Obstacles in the Optimization of Bone Health Outcomes in the Female Athlete Triad
Maintaining low body weight for the sake of performance and aesthetic purposes is a common feature among young girls and women who exercise on a regular basis, including elite, college and high-school athletes, members of fitness centres, and recreational exercisers. High energy expenditure without adequate compensation in energy intake leads to an energy deficiency, which may ultimately affect reproductive function and bone health. The combination of low energy availability, menstrual disturbances and low bone mineral density is referred to as the ‘female athlete triad’. Not all athletes seek medical assistance in response to the absence of menstruation for 3 or more months as some believe that long-term amenorrhoea is not harmful. Indeed, many women may not seek medical attention until they sustain a stress fracture.
This review investigates current issues, controversies and strategies in the clinical management of bone health concerns related to the female athlete triad. Current recommendations focus on either increasing energy intake or decreasing energy expenditure, as this approach remains the most efficient strategy to prevent further bone health complications. However, convincing the athlete to increase energy availability can be extremely challenging.
Oral contraceptive therapy seems to be a common strategy chosen by many physicians to address bone health issues in young women with amenorrhoea, although there is little evidence that this strategy improves bone mineral density in this population. Assessment of bone health itself is difficult due to the limitations of dual-energy X-ray absorptiometry (DXA) to estimate bone strength. Understanding how bone strength is affected by low energy availability, weight gain and resumption of menses requires further investigations using 3-dimensional bone imaging techniques in order to improve the clinical management of the female athlete triad.
For a growing number of female athletes, the desire for athletic success can be associated with a high drive for thinness that may lead to the development of low energy availability, menstrual dysfunction, and low areal bone mineral density (aBMD), a condition collectively referred to as the ‘female athlete triad’. Alone or in combination, the disorders of the female athlete triad can have a negative impact on health and athletic performance. This condition not only affects high-level female athletes but more generally exercising girls and women.
Low energy availability can be caused by high energy expenditure associated with physical exercise and training, inadequate energy intake, or a combination of both. Energy availability is defined as dietary energy intake minus exercise energy expenditure. Energy availability is considered adequate when energy intake is sufficient to maintain normal routine physiological functions in addition to exercise training. When energy availability is too low, the body tends to reduce the amount of energy that is used for physiological functions such as cellular maintenance, thermoregulation, growth and reproduction. Energy conservation, which tends at restoring energy balance, could explain why stable body weight has been reported in amenorrhoeic athletes.[4, 5, 6, 7]
Low energy availability triggers the disruption of the hypothalamo-pituitary-gonadal axis, which leads to menstrual disturbances including amenorrhoea.[1,2,8, 9, 10, 11, 12, 13, 14] Amenorrhoea is defined as having no menses for a minimum period of 3 months. Functional hypothalamic amenorrhoea (FHA) is diagnosed by exclusion of situations (e.g. pregnancy) or medical conditions (e.g. hyperprolactinoma, thyroid diseases) that typically cause the absence of menses.[15,16] Athletic amenorrhoea is a form of FHA observed in athletes who display low energy availability. Amenorrhoea represents the most severe menstrual disturbance along a continuum of abnormalities ranging from luteal phase defects, anovulatory cycles, oligomenorrhoea (irregular and inconsistent menstrual cycles lasting from 36 to 90 days) and amenorrhoea. Irregular menses in athletes (oligo- or amenorrhoea) have been associated with a 2- to 4-fold greater incidence of stress fractures and low aBMD, particularly at the spine.[19, 20, 21, 22, 23, 24, 25, 26]
The prevalence of oligomenorrhea and amenorrhoea in adult athletes across multiple sports has been reported to range from 12% to 79%.[27, 28, 29] In adolescents (<18 years), a high prevalence of oligomenorrhea and amenorrhoea (45–50%) has been reported in sports that emphasize a lean physique, such as ballet dancing and running.[27,30,31] However, regardless of age and type of sports participation, approximately 1 in 5 to 1 in 4 active women present with some form of menstrual disturbance.[32, 33, 34, 35] Notably, the prevalence of oligomenorrhoea and amenorrhoea is more difficult to determine in adolescents since menstrual cycle intervals of >35 days are encountered in 65% of girls during the first 12 months following menarche. Indeed, ovulatory status and menstrual cycle lengths are highly variable for about 5 years in post-menarcheal girls. Menstrual cycles and ovulation are much less variable and cycles are more consistent in length for approximately 20 years in young reproductively mature adults. As women approach menopause, cycle length and ovulatory status become highly variable again for about 10 years.[38,39]
Persistent, irregular menstrual cycles are a warning sign that warrant further medical attention, but may not be perceived as such by athletes. Athletes who display low energy availability, even when accompanied by irregular or absent menses, may not seek medical support until a more obvious symptom, such as a stress fracture, is sustained. In addition, clinicians may not feel confident in treating athletes presenting with amenorrhoea[41, 42, 43] and the use of pharmacological therapies still remains very controversial in this population.[1,41,44] The objective of this review is to investigate current issues in the management of the bone health concerns associated with the female athlete triad.
