August 01, 2005
26 min read

Stress fractures more prevalent in lean athletes

Expert panel discusses etiologic factors, BMD testing and the typical treatment program.

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact


Carol Teitz, MD [photo]Carol Teitz, MD
Orthopaedics and Sports Medicine
University of Washington, Seattle
Member of the Orthopedics Today Editorial Board

John W. O’Kane, MD [photo]

John W. O’Kane, MD
Associate Professor,
Orthopaedics and Sports Medicine
Adjunct Associate Professor,
Family Medicine Medical Coordinator,
Intercollegiate Athletics,
University of Washington, Seattle

Marlene DeMaio, MD [photo]

Marlene DeMaio, MD
Captain, Medical Corps,
United States Navy
Department of Orthopaedic Surgery,
Bone and Joint/Sports Medicine Institute,
Naval Medical Center, Portsmouth, Va.
Assistant Professor,
Department of Surgery,
University of the Health Sciences,
Bethesda, Md.

Sharon L. Hame, MD [photo]

Sharon L. Hame, MD
Assistant Professor,
Department of Orthopaedic Surgery,
David Geffen UCLA School of Medicine,
Team Physician, UCLA Intercollegiate Athletics
Los Angeles

Carol C. Teitz, MD: In which sports are stress fractures most prevalent?

Special concerns about the female athleteJohn O'Kane, MD: The question has been addressed in a number of studies. Populations at high risk for stress fractures include military recruits, distance runners and to a lesser extent sprinters, jumpers, gymnasts, ballet dancers and rowers.

Marlene DeMaio, MD: Editor's note: The views expressed in this article are those of the author and do not reflect the official policy or position of the Department of the Navy, the Department of Defense or the United States government.

Stress fractures are most prevalent in weight-bearing, repetitive loading sports in which leanness is either emphasized, as in cheerleading or ballet, or provide an advantage, such as long distance running. Prospective and retrospective studies of activities involving running, marching and lower extremity loading show the highest incidence of stress fractures. However, overuse of the upper extremity can result in stress fractures of the upper extremity. Be alert to stress fractures in unusual locations.

A very recent study in the American Journal of Sports Medicine (Kiuru MJ, et al, AJSM 2005 33(2):272-276) documented the remarkably high incidence of bone stress reaction in 21 male elite unit military recruits. The investigators evaluated the subjects clinically and by MRI before training, six weeks into training and at the completion of the five-month program. Seventy-five bone stress injuries were diagnosed on MRI. Of these, 30 (40%) were symptomatic. Bone stress injuries were very common. It is interesting to understand exactly what factors take the athlete with bone stress injury on to fracture, when response to the imposed demand is incomplete or fails.

It is important to distinguish between stress reaction and stress fracture. Both are considered fatigue damage to bone. A stress fracture is a partial or complete disruption of the cortex from repetitive loading. A stress injury may have periosteal elevation on radiographs and abnormal signal on bone scan or MRI, but the cortex is intact. Both entities should be associated with repetitive overloads in the patient's history.

Sharon L. Hame, MD: The annual incidence of stress fractures in athletes ranges from approximately 2% to 21%. In the college population, the incidence of stress fractures is consistently highest in track and field athletes, particularly long distance runners.

Overall, stress fractures occur most commonly in the bones of the feet. The single most common bone injured, however, is the tibia. Tennis, volleyball, soccer, basketball players and distance runners have a higher incidence of stress fractures in the long bones such as the tibia while sprinters, hurdlers, jumpers and skaters tend to sustain fractures in the feet. In general, female athletes are more prone to stress fractures than male athletes. In some cases, female athletes' incidence is twice that of males competing in the same sport. Because the overall incidence is higher in track and field, stress fractures should be suspect in males with symptoms who participate in this sport.

Teitz: Do certain sports predispose athletes to stress fractures in specific bones?

O'Kane: Military training, running, jumping sports and ballet classically result in pelvic or lower extremity stress fractures involving the metatarsals primarily but also other bones in foot, the tibia and the femur, including the femoral neck. Gymnasts are prone to lower extremity fractures as well as spondylolysis of the lumbar spine, a stress injury seen in sports involving repetitive spine extension. Rowers are fairly unique in their propensity toward rib stress fractures, although they are reported in golf and tennis and in the first rib of throwers. Throwers and racquet sport athletes more commonly fracture the upper extremity, most often the humerus, ulna or olecranon.

DeMaio: Numerous studies report the association of stress fractures in specific bones related to the sport or activity. Not surprisingly, the distribution of these fractures is related to the type of loading.

Teitz: Are there certain sports in which female athletes are more prone to stress fractures than male athletes?

