Psychology Department

Health Psychology Home Page

Papers written by students providing scientific reviews of topics related to health and well being

  HomeWeight LossAlternative Therapy | Supplements | Eating Disorders | Fitness | About this Page |


Exercise-Induced Amenorrhea and Bone Density Loss

Sarah Clinton

April 30, 2009


In class this semester, we have studied the prevalence and many of the effects of various eating disorders, also touching on the loss of bone density that these often incur. For this paper, I am particularly interested in whether or not exercise-induced amenorrhea results in a significant loss of bone density. We have discussed the fact that in a diagnosed anorectic, either primary or secondary amenorrhea is present and is generally accompanied by a loss of bone density. I will investigate whether or not exercise (which has been shown to increase bone density) will counteract the bone density loss caused by amenorrhea. I will also explore the effects across different age groups and in different athletic pursuits. 


The Female Athletic Triad

The Female Athletic Triad has been thrust into the limelight of the sports medicine scene only fairly recently. In 1992, the American College of Sports Medicine held a two-day convention in which experts discussed the triad of disorders in young female athletes that had just recently begun to be deemed prevalent and serious enough to be described in medical literature (Agostini, R., Drinkwater, B., Nattiv, A., Yeager, K.).

The triad is made up of disordered eating, amenorrhea, and osteoporosis. Disordered eating can take many forms, from anorexia or bulimia to binge eating or an Eating Disorder Not Otherwise Specified, by the criteria of the DSM-IV. Amenorrhea comes in two forms: primary and secondary amenorrhea. Primary amenorrhea is defined as the complete absence of menses in a woman over the age of sixteen, while secondary amenorrhea is defined by the absence of at least three to six menstrual cycles (American Academy of Pediatrics). Osteoporosis, the last disorder in the triad, is characterized by premature bone loss or inadequate bone formation.

Generally, the triad occurs when a young woman, eager to excel in her sport, drops enough weight and/or body fat to incur a loss of menses. This is often, but not always, caused by a pattern of disordered eating, and may also result in the loss of bone density. In addition to the obvious problems this causes, extreme dieting or loss of weight also causes muscle mass to be dropped along with the fat that is lost. This will generally result in the deterioration of performance, causing the opposite result the athlete hopes to achieve in the first place (American Academy of Pediatrics).


Strenuous Exercise as a Cause of Amenorrhea

Strenuous exercise in women has been associated with a high incidence of amenorrhea. The prevalence of amenorrhea is much higher in athletes (with a prevalence rate ranging from 3.4%-66%, depending upon the sport) than in the general female population, in which 2%-5% have reported amenorrhea (Agostini, R., Drinkwater, B., Nattiv, A., Yeager, K.).  The available data suggests that both anorectics and elite female athletes experience alterations in the pulsatile release of GnRH, which in turn causes pertubations in pituitary and ovarian function. This ultimately leads to some type of menstrual dysfunction, often amenorrhea (De Souza, M.J., Metzger, D.).

However, in some cases, weight-bearing exercise has been found to counteract the loss of bone density that occurs in anorectics. An Australian study found that the study participants who reported primary amenorrhea had lower bone mineral density than those that reported secondary amenorrhea. This suggested to researchers that the patients with anorexia nervosa that commenced around puberty were at greater risk for stress fractures, since they would likely have lower bone density. Also, the patients who exercised vigorously were found to have a greater bone mineral density at the proximal femoral sites than the patients that were considered sedentary (Formica, C., Mestrovik, R., Seeman, E., Szmukler, G., Tsalamandris, C.).


Amenorrhea as a Cause of Low Bone Mineral Density

Hypoestrogenism, low levels of estrogen in the body, is associated with amenorrhea and may predispose amenorrheic athletes to the development of premature osteoporosis ( Biller, B., Klibanski, A., Rosenthal, D., Saxe,V., Schoenfeld, D., Saxe V.). Adequate levels of estrogen in the body have been show to slow bone resorption (the process by which cells break down bone and release the encompassed minerals) and improve or maintain bone mass (American Academy of Pediatrics). Thus, it follows that the lack of estrogen that comes with amenorrhea could indeed be a cause of the loss of bone mineral density. If the amenorrhea is caused by extreme amounts of exercise, the strenuous exercise itself could be hailed as the cause of the premature osteoporosis in these cases.

However, high-intensity exercise in certain sports has actually been linked to an increase in bone mineral density in certain skeletal sites, even despite amenorrhea. Elite adolescent ice skaters and gymnasts have been found to have increased bone mineral density in the lower skeleton despite menstrual dysfunction. Girls who begin menarche at a later age and have a lower weight during adolescence have been found to have the lowest bone mineral density when compared with their peers. Also, an increased incidence of stress fractures in dancers has been associated with an older age at the onset menarche (American Academy of Pediatrics).