2. Current Clinical Knowledge, Attitudes and Management of the Female Athlete Triad
Clinicians’ attitudes towards the female athlete triad also vary according to medical specialty.[43,50] For example, although orthopaedic surgeons reportedly suspected eating disorders in 59% of their patient-athletes, discussion of these issues occurred only with 31% of the involved patients. In contrast, family physicians suspected eating disorders in 84% of their patient-athletes and reportedly discussed the problem with 80% of the involved patients.
Physical therapists play an important role in identifying athletes at risk and in managing the female athlete triad due to their expertise in musculoskeletal health and exercise prescription, but their knowledge of the female athlete triad may be lacking. A recent survey (2009) conducted in 205 physical therapists in the US showed that 50% of the survey respondents had treated female athletes for conditions related to the female athlete triad (e.g. stress fractures), yet <25% could accurately list all three components of the syndrome, or had been involved in screening athletes for female athlete triad symptoms. The study concluded that physical therapy curriculums in the US need to better educate physical therapists about the detection, treatment and prevention of the female athlete triad in the future. Updated clinical guidelines have been published by the ACSM and the Medical Commission of the International Olympic Committee (IOC); however, it is not clear whether these guidelines have been well disseminated in the medical and healthcare community. Furthermore, an international coalition with representatives from major organizations dedicated to active women and athletes, the Female Athlete Triad Coalition (http://www.femaleathletetriad.org), has been formed to foster education, research and advocacy with regard to the female athlete triad.
3. Management of Bone Health Issues Related to the Female Athlete Triad
3.1 Assessment of Bone Health
3.1.1 Dual-Energy X-Ray Absorptiometry
First introduced in the late 1980s, dual-energy X-ray absorptiometry (DXA) is the most available and widely used densitometry technique. The ACSM recommends DXA scans for premenopausal women in any of the following situations: (i) oligomenorrhoea or amenorrhoea, present for ≥6 months; (ii) disordered eating or an eating disorder present for ≥6 months; and (iii) the presence of stress fracture or other fracture from minimal trauma. According to the recommendations from the International Society for Clinical Densitometry (ISCD), the appropriate skeletal sites to scan are the lumbar spine and hip in adults,[54,55] with a re-evaluation of aBMD after 12 months, if the aforementioned symptoms persist.
The lumbar spine and forearm should be monitored carefully because they are rich in trabecular bone and submitted to little or no weight-bearing. Not only do amenorrhoeic athletes present with lower aBMD than their eumenorrheic counterparts (which might indicate they did not maximize their exercise-induced skeletal benefits), they can also present with lower aBMD than sedentary women.[19,20,46,63,64] The prevalence of low aBMD (≤−1 standard deviation [SD] compared with the norm in young adults) has been reported to range between 1.4–50% in athletic populations who are considered to be at risk for developing the female athlete triad. Prospective cohort studies in postmenopausal women indicate that the risk of fracture increases by a factor of 1.4–2.6 for each decrease in aBMD by 1 SD. This result suggests that amenorrhoeic athletes who have low aBMD during adolescence or young adulthood are likely to be at higher risk of osteoporotic fracture later in life.
The fact that aBMD values are lower in smaller bones implies that DXA tends to underestimate volumetric BMD in petite individuals. In addition, DXA-derived aBMD is based on the assumption that mass and composition (percentage fat content) of the covering soft tissue are homogeneous in the body. Not only do the thickness of soft tissue and its fat content vary throughout the body, but they are also affected by changes in body weight. Therefore, aBMD results can be affected by body weight loss or body weight gain. For example, an average weight loss of 11.3 ± 6.9 kg over a year in 34 obese subjects was associated with an apparent, but false, 1–2% reduction in spinal aBMD. Little is known about how smaller changes in body weight and/or body composition affect the DXA outcomes; therefore, the error of measurement cannot be predicted. Health professionals must be aware of this issue when interpreting DXA scans in female athletes who may be gaining or losing weight.
