The complex interaction between obesity and puberty and its implications for pubertal timing: a review of recent evidence

Abstract

The childhood and adolescent obesity epidemic remains one of the most important and challenging health issues of the 21st century. Excess body weight affects the immediate health of children and adolescents, being associated with a higher and earlier risk of several non-communicable diseases (NCDs), such as type 2 diabetes and cardiovascular diseases, in addition to adverse psychosocial consequences and impacts on pubertal timing. Evidence dating back to the past century points to a secular trend of earlier puberty in boys and girls in various parts of the world, and this earlier onset has been attributed, at least in part, to obesity, especially in girls. In boys, this relationship is not as linear. Moreover, despite some contradictory results, most recent evidence points to earlier puberty also being associated with an excess of body weight in boys. Several mechanisms have been suggested to explain this possible relationship between puberty onset and obesity. However, the most promising link in this connection is leptin and its interaction with the kisspeptin system, in addition to an important contribution from insulin, which is frequently increased in obesity. The gut microbiota is implicated in the pathophysiology of obesity, which, in turn, is commonly associated with insulin resistance and the development of central precocious puberty, particularly in females. This review summarizes current knowledge on the relationship between obesity and the onset and progression of puberty in both sexes, as well as the pathways potentially involved in the pathophysiology of this association.

Introduction

Childhood and adolescent obesity remains one of the most important and challenging health issues in the 21st century. According to the World Health Organization (WHO), over 390 million children and adolescents aged 5–19 years were overweight in 2022, including 160 million with obesity, a substantial increase from 8% in 1990 to 20% in 2022. This increase occurred similarly among boys and girls and is associated with a higher risk of several NCDs,such as type 2 diabetes and cardiovascular disease (https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight, accessed on October 07, 2025).

Puberty is a complex maturational phenomenon that drives the individual from an immature to a fully mature stage, when the capacity to reproduce is achieved (1). Puberty is influenced by genetic, nutritional, and environmental factors (2). A secular trend toward earlier pubertal timing has been documented worldwide in boys and girls (3, 4, 5, 6). Evidence suggests a link between obesity and the pubertal onset in both boys and girls. However, obesity is only one of many factors that influence the timing of puberty (7).

In girls, an association between overweight and earlier pubertal onset has consistently been observed (6, 8, 9, 10). In contrast, findings in boys are inconsistent. Some studies report an earlier onset associated with overweight (5, 6, 8, 11), while others show no association (10) or even a delay in pubertal onset or completion in individuals with obesity (12, 13, 14). The finding that overweight children, especially girls, tend to mature earlier has led to the hypothesis that the degree of body fat could trigger the neuroendocrine events that lead to the onset of puberty (15). A sexual dimorphism is evident in the negative association between testicular volume and BMI in boys versus a positive association between breast stage and BMI in girls (16).

A recent meta-analysis confirmed that childhood overweight and obesity were associated with a significantly higher risk of precocious puberty in girls (17). In boys, a similar risk was found but without statistical significance (17). Methodological challenges, including the reliable determination of puberty and the use of BMI as a proxy for body fat, complicate epidemiological studies (18). Ethnic–racial background also influences pubertal timing, with studies indicating that children of African descent tend to mature earlier than their peers (4, 19, 20).

Identifying early puberty can potentially lead to preventive health measures because early pubertal onset has been associated with deleterious long-term health outcomes (21). This review will provide an overview of the evidence on the relationship between obesity and puberty, focusing on secular trends and pathophysiological mechanisms. Other factors, such as genetic background, intrauterine and postnatal environment, endocrine disruptors, demographic factors, and socioeconomic status, have also been associated with the timing of puberty onset but are beyond the scope of this review.

Physiology of puberty related to obesity (Fig. 1)

Figure 1.

