Journal Logo

Clinical Sciences: Clinical Case Studies

Normalization of Bone Density in a Previously Amenorrheic Runner with Osteoporosis


Author Information
Medicine & Science in Sports & Exercise: September 2005 - Volume 37 - Issue 9 - p 1481-1486
doi: 10.1249/01.mss.0000177561.95201.8f
  • Free


Young athletes who participate in sports that emphasize a lean body type are at risk of developing disordered eating and its sequelae, menstrual irregularity and low bone density (2,15,18,29). Women who develop this “female athlete triad” may fail to reach their expected peak bone mass and hence increase their lifetime risk of fracture. Weight gain, improved nutrition, and resumption of menses are the only established treatments for reversing bone deficits in amenorrheic athletes (10,11,26,27). Although hormone replacement therapy is widely prescribed (6), limited randomized trials suggest that hormone replacement is not effective in improving bone density in this population (3,27).

The extent to which large bone density deficits can be reversed, even with complete recovery from disordered eating and amenorrhea, is unknown. However, the capacity for skeletal catch-up is believed to diminish with age and duration of amenorrhea. The few longitudinal studies that have followed athletes with osteopenia and osteoporosis show that weight gain and resumption of menses result in increased BMD without complete normalization (10,11,26,27). Women with the female athlete triad share characteristics of women with anorexia nervosa, including low body weight and fat, nutritional inadequacy, prolonged amenorrhea, and severe bone loss, although these features may be less pronounced in the female athlete triad. Several longitudinal studies report persistent osteopenia in young adults who have recovered from anorexia nervosa (8,14,21); these studies, however, may be confounded by the chronic nature of the disease and limited follow-up. In one study of adolescents with anorexia, patients with good recovery demonstrated significant gains in BMD; 4 out of 23 patients with osteopenia achieved normalization of BMD (1). As the majority of bone accrual in women occurs during adolescence (22), researchers have suggested that the window of opportunity for such catch-up is limited to adolescence (1,9). Long-term longitudinal studies of young women with established osteopenia are needed to determine the extent and timing of skeletal changes following recovery from anorexia and the female athlete triad.

We report changes in BMD in an athlete with a history of restrictive eating, primary amenorrhea, and low BMD. Over a span of 8 yr, from ages 22.9 to 30.8 yr, we took 5 measurements of the patient's BMD on the same DXA machine (QDR Hologic 1000, Hologic, Inc.) by the same technician. The longitudinal error for this machine is 0.4%, and the margin of error for the measurement is 2%. We measured lumbar spine (L2-L4) and left proximal femur BMD and BMC each visit and whole body BMD and BMC at the first four. At each visit, we compared measurements in the patient's spine and hip with changes in BMD, weight, use of oral contraceptives, nutritional habits, and training.


Our patient was an elite Caucasian female distance runner referred to our sports medicine clinic for symptoms of prolonged athletic amenorrhea and the female athlete triad. She provided a detailed history of her eating habits, menstrual history, weight, and training history since childhood. She gave informed consent for her clinical data to be presented in this report.

Injury history.

The patient reported no injuries in high school or college. While training for a marathon at age 27.1 she sustained a femoral neck stress reaction (diagnosed by magnetic resonance imaging (MRI)) that prevented her from running for 8 wk. She also had intermittent femoral nerve compression caused by altered biomechanics from a leg-length discrepancy that was successfully treated with a heel lift.

Training history.

Like many elite runners, she started running competitively at age 12, before menarche. She competed year-round throughout high school, college, and after college, until age 25.1. Her personal best time for the marathon was 2 h and 41 min. At the time of presentation to our clinic, she was running 80–90 miles per week.

Weight, menstrual, and bone density history.

Within a year of beginning to run competitively, at age 13 yr, she began to restrict her caloric and fat intake, leading to a failure to gain weight at the pace expected for her age. She maintained an abnormally low weight for her height from ages 13–25 through a combination of restrictive eating and high-mileage training.

