The Physician and Sportsmedicine
Menubar HomeJournalPersonal HealthResource CenterCMEAdvertiser ServicesAbout Us

Managing Common Stress Fractures: Let Risk Level Guide Treatment

Peter Brukner, MBBS; Chris Bradshaw, MBBS; Kim Bennell, PhD


In Brief: The repetitive stresses of sports and exercise can produce an array of stress fractures. Most are uncomplicated, but some, such as femoral neck fractures, carry a higher risk of nonunion or complete fracture. The diagnosis is primarily clinical, but imaging with plain radiographs, scintigraphy, CT, or MRI may provide confirmation if necessary. Treatment of uncomplicated fractures centers on rest and reversing training errors or equipment problems. Management of high-risk fractures is more aggressive. Depending on imaging results, most of these require either surgery or several weeks of non-weight bearing immobilization and rehabilitation.

Most stress fractures are relatively straightforward management problems. Athletes report a history of exercise-related pain, usually well localized and often associated with a recent increase or change in activity. The history and local bony tenderness on the physical examination are often sufficient to make the diagnosis. Uncomplicated stress fractures typically respond to rest from the aggravating activity and removal or modification of risk factors.

Certain stress fractures, however, present diagnostic and treatment challenges: those that are atypical in clinical presentation, inaccessible to physical examination, at high risk for progression to complete fracture, and/or prone to delayed union or nonunion. Specific sites are the neck of the femur, anterior cortex of the middle third of the tibia, medial malleolus, talus and tarsal navicular, and fifth metatarsal.

Stress Fracture Dynamics

Stress fractures, first described in military recruits undergoing basic training, are partial or complete fractures that result from repeated stress that is lower than what is required to fracture the bone in a single load.

Recent studies have shown that the incidence of stress fractures in athletes is higher than previously thought. Our prospective study (1) of 95 track-and-field athletes showed an annual incidence of approximately 20%. This apparent increase is probably related to increased clinical awareness, more sensitive imaging techniques, and increased training loads.

Women reportedly have a higher rate of stress fractures than men. In military studies, female recruits have a higher risk of stress fracture than male recruits, with relative risks ranging from 1.2 to 10 for similar training volumes. In athletic populations, however, a gender difference in stress fracture rates is not as evident (2), perhaps because female athletes are more fit than military recruits.

Stress fractures are most common in the lower extremities, but also occur in non-weight bearing bones, including the ribs and upper extremities, and the pelvis. The most common sites are the tibia, metatarsals, and fibula. A recent study (3) demonstrates a high incidence of tarsal navicular stress fractures, which may be the most common site in certain groups such as sprinters and hurdlers. Sports associated with specific stress fractures include rowing and golf (ribs), baseball pitching (humerus), and gymnastics (spine) (table 1: not shown).

Of the many risk factors for stress fractures that have been proposed, training errors are probably the most important. They include a sudden increase in the quantity or intensity of training, introducing a new activity (eg, hill running), poor equipment (eg, worn-out running shoes), and change of environment (eg, changing surfaces from asphalt to cement). Other risk factors include low bone density, dietary deficiencies, abnormal body composition, menstrual irregularities, and biomechanical abnormalities. The role of these factors and their interrelationships has been examined mostly in cross-sectional studies of female athletes.

Our prospective study (4) showed that in females, lower bone density, a history of menstrual disturbance, less lean mass in the lower limb, a discrepancy in leg length, and a low-fat diet were significant risk factors for stress fractures. In men, no significant risk factors were identified; however, there was a strong trend toward low bone density. Often, a combination of factors is involved, such as a sudden increase in training in an amenorrheic woman who has excessive subtalar pronation.

Clinical Diagnosis

The greatest challenge in diagnosing and treating stress fractures is to identify the contributing factor or factors and modify or eliminate them so that the injury does not recur when the athlete resumes training. Thus, one aim of the clinical assessment is to identify possible risk factors.