Pathways potentially involved in the pathophysiology of obesity and advanced puberty. Leptin stimulates Kiss secretion and HPG axis activation through KNDy neurons in the hypothalamus, which in turn stimulate GnRH neurons. Leptin inhibits NPY and stimulates POMC neurons, and α-MSH promotes an increase in the frequency of GnRH pulses. Leptin effects on KNDy neurons are mediated by mTOR and the inhibition of AMPK. Epigenetic changes associated with increased body weight may reduce hypothalamic SIRT1 levels and increase the Kiss. Leptin stimulates the anterior pituitary the release of LH and FSH through activation of NOS. Elevated leptin levels may inhibit male gonadal function. Higher insulin levels can decrease SHBG and stimulate human adrenal cells, leading to increased adrenal androgens. The intestinal dysbiosis observed in obesity is associated with the development of CPP, related to the activation of the leptin gene. The inflammatory cytokines can affect reproductive function and directly interact with the male reproductive system. *Arcuate nucleus of the hypothalamus in rodents. The figure was created in Adobe XD. Kiss, kisspeptin; HPG, hypothalamic–pituitary–gonadal; KNDy, kisspeptin, neurokinin B, and dynorphin; GnRH, gonadotropin-releasing hormone; NPY, neuropeptide Y; POMC, proopiomelanocortin; α-MSH, α-melanocyte-stimulating hormone; mTOR, mammalian target of rapamycin, AMPK, AMP-activated protein kinase; SIRT1, fuel-sensitive deacetylase; LH, luteinizing hormone; FSH, follicle-stimulating hormone; NOS, nitric oxide synthase; SHBG, sex hormone-binding globulin; MC3R, melanocortin receptor subtype 3; MC4R, melanocortin receptor subtype 4; and CPP, central precocious puberty.

The onset and timing of puberty in humans depend on several factors, including genetic variability, energy balance, multiple neuroendocrine pathways, and hormonal profiles (7, 22) (Fig. 1). Although the contribution of genetic factors to variation in puberty timing is estimated at 60–80% (2), adequate nutrition is an important enabling factor for the normal timing and rhythm of pubertal development (22). The observation of an association between nutritional status and puberty timing is old, and one of the first studies to describe this relationship was the ‘critical weight hypothesis’. This hypothesis defended the concept that a minimum body weight or a minimum fat percentage would be necessary for the development and maintenance of reproductive activity in girls (23, 24).

The hypothalamic–pituitary–gonadal (HPG) axis is transiently activated during neonatal development and early infancy, followed by a period of relative but not absolute quiescence in childhood that ends with reactivation of the gonadotropin-releasing hormone (GnRH) pulse generator in the early second decade of life, consequently leading to thelarche or increased testicular volume (18). GnRH reaches the pituitary gland through the hypothalamic–pituitary portal system, where it activates the secretion of pituitary gonadotropins, luteinizing hormone (LH), and follicle-stimulating hormone (FSH), which, in turn, stimulate the gonads for gametogenesis and production of sex hormones, resulting in the appearance of secondary sexual characteristics (25, 26).

Leptin

Nutritional status and body fat interact with genetic inheritance, thus explaining important individual variability in the body composition. Adipocytes produce adipokines, which modulate central nervous system function, suggesting an ‘adipo–cerebral axis’ (27). The most promising link between an excess of body weight and the timing of puberty appears to be the leptin (7, 27) and its interaction with the kisspeptin system. Leptin links the state of the body’s energy stores to body functions, including body weight control, puberty, and reproduction (1).

Kisspeptin, encoded by the Kiss1 gene, and its receptor, Kiss1R, are essential regulators of puberty onset and gonadotropin secretion. Kiss1 neurons co-express other neuropeptides that control GnRH secretion, such as neurokinin B (NKB), which is stimulatory, and dynorphin A (Dyn), which is inhibitory. Both are co-transmitters and potential autoregulators of Kiss1 neurons (1, 25, 26). Leptin stimulates kisspeptin secretion and, subsequently, the activation of the HPG axis, through the KNDy neurons of the hypothalamus (26). The term KNDy, which stands for kisspeptin, NKB, and Dyn, has been proposed to this population of Kiss1 neurons that appear to be selectively localized to the arcuate nucleus (ARC) or infundibular region in humans (25). The Kiss1 neurons of the ARC are particularly prominent in the metabolic control of the HPG axis and puberty onset (1). A sexual dimorphism has been demonstrated in humans, with women exhibiting higher kisspeptin levels than men (28). Although kisspeptin concentrations do not significantly differ between normal-weight, overweight, and obese patients, leptin levels are significantly higher in obese and overweight individuals compared to normal-weight individuals (28). Kisspeptin is also expressed in peripheral tissues, such as adipose tissue, pancreas, adrenal glands, and gonads of both sexes (27, 28).