3.1.2 Three-Dimensional Imaging Techniques
Bone geometry, volumetric BMD, and most importantly bone strength, cannot be assessed accurately by DXA because this technique fails to provide a 3-dimensional image. Using 3-dimensional imaging technologies is important for populations undergoing bone growth because minor increases in bone size can significantly affect bone strength, despite minor changes in aBMD or bone mass. MRI has been used previously in athletes to investigate the effects of repetitive loading,[74, 75, 76, 77, 78, 79] but it does not measure bone mineral mass and therefore needs to be coupled with DXA.[74,75,77,78]
Importantly, both the DXA and quantitative computed tomography techniques (QCT and pQCT) measure the inorganic component of the bone matrix, i.e. the hydroxyapatite crystals made of calcium and phosphate that give the skeleton its stiffness. The organic component of the bone matrix, which is composed of ∼90% of type I collagen and gives the skeleton its flexibility, also affects bone strength. However, current imaging techniques that are applicable in vivo non-invasively do not provide information on the organic component of the bone matrix.
3.1.3 Bone Turnover Markers
3.1.4 Specific Considerations in the Growing Athlete
Assessing bone health is even more challenging in adolescents because of the constant change in bone mass, size and shape. DXA-derived hip assessment in children and adolescents is not reliable due to significant variability in skeletal development and the lack of reproducible regions of interest. Thus, scans at the spine are preferred. Evaluation of aBMD at the whole body (less the head) is recommended by the ISCD because it has been shown to be associated with fracture risk in children. However, whole body less head aBMD is likely to be normal in young athletes because higher aBMD at loaded sites may mask possible lower aBMD at unloaded sites. In contrast, the distal forearm, for which reference data exist in children,[112, 113, 114] might be a useful site for testing. It is the most common site of fracture in adolescents and it is not loaded in activities such as running or jumping, which account for a significant proportion of children’s physical activity.[115, 116, 117]
In children and adolescents aged 5–19 years, ‘low bone mass’ has been defined by the ISCD as a Z-score ≤−2.0 SD for BMC or aBMD adjusted for age, gender, body size. It can also be helpful to determine if a growth spurt occurred without weight gain, which constitutes a relative weight loss and the assessment of bone age can give an indication of the maturational delay and remaining growth. Normal bone growth can be compromised by a range of diseases but also, to an unknown extent, energy deficiency and hypoestrogenism. Depending upon the age at which bone growth becomes compromised, deficits may occur in limb dimensions (pre-puberty), spine dimensions (early puberty) or volumetric BMD by interfering with mineral accrual (late puberty).
Before DXA scans in growing children or adolescents are interpreted, it is essential to adjust the outcomes to account for differences in body size. Experiments conducted in 150 healthy individuals aged 6–21 years showed that normalizing whole-body DXA bone area for height and BMC for height provided the best measures of bone dimensions and strength as determined by pQCT. DXA BMC for age and aBMD for age were only moderately correlated with pQCT-derived bone strength. Therefore, comparing a child’s whole-body BMC to height-matched reference data provided by the DXA manufacturer is a better approach than looking at the absolute values of aBMD. Adjusting for height implies that a child who is scanned around the growth spurt may lie below the 50th percentile as peak height velocity is achieved 6–12 months earlier than peak in bone mass accrual (i.e. bones grow in length first and increases in bone diameter and bone mineralization lag behind). The different rates of linear growth of bone and bone mineralization cause a relative skeletal fragility around the growth spurt, a time that coincides with the peak incidence of fractures during adolescence.
Concentrations of bone turnover markers in adolescents vary depending on sex, Tanner stage (pubertal stage), height velocity, as well as skeletal mass and rate of bone mineral accrual, which makes the interpretation of the results difficult. Preliminary findings in young athletes with amenorrhoea have been reported in cross-sectional studies[20,104] (table I) and therefore must be viewed with caution.
In summary, despite its limitations, DXA remains the standard method for assessing bone health in amenorrhoeic athletes. Keeping in mind that genetic factors account for 60–80% of the individual variances in aBMD, clinicians can expect aBMD in amenorrhoeic athletes to be higher than the norm or within normal range at loaded skeletal sites, and lower than the norm at non-loaded or moderately loaded sites containing a high proportion of trabecular bone (spine, distal forearm). Growing athletes should be carefully monitored as low energy availability can impact their skeletal development and compromise the attainment of peak bone mass. Increase in skeletal mass slows down at the lumbar spine and femoral neck at 15–16 years in female adolescents. This cut-off point may not apply to young girls whose skeletal maturation has been delayed. Several case studies have reported increases in aBMD after 20 years or even 30 years of age.[21,24,40] However, full recovery in bone strength might not be achieved because bone mineralization in young adults (after completion of longitudinal growth) usually results in increased BMD, not increased bone size. Therefore, the long-term consequences for bone health may be irreversible. Periods of amenorrhoea or oligomenorrhoea during adolescence have been associated with a lower aBMD in adult women and a greater incidence of osteoporosis in post-menopausal women. History of menstrual dysfunction has also been associated with a greater risk of hip fractures and wrist fractures. These findings have not been confirmed in populations of retired athletes and require further investigations, particularly in athletes who typically present with low spinal aBMD, such as runners and ballet dancers.
