Update on Precocious Puberty: Girls are Showing Signs of Puberty Earlier, but Most Do Not Require Treatment
Precocious puberty is one of the most common endocrine disorders seen by primary care physicians and continues to be a major source of concern for both parents and providers. Since the author’s previous review on this subject in Advances In Pediatrics in 2004 [1], there have been several reports that have added to the knowledge base and confirmed that signs of puberty in girls are appearing earlier than in the past in the United States as well as in other countries. In this article, the author reviews what pediatricians should know about the physical findings seen during normal puberty and their hormonal basis. The evidence that signs of puberty are appearing earlier than 30 to 40 years ago, at least in girls, and the major theories about why this might be occurring are also discussed. However, the key point to be made is that pediatricians should be able to recognize the benign causes of early breast and pubic hair development, which are, at least in the United States, far more common than the cases that might benefit from treatment. The current state of knowledge concerning the diagnosis and treatment of central (true) precocious puberty (CPP) and the red flags that might point toward one of the rare causes of gonadotropin-independent precocious puberty (GIPP) are then reviewed.
Review of normal puberty
In girls, the first sign of true puberty is breast development, caused by estrogens produced by the ovaries, under the influence of increased pituitary secretion of the gonadotropin, luteinizing hormone (LH). There is also an increase in follicle-stimulating hormone (FSH), which is more critical for maturation of the immature oocytes. Traditionally, it has been thought that the normal age range for this event, also called thelarche, is between 8 and 13 years; however, as discussed later, there is ample evidence that in many healthy girls, this event can occur before the age of 8 years. Shortly after breasts appear, the pubertal growth spurt starts, and menarche occurs 2 to 3 years later because gonadotropin secretion increases so that there is a midcycle surge of LH, which triggers ovulation.
According to a recent longitudinal study, the appearance of pubic and/or axillary hair follows the appearance of breast tissue in 66% of girls [2]. This occurrence is often confused with the onset of puberty, but has a different hormonal basis, due to an increase in the production of weak adrenal androgens, primarily dehydroepiandrosterone (DHEA), an event known as adrenarche. In the average girl, adrenarche starts between the ages of 9 and 11 years but is not uncommon in girls aged 6 to 8 years, a situation called premature pubarche or premature adrenarche (PA), which is discussed below.
In boys, the earliest reliable sign of puberty is enlargement of the testes (>3 cm3 in volume or >2.5 cm in diameter) under the influence of increased FSH secretion, which occurs at an average age of 11 to 12 years but can occur as early as age 9 years. It typically takes 12 to 18 months after testicular enlargement for testosterone production, under the influence of increasing LH secretion, to increase and cause penile growth and the growth spurt, which in the average boy peaks at 13 to 14 years of age. A recent longitudinal study found that testicular enlargement preceded pubic hair development in 91% of boys [2], but pubic hair may occur several years earlier than genital development because of increased adrenal androgen secretion.
Are girls maturing earlier? If so, how should we define when pubertal development is precocious?
For many years, going back to the classic study of Marshall and Tanner [3] in the 1960s, which was based on a nonrandom sample of 192 white British girls living in a group home, the age at which puberty was considered precocious in girls was by consensus thought to be 8 years. A landmark study in 1997, based on 17,000 girls aged 3 to 12 years examined in 65 pediatric offices participating in the Pediatric Research in Office Settings (PROS) network, confirmed what many had already suspected: that appearance of breasts and pubic hair in girls before age 8 years was common [4]. In 8-year-old girls (defined as from the eighth birthday to the day before the ninth birthday), 37.8% of black and 10.5% of white girls had at least Tanner stage 2 breast development, and a similar percentage had pubic hair. Even for 7-year-old girls, the numbers were high, 15.4% of black and 5% of white girls had breast development. An article written by the author suggested revising the definition of precocious puberty in girls downward to age 7 years in white girls and age 6 years in black girls to take into account this new data [5]. However, many objected to this recommendation on the basis that the PROS study was flawed because of the participants not being a randomized sample. In addition, because the Tanner staging was done mainly by inspection and not palpation, the results may have been biased because of the problems distinguishing breast tissue from fat tissue in chubby girls. However, similar results were found in the National Health and Nutrition Examination Survey (NHANES-III), which from 1988 to 1994 examined pubertal status in 1623 girls between the ages of 8 and 16 years who were scientifically selected to represent the general population [6]. Recently, new data were reported by Biro and coworkers [7] who examined a cohort of 1239 girls aged 7 and 8 years recruited from 3 centers in East Harlem (NY), greater Cincinnati (OH), and the San Francisco Bay area (CA). The investigators found that the proportion of girls in this group with breast development confirmed by palpation may be even higher than in the 2 earlier studies. The results of these 3 studies are summarized in Table 1. Although there are differences in the numbers reported, the key point is that all studies show a high incidence of early breast development and pubic hair in girls aged 7 to 8 years from all regions of the United States, with the highest percentages seen in black girls.