History. A typical clinical feature of a stress fracture is a history of exercise-related localized pain that increases with activity and either abates with rest or persists at a lower level. If training continues, the pain progressively worsens and is brought on with less intense activity.

A comprehensive training history should note recent changes in activity level such as increased quantity or intensity of training and changes in surface, equipment (especially shoes), and technique. It may be necessary to obtain information from the patient's coach or trainer. A dietary history should be taken, with particular attention paid to the possibility of an eating disorder. In women, a menstrual history is necessary, including age of menarche and subsequent menstrual status.

A history of similar or other musculoskeletal injury should be obtained. It is essential to obtain a brief history of the patient's general health, medications, and personal habits to screen for factors that may influence bone health, such as thyroid disease or glucocorticoid or anabolic steroid ingestion. It is also important to ask about the nature of job-related physical activity and to determine how serious the patient is about his or her sport and what significant events are planned for the immediate future.

Physical exam. The most obvious physical feature is local bony tenderness. Obviously, this is easier to locate in bones that are relatively superficial and may be absent in those that are deeper, such as the shaft or neck of the femur. The examiner carefully palpates affected areas, particularly in the foot, where a number of bones and joints in a relatively small area may be affected. Occasionally, local redness and swelling are present. There may also be palpable periosteal thickening, especially in a long-standing fracture. Percussion of long bones may produce pain at distant points.

The physical examination must take into account the potential predisposing factors. A full biomechanical examination must be performed in patients who have a suspected stress fracture of the lower limb, noting evidence of leg-length discrepancy, malalignment (especially excessive subtalar pronation), muscle imbalance, weakness, or stiffness.

The Role of Imaging

The diagnosis of stress fracture is primarily clinical. The classic history of exercise-associated bone pain and typical examination findings of localized bony tenderness highly correlate with stress fracture. However, various imaging techniques are available to the clinician if the diagnosis is uncertain or if the patient is a competitive athlete who wishes to continue training and requires more specific knowledge of his or her condition.

Plain radiographs have poor sensitivity but are highly specific for the diagnosis of stress fractures. A stress fracture can be confirmed by the presence of any of the classic radiographic abnormalities: periosteal bone formation, sclerosis, callus, or a fracture line (figure 1). In most patients who have stress fractures there is no obvious radiographic abnormality unless symptoms have been present for at least 2 to 3 weeks. In some patients, radiographic changes may not appear for 3 months; in still others, radiographic changes never appear.


Bone scans are the most sensitive indicator of bone stress, but have poor specificity. The triple-phase technetium-99m bone scan can differentiate soft-tissue and bony injury. In the first phase, flow images obtained immediately after the intravenous injection of the tracer demonstrate perfusion in bone and soft tissues and may show increased perfusion in acute inflammation. The second phase (the static "blood pool" phase), taken 1 minute after the injection, reflects the degree of hyperemia and capillary permeability of bone and soft tissue. It may also show acute inflammation. The third phase, the delayed image, is taken 3 to 4 hours after injection when approximately 50% of the tracer has concentrated in the bone matrix.

All three phases can be positive (figure 2) in patients who have acute stress fractures. In soft-tissue injuries without bony involvement the first two phases are often positive, but the delayed phase shows no or minimal increased uptake. As the bony lesion heals, perfusion returns to normal, followed by blood pool normalization a few weeks later. Focal increased uptake on the delayed scan resolves last because remodeling continues after pain disappears. As healing continues, uptake intensity on the delayed scan diminishes 3 to 6 months after an uncomplicated stress fracture; some uptake persists longer than 12 months. Consequently, bone scans should not be used to monitor healing--the assessment of healing is a clinical judgment.


