One aim of this study is to determine changes in body composition and hormones that differentiate athletes who stop getting their periods versus those who continue to get their periods and non-athletes. The second aim of this study is to determine whether transdermal or oral estrogen (versus no estrogen) is effective in increasing bone density and improving bone microarchitecture in adolescent athletes who are not getting their periods and are thus estrogen deficient. The investigators hypothesize that transdermal estrogen will be more effective than oral estrogen or no estrogen in improving bone health in amenorrheic adolescent athletes.

As many as 25% of adolescent and young adult endurance athletes develop amenorrhea, and factors that cause amenorrhea to occur in some, but not all, athletes have not been well characterized. Recent data indicate the critical importance of a negative energy balance state and leptin in regulating the Hypothalamic-pituitary-gonadal (H-P-G) axis. However, these factors do not completely account for alterations in this axis, and other contributing factors are unclear. Our preliminary data indicate the importance of low fat mass and fat related hormones in mediating hypogonadism in young athletes. This study will confirm these data and determine whether low fat mass and altered levels of adipokines, such as leptin and adiponectin, and hormones regulated by fat mass, such as ghrelin and peptide YY (PYY), determine alterations in LH pulsatility. A very concerning impact of amenorrhea in athletes is low bone mineral density (BMD). Preliminary data indicate lower BMD in adolescent athletes with amenorrhea (AA) compared with eumenorrheic athletes (EA) and non-athletic controls. The high prevalence of AA in adolescents is particularly concerning, because this population is potentially at greater risk as it is actively accruing bone. Of importance, bone microarchitecture, a better predictor of bone strength than BMD, has not been studied in AA. Because pubertal increases in estrogen are integral to optimizing peak bone mass, and AA is characterized by hypoestrogenism, this randomized study of transdermal estrogen versus oral estrogen or no estrogen will also examine whether estrogen replacement increases BMD and improves bone microarchitecture in adolescent AA 14-21 years old. EA and sedentary controls will be followed without intervention for this period. Despite the prevalent practice of prescribing oral contraceptives in AA, there is a paucity of data regarding benefits of this intervention in teenagers. Because transdermal estrogen, unlike oral estrogen, does not suppress IGF-1, an important bone anabolic factor, we expect effects of transdermal estrogen to exceed those of oral estrogen or no therapy. In addition, preliminary data indicate that low fat mass and alterations in fat related hormones may contribute to decreased bone accrual rates in athletes, and will be confirmed in this study. To summarize, a better understanding of the pathophysiology of reproductive dysfunction is critical to develop therapeutic strategies that will normalize the reproductive axis and bone accrual, and these are the questions that this study aims to answer.

Inclusion Criteria:

• Females 14-21 years old Note: Our pilot data are reassuring in that young women 18-25 years old with hypothalamic amenorrhea are not adversely affected with estrogen use. In fact, in our prospective study, beneficial effects were observed both in young women 18-25 years old using oral estrogen, and in 14-18 year old adolescent girls using transdermal estrogen. We therefore feel that including girls in the 14-21 year age range will not be hazardous to their bone health. In fact, given the lack of data in this age group, it is important to study younger women and teenagers rather than extrapolate data from studies in adults to this younger population. Hormone dynamics differ in teenagers compared with adults, and bone mass accrual is even more dependent on estrogen and IGF-1 in younger than older women who have already achieved peak bone mass.
• Bone age (BA) >15 years Note: 99% of adult height is achieved at a BA of 15 years, thus estrogen replacement will not result in stunting of height potential after this age. Although we could have chosen to include girls with a BA >14 in this study, we are limiting this to girls with a BA of >15 years. This is because 2% of growth potential persists at a BA of 14 years, versus only 1% at a BA of 15 years (~0.6" of potential height (130)). Thus, to avoid potential stunting of growth potential with estrogen replacement, we have chosen to include girls with BA of > 15 years.
• BMI between 10th-90th percentiles for age.
• Amenorrhea (for AA): absence of menses for > three months (74) within a period of oligomenorrhea (cycle length > six weeks) for >six months, or absence of menarche at >16 years.
• Eumenorrhea (EA and controls): > nine menses (cycle length 21-35 days) in preceding year.
• Non-athlete healthy controls will be eligible if weight bearing exercise activity is less than two hours a week and if they are not participating in organized team sports.
• Endurance athletes Note: severity of low BMD and menstrual dysfunction differ by kind of exercise and activity. For example, runners have a higher prevalence of menstrual irregularity than swimmers and cyclists (131). By limiting enrollment to endurance athletes, we will eliminate variability from the type of activity. Endurance training is defined as > 4 h of aerobic weight-bearing training of the legs or specific endurance training weekly, or > 20 miles of running weekly for a period of > 6 months in the last year.

Exclusion Criteria:

• Other conditions that may affect bone metabolism

Athletes with amenorrhea

1 year