In the ARC, neurons expressing neuropeptide Y (NPY) and proopiomelanocortin (POMC) and co-expressing receptor (Rc) of leptin have been found. However, no Rc of leptin has been identified in GnRH neurons. Thus, it seems more likely that leptin influences GnRH secretion indirectly through intermediate neurons such as NPY and the peptide products of the POMC gene (29). Because leptin presumably decreases NPY biosynthesis in the ARC, when NPY release is inhibited, activation of GnRH release occurs (22).

The permissive influence of leptin on the onset of puberty is mediated, at least in part, by the leptin/α-MSH/Kiss1/GnRH signaling pathway. The central effects of α-MSH are mediated by melanocortin receptors (MC3R and MC4R), which are expressed in the hypothalamus. Acute blockade of MC3R/MC4R receptors in rodents resulted in a significant reduction in basal LH levels, and chronic inhibition of central melanocortin signaling during the pubertal transition resulted in a delay in the onset of puberty (30).

Another study showed a strong association between loss-of-function mutation of MC3R and pubertal delay in humans (31). Reproductive impairments in MC4R-deficient mice are largely mediated by the direct action of melanocortins via MC4R on Kiss1 neurons and not their obese phenotype (32).

In the anterior pituitary, leptin directly stimulates LH and FSH release, possibly through activation of nitric oxide synthase in the gonadotrophs (15, 22). The effects of leptin on the gonads are implicated by the expression of leptin receptors in granulosa and theca cells (15, 22) and in Leydig cells (15, 22). Thus, leptin possesses a bimodal action on the HPG axis, depending on its serum levels. In particular, low leptin levels result in HPG axis dysfunction, and leptin administration in low doses may have a permissive, threshold effect on the central networks that regulate gonadotropin secretion. On the other hand, high serum leptin levels in people with obesity may have an inhibitory effect on the gonads (15, 22).

The sexually dimorphic effects of leptin on GnRH secretion and peripheral effects may contribute to the differences observed in male and female puberty. Testosterone inhibits and estrogen stimulates leptin secretion from adipocytes (7). Increased levels of leptin not only compromise testosterone production but also affect spermatogenesis (26). In a large series comparing BMI and reproductive parameters in men, total testosterone and seminal vesicle parameters were negatively correlated with BMI (33). Thus, it would be plausible to assume that obesity constitutes a risk factor for infertility. Obesity further reduces testosterone levels by increasing the activity of aromatase in adipose tissue, which converts testosterone to estrogen (7, 22, 26).

Obesity is associated with a chronic low-grade inflammatory state (26), and adipokines and inflammatory cytokines can affect reproductive function and directly interact with the male reproductive system and impact both pubertal development and fertility in males (15). Besides the inflammation in peripheral tissues, obesity also induces neuroinflammation (34). Glial cells are very reactive to obesogenic diet (35), and gliotransmission may be involved in the advanced puberty in obesity.

mTOR, AMPK, and sirtuins

Experimental studies suggest the contribution of energy/metabolic sensors, such as the mammalian target of rapamycin (mTOR), AMP-activated protein kinase (AMPK), and fuel-sensitive deacetylase (SIRT1) in the control of pubertal timing (1, 22). mTOR is a protein kinase that acts as a cellular sensor of energy status and growth signals and appears to play an important role in the metabolic control of puberty, mediating at least part of the permissive effects of leptin on puberty onset (1). Leptin effects on KNDy neurons are mediated by mTOR (18).