Though the bone scan is virtually 100% sensitive for bone stress, it does not visualize the fracture. On a positive bone scan it may be difficult to precisely locate the fracture, especially in the foot. Other conditions such as osteomyelitis, bony infarct, bony dysplasia, and osteoid osteoma can also produce localized increased uptake. Increased uptake is frequently found at asymptomatic sites of early bone stress, particularly in active patients. Therefore, it is important to balance bone-scan findings with the patient's clinical picture.

Computed tomography (CT) may be useful for identifying conditions that mimic stress fracture on bone scan. CT scans are particularly valuable when a fracture image is needed to make treatment decisions, such as in navicular stress fractures.

Magnetic resonance imaging (MRI) is widely used in the primary investigation of stress fractures. It is as sensitive as scintigraphy, is highly specific, and can visualize soft-tissue changes. The main drawback of MRI, apart from its cost, is that it does not image cortical bone as well as CT.

Bone stress is identified on MRI as marrow edema; a stress fracture typically appears as a fracture line at the cortex surrounded by an intense zone of edema in the medullary cavity (figure 3). These signs are most evident in fat-suppressed views such as the short-T1 inversion recovery sequences.


Treating Uncomplicated Fractures

Basic treatment for patients who have uncomplicated stress fractures is rest from the aggravating activity for 4 to 8 weeks. Patients can maintain fitness during treatment by cycling, water running, or working out on exercise machines. A recent study (5) showed decreased time in returning to sport after a tibial stress fracture with the use of a pneumatic leg brace.

As with any other overuse injury, it is important to identify and correct the stress fracture causes before the patient resumes activity. Corrections may include training modification (after discussion with the coach), the use of orthoses to correct excessive pronation, and hormone supplementation in hypoestrogenic females.

The progress of stress fracture healing is monitored clinically. Patients can resume exercise when healing is evident: when they are pain-free during activities of daily living and there is no local tenderness. Return to activity should be gradual--first walking, then jogging, then running at their normal pace.

High-Risk Fractures: Femoral Neck

Stress fractures of the femoral neck are uncommon, but they may have serious consequences. This stress fracture should be considered in the differential diagnosis when a running athlete has groin pain and reduced hip range of motion. Patients usually report a history of gradual or acute-onset anterior hip, groin, or knee pain that is aggravated by exercise. Night pain may be present. Physical examination reveals pain and restriction at the end of passive hip range of motion. Unlike most other stress fractures, these injuries usually do not cause tenderness on palpation.

Femoral neck fractures are classified as distraction or compression fractures. Distraction fractures often occur in older patients and are radiographically visible at the superior margin of the femoral neck as cortical discontinuity. This fracture requires early recognition because it can progress to complete fracture and displacement. Patients who have nondisplaced distraction fractures require rest in bed or the use of crutches until passive hip movement is pain free and radiographs show evidence of callus formation. Because of the displacement risk, patients should be carefully monitored with serial radiographs every 2 to 3 days for the first week or until pain at rest resolves. Immediate open reduction and internal fixation is indicated if the fracture widens.

Compression fractures of the femoral neck generally occur in younger athletes at the cortex of the lower medial margin of the femoral neck. These stress fractures carry a risk of avascular necrosis and nonunion, and early diagnosis and treatment are essential for successful management. Patients who report exercise-related groin pain and painful restriction of movement at the end of hip range of motion should undergo plain radiography. The fracture appears as an opaque, hazy area of callus. If radiographs are inconclusive, an isotope bone scan should be obtained. Treatment is conservative, with initial rest until passive hip movement is pain free, followed by non-weight bearing on crutches until radiographic evidence of healing. In both types of fractures, weight-bearing exercise is resumed gradually and built up to preinjury level over 6 to 8 weeks.

Anterior Cortex of the Tibia

Most tibial stress fractures occur in the distal third of the bone and respond to rest and a gradual return to weight-bearing activity. Fractures of the middle third of the anterior tibial cortex, however, are a particular concern because they are prone to nonunion.