O'Kane: Early studies on stress fractures focused on military recruits in basic training and found that women were much more likely to sustain stress fractures. Studies in athletes have shown less difference between men and women. Hame et al in 2004 and Arendt et al in 2003 published retrospective reviews of stress fractures at Division 1 universities (Hame et al reported all fractures but separated stress fractures in the analysis). They found track and cross country had the highest incidence for both genders and there was a 1.5- to twofold increased incidence of stress fractures comparing women to men participating in the same sports. The exception was soccer in which men had more stress fractures in Hame's study.

DeMaio: Female cheerleaders, long distance runners, gymnasts, and track and field athletes have higher reported rates of stress fractures than their male counterparts. This is also true for military recruits (both officer and enlisted training programs) and for ballerinas.

Teitz: Are there sports in which we should suspect stress fractures in male athletes?

O'Kane: Track and distance running carry the highest stress fracture risk for males. Limited evidence suggests male soccer players may be at greater risk for stress fracture than their female counterparts. Men's soccer and rugby are both associated with pubic stress fractures. Tibial stress fractures are reported in men's ice hockey.

DeMaio: While stress fractures have been reported in female weight lifters, there appears to be more stress fractures in this group of male athletes.

Teitz: Which history and physical exam features do you find especially useful in making the diagnosis of stress fracture? (Any clinical pearls?)

O'Kane: Stress fractures usually start as manageable, poorly localized pain with activity, then progress to limiting, focal pain with tenderness. With progression, the pain first prevents sport activity and then progresses to pain with normal activities and at rest.

DeMaio: The questions in the history are to determine the primary reason for the stress fracture, what other factors contributed to the development of the fracture, and to see if the patient fits in a special patient category. The usual reason for stress fracture is overuse, overload or overtraining. Numerous studies document either poor physical fitness or low level of physical activity prior to stress fracture. "Poor" can be a relative term. For example, a high school cross country runner takes the summer off before his freshman year in college. The training program there is more rigorous and he is coming off a period of rest. Athletes who have increased their training, have resumed training after prolonged rest or have never trained for this activity are at particular risk. Athletes who are not fit and overweight and embark on a rigorous training program are also prone to stress fracture. The inciting activity or activities must be determined. For both the male and female athlete in whom a stress fracture is suspected, obtain a detailed description of the training schedule, particularly any changes in the training regimen or footwear and newness of the activity.

Determine the site of pain and its relationship to the inciting activity. Does the pain remain at that site or does it radiate? Are there any other sites of bone pain? If the patient had a previous stress fracture, are these symptoms the same? Is it in the same location? Contributing factors must also be identified.

Occasionally, a patient will present with a completed fracture who describes antecedent bone pain while training prior to the fractures. This is an interesting subset of patients who have a stress fracture but only present to you after they complete the fracture. These patients may fracture unusual bones, such as the olecranon, the scaphoid and the medial malleolus. The same points of the history apply.

A wide variety of stress fracture locations have been reported in the literature. Even the sternum has been reported as having a stress fracture in a weight lifter. So, if the symptoms point to a stress fracture, then it should be worked up as such.

Hame: A stress fracture usually presents itself as activity-related pain that gradually gets worse over time. If not treated, pain may eventually occur at rest. A history of a previous stress fracture is also important since the risk of having a second stress fracture is high. A complete menstrual history and diet history are also important. On physical examination, localized pain with palpation at the site of the stress fracture is a good indication of a stress fracture. Edema may occur at the site. The athlete may have pain with percussion distal to the site of fracture or when a tuning fork is used. A hop test may also elicit pain at the fracture site. Have a high index of suspicion for sacral and hip stress fractures in runners complaining of hip, groin or low back pain.

“I have been impressed by the number of young women with abnormal menstrual cycles whose sisters or cousins have also had stress fractures.”
—Marlene DeMaio

Teitz: How might these features vary from one bone to another?

O'Kane: Metatarsal, tarsal navicular, tibial, and forearm stress fractures tend to present with localized pain and tenderness over the fracture site. Femoral stress fractures often present with less localized pain anywhere from the hip to knee with no focal tenderness on palpation. Pelvic pain with pelvic stress fractures may be vague as well. Lumbar spine stress fractures tend to have more localized pain at the base of the spine on the affected side. The pain increases with lumbar extension, particularly single-legged extension standing on the ipsilateral leg. Rib stress fractures present with regional pain about the size of the patient's palm located in the mid to upper rib cage, more axillary than central. Point tenderness is not always elicited. Proximal humeral stress fractures may present with vague shoulder pain and no tenderness.

DeMaio: Depending how late in the process the patients present, the patient may have pain with activities of daily living. In other athletes, they may have to do the inciting activity for the pain to occur.

