Identification and Management of 2 Femoral Shaft Stress Injuries

Identification and Management of 2 Femoral Shaft Stress Injuries Marc D. Weishaar, PT, DSc, SCS 1 Danny J. McMillian, PT, DSc, OCS 2 Josef H. Moore, PT, PhD, SCS, ATC 3 Study Design: Resident s case problem. Background: Although femoral shaft stress fractures in the general population are rare, they are more common among endurance athletes and military recruits. Such individuals presenting with a complaint of hip, thigh, or knee pain should raise suspicion for femoral shaft stress injury. A United States Military Academy cadet presented to West Point s Physical Therapy-Sports Medicine clinic with a complaint of thigh pain related to training with the local marathon team. A second cadet presented to the same clinic during Cadet Basic Training with a complaint of vague but increasing hip, thigh, and knee pain. Diagnosis: Both cadets were suspected of having femoral stress injuries, based on clinical exams, and both diagnoses were confirmed with diagnostic imaging. The 2 cadets were both treated conservatively with progressive rehabilitation once healing was confirmed with radiographs. They both responded favorably to conservative management and returned to full athletic activity at approximately 12 weeks. Discussion: Symptoms from a femoral shaft stress fracture can be vague and mimic those of other etiologies. Providers should consider a broad differential diagnosis, to include femoral shaft stress fracture, when treating endurance athletes and military recruits with anterior hip, thigh, or knee pain. Proper imaging confirms the diagnosis and sequential radiographs assist in rehabilitation planning. J Orthop Sports Phys Ther 2005;665-673. Key Words: differential diagnosis, femur, overuse injury, stress fracture Stress-related bone and soft tissue injuries occur in response to an excessive progression of repetitive loading activities such as running and marching. When bone is subjected to repetitive compressive or tensile loads, remodeling occurs as a normal physiologic response. Stress injuries and stress fractures develop when the extent of the microdamage exceeds that of the remodeling process. The incidence of stress fractures is as high as 21% among track and field athletes, 2 and as high as 31% among military recruits, 10,11 demonstrating the risk for osseous injuries among individuals performing high levels of endurance training. Although the tibia is most susceptible to these types of injuries, the femur accounts for 1 Officer in Charge, Cadet Physical Therapy Clinic, USAF Academy, CO. 2 Chief of Physical Therapy, Mannheim Health Clinic, Mannheim, Germany. 3 Associate Professor and Director, US Military-Baylor University Post-Professional Sports Medicine- Physical Therapy Doctoral Program, West Point, NY. At the time these cases were managed Mark Weishaar and Danny McMillian were residents in the US Military-Baylor University Post-Professional Sports Medicine-Physical Therapy Doctoral Program, West Point, NY. The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the Department of the Air Force, or the Department of Defense. Address correspondence to Marc Weishaar, Cadet Physical Therapy Clinic, 4102 Pinion Drive, Suite 100, USAF Academy, CO 80840. E-mail: marc.weishaar@usafa.af.mil. approximately 5% to 7% of all stress fractures. 9,12 While the femoral neck has a higher incidence of stress-related injuries, the femoral shaft is also at risk of developing a stress fracture. The incidence of femoral shaft stress fractures is not well documented. One study by Orava 14 reported a 3.5% incidence of femoral shaft involvement among 200 stress fractures, while Johnson et al 8 reported a 20% incidence of femoral shaft involvement among 34 stress fractures, both in athletic populations. Femoral shaft stress fractures in the general population are rare, but they are not uncommon among distance runners and military recruits. Such individuals presenting with a complaint of hip, thigh, or knee pain should raise suspicion for a femoral stress injury. Although the location of a stress injury may not always correlate with the area of pain, femoral shaft stress injuries will typically cause vague pain in the thigh and may radiate to the hip or knee. Hip range of motion will typically not be limited, but may be painful. This is in contrast to femoral neck stress fractures, which often involve loss of hip motion due to muscle guarding. 11 Approximately 20% of patients with a femoral shaft stress fracture will demonstrate an antalgic gait and 70% will have pain hopping on the affected limb. 5 The fulcrum test has been Journal of Orthopaedic & Sports Physical Therapy 665