Nonunion is demonstrated radiographically by the "dreaded black line" at the anterior cortex (figure 4), which suggests bony resorption at the fracture site. At this point, the patient may have been symptomatic for some months, but frequently the symptoms are minimal and the athlete is able to continue activity. There have been a number of reports of patients with this fracture who developed acute transverse fractures of the tibia with continued exercise.


This fracture should be treated aggressively with either immobilization in a non-weight bearing cast for about 3 to 6 months or early surgery. We favor insertion of an intramedullary rod, which enables a return to sports after 6 to 8 weeks.

Medial Malleolus

Stress fractures of the medial malleolus occur in distance runners; patients typically report medial ankle pain with local tenderness. The fracture line is often vertical from the tibial plafond (the smooth, concave articular surface of the distal end of the tibia) and medial malleolus, but may arch obliquely from the distal tibial metaphysis. Shelbourne et al (6) recommend that patients who have a visible fracture line on radiography undergo open reduction and internal fixation. Fractures that are positive on bone scan and negative on radiographs can be treated with immobilization for about 6 weeks in a pneumatic brace.


Stress fractures of the body of the talus extend into the subtalar joint and are prone to nonunion. We described four patients--a footballer (Australian rules football), a triathlete, a pole vaulter, and a recreational athlete--with lateral-body stress fractures who had signs and symptoms of severe sinus tarsi syndrome with gradual onset of lateral ankle pain (7). Plain radiographs did not demonstrate the fracture in any of the cases, but a characteristic appearance of localized increased uptake was seen on bone scans. Recommended treatment is immobilization in a non-weight bearing cast for 4 to 6 weeks if the fracture is diagnosed early. Excision of the lateral process of the talus through the fracture line is recommended for longstanding symptomatic fractures.

Tarsal Navicular

Stress fractures of the tarsal navicular bone are common, especially among runners, jumpers, hurdlers, and basketball players. Symptom onset is usually insidious; patients have increased foot pain after sprinting, jumping, or running. These fractures are often poorly recognized, possibly because the patients' symptoms are vague, particularly in the early stages of injury. The pain typically radiates along the medial longitudinal arch or along the dorsum of the foot. Symptoms abate rapidly with rest but recur when the patient resumes activity. Without correct diagnosis and definitive treatment, patients who have navicular stress fractures usually suffer prolonged disability.

Examination reveals tenderness over the proximal dorsal aspect of the navicular--the "N" spot (figure 5) (8,9). Typical scintigraphic and CT appearances are shown in figure 6.


Navicular fractures that are treated with rest and a gradual return to activity have a high incidence of delayed union and nonunion. Recommended treatment of a navicular stress fracture is immobilization in a non-weight-bearing cast for at least 6 weeks. Even complete navicular fractures have been shown to heal with this regimen. After 6 weeks the patient's cast is removed and healing is assessed by palpation of the "N" spot. X-rays or isotope bone scanning should not be used to monitor progress of the fracture because radiographic signs of healing lag well behind clinical healing. A 6-week rehabilitation program, consisting of joint mobilization, muscle strengthening, and gradual return to activity, can begin when the patient has no pain on palpation of the "N" spot. As with other stress fractures, possible risk factors, such as excessive subtalar pronation, should be identified and treated with orthoses or other measures.


Fifth Metatarsal

Diaphyseal stress fractures of the fifth metatarsal are sometimes mistakenly called Jones fractures. To avoid confusion, the term "Jones fracture" should be reserved for an acute fracture of the fifth metatarsal at the level of the articular facet between the fourth and fifth metatarsals that does not extend distally (10).

Stress fractures of the diaphyseal shaft of the fifth metatarsal occur in the proximal 1.5 cm of the shaft. Torg et al (11) described three subtypes of diaphyseal stress fractures: acute (or early), delayed union, and nonunion (Torg types 1, 2, and 3, respectively). What Torg et al described as "acute" was an early stress fracture with periosteal reaction that represents healing of an incomplete fracture. Patients who have acute fractures report no prior history of pain in the fracture region. X-ray confirms the fracture and shows no medullary sclerosis. Acute fractures have a good prognosis with immobilization in a nonwalking, short leg cast for 6 weeks, after which healing is assessed clinically and radiographically. Patients can then gradually resume activity.