For the physical exam, there is usually discrete bone pain. For long bones, the fulcrum test is helpful. Percussion over the tibia is usually enough to cause severe pain at the stress fracture site. A log roll test for the hip is usually helpful. Extension or hyperextension of the spine is helpful there.

Teitz: Which etiologic factors should be considered when taking a history from a patient with a suspected stress fracture?

O'Kane: Normal bone metabolism involves ongoing bone formation and resorption. Stress fractures result when bone resorption outpaces bone formation, resulting in weakness and eventually insufficient bone strength to accommodate the applied loading. Factors contributing to stress fracture are those that increase bone resorption, hamper bone formation or increase bone loading more rapidly than the bone can metabolically respond to the increased stress. Rapid increases in training volume and insufficient recovery time can lead to stress fractures despite "normal" bone metabolism. Factors that adversely affect bone metabolism and lower bone density can result in stress fractures in the setting of "normal" loading. These include but are not limited to hypoestrogenic states in females (athletic amenorrhea or post-menopause), hyperparathyroidism, hyperthyroidism, hyperprolactinemia, adrenal and pharmacologic glucocorticoid excess, malnutrition, vitamin D or calcium deficiency, and many chronic disease states.

Studies in healthy athletes have generally not identified stress fracture risk factors for males. Bennell et al in 1996 prospectively followed 53 female athletes for a year and identified risk factors, including low bone mineral density, menstrual irregularity, a low-fat diet and less lean muscle mass in the lower limb.

DeMaio: The etiological factors to be considered include correctable and non-correctable factors.

Correctable factors may include footwear and nutrition. Poor footwear or a change in footwear has been associated with stress fractures. Nutritional status is important. Many athletes do not ingest enough calories and are catabolic. Similarly, ask if the patient eats a "balanced diet." Some vegetarians, particularly vegans, may not be ingesting enough calcium. Disordered eating is associated with stress fractures. Questions to rule out bulimia and anorexia are essential. Besides the catabolism that results, secondary endocrinopathies may result. Lowered levels of testosterone in men and lowered levels of estrogen in women can result in lowered bone mass. However, most athletes will reveal eating habits with inadequate calories, inadequate protein, and inadequate vitamins and calcium. While disordered eating and true eating disorders are more common in women, they do occur in men, particularly long distance runners.

Special patient categories include those with previous stress fractures, the female triad, disabled athlete and other syndromes. A history of previous stress fractures is important. Often the patient has the same symptoms. Determine if other bones were involved, what the inciting activity was, and what and how effective was the previous treatment. Ask the patient if anyone in the family has had stress fractures, kidney disease or metabolic bone disease.

For females, a menstrual history, pregnancy history and use of hormones are required. The age of onset of menses, cycle frequency and duration, and any history of amenorrhea or oligomenorrhea are also important. I have been impressed by the number of young women with abnormal menstrual cycles whose sisters or cousins have also had stress fractures.

I have a form that I use for the history, so I can make sure I ask each of the points.

Disabled athletes are competing in more sports and at very high levels of activity. Little information on stress fractures has been published on this group of athletes. A scapular fracture has been reported in an above-knee amputee (Fink-Bennet DM, et al. Clin Nucl Med. 1984 9(8):430-434). The mechanism of loading must be taken in context to this athlete. Bone pain associated with overuse or new training should raise the possibility of stress fracture. For the paralyzed athlete, warmth, swelling and erythema may be the only signs. Athletes with muscle disorders, particularly those with muscle weakness (myotonic dystrophy, facial-scapulo-humeral dystrophy), may be more at risk for stress fracture due to their muscle disorder.

Patients with a family history of metabolic bone disease or those with several stress fractures and no clear history of overuse are worth mentioning. A stress fracture associated with low levels of training may be the first sign of metabolic bone disease. A patient with several stress fractures and no clear history of overuse or trauma may have metabolic bone disease or osteogenesis imperfecta.

Hame: There are many etiologic factors that should be considered when taking a history from a patient with a stress fracture. These factors include biomechanical factors, bone geometry, muscle mass and strength, changes in training and training errors. There are additional factors associated with female athletes, including bone mineral density, menstrual history and nutritional habits.

“Female athletes diagnosed with the female athlete triad are at risk for developing stress fractures.”
—Sharon Hame

Female athletes diagnosed with the female athlete triad are at risk for developing stress fractures. Research in this area has shown that these athletes have decreased bone density with decreased markers for bone formation. It has been suggested that this component of the triad is interrelated with mechanisms and metabolic pathways of the other components.