AMPK, the main energy sensor at the cellular level, is a serine/threonine kinase, which is activated under conditions of energy depletion to generate ATP (1, 36). In the ARC, AMPK is activated by ghrelin and inhibited by leptin. Unlike AMPK, mTOR is activated under conditions of energy excess to promote anabolic pathways (1). Kiss1 neurons in the ARC express AMPK, and activation of this protein kinase reduced the expression of Kiss1 in rodents. This neuroendocrine pathway provides a molecular link between negative energy balance and delayed puberty (1, 36). Persistent negative energy balance associated with malnutrition, which is known to reduce circulating leptin levels, would inhibit hypothalamic Kiss1/kisspeptin expression, as a major mechanism for suppression/delay of puberty onset, by reciprocal alteration of the mTOR/AMPK system, causing an inhibition of mTOR and activation of AMPK signaling, the latter presumably in Kiss1 neurons of the ARC (1).

Sirtuins have also emerged as metabolic modulators activated by energy deficits, and their hypothalamic content is negatively correlated with Kiss1 expression (1). In obese rodents, hypothalamic SIRT1 levels are reduced as Kiss1 expression increases. In contrast, malnutrition elevates SIRT1 and suppresses Kiss1, delaying puberty. Thus, SIRT1 may be a fundamental epigenetic factor that links nutritional status and puberty timing through the regulation of Kiss1 transcription in the ARC (1, 37). The global SIRT1 overexpression, as well as in the ARC only, leads to delayed puberty onset in female rodents, without affecting body weight (37). However, the definitive role of these cellular energy sensors in obesity-induced pubertal alterations requires further elucidation (1).

Insulin

Another molecule strongly implicated in pubertal onset and development associated with obesity is insulin (7, 16). In peripubertal obesity, insulin resistance (IR) and compensatory hyperinsulinemia increase androgen production in the ovaries and adrenal glands and decrease hepatic sex hormone-binding globulin (SHBG). This cascade increases the bioavailability of androgens and sex steroids, which may, in turn, promote pubertal development by acting peripherally or centrally on the HPG axis (7, 15, 22). Early rapid weight gain has been associated with elevated leptin, IR, and increased adrenal androgen (15). Potential mechanisms of obesity-associated hyperandrogenemia include IR, enhanced androgen production in an expanded fat mass, and potential effects of abnormal adipokine/cytokine levels (22). An increased activity of 5α-reductase, an enzyme that converts testosterone (TS) to the highly potent androgen dihydrotestosterone, is related to an expanded fat mass and represents another mechanism by which obesity may promote hyperandrogenism (22).

Higher levels of basal insulin and IR were observed in obese prepubertal children compared to overweight and normal-weight children (28), which may affect sex hormone concentrations. It has been proposed that, as in other forms of premenarchal hyperandrogenism, such as congenital adrenal hyperplasia, the hypothalamic pulse-generating unit may eventually be stimulated or derepressed by the elevated androgens in obese girls. This androgen production in prepubertal children can be induced by IR as suggested by the association between IR and TS, even independent of BMI or fat mass (16). After androgens allow for the onset of central puberty, the insulin-stimulated ovaries, uterus, and breast tissue exposed to estrogen may respond more robustly, leading to a more rapid pubertal progression and earlier menarche (16). IR, independent of adiposity, also showed sexually dimorphic associations. In girls, higher IR level were associated with increased ovarian volume and TS concentrations, regardless of BMI-z or fat mass. Conversely, in boys, also independent of BMI-z, there was a negative association between IR and testicular volume and Tanner stage of pubic hair. Obesity was associated with a smaller testicular volume and lower TS levels in boys, but not with ovarian volume in girls (16). Obese girls often present with hyperandrogenism, characterized by lower SHBG levels and higher free androgen levels to normal-weight girls, but no differences in total TS levels (38). A decreased SHBG level was also observed in obese boys, but unlike girls, they had lower levels of total T to normal-weight boys (38). In addition, leptin can stimulate 17,20-lyase activity in human adrenal cells and consequently increase adrenal androgens (39).

