Similarly, genital maturation in boys is divided into five stages from stage 1 (prepubertal) in which the penis, testes, and scrotum are about the same size and proportions as in early childhood to stage 5 (fully mature, see Table 4 and Fig. 1). In a manner similar to that in girls, pubic hair is divided into stages 1 (none) to 5 (dark, coarse mainly in the triangular distribution but may spread to the medial thighs or vertically). Adolescent boys may attain adult height and mature sexual maturation years before acquiring their adult body composition.

Testicular maturation (size) is conveniently measured by comparison to a Prader orchidometer, which is a set of ellipsoids roughly of the size and shape of testes from prepubertal to fully adult (Fig. 5). Other available methods include rulers (semimajor and semiminor axes), calipers, and ultrasound. The normal prepubertal testis is 1–2 mL and the fully mature one 15–25 mL. One considers puberty to have begun when the testes are >3 mL. There is a rough correlation between testicular size and the Tanner stages for genital and pubic hair maturation. The common practice is to denote each parameter individually.

The pubertal growth spurt can be divided into three stages: the stage of minimal growth velocity just before the spurt (takeoff velocity), the stage of most rapid growth or peak height velocity (PHV), and the stage of decreased velocity and cessation of growth at epiphyseal fusion. Boys reach PHV approximately 2 years later than girls and are taller at takeoff; PHV occurs at stages 3 to 4 of puberty in most boys and is completed by stage 5 in more than 95 % of boys (Largo, Gasser, Prader, Stuetzle, & Huber, 1978; Tanner, Whitehouse, Marubini, & Resele, 1976). The mean takeoff age is 11 years, and the PHV occurs at a mean age of 13.5 years in boys. The total height gain in boys between takeoff and cessation of growth is approximately of 31 cm (Abbassi, 1998). The mean height difference (boys taller than girls) between adult height men and women is 12.5 cm.

In physiologically mature individuals, the secretion of gonadal steroids (testosterone in boys and estradiol in girls) is controlled by the gonadotropic hormones, luteinizing hormone (LH), and follicle-stimulating hormone (FSH), secreted by the pituitary gland. LH and FSH secretion are themselves regulated by gonadotropin-releasing hormone (GnRH) which is produced in the hypothalamus. In the adult, GnRH is secreted in tightly regulated bursts (pulses) every 60–120 min throughout the day (Knobil, 1980). This pulsatility is required for normal gonadotropin secretion and gonadal steroid production. In the circulation, estradiol and testosterone travel not only to target organs but also to the CNS, where they act to suppress production of GnRH, LH, and FSH, thus establishing a negative feedback loop.

In the prepubertal child, the situation differs markedly from the adolescent and adult. During early infancy, the HPG axis is active, but it then becomes suppressed until the approach of puberty. This suppression is known as the juvenile pause. During this stage, GnRH and gonadotropin secretion are highly suppressed by the low concentrations of circulating gonadal steroids (Sisk & Foster, 2004). With currently available assays, the serum concentrations of testosterone and estradiol are usually unmeasurable, but they circulate at levels below those necessary for the physical changes of puberty. However, estradiol levels are sufficient to add to the growing difference in fat mass in boys and girls between the ages of approximately 6 years and the onset of the external signs of pubertal maturation. The heightened negative feedback keeps sex hormone concentrations low and prevents pubertal maturation, although GnRH continues to be produced in small, irregular, and widely spaced pulses (Sisk & Foster, 2004). Gonadotropins are also secreted, again at very low rates that may be difficult to detect in routine clinical laboratory assays.

In late childhood, before any physical evidence of puberty is present, the intensity of the negative feedback diminishes. Although the mechanism for this process is poorly understood, GnRH pulse amplitude and frequency both increase as the sensitivity to sex steroids decreases (Dunkel, Alfthan, Stenman, Tapanainen, & Perheentupa, 1990). The increased GnRH pulsatility first occurs primarily at night, with resulting diurnal variation in gonadotropin secretion and waxing and waning levels of sex hormones over the course of the day (Fig. 6). The increasing circulating sex-steroid levels directly lead to the physical and behavioral alterations noted during pubertal maturation. As puberty progresses, GnRH pulses become more consistent throughout the day, and the troughs and peaks in sex hormone concentrations tend to even out somewhat. However, the diurnal testosterone variation in boys persists into adulthood, although with a smaller day/night difference (Manasco et al., 1995). For this reason, measurement of testosterone in adolescent and young adult males is best performed in the AM, when levels tend to be at their highest. In girls, serum estradiol concentrations increase from <5 pg/mL (∼18 pmol/L) before puberty up to >100 pg/mL (∼370 pmol/L) in a post-menarchal girl, varying greatly over the menstrual cycle. Testosterone concentrations in prepubertal males generally are below 10 ng/dL, (∼0.3 nM) and in young adults they are above 300 ng/dL (∼10 nM).