The usefulness of estrogen replacement in the treatment of menstrual dysfunction and decreased bone mineral density is unclear. Early cross-sectional investigations showed an association between improved BMD and estrogen use. Recently, however, weight gain was found to be more effective than estrogen therapy. Improving diet and allowing for weight gain seems to allow for improved regulation of hormones and reversal of menstrual irregularity.

If an orthopedic surgeon diagnoses a female athlete triad, then all aspects of the disorder should be addressed. A complete physical examination should be done including vital signs and general nutritional status. The patient should be referred for a complete gynecological examination and psychological assessment. Laboratory testing should include serum electrolytes, glucose levels and a CBC. Additional testing should include a urine pregnancy test, TSH, LH and FSH levels. Estradiol levels may also be helpful. Bone markers to be tested should include N-telopeptide and serum osteocalcin. Radiologic studies should include bone mineral density testing. The orthopedic surgeon should work closely with the athlete's primary care physician and mental health professional to ensure the proper treatment of the athlete.

Teitz: What is the latest information on the female athlete triad and its relation to stress fractures?

O'Kane: The female athlete triad is the combination of amenorrhea, disordered eating and osteoporosis. Calorie intake that is insufficient to meet metabolic and training demands results in hypothalamic disruption of normal LH pulse frequency with subsequent anovulation, amenorrhea and hypoestrogenemia.

It is not training intensity but the lack of available energy for training, particularly carbohydrates, that is primarily responsible for hypothalamic dysfunction. Most athletes with the triad do not have anorexia or bulimia but are characterized as having disordered eating, implying their calorie intake does not meet their caloric expenditure. Adolescent athletes often report that they eat "as much or more" than their peers. They fail to recognize that because of their training, their caloric requirements are much higher.

Athletes with the triad are more likely to be osteopenic than osteoporotic, and some authorities have recommended changing osteoporosis to osteopenia in the definition of the female athlete triad. Even normal-range bone density can be falsely reassuring because athletes should have above-average bone density at high-load sites (may vary depending on sport). An osteopenic runner may have a normal calcaneal bone mineral density but have low density of the lumbar spine with a dual-energy X-ray absorptiometry (DEXA) scan. For this reason, a DEXA scan Z-score (age matched) of the lumbar spine and proximal femur is a much more appropriate assessment of BMD than a calcaneal scan.

Age at menarche plays an important role in stress fracture risk. Several studies have shown that delayed menarche increases the stress fracture risk. Peak bone density occurs between ages 25 and 30, and peak bone acquisition occurs in females aged 11 to 14. It is not surprising that calorie deficits pushing menarche beyond the years of peak bone acquisition, could result in lower bone density and increased risk of fracture.

DeMaio: I think Darryl Thomas and Dean Taylor did a terrific review of this syndrome in the Sports Medicine OKU-3. It is a comprehensive overview.

The latest information on the female athlete triad and stress fractures includes a study just published in Medicine and Science in Sports and Exercise by Torstveit and Sundgot-Borgen (37(2):184-193, Feb 2005). The authors administered a detailed questionnaire to all of the elite athletes in Norway (ages 13-39, N=938) and non-athlete controls (n=900). A total of 669 (88%) and 607 (70%) controls completed the questionnaire. A higher percentage of controls (69.2%) was classified as being at risk for the female triad compared to the athletes (60.4%) with P<.01. The risk for being classified for the female triad was found to be highest in the athletes competing in leanness sports: athletes in leanness sports, 70.1%; nonathlete controls, 69.2%; nonleanness athletes, 55.3% (P<.001). Athletes had more menstrual irregularities and stress fractures compared to controls (P<.05). I think this study illustrates that the risk of being classified in the female triad is quantifiable and not low (69.2%). Basically, we need to have a high level of suspicion of the female triad in the female athlete with a stress fracture. Also, athletes in nonleanness sports have a lower risk for being in the female triad classification but not substantially less.

Teitz: What is the evidence, if any, that oral contraceptives positively affect bone density in the patient with the female athlete triad and a stress fracture?

“Oral contraceptives may be a ‘better than nothing’ solution for athletes with hypothalamic amenorrhea who will not or cannot correct their energy imbalance.”
—John O’Kane

O'Kane: A deceleration in the normal rate of bone acquisition occurs after only three months of absent menses. After three years of amenorrhea, normal peak bone density may not be achieved despite reversing the caloric deficit with resumption of normal menses. In the setting of hypothalamic amenorrhea, normalization of bone acquisition occurs only following the resumption of caloric intake sufficient to reverse the caloric deficit. Restoring menses with oral contraceptives (OCs) decreases bone loss rate in the setting of hypothalamic amenorrhea, but normal bone acquisition does not resume. Oral contraceptives may be a "better than nothing" solution for athletes with hypothalamic amenorrhea who will not or cannot correct their energy imbalance, but the clinician should focus on addressing the calorie-exercise imbalance not fixing the amenorrhea.