Gut microbiota

The gut–brain axis (GBA) is implicated in the pathophysiology of obesity through bidirectional signaling involving metabolic, endocrine, and neural mechanisms (40). The gut microbiota (GM) is thought to contribute by modulating appetite-related hormones, such as leptin, ghrelin, and glucagon-like peptide 1 (GLP-1) (40). The human intestinal tract is colonized by a large number of microorganisms (approximately 10–100 trillion), collectively known as the ‘microbiota’ (41). While the connection between the gut microbiota–brain axis and the neuroendocrine system is not fully understood, it may involve bacterial metabolites, such as short-chain fatty acids (SCFAs), which can enter circulation and directly influence the brain (40, 41). A growing amount of evidence suggests an association between intestinal dysbiosis observed in obesity and the development of central precocious puberty (CPP), particularly in females, although the causal mechanisms remain to be elucidated (41, 42, 43, 44, 45).

Findings regarding the role of SCFAs are conflicting. Some studies have associated elevated levels of SCFAs with CPP development through the activation of the leptin gene and, consequently, the HPG axis (41, 42). In girls with idiopathic CPP (ICPP), a greater diversity of GM known to produce SCFAs was found compared to controls (42). Conversely, other research has linked reduced colonic SCFA levels to early puberty associated with obesity (43). In one animal model, supplementing SCFAs, such as acetate, propionate, and butyrate, to a high-fat diet (HFD) reversed signs of precocious puberty (PP) by reducing hypothalamic GnRH and delayed the development of the gonadal axis, involving decreased expression of the Kiss1 gene (43).

Animal models further support a link. Rats transplanted with intestinal flora from obese patients and obese patients with PP presented abnormal activation of the Kiss1/GnRH pathway and accelerated gonadal development (44). Other studies have demonstrated that PP and obesity-related precocious puberty (OPP) are associated with reductions in GM diversity and richness (45). In another study, a high-fat diet was associated with intestinal dysbiosis in rats with OPP (43). Beyond the HPG axis, the GM may influence estrogen levels. The microbiome metabolizes estrogens via the enzyme β-glucuronidase, which deconjugates estrogen. Unconjugated estrogen can enter the circulating blood and becomes metabolically active by acting on estrogen receptors (41).

Increasing evidence suggests that GM dysbiosis may play a significant role in the pathogenesis of PP in children. However, whether the GM can be a direct causative factor remains unknown (46).

Neuroglia

The past decades of neuroscience research have brought to light the importance of astrocytes in regulating the development and function of neural circuits. Astrocytes are intimate partners of kisspeptin and GnRH neurons, which control fertility in mammals (47). Positioned at the blood vessels–neurons interface, glial cells integrate peripheral information to neuronal networks for the neuroendocrine control of body metabolism, while microglia coordinate hypothalamic inflammatory responses (34).

A possible interaction between a HFD and PP was demonstrated in a recent experimental study in rodents. The anticipation of puberty onset was associated with an increase in astrocyte apposition on GnRH neurons in females during the pubertal transition, suggesting that puberty anticipation induced by a HFD appears to be, at least partially, mediated by an increase in the morphological association between astrocytes and GnRH neurons, which may increase the excitability of GnRH neurons (35). The regulatory role of astrocytes is further supported by the presence of kisspeptin receptors in these cells in the ARC and anteroventral periventricular nucleus (48). Notably, specific deletion of the kisspeptin receptor in astrocytes prevented HFD-induced precocious puberty in female mice without affecting body weight gain (48). This highlights the involvement of astrocytic kisspeptin signaling in mediating the effects of obesity on pubertal timing.