3.2 Pharmacological Treatments and Current Issues
A recent literature review investigated the different pharmacological strategies that have been used to treat impaired bone health in women with FHA. The most common intervention consists of treating the hypoestrogenism, either with the oral contraceptive pill (OCP) or other forms of estrogen therapies.
3.2.1 Estrogen Therapy
In 1989, the American Academy of Pediatrics recommended that estrogen supplementation in amenorrhoeic adolescents should only be considered if the individual is 3-years post-menarche and older than 16 years of age. This position has been endorsed by other authors and organizations.[1,118,133] Some state that supplementation could be permitted at a younger age if the athlete has previously sustained a stress fracture.
However, the use of OCP and other forms of estrogen therapy in adolescent females and adults with anorexia nervosa and FHA remains controversial.[49,135, 136, 137] Longitudinal cohort studies have reported either an increase in aBMD[138, 139, 140] or a reduction of bone loss in athletes taking OCP (0.020–0.035 mg of ethinyl estradiol + a progestogen) when compared with athletes not taking OCP. In the only large randomized controlled trial ever conducted, the effects of 2 years of OCP treatment (0.030 mg ethinyl estradiol and 0.3 mg norgestrel) on aBMD in both oligo/amenorrhoeic runners and eumenorrheic runners were inconclusive. The results of this study were confounded by the fact that the women who dropped out from the OCP group were more likely to be amenorrhoeic and to practice disordered eating. Anecdotal evidence from case studies,[21,40,142] and small prospective or retrospective studies (n < 10),[23,143] have also provided contradictory findings on the effects of OCP in amenorrhoeic athletes. Two randomized trials failed to detect an effect on aBMD in 24 amenorrhoeic ballet dancers and 34 oligo/amenorrhoeic runners, when given low doses of estrogens. Many studies that incorporate OCP as a form of treatment are limited by high withdrawal rates, noncompliance to treatment, weight gain during treatment[21,40,45] and spontaneous resumption of menses in controls,[45,46] making it difficult to draw definitive conclusions regarding the effects of estrogen therapies on aBMD in amenorrhoeic athletes.
Similarly, the efficacy of estrogen treatment in preventing stress fractures in athletes remains unknown. A stress fracture is a partial or complete bone fracture that is caused by repetitive loading and consequent microtraumas to the bone. Although the magnitude of stress applied to the bone is lower than the stress required to fracture the bone in a single loading, repeated microtraumas can eventually result in bone fracture if microtraumas accumulate faster than they heal. Three prospective cohort studies, one in athletes, and two in military recruits,[146,147] failed to show any protective effect of OCP on the incidence of stress fractures in active women, while a cross-sectional study and a case-control study reported a lower use of OCP in athletes who had sustained a stress fracture. In the only randomized controlled trial conducted, randomization to OCP tended to be associated with a lower incidence of stress fractures — 18 stress fractures occurred over 2 years, 6 stress fractures were sustained by runners randomized to OCP and 12 stress fractures were sustained by control runners.
The evidence supporting the use of estrogen treatment for stress fracture prevention in athletes remains inconclusive. Current studies are limited in their findings due to the use of self-reported non-documented stress fractures, potential confounders such as body weight or training volume, small sample size, a relatively low number of stress fractures[45,145,148,149] and poor compliance to placebo or treatment with OCP. Importantly, oligo/amenorrhoea was not an inclusion criterion in most of these studies.[45,145, 146, 147,149]
Different factors could explain the lack of efficacy of estrogen therapy on aBMD and stress fractures risk. Estrogen therapy is used in postmenopausal women to prevent the hypoestrogenism-induced increase in bone resorption.[150, 151, 152] Estrogen replacement has also been shown to have positive effects on aBMD in young women with primary ovarian insufficiency. However, bone resorption is not necessarily elevated in amenorrhoeic athletes (table I), in which case estrogen therapy is unlikely to have any further anti-resorptive effects.