DeMaio: Oral contraceptives are estrogen based and progesterone based. There is some concern that use of progesterone-based OCs may not be the best choice for contraception in the athlete with stress fracture because the osteoblast responds to estrogen and serum estrogen is lower in this patient group. Estrogen-based OCs are associated with healing of stress fractures. However, there is no compelling evidence that estrogen-based OCs increase bone density. Estrogen-based OCs seem to maintain the bone density not increase it. The change in serum estrogen levels and the return of menses seems to be the basis for this. Few studies in the premenopausal woman have been done specifically looking at this problem.

One study (Lappe JM, et al. Osteoporosis Int. 2001;12(1):35-42) documented use of depo-medroxyprogesterone acetate (DMPA) as one of several risk factors associated with the development of stress factures in female Army recruits. This study did not specifically look at the role of DMPA but it notes DMPA as a contributor to stress facture risk. For this reason and because the role of estrogen in the maintenance and production of bone mass is known, it is probably best to select estrogen-based HRT. For patients with non-healing stress fractures on DMPA, I have consulted an endocrinologist with an interest in bone metabolism.

Teitz: When an orthopedic surgeon diagnoses the female athlete triad, what should be done next?

O'Kane: Female athlete triad is a diagnosis of exclusion after a number of other issues resulting in osteopenia and amenorrhea are ruled out. Athletes in whom the triad is suspected should be referred to a physician with expertise in this area. Orthopedic surgeons should have familiarity with the triad so at-risk athletes can be identified and appropriately referred.

DeMaio: Each aspect of the diagnosis must be addressed. How each orthopedic surgeon handles this depends on his practice and available consultants. The orthopedic surgeon may manage the consequences of the osteoporosis, but he is obligated to either "triage" or refer regarding the other two components of the triad (disordered eating and oligomenorrhea).

Measure the patient's height and weight and calculate her percent body fat. Get more information such as highest and lowest weight, recent weight loss or gain (intentional or not), use of vomiting/laxatives/diet supplements, and eating patterns/habits. Depending on the degree of the disordered eating, referral to a dietician or nutritionist may suffice. You can ask the patient to start a food diary and continue it for three days. This will be helpful when she arrives at the dietician or nutritionist. If the patient appears grossly malnourished, you can order baseline laboratory studies including BUN, creat, WBC, LFTs, serum calcium, serum phosphate, serum protein and albumen. If you identify a serious eating disorder, the patient should be referred to a psychiatrist or other health professional with expertise in this area.

For irregular or absent periods, get more information like age at the onset of menarche, length and frequency of menses, pregnancy/births, use of birth control pills and last menstrual period. If the patient has secondary amenorrhea, a urine pregnancy test should be ordered. If the patient has recurrent stress fractures, multifocal stress fractures, or is severely malnourished, the laboratory tests listed above and serum estrogen, FSH, LH, thyroid function tests should be ordered. The patient should be referred to an endocrinologist or gynecologist with an interest in this syndrome. If the patient is oligomenorrheic ask additional questions about the duration of the irregularity, its onset and past occurrence. These patients should be followed by their primary care provider. Both groups of patients may require HRT if they have four periods or less a year. The management of HRT is usually by the endocrinologist, gynecologist or primary care provider.

These patients take some time in the clinic; it is best to schedule for this (if possible). It is also good to know your consultants in your area who have an interest in this patient population.

Teitz: In which patients with stress fractures would you obtain a bone density study?

O'Kane: Athletes with more than one stress fracture or with other issues raising the likelihood of low bone density (amenorrhea, low weight or an eating disorder) should be referred for a DEXA scan.

DeMaio: Bone density tests are indicated in patients with recurrent stress fractures, multifocal stress fractures, known eating disorders and the female triad. A bone density study is also indicated in athletes with amenorrhea for one year and with poor nutritional status. It should be considered in athletes with amenorrhea of six months or more. While bone density tests are less reliable in very young athletes, as they are still laying down bone, it is important to get a baseline in this group. Similarly, it is part of the work up to support the diagnosis of stress fracture.

In general, the bone density tests are done 12 months apart. Follow-up bone density tests are helpful when low bone density is identified and treatment to correct this problem has started. The effectiveness of the treatment can be monitored.

Hame: An athlete who presents with multiple stress fractures or any risk factors for the female athlete triad should undergo a bone density test. There is a high correlation between low bone density and stress fracture formation. A standard DEXA scan on a mature patient should include the hip and the spine. This is what is performed at our institution. For adolescents and children, a DEXA of the whole body and spine is recommended. Since female athletes may present with the female athletic triad in adolescence, the whole body and spine would be the optimum choice for accuracy of evaluation of bone density.