She was primary amenorrheic until age 23. Her previous doctor suggested that she begin oral contraceptives during her freshman year of college, but she declined because she was concerned how it would affect competition. She was eventually evaluated by her primary doctor for low bone density at age 22.9. Her weight at that time was 48.6 kg (BMI 15.8 kg·m−2). Her T scores were −2.50 in the lumbar spine and −1.54 in the hip (Table 1). A value greater than −1 standard deviation (SD) is considered normal, between −1 SD and −2.5 SD indicates osteopenia, and less than −2.5 SD indicates osteoporosis. Interestingly, her bone density measurements at this time were roughly the values that would be expected for an average 13-yr-old, corresponding to the age of onset of disordered eating.

Changes in BMD and body composition over 8 yr of observation.

Because of her low bone density, she agreed to start on a regular calcium supplement of 1500 mg·d−1 and estrogen replacement in the form of oral contraceptives, which induced normal menses. Her medication contained 0.3 mg of norgestrel and 0.03 mg of ethinyl estradiol for 21 d of a 28-d cycle.

One year later, at age 24.1, she had increased her weight by only 0.9 kg (BMI 16.1 kg·m−2). A repeat bone density measurement revealed little change in BMD, with T scores of −2.48 in the lumbar spine and −1.62 in the hip. She continued to run competitively over the next year, and a repeat bone density at age 25.1 was essentially unchanged. At this point, she became concerned about her long-term health and made a decision to give up competing at an elite level, to reduce her weekly mileage, and to gain weight. Over the next 4 months, her weight increased from 50.4 kg (BMI 16.4 kg·m−2) to 55.7 kg (BMI 18.1 kg·m−2). Repeat bone density scores improved slightly, with T scores of −2.42 in the lumbar spine and −1.44 in the hip (Table 1). In the same four months, her total body BMC also increased from 1687 g to 1815 (7.6%).

During the next 5.4 yr, the patient continued to gain weight and to establish better eating habits by increasing her caloric and fat intake and adopting a less restrictive diet. She gained weight rapidly in the year following the cessation of competitive running and gained weight more gradually thereafter (Fig. 1). She continued to run 20–50 miles per week. She discontinued oral contraceptives at age 27.6 and her menstrual cycles have remained normal. By age 30.8, her weight had increased to 65.5 kg (BMI 21.3 kg·m−2), and her bone density measures dramatically improved with T scores of −0.63 in the lumbar spine and −0.33 in the hip (a 25.5% change in the spine and a 19.5% change in the hip from the baseline bone density studies) (Table 1, Fig. 1). Figures 2 and 3 document the dramatic change in mineralization of the patient's bone.

FIGURE 1— Changes in BMD at the lumbar spine (L2-L4) and total proximal femur with changes in body mass index (BMI) from ages 22.9 to 30.8 yr. The arrows indicate the time points at which oral contraceptives were started and stopped and when weight gain was initiated. Dashed lines indicate a possible course of bone change during the 5 yr in which bone density was not measured. Weight measurements from ages 26–29 are based on the patient's recall.
FIGURE 2— Change in total proximal femur BMD g·cm−2 (left) and lumbar spine (L2-L4) BMD g·cm−2 (right) from ages 25.4 to 30.8. Shaded areas represent the 95% confidence interval of bone density values over time.
FIGURE 3— The DXA machines spine image provides a visual representation of mineralization from age 25.4 (:
left ) to age 30.8 ( right ). (Please note that this technique is not considered accurate enough for diagnostic purposes.)


This case shows that large gains in BMD are possible for women who have recovered from the female athlete triad and that this recovery may occur into the third decade of life. Our patient significantly recovered bone density with improved nutrition, resumption of menses, and weight gain, despite a 12-yr history of restricted eating and amenorrhea. Recovery may have taken as long as 5 yr. For this athlete, estrogen replacement therapy and calcium supplementation were ineffective in improving BMD without weight gain. Continued weight-bearing exercise may have contributed to the patient's improved BMD. Studies have shown that there is an increase in relative risk of fracture for each decrease in BMD of one standard deviation (17).With the large gains in BMD demonstrated by our patient (Fig. 4), she is now projected to have only a slightly increased risk of fracture later in life, compared with the general population. In contrast, her initial BMD measurements predicted an eightfold increase in lifetime risk of vertebral fracture and a fourfold increase in lifetime risk of hip fracture (17).

FIGURE 4— Theoretical comparison of spinal bone accretion and loss in a normal woman and in a female athlete whose skeletal development was interrupted at age 13 by undernutrition and delayed menarche. Skeletal development resumes at age 25 with weight gain and restoration of nutritional adequacy.