Secular trend of pubertal onset in girls

Numerous factors have been proposed to explain the secular trend toward earlier female pubertal maturation, with increased BMI emerging as a principal contributor (49). An overview of the age of thelarche in various countries over the years is presented in Fig. 2 (3, 4, 6, 8, 50, 51, 52). In the pioneering studies of Marshall and Tanner in the 20th century, the age at thelarche in British girls was 11.2 years (50). North American data from the Third National Health and Nutrition Examination Survey (NHANES III) revealed a mean age at thelarche of 10.2 years in normal-weight girls, with overweight status being associated with earlier thelarche and menarche (51). A subsequent large-scale American study of over 17,000 girls reported a similar age at thelarche of approximately 10 years, noting that African American girls entered puberty 1–1.5 years earlier and initiated menses 8.5 months sooner than white girls (4). This racial disparity was corroborated by a longitudinal US study, where African American participants began puberty at a mean age of 9.6 years compared to 10.2 years for white participants (19).

Figure 2.

Age of thelarche in the world in the past 50 years. The data in this graph have been compiled for this review from several historical sources.

The association between adiposity and pubertal timing is a globally observed phenomenon. A Danish longitudinal study demonstrated that childhood overweight and obesity were associated with earlier pubertal milestones in girls, a finding that persisted after sibling-matched analysis, suggesting a causal relationship. In this cohort, thelarche occurred approximately 9.5 and 15.2 months earlier in overweight and obese girls, respectively, compared to their normal-weight peers by 10 years (6) (Fig. 3). An Italian study reported a mean age at puberty onset of 10.3 years in normal-weight girls versus 9.9 years in girls with obesity (8) (Fig. 3). Similarly, among girls of the same age and race, there was almost twice the chance of having entered puberty for each point increase in BMI-z (16). In Brazil, a cross-sectional study reported thelarche at 9.8 years in a sample with a high prevalence of overweight and obesity (31.3%) (52).

Figure 3.

Age of thelarche according to nutritional status. The data in this graph have been compiled for this review from several historical sources.

This trend is also evident in Eastern countries. A recent Chinese study found that overweight and obese girls reached puberty approximately 1.6 years earlier than normal-weight girls (9) (Fig. 3). In Egypt, thelarche occurred at a mean age of 9.9 years in overweight girls compared to 10.3 years in normal-weight girls (10). Interestingly, the Copenhagen study documented a one-year decline in the age at thelarche over a 15-year period, from 10.9 years in 1991 to 9.9 years in 2006, a change that could not be explained by differences in BMI between the cohorts (3).

A meta-analysis based on studies from Europe, Middle East, Asia, United States, and Africa suggested a significant downward trend of 0.24 years in the age of onset of thelarche per decade, from 1977 to 2013, with earliest onset in the United States (8.8–10.3 years) and latest onset in Africa (10.1–13.2 years) (21).

Evidence that the timing of puberty is associated with metabolic risk is abundant and has been observed across different ethnic backgrounds, where part of this association has been attributed to increased adiposity (53).

Secular trend of menarche

Menarche is often used as a marker of the moment of puberty (19), and in many cases, it is considered an important parameter of pubertal development, which may reflect socioeconomic and nutritional changes in a population (54). Historically, the age at menarche (MA) has declined significantly, from approximately 17 years in the early 19th century to around 13 years by the mid-20th century in Europe and the USA (7, 18, 55). While this trend was pronounced in the first half of the 20th century, the rate of decline has attenuated since 1960, with some populations showing stabilization or even a slight increase in MA (49). This downward trend in MA is largely attributed to improvements in nutrition, hygiene, and living conditions of the population and appears to have decreased because of increased stability in socioeconomic conditions (7, 55).

Some studies have shown that the decrease in puberty onset was not accompanied by the same magnitude as the decrease in MA, as in the Danish study: while the age at thelarche decreased by 12 months between its 1991 and 2006 cohorts, the corresponding decrease in MA was only 3 months (from 13.4 to 13.1 years). The earlier breast development was not associated with elevated gonadotropins levels, suggesting a potential gonadotropin-independent estrogenic mechanism rather an activation of the HPG axis (3), which may represent a plausible explanation for the more modest reduction seen in MA in relation to age at thelarche.

The anticipation of MA, which has been described over the past few decades, has been largely attributed to the increasing prevalence of childhood and adolescence obesity (54). Substantial evidence from both cross-sectional and longitudinal studies supports a strong association between excess body weight and earlier menarche (6, 9, 54, 56) (Fig. 4).