Amenorrhoea in athletes is associated with a range of disturbances in hormones and nutrients including a decrease in total tri-iodothyronine, leptin, insulin, IGF-1/IGF-binding protein-1, glucose, luteinizing hormone pulsatility, follicle-stimulating hormone, estradiol and progesterone, as well as an increase in growth hormone and cortisol.[59,61,102,103,154,155] Estrogen therapy is unlikely to normalize the metabolic factors that impair bone formation, which might explain its lack of efficacy in improving aBMD or reducing bone loss. Specific concerns have also been raised regarding exogenous estrogen administration in athletes with amenorrhoea. In women with FHA, OCP use might have a detrimental effect on androgen secretion, and this could ultimately be detrimental for aBMD. In growing athletes, exogenous estrogen may induce premature closure of the epiphyses and compromise the attainment of full length of long bones.[92,157]
3.2.2 Bisphosphonates and Other Anti-Osteoporotic Therapies
Bisphosphonates, which markedly reduce bone turnover, have emerged as one of the leading effective treatments for postmenopausal and other forms of osteoporosis. Bisphosphonates adhere to the bone surface, impair osteoclast function and induce apoptosis by inhibiting a key enzyme in the mevalonate pathway. Using bisphosphonates for preventing or treating stress fractures in female athletes has also been studied. Findings indicated that risedronate did not reduce the incidence of stress fractures in military recruits, but the study suffered from a 60–70% dropout rate. In another trial using bisphosphonates as a treatment for stress fracture, a weekly dose of intravenous pamidronate over a 5-week period permitted four of five athletes with tibial stress fractures to return to their previous training regimen within 1 week of initiating the intervention, but fracture recurrence was not reported. Most importantly, long-term effects of bisphosphonates are unknown and side effects (stomach pain and bloating) have been reported in athletes. The Medical Commission of the IOC does not approve of the use of bisphosphonates in premenopausal women due to the long half-life of bisphosphonates in bones (up to 10 years) and their potential teratogenic effect on the fetus during future pregnancies.[13,134]
3.2.3 Other Pharmacological Therapies
Other therapies, including recombinant human IGF-1 (rhIGF-I)[162, 163, 164] and androgens[165,166] have been tested in amenorrhoeic women with anorexia nervosa. Due to the less extreme pathophysiology of the female athlete triad, these treatments have not been tested in female athletes with amenorrhoea. Recombinant human leptin (rhLeptin) may improve markers of bone formation in women with FHA.
Although calcium and vitamin D supplementation are frequently prescribed for amenorrhoeic athletes, these nutrient supplements have never been prospectively assessed as an intervention using aBMD as an outcome variable. Currently there is no consensus as to the appropriate dosage and form of calcium and vitamin D supplementation for this population. The Medical Commission of the IOC recommends a calcium intake of at least 1500 mg/day, whereas the ACSM guidelines indicate 1000–1300 mg/day.
Low levels of vitamin D are a concern worldwide.[169,170] The criterion to define vitamin D deficiency varies, although serum 25-hydroxyvitamin D concentrations below 50 mmol/L (20 ng/mL) are commonly used. This threshold is largely debated and some experts have suggested that serum 25-hydroxyvitamin D levels above 75 nmol/L (30 ng/mL) maximize the health benefits of vitamin D.[169,171] Athletes engaged in indoor sports are at greater risk of vitamin D deficiency, which may affect their muscle function, bone strength and performance.[173, 174, 175] An appropriate dose of vitamin D is 400 IU to 800 IU per day, which combined with an adequate calcium dosage, may aid in the reduction of stress fractures.
In summary, none of the pharmacological strategies have demonstrated efficacy in correcting bone health abnormalities, including low aBMD in oligo- and amenorrhoeic athletes. Although low-dose oral contraceptive (<35 μg of estrogen per day) has been suggested to be an appropriate treatment in amenorrhoeic athletes, evidence to support such a conclusion is weak. The criteria for initiating estrogen therapy, the optimal mode of therapy (estrogen alone or combined with a progestogen) and dosing schedule have not been defined. Further studies are needed to clarify the potential risks and benefits of estrogen therapy in amenorrhoeic athletes. More research is also needed to define optimal vitamin D levels and to quantify the best combinations of calcium and vitamin D dosages necessary to achieve a protective effect on bone health.