shown to have high clinical correlation in assessing for femoral shaft stress injuries. 8 To perform the fulcrum test, the examiner s arm is placed under the sitting patient s thigh (femoral shaft) and a downward pressure is applied over the anterior aspect of the distal end of the femur. An exacerbation of sharp thigh pain and apprehension is considered a positive finding and should raise suspicion for a femoral shaft stress injury (Figure 1). 8 The sensitivity, specificity, and predictive values for the fulcrum test have not been established. Considering that most femoral shaft stress injuries occur in the proximal third of the bone, 7 the practitioner performing the fulcrum test should ensure that the primary stress of the test (directly above the inferior arm) is directed appropriately. Because stress fractures occur as a result of repeated loading, a primary risk factor is a sudden increase in training volume, especially for relatively inactive individuals. Other risk factors include menstrual disturbances, caloric restriction, lower bone density, muscle weakness, and leg length differences. 3 A thorough history identifies risk factors for developing a stress injury, and should include determination of the approximate weekly running mileage. Clement et al 5 reported minimum weekly mileages for runners who developed femoral stress injuries. The possibility FIGURE 1. The fulcrum test as described in Johnson et al. 8 The examiner s arm is placed under the sitting patient s thigh (femoral shaft) and a downward pressure is applied over the anterior aspect of the distal end of the femur. An exacerbation of sharp thigh pain and apprehension is considered a positive finding and should raise suspicion for a femoral shaft stress fracture. of such injury should be considered when assessing recreational runners running over 56 km (35 miles) per week, marathon runners running over 72 km (45 miles) per week, and competitive runners running over 112 km (70 miles) per week. Other concerns are restrictive diets, history of previous stress injuries, use of corticosteroids (which alter bone metabolism), and a history of amenorrhea or oligomenorrhea. According to Bennell et al, 3 female athletes who develop stress fractures have significantly less total body bone mineral content, lower lumbar spine and foot bone density, less lean mass in the lower limbs, a later age of menarche, fewer menses per year since menarche, a lower-fat diet, and a leg length discrepancy of more than 0.5 cm as compared to female athletes without stress fractures. 3 Early detection of a femoral shaft stress injury may prevent the injury from progressing to a complete, displaced fracture requiring surgical intervention. The following 2 case reports depict a pattern of physical activity and presenting symptoms that should be recognized when assessing endurance athletes and military recruits with hip, thigh, or knee pain. DIAGNOSIS (CASE 1) Examination/Evaluation The first patient was a 19-year-old male military cadet on an intercollegiate club marathon team at the United States Military Academy, who came to our physical therapy-sports medicine clinic with a complaint of diffuse right thigh pain. He was in good general health, with no other reported medical problems. His mild thigh pain had been present for 9 days and occurred only while running. During the initial evaluation, he presented with a normal gait and normal lower extremity range of motion. His lower extremity strength was rated as normal (5/5) bilaterally, but he did have mild and vague right thigh discomfort with resisted right knee extension. There was no swelling present throughout the right leg. There was a mild but notable tenderness of the right medial thigh. The fulcrum test (Figure 1) was negative. The initial diagnosis was a quadriceps strain based on reproduction of the patient s symptoms with quadriceps muscle strength testing, normal hip range of motion, and a negative fulcrum test, all suggesting no femoral stress injury. He was given a light stretching program for the quadriceps, hip flexors, and hamstrings, and instructed to reduce his training to allow for pain-free activity. The cadet was instructed to hold each stretch for 30 seconds and repeat each stretch 3 times in a standing position 3 times a day. The primary reason to decrease his activity level was to allow for tissue healing. The stretching program was used to optimize tissue mobility while healing. The cadet was also instructed to follow up in 1 week 666 J Orthop Sports Phys Ther Volume 35 Number 10 October 2005

FIGURE 2. Initial radiographs of the patient in the first case report with a subtle periosteal reaction found along the medial midshaft of the right femur consistent with a stress reaction. to be reassessed and to progress to a lower body strength program if his symptoms had improved. The cadet failed to follow-up in 1 week, but did return to the clinic 2.5 weeks later secondary to an increase in right thigh pain after completing a marathon. At that time, we discussed his running program over the previous 3 months. He had been running approximately 64 km (40 miles) per week at the time of his symptoms. He had been previously running 96 km (60 miles) per week 2 months prior, but had tapered off in preparation for the marathon. He reported his diet as balanced with no restrictions. His follow-up