Teitz: What is your preference for type/location of bone density study (eg, spine, hip, wrist, heel, etc.)?

O'Kane: A DEXA scan provides the best estimate of bone density with minimal radiation. Different protocols will provide bone density at different sites. Vertebral and trochanteric density are frequently used in the literature and provide a good assessment. Assessing only the calcaneus, which is fairly common, often leads to falsely elevated assessments in runners because of the osteogenic response from heavy loading. Evaluating only the dominant wrist in a tennis player could result in the same overestimate.

DeMaio: A DEXA scan is my preference. The WHO criteria for osteopenia and osteoporosis is based on this study. This study evaluates the bone density of the lumbar spine and hip.

Teitz: What is the evidence that osteopenia or osteoporosis on a bone density study is associated with a current stress fracture?

O'Kane: A number of studies have demonstrated that in female athletes, low bone density, low caloric intake, menstrual irregularity, and a history of prior stress fracture are risk factors for current stress fracture. In male athletes, the risk factors are less clear.

DeMaio: Most well-nourished young patients will not have T scores and Z scores far outside the norm. As you know, the T score is the comparison of the patient's bone density to the peak bone density of the young, healthy controls. We conclude that these patients had "normal" bone density and sustained their stress fracture from overuse, with or without other contributing factors (low protein intake, low calcium and vitamin D intake, oligomenorrhea). However, there is a group of patients where the bone density is low by T score and Z score. Here, a bone density test is very helpful as a baseline and for future monitoring.

I found one retrospective study (Lauder TD, et al. Arch Phys Med Rehabil. 2000 81(1):73-79) that reviewed the BMD in 27 active duty women with stress fractures within two years and compared them to 158 female controls. The females with stress fractures had lower BMD and higher activity levels.

One study evaluated the bone mineral density using CT densitometry, trace mineral analysis and bone histomorphology from the iliac crest in 13 military recruits with femoral neck hip stress fracture undergoing ORIF and a control group of 15 undergoing iliac crest bone grafting for scaphoid fracture. The patients with the femoral neck fracture had lower bone mineral density on the CT but there were no significant differences between the two groups by bone histomorphometry (Muldoon MP et al. J Orthop Trauma. 2001 5(3):181-185).

Using bone density tests to predict future stress fractures in athletes, malnourished athletes, and female triad athletes is not established. I have not seen a study with baseline bone density tests and the development of stress fractures.

Teitz: What are your indications for obtaining imaging studies of the bone with a suspected stress fracture?

O'Kane: Imaging should be ordered when the results of the imaging are likely to change your management. If an athlete has a suspected stress fracture with low potential for significant morbidity (mid-shaft metatarsal) and she is willing to stop training for six weeks, slowly resuming training as symptoms allow, imaging beyond plain X-ray is not needed. Nucleotide bone scan or MRI is necessary when an athlete will only stop training if absolutely necessary, or if the athlete needs to return to training as soon as possible. Suspected stress fractures with the potential for significant morbidity should also have early imaging. They include tarsal navicular, proximal fifth metatarsal, talus, anterior cortex of tibia and femoral neck.

“Patients with recurrent stress fracture and those with signs and symptoms of multifocal stress fracture should get whole-body nuclear imaging. ”
—Marlene DeMaio

DeMaio: While X-rays are required for patients with bone pain, they are usually negative early in the natural history. It may take up to three months for the radiograph to be positive. X-rays are necessary to rule out other less common causes of bone pain, particularly tumors. If the X-ray is negative, the next step is a three-phase Tc-99 bone scan. This study is considered the "gold standard" for stress fractures. The history and physical determine the working diagnosis; therefore, the lack of specificity of the bone scan is not so much an issue for this group of patients.

Patients whose history is not clearly a stress facture may require other imaging and tests. For example, a patient with leg pain may have periostitis, compartment syndrome or stress fracture. I have had three patients with stress fracture and exertional compartment syndrome. Patients with recurrent stress fracture and those with signs and symptoms of multifocal stress fracture should get whole-body nuclear imaging.

Hame: If an athlete has a suspected stress fracture, then I obtain radiographs of the injury site. If the radiographs are positive for a stress fracture then no further studies are ordered. If the radiographs are normal and the athlete's symptoms persist, I then order a MRI of the region. MR imaging with T1-weighted sequences and STIR or T2 images are helpful. Fat suppression appears to increase image sensitivity and specificity. Early stress changes such as bone marrow edema can be visualized. In addition, soft tissue injuries including periosteal edema, muscle injury, tendon injury and avascular necrosis can be diagnosed. Stress fractures may also be graded on MRI, which allows for determining return to play issues.