During the prepubertal years, the ovaries gradually increase in size, and the number of small follicles increases (Bridges, Cooke, Healy, Hindmarsh, & Brook, 1993; Peters, Byskov, & Grinsted, 1978). In response to increased GnRH pulses primarily during the nighttime, gonadotropin secretion increases, particularly FSH. The increased serum concentration of FSH leads to increased follicular recruitment. Acting via its receptor, LH promotes androgen production, primarily androstenedione, from the ovarian thecal cells. FSH stimulates aromatase activity in the granulosa cells and leads to estradiol production. Before menarche, estradiol levels peak approximately 12 h after the diurnal peaks in gonadotropin secretion (Norjavaara, Ankarberg, & Albertsson-Wikland, 1996).

As puberty progresses, a cyclic pattern emerges in estradiol secretion (Rosenfield, Cooke, & Radovick, 2008). In this setting, estradiol comes from the small antral follicles developing midway through pubertal maturation. As estradiol concentrations increase and promote endometrial proliferation, anovulatory bleeding may occur following cyclic declines in estradiol. True ovulatory cycles do not begin until after the HPG axis matures to the point at which the mid-cycle LH surge takes place.

Before ovulation (the follicular phase), gradually increasing FSH and LH production prompt the recruitment of small follicles, which begin to secrete estradiol. With increasing stimulation, the follicle with the greatest FSH sensitivity becomes dominant and begins to produce progesterone as well as additional estradiol. Estradiol causes proliferation of the endometrial lining of the uterus. Although in boys, sex steroids consistently have a suppressive effect on GnRH and gonadotropin secretion, such is not the case for girls. In the mid-part of the menstrual cycle, increasing serum concentrations of estradiol, acting synergistically with progesterone, cause a dramatic increase in LH secretion, with levels approaching 60 IU/L. In the dominant (ovulatory) follicle, alterations in gene transcription in response to the LH surge result in the formation of inflammatory mediators and disruption of cellular integrity, causing rupture of the follicle and ovulation (Russell & Robker, 2007). After ovulation (the luteal phase), estradiol concentrations initially decline, but as the corpus luteum forms, both estradiol and progesterone levels increase, resulting in endometrial maturation. In the absence of embryonal hCG, however, the corpus luteum has a limited lifespan, and hormone production declines. With falling estradiol and progesterone levels, the endometrium becomes ischemic and sloughs off, resulting in menstrual bleeding. The high FSH levels observed during the follicular phase and at the end of the luteal phase cause the initial recruitment of another wave of small follicles, one of which will become the dominant follicle 2½ cycles later (Gougeon, 1996).

The timing of puberty is a complex trait, and in the general population, it has a Gaussian distribution. There are many influences on the regulation of pubertal timing, including nutritional factors, environmental influences, and genetic input. Complex traits often demonstrate a high degree of genetic regulation, and it is estimated that between 50 and 80 % of the variance of normal pubertal timing is explained by genetic factors (Palmert & Boepple, 2001). Efforts to understand the genetics of pubertal timing have led to the discovery of many genes that are clearly necessary for pubertal maturation. Many of these, when mutated, cause specific syndromes of delayed or absent puberty. However, it is not clear that these genes individually or as a group explain much of the variability of the onset of normal puberty.