We believe this to be the first published report to document reversal of BMD values from osteoporotic and osteopenic ranges to normal values in a woman recovered from the female athlete triad. One previous report demonstrated recovery from osteopenia in four adolescents who had overcome anorexia (1). Whether or not the majority of women with osteopenia or osteoporosis due to athletic amenorrhea or anorexia can return to normal bone density values is controversial. Population-based studies of women with a history of anorexia show increased lifetime fracture risk with ongoing compromise of bone density and bone quality (16,24). However, population-based studies cannot control for confounders that may explain the observed increase in fracture risk. Cross-sectional studies examining BMD in women who had formerly had anorexia or elite athletes up to 25 yr after diagnosis or cessation of competition show mixed results, including normal BMD values for age (12,23,28), moderately reduced BMD (7), and unexpectedly high proportions of osteopenia and osteoporosis (5,24,25). Cross-sectional studies, however, are difficult to interpret because they lack baseline and interim BMD measurements. Short-term longitudinal studies suggest that bone deficits in anorexia (8,14,21) and the female athlete triad (10,11,26,27) are at least partially irreversible, unless duration of illness is short and recovery takes place during adolescence (1,9). Our case study shows that the window of opportunity for bone density recovery may be longer than previously expected, and that prognosis may be better than previously thought if full nutritional and weight recovery are achieved. Skeletal recovery may take longer than the typical 1- to 2-yr duration of longitudinal studies.

Normalization of bone density in this patient may represent the completion of interrupted skeletal maturation rather than the reversal of bone loss from an already mature skeleton (Fig. 4). The patient initiated disordered eating before menarche, and her BMD at age 22.9 was similar to the BMD expected for perimenarcheal girls (22). From ages 22.9 to 25.1 yr, she failed to gain bone, but did not lose bone. We speculate that this patient suppressed, but did not bypass, the rapid skeletal mineralization that normally occurs in adolescent women around the time of puberty; when nutritional reserves finally became adequate at age 25.1, normal skeletal development resumed, albeit 12 yr later than expected. This patient's decision to improve nutrition and gain weight occurred only 2 yr after menarche. Other studies have reported more substantial gains in bone in women who developed anorexia nervosa before menarche (9) or in amenorrheic athletes with substantially delayed menarche (27). Thus, our case report is compatible with the hypothesis that the window of opportunity for major skeletal gains is limited to a few years after menarche. The opportunity for catch-up in BMD may be defined by skeletal rather than chronological age.

A recent case report in this journal (30) presented yearly BMD measurements of a female runner with disordered eating, primary amenorrhea until age 24.8 yr, osteopenia of the spine, and osteoporosis of the hip. Similar to our case report, the young woman's skeleton did not respond to oral contraceptives, but did respond to weight gain. In contrast to this report, however, weight gain was initiated at age 33.5 and was not as extensive (final BMI: 17.6 kg·m−2). Despite significant gains in BMD from her lowest values, this woman remained osteoporotic at the hip and osteopenic at the spine, and her prognosis for a full recovery of BMD was uncertain. The window of opportunity for achieving peak bone mass may only extend into the third decade of life; healthy young women gain small amounts of bone during the third but not fourth decade of life (19). Recovery of BMD may also require more substantial weight gain; our patient gained 33.6% in body weight from her initial presentation. Thus, the importance of early and aggressive intervention must be emphasized.

For our patient, oral contraceptives, in the absence of weight gain, proved an ineffective treatment for osteopenia and osteoporosis. Over the past decade, estrogen replacement therapy in the form of oral contraceptives has been the most commonly prescribed therapy for athletic amenorrhea or anorexia nervosa. In a 1995 survey, 92% of sports medicine doctors and family physicians reported prescribing estrogen replacement to treat athletic amenorrhea (6). Similarly, in a 2000 survey, 77.6% of pediatricians, sports medicine doctors, and general practitioners reported prescribing hormone replacement therapy to reduce the risk of osteopenia in teenage girls with anorexia nervosa (20). However, evidence from clinical trials suggests that oral contraceptives are ineffective in improving BMD in patients with anorexia (4,13) and athletes with amenorrhea (3,27) who have osteopenia. Epidemiologic studies and animal models demonstrate that an energy imbalance is the primary cause of athletic amenorrhea (2,15,29); thus, it may not be surprising that treating the nutritional deficiencies present in the female athlete triad would be more effective than treating the hormonal deficiencies. It may even be prudent to recommend against the use of hormone replacement therapy because women may falsely believe that their bones are being protected and may delay weight gain (20). Additionally, the spontaneous return of menses is a good benchmark of improved nutrition (2). In our case study, the patient waited 2 yr after the discovery of osteopenia and osteoporosis to gain weight; she chose to gain weight after it was apparent to her that hormone replacement was ineffective.