Teitz: What type of imaging do you prefer and why?

O'Kane: Many studies have demonstrated that MRI has equivalent sensitivity to bone scan and the MRI does not require injection, involves no radiation, and provides some additional information. In 2003, Arendt et al proposed a stress fracture grading system using MR findings ranging from bone edema on STIR imaging to abnormal radiographs with a visible line on T1 and T2 imaging to gauge treatment intensity and predict time to return. MRI also provides soft tissue information that may reveal the cause of pain while ruling out stress fracture. The advantages of bone scan are lower cost (regionally dependent) and a wider field of imaging. There are reports in runners where pain from a mid-shaft femoral stress fracture is localized more in the hip, and hip MR imaging misses the fracture. For this reason in runners with hip/thigh pain and stress fracture in the differential, a bone scan may be a better first step followed by CT to evaluate high-risk positive findings (femoral neck).

Suspected spondylolysis should be assessed with a lumbar lateral plain film to rule out spondylolisthesis then a SPECT scan of the lumbar spine. MR may reveal spondylolysis but lacks the sensitivity of SPECT scanning.

DeMaio: I prefer the three-phase Tc-99 bone scan. CT scans are helpful for stress fractures of the pars, the navicular and the sacrum. For these fractures the CT better defines the cortical extent and can be helpful in determining the role of surgery.

For patients with localized pain and a history suspicious for neoplasm and not completely consistent with stress fracture, an MRI is the most sensitive and specific.

Patients with a hip stress fracture should be imaged with an MRI to see the extent and location of the fracture. While tension side stress fractures are treated surgically, compression side fractures with a large injury zone extension to the superior aspect of the neck should be considered for surgery.

MRI has good intra- and interobserver agreement which improves using T2W and STIR images (Ahovuo JA, et al. Magn Reson Imaging. 2002;20:401-406).

Teitz: What are the typical components of your treatment program for an athlete with a lower limb stress fracture? Please consider the following in your reply:

  1. Immobilization
  2. Non-weight-bearing
  3. Your experience, if any, with electrical bone stimulators in the treatment of stress fractures
  4. The typical components of your return to play program
  5. The appropriate settings for operative intervention.

O'Kane: Low-risk stress fractures require sufficient rest, immobilization and modified weight bearing to eliminate all pain. At that point slow progress back to activity is possible. Immobilization can help a foot stress fracture by allowing earlier pain-free weight-bearing, while a femoral shaft fracture may require longer use of modified weight-bearing but not be a candidate for immobilization. In general, it takes four to six weeks to get symptoms under control, then six weeks progressing through low-level training advancing as symptoms allow to get back to full participation. High-risk stress fractures require longer periods of rest, immobilization and non-weight-bearing, and possibly surgery depending on the fracture. These injuries should be managed by a physician familiar with the injury.

Bone stimulators have demonstrated some utility in decreasing time to radiographic and clinical healing in certain fractures, and in some studies they result in accelerated osteogenesis in lab models. There are reports of bone stimulators being used when treating recalcitrant stress fractures, some with eventual healing. The literature lacks evidence for using bone stimulators in routine stress fracture management.

Return to play is guided by clinical response. Training should slowly progress with increasing load and volume, at a rate that results in no recurrence of symptoms. The MR grade of fracture (Arendt et al) can assist in the predicted pace of progression. Generally low-risk fractures need to rest for three to six weeks followed by increased training in small increments every few days provided no pain occurs.

Surgery is indicated when stress fractures fail to heal despite appropriate management. The indications vary with every fracture and patient, although scenarios that more often result in surgery include: Complete femoral neck fractures or those involving to a significant extent the superior aspect of the femoral neck, complete tarsal navicular fractures, fractures of the proximal fifth metatarsal with delayed healing or a strong desire to return quickly, anterior cortical fractures of the tibia with radiographic evidence of bone resorption ("dreaded black line") and complete talar neck fractures.

DeMaio: Before considering the orthopedic aspects of treatment, I address the nutritional and medical factors first, so I don't overlook them. Ideally, consultation with a dietician or nutritionist to evaluate the caloric intake and distribution of the calories is the best. If this is not an option, then the patient needs instruction on the number of calories, snacks and protein intake. Calcium and vitamin D along with a multivitamin should be prescribed. Appropriate referrals for eating disorders should be made.

Treatment depends on the extent of the stress fracture, duration of the fracture, response to previous treatment (if any), activity demands of the patient and fracture site. Certain sites with a predilection for nonunion, delayed union, and completion and or displacement may meet the indications for acute or early surgery.