Table 1 Proportion of 7- and 8-year-old girlsa with breast development and pubic hair from 3 recent US studies
Another way of looking at pubertal trends in girls is to calculate the mean age of 2 pubertal milestones: thelarche and menarche. The PROS and NHANES-III studies found that the mean age of thelarche, 10.0 to 10.3 years in white girls and 8.8 to 9.5 years in black girls, was earlier than that in previous studies done in the United States and the United Kingdom, which had indicated that the mean age was closer to 11 years. Until recently, most studies from Europe and other industrialized countries had not found any decrease in the mean age of breast development from the 1960s to the 1990s. However, the Copenhagen Puberty Study done from 2006 to 2008 found that girls attained stage 2 breast development at a mean age of 9.86 years, a full year earlier than the previous study done in 1991 to 1993 by the same research group, using the same methodology [8]. Even when adjustments were made for body mass index (BMI), which changed little between the 2 cohorts, the difference remained significant. Furthermore, a recent study on more than 20,000 healthy girls living in China found that the median age for Tanner stage 2 breast development was only 9.2 years, with the median age for Tanner 2 pubic hair significantly later at 11.16 years and the median age at menarche 12.27 years [9]. Despite the rather dramatic declines reported for the mean age of thelarche, changes in the mean age at menarche have been more modest. In the United States, the PROS study found no decrease in age at menarche when compared with an earlier study in 1966 to 1970 (both 12.7–12.8 years), whereas in the NHANES III study, age at menarche decreased slightly to 12.5 years [6]. A more recent NHANES survey done from 1999 to 2004 found the mean age at menarche for women born from 1980 to 1984 to be 12.4 years [10]. In the Copenhagen study, the mean age at menarche declined from 13.42 years in 1991 to 13.13 years in 2006, again less than the magnitude of the decline in the mean age at thelarche [8]. It is also noteworthy that the mean interval from thelarche to menarche was (13.13−9.82) 3.3 years, much longer than the mean interval of 2.3 ± 1.1 years reported by Marshall and Tanner [3]. This finding suggests either 1) that it now takes long for most girls to progress from thelarche to menarche than in the past or 2) that a significant number of contemporary girls have a nonprogressive or slowly progressive variant of puberty and may not progress at all for a year or two after the first sign of breast development [11].
The key question pediatricians struggle with remains this: given that many studies show a trend toward earlier breast (as well as pubic hair) development, at what age should signs of puberty in girls be considered abnormal and thus trigger a referral to an endocrinologist. As noted earlier, a 1999 review suggested that the age limits for when puberty is considered precocious should be decreased to 7 years in white and 6 years in black girls [5]. Although the proportion of 7-year-old girls with signs of puberty is quite high, most specialists and pediatricians still feel more comfortable with the traditional cutoff of 8 years, even though this cutoff labels many girls as abnormal who are undoubtedly at the lower end of the current normal range for pubertal maturation. The author believes that rather than focusing on one age as separating normal from abnormal, the rate of progression of pubertal maturation in deciding how concerned to be about a particular child should be taken into account. For example, in a 7.5-year-old girl with a small amount of breast tissue (Tanner stage 2) on examination and no evidence of a growth spurt, pubertal maturation may progress slowly and often the findings remain the same 6 months later; this girl does not benefit from treatment to suppress puberty (see later). On the other hand, an 8.25-year-old girl who is at Tanner 3 for breast development and has grown 3–4 inches in the past year is clearly progressing more rapidly and needs prompt evaluation. In addition, pediatricians should be more concerned about girls with early and progressive breast development with or without pubic hair than girls who only have pubic hair because girls in the latter group rarely have a condition requiring intervention.
Are boys maturing earlier than in the past and when should they be referred?