3.3 Non-Pharmacological Approaches
3.3.1 First-Line Strategy: Increasing Energy Availability
The recommended first-line treatment strategy for athletes with oligo- and amenorrhoea consists in increasing energy availability by increasing dietary intake, decreasing energy expenditure, or both. The strongest evidence that supports bone health improvements resulting from implementing this strategy comes from longitudinal data obtained in girls or women with anorexia nervosa. Both components of recovery (body weight gain, resumption of menses) are thought to have independent and additive effects on bone health.[136,178, 179, 180, 181, 182, 183] For instance, a 38% increase in body weight over 3 months was associated with significant increases in aBMD in anorexic women, despite persistence of amenorrhoea (+2.6% and +1.1% at the hip and spine, respectively). Increases in spinal aBMD were found to be conditional on resumption of menses, whereas hip aBMD was found to be more responsive to weight gain.[181,184]
Similar studies looking at partial or full recovery (body weight gain and/or resumption of menses) in female athletes are scarce. A 2-year follow-up in five amenorrhoeic dancers showed that in the first year the increase in spinal aBMD was correlated to the increase in caloric intake and in the second year, spinal aBMD increase was correlated to weight gain. The two dancers who had the largest increases in spinal aBMD during the first year (+18.8 and +20%) also had weight gain (+4.1 and +3.6 kg) and three menses during the year. A study in runners showed that when an injury or illness forced athletes to decrease their mileage and participate in activities other than running, the athletes’ menses resumed and weight gain (+1.9 kg) occurred. When the athletes were reassessed at 15.5 months, spinal aBMD had increased by 6.3% compared with baseline (p < 0.01). Even though aBMD may increase with weight gain and/or resumption of menses after 20[21,24] and 30 years of age, athletes may fail to normalize aBMD.[25,62,142,185] In a study that included both dancers and nondancers (n = 50; 19 with amenorrhoea), the seven participants who resumed menses over the 2-year follow-up showed a 17% increase in spinal aBMD (vs 4% in those who remained amenorrhoeic), but their aBMD remained below aBMD in the eumenorrhoeic participants. Further investigations using 3-dimensional imaging techniques are needed as DXA-derived outcomes are influenced by changes in body weight and body composition.
Two studies provide preliminary data on the effectiveness of an intervention to increase energy availability in small groups of athletes (n = 4–7).[187,188] A decrease in training load by 43% in four amenorrhoeic female runners resulted in a 5% increase in body weight, increased estradiol levels to within normal range, resumption of menses, and a 6% increase in spinal aBMD over 15 months. The three other participants who did not decrease their training load remained amenorrhoeic. Although adding a day of rest in the training schedule sounds like a reasonable treatment approach in active amenorrhoeic athletes, it was found that the benefits may be offset by an increase in training volume on alternate days. This study, where the intervention combined a reduction in energy expenditure and an increase in energy intake, did not include bone status as a means to validate treatment outcomes in amenorrhoeic athletes. Experiments in female monkeys showed that the number of days required for amenorrhoeic exercising monkeys to resume menses was directly proportional to the number of extra calories consumed per day. For example, animals that ate more daily calories recovered menses more quickly.
Further trials on larger samples are needed to determine the relationship between the increase in energy availability and the timing in which menses and bone restoration may return. It is important to mention that poor diet quality (deficiencies in minerals and vitamins, e.g. iron and vitamin D, respectively) often results from low energy availability. This aspect should not be overlooked when strategies to increase energy availability are employed.
3.3.2 Other Non-Pharmacological Strategies to Improve Bone Health
Instructing an athlete to participate in resistance training by changing a work-out routine from cardiovascular to weight-training is one of the recommended strategies reported by American Physical Therapy Association to treat the female athlete triad in an effort to reduce energy expenditure and increase aBMD.[51,52] Progressive high-intensity resistance training in premenopausal women was shown to increase absolute aBMD at the lumbar spine[190,191] and at the total forearm (although exercises with impacts might be more beneficial at that skeletal site). Part of these skeletal adaptations are due to the fact that a new form of exercise such as resistance training introduces variety in the loading pattern of the skeleton, which is beneficial for bone strength. The effects of resistance training alone or as part of a multidisciplinary exercise programme have been investigated in anorexic girls and women.[194, 195, 196, 197] Overall, resistance training improved body composition, muscle strength and psychological well-being, but the effects on aBMD were not investigated. Resistance training has been suggested as an alternative to weight-bearing exercises for stress fracture prevention in athletes participating in high-impact sports. Resistance exercise programmes to improve bone health typically include three sets of 8–12 repetitions of 6–10 exercises at ∼80% one-repetition maximum (1RM: maximum weight that can be lifted once), with 1–2 minutes of rest between exercises and a progressive increase in intensity. The recommended frequency of performing resistance exercises is three times/week, as resistance training programmes performed only 1–2 times/week[198,199] have been shown to be less effective in premenopausal women.