Figure 4.

Age at menarche according to nutritional status. The data in this graph have been compiled for this review from several historical sources.

Pioneering epidemiological studies from the 1970s suggested a critical minimum weight for the occurrence of menarche of around 48 kg (23). Similarly, a recent study in Chinese girls found no significant weight difference between those with normal and early menarche, supporting the necessity of a critical weight threshold. In this cohort, with a mean MA of 10.7 years, a 1-standard deviation increase in BMI was associated with a nearly twofold greater risk of early menarche. After adjustments for potential confounders, overweight and obese girls had 2.9 and 8.9 times higher odds of early menarche, respectively, compared to their normal-weight peers (57). Another Chinese cross-sectional study with almost 3,000 girls reported a mean MA of 12.4 years, with girls with obesity and overweight experiencing menarche at 11.4 and 11.7 years, respectively, compared to 12.5 years in normal-weight girls (9) (Fig. 4). In an Italian longitudinal study, prepubertal BMI was inversely associated with age at puberty onset and with MA. In this study, the mean age at menarche was 12.4 years in normal-weight girls and 11.5 in obese children (8). In the USA, data from NHANES III survey (1988–1994) showed an MA in girls with normal weight of 12.6 years and in those with overweight of 12.1 years (51).

Subsequently, in an American prospective longitudinal study in the first decade of the 2000s, the MA was 12.2 years, showing a minimal decrease in the 20 years prior to the study. Body weight was associated with MA: overweight and obese girls reached menarche 0.3 and 0.5 years earlier, respectively, than their normal-weight counterparts, while underweight girls were delayed by 0.5 years. The study also highlighted racial differences, with black girls reaching menarche earlier than white girls (12.0 vs 12.7 years) (56). Similarly, a recent Danish longitudinal cohort reported a mean MA of 13.1 years, which occurred 4.2 months earlier in overweight girls and 5.7 months earlier in those with obesity (Fig. 4) (6). In a study conducted with more than 1,000 girls from the Brazilian Amazon region, born in the first decade of the 2000s, the average MA was 11.9 years, with obese girls menstruating approximately 0.6 years before normal-weight girls and 1.3 years before thin girls (54) (Fig. 4).

Secular trend of pubertal onset in boys

A secular trend toward an earlier onset of male sexual characteristics has also been observed in several places around the world, with much of this anticipation linked to the rising prevalence of obesity (5, 6, 8, 11). However, unlike what is observed in girls, the relationship between pubertal timing and body fat may not be linear and, at times, contradictory, with some studies suggesting an inverse relationship with obesity (12, 13, 14, 16). The age of gonadarche in boys in various countries is presented in Fig. 5 (5, 6, 8, 12, 20, 58).

Figure 5.

Age of gonadarche in boys in the past 50 years. The data in this graph have been compiled for this review from several historical sources.

Defining the onset of male puberty presents significant challenges compared to female puberty. Boys lack a clear, self-reported milestone equivalent to thelarche or menarche. In addition to increased testicular volume, other indirect markers are used as milestones in male pubertal development, such as voice break and peak growth velocity, which can contribute to variability in data analysis.

In the USA, an analysis of cross-sectional data from NHANES III indicated that American boys experienced gonadarche earlier than their European counterparts. In that study, the mean age for the initiation of genital development was 10.1 years for white boys and 9.5 years for African American boys (20) (Fig. 5).

About 10–20 years after this study, a large, racially diverse American sample found no significant change in the age of gonadarche during this period in the USA but revealed variations of both onset and completion of puberty according to weight category. For white boys, the mean age of gonadarche was 10.0 years for those with normal weight, 9.3 years for those overweight, and 9.4 years for those with obesity. Pubertal completion occurred at 15.2, 14.5, and 15.4 years for the normal-weight, overweight, and obese groups, respectively. These findings led to the conclusion that puberty began earlier in overweight boys but was completed later in obese boys, although these patterns were not consistent across all pubertal stages or racial and ethnic groups (12).