This study is limited to a single patient and cannot be used to make treatment recommendations. However, this case shows that substantial gains in BMD after recovery from the female athlete triad are at least possible. Further prospective DXA studies are needed on groups of women with anorexia or women with female athlete triad with osteopenia; these studies should measure BMD both during treatment and for several years after recovery.

This case report indicates that the critical component of treatment for low bone density in the female athlete triad is improved nutrition, weight gain, and resumption of menses. We suggest that these relatively simple interventions can significantly impact bone density, and more effectively than hormone replacement therapy.

Convincing competitive athletes to gain weight is often a formidable challenge. When faced with the diagnosis of osteopenia or osteoporosis, and the possibility of its persistence through lifetime, they may be motivated to change behavior if there is hope for reversal of these morbid conditions.


1.Castro, J., L. Lazaro, F. Pons, I. Halperin, and J. Toro. Adolescent anorexia nervosa: the catch-up effect in bone mineral density after recovery. J Am Acad Child Adolesc Psychiatry. 40:1215–1221, 2001.
2.Cobb, K. L., L. K. Bachrach, G. Greendale, et al. Disordered eating, menstrual irregularity, and bone mineral density in female distance runners. Med Sci Sports Exerc. 35:711–719, 2003.
3.Gibson, J. H., A. Mitchell, J. Reeve, and M. G. Harries. Treatment of reduced bone mineral density in athletic amenorrhea: a pilot study. Osteoporos Int. 10:284–289, 1999.
4.Grinspoon, S., L. Thomas, K. Miller, D. Herzog, and A. Klibanski. Effects of recombinant human IGF-I and oral contraceptive administration on bone density in anorexia nervosa. J Clin Endrocrinol Metab. 87:2883–2891, 2002.
5.Gross, G., J. D. Russell, P. J. V. Beumont, et al. Longitudinal study of patients with anorexia nervosa 6 to 10 years after treatment. Impact of adequate weight restoration on outcome. Ann N Y Acad Sci. 904:614–616, 2000.
6.Haberland, C. A., D. Seddick, R. Marcus, and L. K. Bachrach. A physician survey of therapy for exercise-associated amenorrhea: a brief report. Clin J Sport Med. 5:246–250, 1995.
7.Hartman, D., A. Crisp, B. Rooney, C. Rackow, R. Atkinson, and S. Patel. Bone density of women who have recovered from anorexia nervosa. Int J Eat Disord. 28:107–112, 2000.
8.Hotta, M., T. Shibasaki, K. Sato, and H. Demura. The importance of body weight history in the occurrence and recovery of osteoporosis in patients with anorexia nervosa: evaluation by dual X-ray absorptiometry and bone metabolic markers. Eur J Endocrinol. 139:276–283, 1998.
9.Jagielska, G., T. Wolanczyk, J. Komender, C. Tomaszewicz-Libudzic, J. Przedlacki, and K. Ostrowski. Bone mineral content and bone mineral density in adolescent girls with anorexia nervosa—a longitudinal study. Acta Psychiatr Scand. 104:131–137, 2001.
10.Jonnavithula, S, M. P. Warren, R. P. Fox, and M. I. Lazaro. Bone density is compromised in amenorrheic women despite return of menses: a 2-year study. Obstet Gynecol. 81:669–674, 1993.
11.Keen, A. D., and B. L. Drinkwater. Irreversible bone loss in former amenorrheic athletes. Osteoporos Int. 7:311–315, 1997.
12.Khan, K. M., R. M. Green, A. Saul, et al. Retired elite female ballet dancers and nonathletic controls have similar bone mineral density at weight-bearing sites. J Bone Miner Res. 11:1566–1574, 1996.