presentation revealed a mildly antalgic gait. The fulcrum test of his right thigh was positive for pain in the mid thigh. He had no pain with deep squat, but single-leg hopping did cause moderate thigh pain. Although there was no acute traumatic injury, ligament and menisci tests were performed to help determine if there was any chronic internal derangement to the knee that may have been contributing to his symptoms. All ligament (Lachman, varus/valgus stress, and posterior drawer) and meniscus (McMurray and Apley) special tests were negative. The cadet was placed on crutches with no weight bearing on his right lower extremity and sent for plain radiographs to rule out a femoral stress reaction or stress fracture. Radiographs did not reveal a lucent fracture line, but did show a subtle periosteal reaction along the medial aspect of the midshaft of the right femur, consistent with a stress reaction (Figure 2). Because magnetic resonance imaging (MRI) was readily available in our facility and neither the patient nor the facility incurred additional costs for the use of the imaging study, we ordered an MRI to distinguish between a grade III and a grade IV stress injury. The MRI demonstrated a grade IV stress fracture at the same location of the femoral shaft as the periosteal reaction (Figure 3). Intervention The cadet was instructed to use crutches for 2 weeks, with a recommendation to begin toe-touch weight bearing and progress to weight bearing as tolerated, based on symptoms. The first 2 weeks of rehabilitation consisted of active range of motion and stretching exercises for the hip, knee, and ankle, low-resistance stationary biking and easy freestyle swimming for aerobic conditioning, non weightbearing lower body therapeutic exercises (4-way leg raises), and upper body weight lifting. At 2 weeks after his diagnosis, he demonstrated normal ambulation without crutches and was progressed to light weight-bearing lower body therapeutic exercise (supine bridges) without an increase in pain. Follow-up radiographs at 5 weeks noted smooth periosteal thickening, which corresponded to the location of the stress fracture (Figure 4). Because the radiographs demonstrated evidence of healing, the rehabilitation program for this patient was progressed to light weight-bearing strength training (wall squats with body weight only, standing 4-way leg kicks with no resistance, and standing heel raises), and lowimpact aerobic activity (elliptical trainer). J Orthop Sports Phys Ther Volume 35 Number 10 October 2005 667

hopping, and normal lower extremity strength. He was discharged from physical therapy at 12 weeks. FIGURE 3. Magnetic resonance imaging of the patient in case 1, demonstrating moderate to severe periosteal edema. There is a linear band of low signal within the bone marrow edema consistent with a grade IV stress fracture of the right mid femoral shaft. With no increase in pain after 1 week of the added weight-bearing exercises, a slow progressive walk-torun program starting on a treadmill was added at 6 weeks (Table 1). 13 At 10 weeks, he was allowed to progress to 5-km (3-mile) runs 3 times per week. At 12 weeks, he was progressed to 6- to 13-km (4- to 8-mile) runs 6 times per week, with no reported complaints of thigh pain. At 10 weeks, the cadet had a negative fulcrum test, no pain with single-leg DIAGNOSIS (CASE 2) Examination/Evaluation The second patient was an 18-year-old male cadet who reported to the same physical therapy-sports medicine clinic with complaint of increasing anterior hip, thigh, and knee pain that began 2 weeks earlier during Cadet Basic Training (CBT) at the United States Military Academy. The 6-week CBT is designed to provide a rapid transition to military life. Cadets are physically challenged with daily physical fitness training, long foot marches, mountaineering, rifle marksmanship, and tactical maneuvers. CBT is the first training activity upon arrival at the United States Military Academy and takes place the summer before the freshman year. The cadet described was physically active in water polo in high school and had completed all training during the first 5 weeks of CBT. The initial evaluation occurred approximately 12 hours after completing a 21-km (13-mile) foot march. He reported symptoms of increasing anterior hip, thigh, and knee pain that had begun 2 weeks earlier. He reported no incident of trauma. FIGURE 4. Follow-up radiograph of the patient in case 1 at 5 weeks, with smooth periosteal thickening, which corresponds to the location of the stress fracture and consistent with interval healing. 668 J Orthop Sports Phys Ther Volume 35 Number 10 October 2005