Plyometric exercises, characterized by high-intensity, explosive muscular contractions, are another form of exercise that can be substituted to usual training routines to improve bone health. Interventions including various jumping exercises have been shown to improve aBMD at the hip and spine in girls.[200,201] The osteogenic effects varied with stages of puberty, and may sometimes be masked by the overwhelming effect of growth on aBMD. In adults, studies have reported positive effects of plyometrics on aBMD at the hip after only 6 months of training, or hip and spine after 18 months of training. To target non-weight-bearing skeletal sites that do not benefit from jumping exercises, upper body plyometric exercises can be performed (e.g. throwing). The main limitation of plyometrics is that these exercises are not recommended when returning from a stress fracture injury. Plyometrics can be combined with resistance training to maximize skeletal benefits.
3.3.3 Specific Considerations for Stress Fractures
Experience has shown that convincing athletes to increase their energy intake and/or reduce their training volume can be very challenging. A common training error is failing to incorporate training periodization, the allowance of active rest cycles during a weekly training regimen. This training error can lead to stress fractures.
The stress fracture rate in athletes/military recruits with menstrual dysfunction is 2- to 4-fold higher than in athletes/military recruits with normal menstrual cycles. Due to the role of estrogen receptor alpha in mechanotransduction, hypoestrogenism may impair the capacity of the osteocytes to respond to repetitive loading, inducing an accumulation of micro-damage, stress reaction and eventually stress fracture.[207,208] The adverse effects on bone may not always translate to findings of low aBMD on a DXA scan because long-term repetitive loading may compensate for the negative effect of hypoestrogenism. Findings are contradictory as to whether a relationship between stress fractures and low aBMD exists.[209, 210, 211, 212, 213, 214, 215] Bone geometry and estimates of bone strength, rather than aBMD, might be more relevant parameters when assessing stress fracture risk.[213,216, 217, 218, 219, 220] For example, individuals with ‘slender bones’ might be more susceptible to stress fractures when long bones are subjected to extreme loading, such as in military training. The incidence of stress fractures may also be affected by poor muscle strength and training-induced muscle fatigue. Diminished muscular support surrounding the joints may cause increased tensile forces and shear stresses on the bone and joint surfaces, which can potentially, over time, lead to injury. Finally, alterations in trabecular bone microarchitecture, possibly caused by hypoestrogenism, have been reported in fractured female athletes and might contribute to the aetiology of stress fractures.
Amenorrhoeic athletes who have sustained a stress fracture need to recognize the time allowance that is required for proper tissue healing and the avoidance of future bone fractures. A restriction in the amount and intensity of exercise necessary to heal bone may be a potential source of frustration and depression for the athlete.[51,52,224] To avoid focusing on activity restriction, athletes can be directed to perform non-weight-bearing activities, and resistive, balance or coordination exercises using variations in the applied mechanical loads.[224,225]
Cross-training, i.e. replacing a training session in the athlete’s specialty (e.g. running) with a dissimilar mode of exercise (e.g. cycling), can provide physiological and psychological benefits to the athlete while reducing or modifying the biomechanical demand placed on the musculoskeletal system, and perhaps, the energetic demand, depending on the exercise volume of the session.[51,52] A reduction in energy expenditure without reducing the number of training sessions can also be achieved by having the athlete work on technical and tactical skills, video analysis of sport skills and performance, sport visualization and cognitive techniques.
4. Conclusions and Perspectives
The female athlete triad has drawn much attention from the research community in the past 30 years. Low energy availability and menstrual dysfunction place female athletes at greater risk for sustaining stress fractures, as well as osteoporotic fractures later in life. Evidence regarding the efficacy of treatment to prevent fractures is lacking because conducting trials with bone fracture as a major outcome typically requires large sample sizes and long-term follow-up studies. Previous studies in anorexic girls/women and female athletes indicate that an increase in energy availability through refeeding and/or decreasing energy expenditure is sometimes accompanied by positive outcomes for aBMD, i.e. the maintenance or increase in aBMD. This strategy should prevail over pharmacological therapies whose efficacy has not been proven. Evaluation of bone health in exercising women has relied on the measurement of aBMD by DXA, which has inherent limitations particularly in growing subjects. Further investigations using 3-dimensional imaging techniques are needed to clarify the effects of low energy availability on bone strength, and the capacity to recover bone strength with weight gain and/or resumption of menses. Advancing research and knowledge in bone health related to the female athlete triad is essential to ensure that all women can receive the physical and psychological benefits that are associated with exercise participation.