For many years, the definition of precocious puberty in boys has been signs of maturation starting before the age of 9 years, but there have been relatively few studies on whether the mean age at the onset of puberty in boys has decreased over the past 30 to 40 years. One report based on the NHANES III study seemed to indicate that stage 2 genital growth, a more reliable sign of puberty in boys than pubic hair, was occurring much earlier than earlier studies had found, with a mean age of about 10 years [12]. However, the accuracy of sexual maturation ratings in this survey has been questioned. A study of male puberty in the United States undertaken by the PROS network is nearing completion, and these data should be available in the near future. In the Copenhagen Puberty Study, it was recently reported that compared with the period from 1991 to 1993, boys examined during 2006 to 2008 had slightly earlier attainment of testicular volume of greater than 3 cm3, which correlated with an increase in BMI [13]. Unlike the situation for girls, it is generally agreed that endocrinologists have observed no increase in the past 40 years in the number of boys referred for true precocious puberty. At present there is no reason to change the age at which pubertal development in boys is considered precocious, which should remain 9 years.
What can we say about why girls are starting puberty earlier?
This topic was discussed at length in the author’s article in 2004 for this series, and in this section, the author mainly highlights recent articles that bear on the subject. The relationship between increased body fat (typically measured as elevated BMI for age) and earlier puberty has been summarized in a recent review [14]. Multiple studies have documented that higher BMI (which is a good surrogate measure of increased body fat) is correlated with earlier onset of breast development, pubic hair, and menarche. Although the precise cause of this relationship is not known, one possibility is that higher levels of the fat cell–derived protein leptin, seen in children with increased body fat, allow puberty to start earlier. Leptin is required for normal gonadotropin secretion because mice and humans lacking the ability to produce or respond to leptin fail to make LH and FSH, and providing leptin to leptin-deficient mice results in the onset of puberty. One area of debate is whether earlier puberty itself is the cause of the increase in body fat or if an increase in body fat precedes and causes the earlier appearance of breast development. A recent article by Lee and colleagues [15] presented findings that favor the latter interpretation. The study included 354 girls from the National Institute of Child Health and Human Development Study of Early Child Care and Youth Development for whom there was longitudinal data on BMI at ages 36 and 54 months and grades 1, 4, 5, and 6, with pubertal assessments by physical examination in grades 4 to 6. The investigators found that elevated BMI standard deviation scores at ages 36 and 54 months and in grade 1 and the rate of change in BMI from 36 months to grade 1 were associated with increased odds of earlier puberty (defined as Tanner ≥2 breasts in grade 4 or Tanner ≥3 breasts in grade 6). This finding suggests that well before the onset of puberty, as early as 3 years of age, the presence of increased BMI is an important factor in determining the age of onset of puberty in girls in the United States. It is curious that the same relationship between increased BMI and the timing of puberty has not consistently been found in boys. A recent study from the same group found that boys who were in the highest BMI trajectory through 11.5 years had a significantly higher chance of remaining prepubertal than boys in the lowest BMI trajectory [16], suggesting that increased body fat does not promote earlier puberty in boys, at least in the United States; this contrasts with the recent findings in Danish boys.
Many clinicians and investigators have proposed that chemicals in the environment with weak estrogenlike properties, including certain pesticides, phthalates, bisphenol A, and plant-derived phytoestrogens, or low levels of estrogens itself in the food supply are driving the trend toward earlier puberty in girls [17,18]. The investigators of one review pointed out that prepubertal girls are extremely sensitive to low levels of exogenous estrogens and that exposures previously thought to be safe based on less-sensitive measures of estradiol levels in young children may in fact have important effects on growth and pubertal development, even at levels below the detection limits of most assays [17]. However, except in a few isolated cases, evidence for specific exposures linked to earlier onset of puberty has been difficult to come by. Research on such exposures in girls with earlier pubertal development is ongoing, but it is recognized that examining blood or urine levels of these chemicals in a 7- to 8-year-old girl may miss a potentially higher level of exposure to one or more of these chemicals in the critical perinatal period.
Common variations of pubertal development
Pediatricians confronted with a child with early breast development or pubic hair need to know how to distinguish the common benign normal variations from children who need prompt referral for possibly treatable conditions, as well as the limited usefulness of hormone testing and radiography in evaluating such children.
Premature adrenarche