TABLE 1. Walk-to-run program (all phases measured in miles). From Moore JH, Ernst GP. 13 Phase Mileage* 1 Walk 2 miles at your own pace 2 Progress to walking 2 miles in 35 min 3 Walk 1 4, run 1 4, walk 1 4, run 1 4 4 Walk 1 4, run 1 4, walk 1 4, run 1 4, walk 1 4, run 1 4, walk 1 4, run 1 4 5 Walk 1 4, run 1 2, walk 1 4, run 1 2, walk 1 4, run 1 2 6 Walk 1 4, run 3 4, walk 1 4, run 3 4 7 Walk 1 4, run 1, walk 1 4, run 1 8 Walk 1 4, run 1, walk 1 4, run 1, walk 1 4, run 1 General Guidelines 1. Use brand name running shoes and not court or cross trainers. 2. Begin at an easy pace on level surfaces, no hills until at least 3 to 5 weeks after phase 8. 3. Stop if increased pain, swelling, or stiffness is noted, especially while running and if symptoms are present by the next morning. Do not resume running until cleared by provider. 4. Do not run more than 3 times per week and do not run daily until 3 to 5 weeks after phase 8. 5. Try each phase at least twice. Progress to the next phase if no increase in pain, swelling, or stiffness. 6. After phase 8 gradually begin to increase running without walking. 7. All increments for walk-to-run progression are based on miles. 8. Following phase 8, do not increase distance or pace more than 10% per week. * 1 4 mile =.40 km, 3 4 = 1.21 km, 1 mile = 1.61 km. The cadet arrived at the physical therapy-sports medicine clinic demonstrating a mildly antalgic gait. He was able to fully squat and shift weight side-to-side without symptoms. Single-leg stance on the right was mildly symptomatic, with a negative Trendelenburg sign. The ankle, knee, and hip active and passive ranges of motion were bilaterally equal and within normal limits. There were no signs of effusion throughout the entire lower extremity. Knee stability was normal and there was no tenderness to palpation. The fulcrum test on the right was positive for moderate thigh pain. The cadet also reported mild pain with firm, steady manual pressure applied to the proximal right thigh over the shaft of the femur. Soft tissue palpation of the right proximal thigh and groin was unremarkable. The first imaging studies ordered were radiographs of the right hip and pelvis. These views were ordered to provide imaging of 2 common sites of stress fractures in new soldiers the pubic ramus and the femoral neck. There was no evidence of a pubic ramus or femoral neck fracture; however, the lower margin of the anterior-posterior view of the pelvis showed a periosteal reaction of the medial aspect of the femur (Figure 5). To better view this area, frontal FIGURE 5. Initial radiographs of the patient in case 2, demonstrating a periosteal reaction of the medial aspect of the right femur. J Orthop Sports Phys Ther Volume 35 Number 10 October 2005 669

tion of the proximal femoral diaphysis. The bony trabecular pattern within the medullary cavity at the same level was somewhat indistinct. No lucent fracture line through the femur was seen. The same day, MRI showed marked periosteal and marrow edema of the distal portion of the proximal third of the right femur. No distinct fracture line was seen. The diagnosis was a grade III stress injury of the distal portion of the proximal one third of the right femur (Figure 6). FIGURE 6. Magnetic resonance imaging of the patient in case 2, demonstrating marked periosteal and marrow edema of the distal portion of the proximal third of the right femur. No distinct fracture line seen. The radiographic diagnosis was a grade III stress injury of the distal portion of the proximal one third of the right femur. and lateral views of the proximal to mid right femur were ordered. These views demonstrated abnormal cortical lucency with overlying fluffy periosteal reac- Intervention After the initial clinical and imaging examinations, the cadet was excused from the last week of field training exercises and instructed on touch weightbearing ambulation using crutches. He was also instructed in range of motion exercises for the ankle, knee, and hip. He remained at the field site during this time, participating only in tactical classes. Upon return from the field training exercise, the cadet began daily physical therapy. The first week of physical therapy consisted of stationary biking with minimal resistance, swimming using a flutter kick only, and seated upper body resistance training. Anterior and posterior stretches of the hip and thigh were also initiated at this time. Two weeks after the initial visit, he tolerated supine and side bridging exercises and performed these to fatigue without pain of the right hip or thigh. Radiographs were repeated at 2 weeks, with little change noted from the initial films. Therapeutic management did not change. Radiographs were repeated 6 weeks after the initial visit. At that time, sclerosis and coalescence of the periosteal reaction was noted, consistent with interval healing. An oblique lucency was seen traversing the periosteal reaction, but was considered less prominent than that observed on the films taken 4 weeks previously (Figure 7). The cadet was allowed to walk without crutches after review of the radiographic report at 6 weeks. Because cadets walk up to 0.4 km (0.25 miles) between classes, a formal walking program was delayed until he reported no symptoms with day-to-day activities. At this time, he also began body weight squats (alternating between free stance and the use of an exercise ball for wall squats) and heel raises. Two weeks later, he reported that neither walking around campus nor initiation of body weight resistance exercises had caused hip or thigh pain. At that time, he was instructed on the progression of a systematic walk-to-run program that was completed without setbacks over the next 4 weeks (Table 1). 13 The final radiographs were taken at 10 weeks after the initial visit, showing increased sclerosis consistent with further interval healing. At 10 weeks, the cadet had a negative fulcrum test, no pain with single-leg hopping, and normal lower extremity strength. He was discharged from physical therapy at 10 weeks and 670 J Orthop Sports Phys Ther Volume 35 Number 10 October 2005