No funding was used to assist in the preparation of the manuscript. The authors have no conflict of interest.
- 14.Zanker C, Hind K. The effect of energy balance on endocrine function and bone health in youth. In: Daly RM, Petit MA, editors. Optimizing bone mass and strength:the role of physical activity and nutrition during growth. Basel: Karger, 2007: 80–101Google Scholar
- 15.The Practice Committee of the American Society for Reproductive Medicine. Current evaluation of amenorrhea. Fertil Steril 2004 Sep; 82 Suppl. 1: S33–9Google Scholar
- 24.Hind K. Recovery of bone mineral density and fertility in a former amenorrheic athlete. J Sports Sci Med 2008; 7: 415–8Google Scholar
- 35.Mudd LM, Fornetti W, Pivarnik JM. Bone mineral density in collegiate female athletes: comparisons among sports. J Athlet Training 2007; 42 (3): 403–8Google Scholar
- 39.Treloar AE, Boynton RE, Behn BG, et al. Variation of the human menstrual cycle through reproductive life. Int JFertil 1967 Jan-Mar; 12 (1Pt2): 77–126Google Scholar
- 43.Troy K, Hoch AZ, Stavrakos JE. Awareness and comfort in treating the female athlete triad: are we failing ourathletes? Wisconsin Med J 2006; 105 (7): 21–4Google Scholar
- 44.The IOC Medical Commission Working Group Women in Sport. Position stand on the female athlete triad [online]. Available from URL: http://multimedia.olympic.org/pdf/en_report_917.pdf [Accessed 2009 Sep 27]
- 53.Adams J, Bishop N. DXA in adults and children. In: Rosen CJ, Compston JE, Lian JB, editors. Primer on the metabolicbone diseases and disorders of mineral metabolism. 7th ed. Washington, DC: American Society for Bone andMineral Research, 2008: 152–8Google Scholar
- 78.Greene DA, Naughton GA, Briody JN, et al. Bone and muscle geometry in female adolescent middle-distancerunners. Pediatr Ex Sci 2005; 17: 377–89Google Scholar
- 80.Cann CE, Cavanaugh DJ, Schnurpfiel K, et al. Menstrual history is the primary determinant of trabecular bone density in women [abstract]. Med Sci Sports Exerc 1988; 20 Suppl.2: S59Google Scholar
- 92.Schneider P, Biko J, Schlamp D, et al. Comparison of total and regional body composition in adolescent patientswith anorexia nervosa and pair-matched controls. Eating Weight Disord 1998; 3: 179–87Google Scholar
- 96.Ducher G, Bass SL, Karlsson MK. Growing a healthy skeleton: the importance of mechanical loading. In: Rosen CJ, Compston JE, Lian JB, editors. Primer on the metabolicbone diseases and disorders of mineral metabolism. 7th ed. Washington, DC: American Society for Bone and Mineral Research, 2008: 86–90CrossRefGoogle Scholar
- 110.Bass S, Daly R, Blimkie CJ. Growing a healthy skeleton: exercise — the primary driving force. In: Hebestreit H, Bar-Or O, editors. The encyclopaedia of sports medicine:the young athlete. Oxford: Blackwell Publishing, 2008: 112–26Google Scholar
- 132.American Academy of Pediatrics. Committee on Sports Medicine. Amenorrhea in adolescent athletes. Pediatrics 1989 Sep; 84 (2): 394–5Google Scholar
- 133.Rogol AD. Delayed puberty in girls and primary and secondary amenorrhea. In: Hebestreit H, Bar-Or O, editors. The encyclopaedia of sports medicine: the young athlete. Oxford: Blackwell Publishing, 2008: 227–42Google Scholar
- 157.Warren MP, Perlroth NE. The effect of intense exercise on the female reproductive system. J Endocrinol 2001; 170Google Scholar
- 169.Holick MF, Chen TC. Vitamin D deficiency: a worldwide problem with health consequences. Am J Clin Nutr 2008; 87 Suppl.: 1080S–6SGoogle Scholar
- 188.Lindberg JS, Powell MR, Hunt MM, et al. Increased vertebral bone mineral in response to reduced exercise inamenorrheic runners. West JMed 1987 Jan; 146 (1): 39–42Google Scholar
- 214.Grimston SK, Engsberg JR, Kloiber R. Bone mass, external loads, and stress fracture in female runners. J Appl Biomech 1991; 7 (3): 293–302Google Scholar