Decreased Bone Density in Athletes: Does Age Matter?

            Although one of the many benefits of exercise is increased bone mineral density, adult female athletes that have amenorrhea have been shown to have low bone-mineral density (Joffe, A.). To determine whether or not this held true for adolescents, researchers compared the bone mineral density, body composition, and insulin-like growth factor in twenty-one adolescent athletes with amenorrhea, eighteen athletes without amenorrhea, and eighteen non-athlete controls that did not have amenorrhea.

            Athletes with amenorrhea had significantly lower bone mineral density z-scores at the spine and whole body than either eumenorrheic athletes or controls and significantly lower hip bone-mineral density z-scores than eumenorrheic athletes. Lean mass did not differ among the three groups, but athletes with amenorrhea had lower bone mineral density  z-scores than eumenorrheic athletes and lower insulin-like growth factor levels than controls. Markers of bone turnover were significantly lower in amenorrheic athletes than in controls (Joffe, A.). This would seem to show that in adolescence as well as adulthood, amenorrhea counteracts the bone mineral density-increasing attributes of exercise.

            In a study of 45 prepubertal gymnasts and 35 prepubertal controls, bone mineral density in the legs, spine, and arms was higher in the gymnasts than in the controls. After a twelve-month follow-up to the study, it was found that the gymnasts’ bone mass had increased more than the controls.’ Additionally, when the gymnasts’ bone mineral density was checked after they had retired (at the time of this study, the former elite gymnasts were between the ages of eighteen and thirty-five and had been retired for eight years), bone mineral density was still higher at all sites except the skull (Bass, S., Bradney, M., Delmas, P., Harding, A., Hendrich, E., Pearce, G., Seeman, E.). This study shows evidence that strenuous exercise in females before they reach puberty could help increase bone mineral density, and even protect against premature osteoporosis later in life.


Does the Type of Sport Make a Difference?

A higher percentage of athletes competing in sports that emphasize thinness and/or require a certain weight or body composition reported a greater amount of menstrual dysfunction than athletes competing in sports focusing less on such factors (Sundgot-Borgen, J., Torstveit, M.). A study done on female endurance athletes to test whether or not vitamin K or estrogen supplements could counteract the effects of amenorrhea revealed an “unexpectedly high” amount of bone density loss in all of the athletes studied. Vitamin K was not shown to have any effect on decreasing the amount of bone density that was lost, and estrogen supplementation, though slightly beneficial, did not prevent bone density loss (Braam, J., Brouns, F., Geusens, P., Knapen, M., Vermeer, C.).  

In a comparative study of college runners and gymnasts and non-athlete college women, gymnasts and controls were found to have a higher bone mineral density than did runners. Although both groups of athletes were found to have similar values for body fat percentage (which was significantly lower than in the controls), muscle strength was much higher in the gymnasts than in the runners or the controls. In both groups of athletes, the women that were menstruating regularly had slightly higher values when bone mineral density was tested (Gillis, D., Marcus, R., Robinson, T. Shaw, J., Snow, H.,  Taaffe, D.,).


Stress Fractures

            Another indicator that exercise-induced amenorrhea causes a decrease in bone mineral density is the startling number of stress fractures female amenorrheic athletes sustain compared with their non-amenorrheic counterparts. In a study done on female Marine recruits going through boot camp, results showed that among measures of menstrual status and oral contraceptive use, only women who reported no menses or those that reported secondary amenorrhea within the last year were at greatly increased risk for stress fractures (Brodine, S., Macera, C., Rauh, M., Shaffer, R, Trone, D.,).

Menses in the past


Overall Stress

Fracture Incidence %

Pelvic of Femoral

Fracture Incidence %











This table, a relevant excerpt from the full table formulated from the study described above, shows just how much of a risk amenorrhea creates in female athletes/recruits for obtaining a stress fracture.

Discussion of Cited Studies

            Braam study: There were 115 participants in this prospective cohort study. While this seems a big enough number to obtain a trend from the results, there was no control group in the research design, so there was no stable group to compare the results to. Although one group was given a placebo instead of the Vitamin K pill, they were given estrogen supplementation instead. Neither estrogen supplementation nor Vitamin K pills worked to stop the decrease in bone mineral density, and the loss was much higher than expected.

            Tortsveit Study: This study included 1,276 elite athletes and controls. This is a significantly large enough number of participants to gauge a trend. The study was conducted by giving detailed surveys to the respondents and gleaning data from there. To assume that the results of this study have internal and external validity, we must also assume that the respondents answered truthfully and accurately.