Immobilization is helpful for stress fractures of the lower extremity including the patella, tibia, fibula, ankle and foot. Immobilization rests the limb and provides load sharing. It may be used as the primary mode of treatment for lower extremity fractures other than the hip and femur.

A knee immobilizer can be used for proximal tibial fractures for short periods and with range of motion. A long leg splint (Air Cast) provides good relief for the mid- and distal tibial stress fractures. A boot is helpful for the distal tibia, medial malleolus, calcaneal and foot fractures.

There are some fractures with a high rate of completion (intracapsular hip stress fractures, femoral stress fractures and medial malleolar stress) where surgery should be considered as primary treatment.

Nonweight-bearing and toe-touch weight-bearing is used acutely until the patient's pain decreases. Then progressive weight-bearing may begin. Nonweight-bearing and toe-touch weight-bearing is the treatment for compression side non-displaced intracapsular hip fractures.

Subchondral hip stress fractures are also treated with nonweight-bearing. This diagnosis seems to be distinct from AVN (Song WS, et al. JBJS. 2004;86A:1917-1924 and Visuri T. Acta Orthop Scand. 1997;68(2):138-141).

Acetabular stress fractures are, not surprisingly, also treated with nonweight-bearing. Other associated fractures occur with this unusual fracture, such as femoral neck stress fracture and inferior pubic ramus fracture (Williams TR. Skeletal Radiol. 2002;(5):277-281).

If the patient's pain is controlled in a brace or walker, then weight-bearing is permitted. Use pain as a guide.

During the period of nonweight-bearing and activity restriction, other types of aerobic training should be started. This may include running in the water (aqua-jogging), swimming, aerobic weight training, upper extremity cycle, and other nonweight-bearing activities (sit-ups, floor exercises, etc.).

One prospective study (Rue JP. Orthopedics. 2004.27(11):1192-1195) showed no difference in the healing time between pulsing electromagnetic fields (PEMF) and traditional treatment for tibial stress fractures in midshipmen. Another study in navicular stress fractures (Saxena A. J Foot Ankle. 2000;39(2):96-103) used PEMF in some of the patients; about 27% of the these patients went on to have ORIF.

There is no clear evidence that this treatment option is effective, nor is there evidence that it delays healing. I do not use them in the treatment of stress fractures.

There must be resolution of pain with activities of daily living, training, and practice. Generally, athletes need six to eight weeks for recovery. Stress fractures involving the hip, femur, distal tibia and the navicular often need longer.

Surgery is indicated for non-healing stress fractures and those fractures where completion of the fractures is disastrous. In the acute setting, tension hip stress fractures, some compression hip stress fractures, some femoral stress fractures, some patellar stress fractures (rare), some medial malleolar stress fractures, and some base of the fifth metatarsal stress fractures require ORIF. Nonunion of tibial stress fractures (Chang PS, et al. AJSM. 1996;24(5):688-92) and pars fractures generally require surgery, particularly if the athlete wants to continue competing. Symptomatic nonunions in other sites may also require surgery once all correctable factors have been treated and the patient has failed a period of true rest. For the fifth metatarsal, nonweight-bearing for four weeks followed by four weeks of partial WB usually results in healing.

Patients who present with a completed stress fracture may need special fixation. For example, medial tibial stress reactions that complete may require low profile plates as opposed to the more common distal tibial stress fracture in which an intramedullary rod is indicated. Consideration of bone graft should be given to some fractures in this category, such as olecranon stress fractures that complete.

Hame: The treatment of the athlete with a stress fracture depends on the sport and the level of the competition.

I treat most lower extremity stress fractures with a protective device such as a walking boot. Athletes are allowed to weight-bear as long as they are painfree. Athletes who remain symptomatic will require crutches and limited weight-bearing.

To avoid stiffness I allow range-of-motion exercises. Athletes are encouraged to continue upper body strengthening and to maintain aerobic capacity with an arm ergometer or in the swimming pool. I also advise pool therapy for exercise and gradual weight-bearing exercises. Bone stimulators may be helpful in fractures that appear to be healing slowly, but I do not use them routinely in this population. Once pain-free ambulation has been achieved, a gradual return to sport is initiated. Athletes must be counseled on the expectation of the recovery time and return to full participation. Some fractures require operative intervention. Stress fractures in the superior portion of the femoral neck are at risk for displacement and, therefore, require internal fixation. Fifth metatarsal and tarsal navicular fractures also have demonstrated earlier return to sport with internal fixation. The treatment of tibial stress fractures is controversial especially those with cortical compromise. Prophylactic intramedullary nailing of the tibia may be helpful in some of these athletes.