In 2003, the kisspeptin/GPR54 system was discovered. It is a key regulatory system in the initiation of increased GnRH pulsatility at the onset of puberty (Seminara et al., 2003). Kisspeptin is produced by hypothalamic neurons and interacts with its receptor, GPR54, on GnRH-secreting neurons. In humans, inactivating mutations are associated with pubertal delay, and an activating mutation has led to precocious puberty (Teles et al., 2008). Exogenous administration of kisspeptin increases the pulse amplitude of GnRH. Short-term administration of kisspeptin to healthy male and female volunteers increases gonadotropin concentrations, although chronic dosing appears to induce tachyphylaxis (Dhillo et al., 2005, 2007; Jayasena et al., 2009). The kisspeptin/GPR54 system may also be involved in the negative and positive feedback effects of sex steroids on GnRH secretion. Although it is clear that kisspeptin plays a role upstream of GnRH, the factors regulating its release are not presently fully understood (Kauffman, Clifton, & Steiner, 2007).

Both inhibitory and stimulatory factors in the central nervous system are involved in the initiation of puberty. Gamma-aminobutyric acid (GABA)-secreting neurons play an inhibitory role and may be involved in the temporary suppression of the HPG axis during childhood. Suppression of GABAergic neuronal input may lead to the initiation of puberty (Terasawa & Fernandez, 2001). Additional factors may include increased glutamate stimulation of N-methyl-d-aspartate (NMDA) receptors, which are stimulatory to GnRH release, and neuropeptide Y-expressing neuron activity (Gamba & Pralong, 2006). Leptin, acting through its receptor, links nutritional status to puberty, and sufficient levels of leptin are thought to be required to initiate puberty (Farooqi et al., 2002).

One must have sufficient energy to support body growth and to store energy, predominantly as fat, for longer-term energy requirements such as the menstrual cycle, pregnancy, and lactation (Loucks, 2007). Thus, energy availability is one of the factors that contribute to the regulation of timing of puberty and the menstrual cycle. Energy availability is defined as the difference between energy intake and energy expenditure, and it may be decreased by either lowering food intake or increasing caloric expenditure through exercise. Loss of energy availability leads to alterations in puberty and the menstrual cycle by decreasing the frequency of GnRH pulses, thus lowering mean serum levels of gonadotropins, decreasing ovarian follicle formation, and decreasing estradiol secretion (Gordon, 2010). This phenomenon may be mediated by loss of body fat and decreases in circulating leptin, considered permissive for normal hypothalamic-pituitary-ovarian activity. Younger women and adolescents are particularly prone to hypothalamic amenorrhea as a result of decreased energy availability. Loss of energy availability is reflected clinically by decreases in percent body fat and, to a lesser extent, body weight and BMI. In athletes, weight and BMI may not accurately reflect body fat stores due to the higher density of muscle. Restoration of energy availability results in increased body weight. Return of body weight to >90 % of ideal usually, but not always, results in resumption of menses in amenorrheic patients (Golden et al., 1997). Whether weight loss occurs from excessive exercise or by severe caloric restriction, as seen in starvation and anorexia nervosa, the hypothalamic effects appear the same.

Growth hormone release is intermittent (“burst-like” or pulsatile) at all ages and is magnified in low energy states, such as malnutrition (Soliman et al., 1986) and anorexia nervosa (Argente et al., 1997; Scacchi et al., 1997). Once a regular day/night cycle is established, there is a major burst of GH release 45–90 min following the onset of deep sleep (Martha, Gorman, Blizzard, Rogol, & Veldhuis, 1992). Generally, more than 75 % of the total daily production of GH accompanies sleep during the nighttime hours. However, this amount is dampened to a variable degree by the amount of body fat and its distribution in the subcutaneous vs. abdominal/visceral regions. There are significant increases in GH release during the newborn period, possibly related to a state of transient nutritional insufficiency (see Anorexia Nervosa, below) (Wright, Northington, Miller, Veldhuis, & Rogol, 1992) and more marked increases during adolescence (see below) (Martha et al., 1992).

Changes in GH release during pubertal maturation are large. There is an approximately threefold increase in the amount of GH released, which translates into a similar increase in serum IGF-I concentrations (Martha et al., 1992; Martha, Rogol, Veldhuis, & Blizzard, 1996; Roemmich et al., 1998). These increases are tied to the rising levels of sex steroids, but toward the end of pubertal maturation the GH secretory alterations return virtually to the prepubertal state (Kerrigan & Rogol, 1992). Variation in body composition and the regional distribution of fatness also influence GH release during puberty and into adulthood (Roemmich et al., 1998; Roemmich & Rogol, 1999; Veldhuis et al., 2005; Veldhuis, Roemmich, & Rogol, 2000).