            Formica study: This was an Australian study of sixty-five patients that had been diagnosed with anorexia nervosa. This group is large enough to obtain a trend, but perhaps not large enough to be representative of other populations. Patients either took oral contraceptives, exercised vigorously, or did neither (the controls).  These results could be considered significant for my topic, as the results for the three groups displayed significant data for the effects of exercise and amenorrhea, and there was a control group that data could be compared to.

            Brodine study: This study included 2,962 women that were going through Marine boot camp. This number is large enough to obtain trends from. Normally, data would be atypical, since the fitness level, among other things, of women in boot camp differs significantly from the rest of the women in the population. However, these women have a level of activity similar to elite athletes, and thus this study was relevant to my paper. The part of the study that I was interested in for my project was collected by surveying the women about their mental history and then comparing these results to the number of stress fractures the women incurred. The closest thing to a control in this study would be the women that were not amenorrheic, although all women underwent the same amount of training.

            Biller study: Thirty-six women total participated in this study, which is a very small number to be considered statistically significant. The control group was the twelve women who had regular menstrual cycles. The results compared the bone densities of the control group, the women with amenorrhea, and the women with hypothalamic amenorrhea.

            Journal Watch study: There were fifty-seven participants in this study. Again, this was enough to show a trend among the group, but not necessarily to extrapolate to the rest of the population. There was a sufficient control group to compare the results to. Researchers compared the levels of bone mineral density in non-athletes who did not have amenorrhea, athletes with amenorrhea, and athletes that menstruated regularly.


            As a result of reading and analyzing these studies, I have come to several conclusions concerning amenorrhea and its relation to exercise and bone density. First of all, exercise-induced amenorrhea does cause a loss of bone mineral density in female athletes. However, this effect is worse in those with primary amenorrhea than in those with secondary amenorrhea. Also, strenuous exercise before the onset of puberty has been shown to be very beneficial for increasing bone mineral density and for protecting against the effects of premature osteoporosis due to amenorrhea after puberty has set in. Decreases in bone mineral density have been shown to be more extreme in endurance athletes, like runners, and in athletes that have to meet certain weight/body composition requirements, such as ballet dancers, than in athletes whose sports require an abundance of muscle, like gymnasts.






Agostini, R., Drinkwater, B., Nattiv, A., Yeager, K. “The female athlete triad: disordered eating, amenorrhea, osteoporosis.” American College of Sports Medicine. (April 1993).


Bass, S., Bradney, M., Delmas, P., Harding, A., Hendrich, E., Pearce, G., Seeman, E. “Exercise Before Puberty May Confer Residual Benefits in Bone Density in Adulthood: Studies in Active Prepubertal and Retired Female Gymnasts.” Journal of Bone and Mineral Research. (March 1998)


Biller, B., Klibanski, A., Rosenthal, D., Saxe,V., Schoenfeld, D., Saxe V. “Effects of prolactin and estrogen deficiency in amenorrheic bone loss.” National Center for Biotechnology Information.  (July 1988)


Braam, J., Brouns, F., Geusens, P., Knapen, M., Vermeer, C. “Factors Affecting Bone Loss in Female Endurance Athletes.” The American Journal of Sports Medicine. (2009)


Brodine, S., Macera, C., Rauh, M., Shaffer, R, Trone, D. “Predictors of Stress Fracture Susceptibility in Young Female Recruits.”  The American Journal of Sports Medicine. (2006)


DE SOUZA, M.,  METZGER, D.Reproductive dysfunction in amenorrheic athletes and anorexic patients: a review.” Medicine and Science in Sports and Exercise. (September 1991)


“Exercise plus Amenorrhea Equals Thinner Bones.” Journal Watch. Pediatrics (June 2008).


Formica, C., Mestrovik, R., Seeman, E., Szmukler, G., Tsalamandris, C. “Osteoporosis in anorexia nervosa: the influence of peak bone density, bone loss, oral contraceptive use, and exercise.” National Center for Biotechnology Information. (December 1992).


Official Journal of the American Academy of Pediatrics. “Medical Concerns in the Female Athlete.” Pediatrics, Vol. 106, No. 3 (September 2000).


Sundgot-Borgen, J., Torstveit, M. “Participation in leanness sports but not training volume is associated with menstrual dysfunction: a national survey of 1276 elite athletes and controls.” British Journal of Sports Medicine. (2005).





Psychology Department

The Health Psychology Home Page is produced and maintained by David Schlundt, PhD.


VuLogoVanderbilt Homepage

Return to the Health Psychology Home Page
Send E-mail comments or questions to Dr. Schlundt