits during the ages 0 and 9,17 and that some of these visits are specifically focused on the assessment of pubertal development, it is unlikely that there were undiagnosed cases of PP in our area. This is supported by the fact that the PP rates observed in our study are similar to those from former studies.1,3 Of note, this condition is primarily diagnosed clinically, so it is possible that some cases remain undetected, which could result in an underestimation of the true incidence rate. Although PP is defined as the onset of puberty in girls aged <8 years and boys aged <9 years,1 the incidence rate of PP in our study was calculated based on the population ≤9 years of age. This imprecision arises from the fact that the data provided by the Spanish National Institute of Statistics are grouped in 5-year intervals (e.g., 0–4 years and 5–9 years), which may have contributed to an underestimation of the actual incidence rate of PP, particularly in girls. Furthermore, patients with early puberty could not be considered cases of NVP because they were not always referred for Pediatric Endocrinology assessment, as they are sometimes considered within the normal range of puberty by primary care pediatricians. Moreover, when such patients were referred, the GnRH test was often not performed. All patients included were exposed to air pollution to varying degrees. The inability of obtaining a non-exposed population limits the possibility of attributing observed effects specifically to one or more pollution parameters. Additionally, the measured concentrations of air pollutants represent their average levels in a given region (without accounting for sudden spikes in air pollutant concentrations) but may not accurately reflect individual exposure. Consequently, it may be inappropriate to infer individual risk based solely on associations observed at the community levels. The analyses assessing the association between the different air pollutants and the diagnosis of PP were univariate. Therefore, we cannot entirely exclude the possibility of confounding by a third variable. In our sample, metabolic and hormonal covariates such as BMI, TSH and FT4 were not associated either with exposure (air pollutants; data not shown) or with the outcome (CPP [vs. NVP]; Table 1), suggesting that their potential role as confounders is limited. Nonetheless, unmeasured covariates such as exposure to other endocrine-disrupting chemicals,11 dietary habits,8,9 screen time related to the use of digital devices,10 or socioeconomic status, which might influence both exposure to air pollution28 and the timing of pubertal onset,29 could represent a source of residual confounding. Of note, the study period (2021–2024) was influenced by the COVID-19 pandemic. Despite the reduction in industrial activity and the increase in population body weight during this time,30 several studies have reported a rise in the incidence rate of precocious puberty diagnoses during and immediately after the pandemic,30 which may be associated with BMI or increased use of electronic devices.30 Lastly, most of the associations observed in our study were not statistically significant, likely due to the small number of cases of PP and NVP, which may have resulted in type 2 errors and indicate that the study may have been underpowered to detect true differences. That said, ecological studies are inherently hypothesis-generating, and our study may be useful in prompting future investigations with more detailed evaluations of exposure to air pollutants.

Our study suggests a possible link between air pollution and PP, particularly regarding exposure to CO. Given the growing concern about the rising incidence of PP worldwide and its multifactorial etiology, larger studies are needed to clarify the potential role of environmental pollutants (especially CO, which has not been previously studied) in pubertal development.

This study was conducted in full compliance with the principles outlined in the Declaration of Helsinki and the Committee on Publication Ethics (COPE) and was approved by the regional ethics committee of the Principality of Asturias (CEImPA) (Asturias, Spain) code CEImPA 2024.529. Furthermore, in full compliance with the national legislation and institutional guidelines, written informed consent of study participants was deemed unnecessary. All data were anonymized to ensure participant confidentiality, and any personal identifiers were removed to maintain privacy.

TGV was supported by a Río Hortega research contract (CM24/00080) from Instituto de Salud Carlos III. The remaining authors received no specific funding whatsoever.

None declared.

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.