Using the NHANES III database, significant associations were shown between early sexual maturation and obesity for both boys and girls, but these associations were in opposite directions. Boys with early maturation had lower BMI and skinfold thickness measurements than their peers (14). Another American study supported this, finding that a higher BMI during early and middle childhood was not associated with earlier pubertal onset and may, in fact, be linked to a later onset (13). Adiposity has also been inversely related to testicular volume, with obese boys showing a significantly smaller testicular size after a bone age (BA) of 13 years compared to non-obese boys, suggesting delayed puberty despite similar skeletal maturation (16).

In Europe, a marked secular trend toward earlier male puberty was not evident from the mid-1960s to the late 1990s (18). In a British study of the 1970s, the age of the gonadarche was approximately 11.6 years (58). However, more recent data from Denmark, comparing cohorts from 1991 to 1993 and 2006 to 2008, revealed that pubertal onset had advanced by approximately three months and puberty completion was reduced by approximately 12 months in the 2006–2008 cohort. This study demonstrated that age-adjusted BMI was negatively associated with age at onset of testicular growth. In addition, this reduction in age at pubertal onset did not remain significant after adjustment for BMI, indicating that BMI, at least in part, explained these findings. In that same period in Denmark, the prevalence of overweight and obese boys nearly doubled, rising from 5.8% overweight and 5.3% obese in the 1991 cohort to 9.7% overweight and 9.4% obese in the 2006 cohort (5). A later Danish longitudinal study confirmed this trend, reporting a gonadarche age of 10.9 years in normal-weight boys, with puberty occurring 3.1 months earlier in overweight boys and 3.5 months earlier in obese boys (6).

A Swedish study showed that the age of peak height velocity (PHV) became progressively earlier by 1.5 months per decade for each decade of increase in birth year over a 50-year period, a trend partly attributed to increasing childhood BMI. However, BMI could not fully explain these findings, as the trend persisted even after adjusting for BMI, suggesting that other factors are also involved (11). More recently, a large longitudinal study revisited the ‘critical weight hypothesis’, demonstrating that prepubertal body weight is a more robust inverse predictor of pubertal timing than prepubertal BMI in boys, with body weight being associated with age at PHV independently of BMI (59). In another longitudinal study, the mean age of puberty onset was 12.1 years in the normal-weight group and 11.7 years in the obese group, both starting and completing puberty earlier (8). In Chile, obesity at 4–7 years presented a significant inverse association with precocious gonadarche in a longitudinal study. Age at testicular enlargement in normal-weight, overweight, and obese boys was 11.2, 10.9, and 10.7 years, respectively (60). In contrast, a cross-sectional study in Egypt found the age of gonadarche to be 11.3 years, with no significant relationship to BMI (10).

Conclusion

Evidence has been accumulating over the past few decades to support the early onset of male and female puberty, which is at least partly associated with excess weight, especially in girls but also in boys. Therefore, the observation that obesity is associated with delayed puberty in males is becoming increasingly unlikely, although a delay in puberty completion may be considered in boys. With the growing epidemic of obesity in children and adolescents and robust evidence of the impact of this association on the onset and progression of puberty, it is essential to understand the mechanisms that underlie this pathophysiological basis and prioritize measures to prevent excess weight, especially during this developmental window, and thus reduce risk factors for reproductive and metabolic diseases.

Declaration of interest

SRRA is a Senior Editor of Endocrine Connections, and KCO and LLKE are reviewers of Endocrine Connections. They were not involved in the review or editorial process for this paper. There are no other conflicts of interest relevant to this article to declare.

Funding

SRRA was supported by FAPESP (Sao Paulo State Research Funding Agency), CNPq (Brazilian National Council for Research), and FAEPA-HCFMRP-USP. LLKE was supported by FAPESP (grant # 18/18071-5), CNPq, and FAEPA-HCFMRP-USP.

Author contribution statement

KCO performed the literature review and wrote the initial draft of the manuscript. SRRA and LLKE critically reviewed and edited the manuscript. All authors reviewed and approved the final draft.