Most diaphyseal fractures occur acutely. Patients initially present with an apparent acute fracture, but when questioned they describe a variable period of pain at the fracture site with activity. X-ray shows a fracture line and, depending on the duration of the symptoms, may show medullary sclerosis. These fractures are prone to delayed union and nonunion, though most will heal in about 3 to 6 months with prolonged cast immobilization. We recommend surgery for patients who cannot tolerate prolonged immobilization.

Torg type 3 diaphyseal stress fractures are radiographically characterized by established nonunion with complete intramedullary obliteration. These patients require surgery--screw fixation or bone grafting. Patients typically return to activity in 6 weeks following screw fixation and 12 weeks after bone grafting.

Minimize Downtime

Diagnosing and treating stress fractures hinges on the back-to-basics approach of a thorough history and office exam. Physicians who take the extra step of identifying and correcting training and equipment errors, along with the medical conditions that can contribute to stress fractures, will help active patients avoid recurrence.


  1. Bennell KL, Malcolm SA, Thomas SA, et al: The incidence and distribution of stress fractures in competitive track and field athletes: a twelve-month prospective study. Am J Sports Med 1996;24(2):211-217
  2. Bennell KL, Brukner PD: Epidemiology and site specificity of stress fractures. Clin Sports Med 1997;16(2):179-196
  3. Brukner P, Bradshaw C, Khan KM, et al: Stress fractures: a review of 180 cases. Clin J Sports Med 1996;6(2):85-89
  4. Bennell KL, Malcolm SA, Thomas SA, et al: Risk factors for stress fractures in track and field athletes: a twelve-month prospective study. Am J Sports Med 1996;24(6):810-818
  5. Swenson EJ Jr, DeHaven KE, Sebastianelli WJ, et al: The effect of a pneumatic leg brace on return to play in athletes with tibial stress fractures. Am J Sports Med 1997;25(3):322-328
  6. Shelbourne KD, Fisher DA, Rettig AC, et al: Stress fractures of the medial malleolus. Am J Sports Med 1988;16(1):60-63
  7. Bradshaw C, Khan K, Brukner P: Stress fracture of the body of the talus in athletes demonstrated with computer tomography. Clin J Sport Med 1996;6(1):48-51
  8. Brukner PD, Khan KM: Clinical Sports Medicine. Sydney, McGraw-Hill Book Co, 1993
  9. Khan KM, Brukner PD, Kearney C, et al: Tarsal navicular stress fractures in athletes. Sports Med 1994;17(1):65-76
  10. Jones R: Fractures of the base of the fifth metatarsal bone by indirect violence. Ann Surg 1902;35:697-700
  11. Torg JS, Balduini FC, Zelko RR, et al: Fractures of the base of the fifth metatarsal distal to the tuberosity: classification and guidelines for non-surgical and surgical management. J Bone Joint Surg (Am) 1984;66(2):209-214

Dr Brukner is the clinic director, Dr Bradshaw is a sports physician, and Dr Bennell is a physiotherapist at Olympic Park Sports Medicine Center in Melbourne, Australia. Dr Bennell is a lecturer in the School of Physiotherapy at the University of Melbourne, Drs Brukner and Bradshaw are fellows of the Australian College of Sports Physicians. Address correspondence to Peter D. Brukner, MBBS, Olympic Park Sports Medicine Centre, Swan St, Melbourne, 3004, Victoria, Australia; e-mail correspondence to



The McGraw-Hill Companies Gradient

Copyright (C) 1998. The McGraw-Hill Companies. All Rights Reserved