13.Klibanski, A., B. M. Biller, D. A. Schoenfeld, D. B. Herzog, and V. C. Sace. The effects of estrogen administration on trabecular bone loss in young women with anorexia nervosa. J Clin Endocrinol Metab. 80:898–904, 1995.
14.Kotler, L., L. Katz, W. Anyan, and F. Comite. Case study of the effects of prolonged and severe anorexia nervosa on bone mineral density. Int J Eat Disorders. 15:395–399, 1994.
15.Loucks. A. B. Introduction to menstrual disturbances in athletes. Med Sci Sports Exerc. 35:1551–1552, 2003.
16.Lucas, A. R., L. J. Melton Jr., C. S. Crowson, and W. M. O'Fallon, Long-term fracture risk among women with anorexia nervosa: a population-based cohort study. Mayo Clin Proc. 74:972–977, 1999.
17.Marshall, D., O. Johnell, and H. Wedel. Meta-analysis of how well measures of bone mineral density predict occurrence of osteoporotic fractures. BMJ. 312:1254–1259, 1996.
18.Otis, C.L., B. B. Drinkwater, M. Johnson, A. Loucks, and J. Wilmore. American College of Sports Medicine position stand: the female athlete triad. Med Sci Sports Exerc. 29:i–ix, 1997.
19.Recker, R. R., K. M. Davies, S. M. Hinders, R. P. Heaney, R. M. Stegman, and D. B. Kimmel. Bone gain in young adult women. JAMA. 268:2403–2408, 1992.
20.Robinson, E., L. K. Bachrach, and D. K. Katzman. Use of hormone replacement therapy to reduce the risk of osteopenia in adolescent girls with anorexia nervosa. J Adoles Health. 26:343–348, 2000.
21.Soyka, L. A., M. Misra, A. Frenchman, et al. Abnormal bone mineral accrual in adolescent girls with anorexia nervosa. J Clin Endocrinol Metab. 87:4177–4185, 2002.
22.Theintz, G., B. Buchs, R. Rizzoli, et al. Longitudinal monitoring of bone mass accumulation in healthy adolescents: evidence for a marked reduction after 16 years of age at the levels of lumbar spine and femoral neck in female subjects. J Clin Endocrinol Metab. 75:1060–1065, 1992.
23.Valla, A., I. L. Groenning, U. Syversen, and A. Hoeiseth. Anorexia nervosa: slow regain of bone mass. Osteoporosis Int. 11:141–145, 2000.
24.Vestergaard, P., C. Emborg, R. K. Stoving, C. Hagen, L. Mosekilde, and K. Brixen. Fractures in patients with anorexia nervosa, bulimia nervosa, and other eating disorders—a nationwide register study. Int J Eat Disord. 32:301–308, 2002.
25.Ward, A., N. Brown, and J. Treasure. Persistent osteopenia after recovery from anorexia nervosa. Int J Eat Disord. 22:71–75, 1997.
26.Warren, M. P., J. Brooks-Gunn, R. P. Fox, C. C. Holderness, E. P. Hyle, and W. G. Hamilton. Osteopenia in exercise-associated amenorrhea using ballet dancers as a model: a longitudinal study. J Clin Endocrinol Metab. 87:3162–3168, 2002.
27.Warren, M. P., J. Brooks-Gunn, R. P. Fox, et al. Persistent osteopenia in ballet dancers with amenorrhea and delayed menarche despite hormone therapy: a longitudinal study. Fertil Steril. 80:398–404, 2003.
28.Wentz, E., D. Mellstrom, C. Gillberg, V. Sundh, I. Gillberg, and M. Rastam. Bone density 11 years after anorexia nervosa onset in a controlled study of 39 cases. Int J Eat Disord. 34:314–318, 2003.
29.Williams, N. I., D. L. Helmreich, D. B. Parfitt, A. Caston-Balderrama, and J. L. Cameron. Evidence for a causal role of low energy availability in the induction of menstrual cycle disturbances during strenuous exercise training. J Clin Endocrinol Metab. 86:5184–5193, 2001.
30.Zanker, C. L., C. B. Cooke, J. G. Truscott, B. Oldroyd, and H. S. Jacobs. Annual changes of bone density over 12 years in an amenorrheic athlete. Med Sci Sports Exerc. 36:137–142, 2004.


©2005The American College of Sports Medicine