completed all cadet physical requirements without further symptoms the following semester. DISCUSSION These 2 cases demonstrate the typically vague symptoms with which endurance athletes might present when suffering from a femoral shaft stress injury or stress fracture. Clinicians must realize that the location of a femoral stress injury does not always correlate with the area of pain. Such injuries will typically cause vague pain in the thigh and may radiate either proximally into the anterior hip or distally to the knee. Hip range of motion will usually not be limited, but might be painful. Full range of motion of the hip with femoral shaft stress injury is in contrast to femoral neck stress injury, in which loss of hip motion is common, especially that of internal rotation. 4 The clinical implication is that individuals presenting with symptoms of femoral stress injury who have full passive hip range of motion should receive imaging studies of the femoral shaft, not just the standard hip views. While the patient s location of pain may be vague, the history of a recent increase in repetitive loading activities is a clear feature of femoral shaft stress fractures in athletes. Both cases described here fit this pattern. The cadet in the first case report had been running up to 96 km (60 miles) per week prior to FIGURE 7. Follow-up radiograph of the patient in case 2 at 6 weeks, showing sclerosis and coalescence of the periosteal reaction consistent with interval healing. developing his symptoms. According to Clement et al, 5 marathon runners with a minimum weekly distance of 72 km (45 miles) are at an increased risk for developing femoral stress injuries. The cadet described in the second case report was very active before entering military service; however, his primary sport was water polo, a non weight-bearing activity. He reported little weight-bearing exercise before beginning the rigorous marching and running requirements of the military academy. Thus, when obtaining the patient s activity history, it might be useful to distinguish between weight-bearing and non weight-bearing activities. When there is clinical suspicion for a femoral stress fracture, imaging studies should begin with plain radiographs, although they will often be normal. Conventional radiographs, when positive, demonstrate findings of periosteal bone formation (callus) with sclerosis. These signs may be evident as early as 3 weeks or as late as 3 months after the initial symptoms of pain are present. 12 Clement et al, 5 in a study of 71 athletes, found plain radiographs to be diagnostic of a femoral stress fracture only 24% of the time. In the absence of positive plain radiographs and where clinical suspicion is still present for a stress fracture, bone scintigraphy or MRI is recommended. Both are effective in evaluating stress injuries of the femur. Although a bone scan is more cost effective and can evaluate a larger area, patients are exposed to radiation. MRI has comparable sensitivity and superior specificity compared to the bone scan in detecting osseous abnormalities. 6,15 Shin et al 15 prospectively evaluated the accuracy of MRI in 19 military subjects engaged in rigorous endurance training. All subjects had hip pain, negative radiographs, and bone scans consistent with femoral neck stress fractures (2 subjects had bilateral hip pain and positive bone scans). This study included 22 hips because 1 subject had unilateral hip pain, but a positive bilateral hip bone scan. Subjects underwent MRI and 6-week follow-up plain radiographs. Follow-up radiographs at 6-weeks, which demonstrated healing callous, were considered pathognomonic for healing femoral neck stress fractures. MRI revealed 15 (68%) hips with femoral neck stress fractures all confirmed with positive 6-week radiographs and 7 (32%) hips with other diagnoses all with negative 6-week radiographs. This study demonstrated 100% sensitivity and specificity with MRI. It also demonstrated 100% positive predictive value for MRI, which is the percentage of patients with positive MRI who also had radiographic evidence of a stress fracture at follow-up. The positive predictive value for the bone scan in this study was 68%. Although this study did demonstrate diagnostic benefits of MRI over bone scan, the sample size must be considered a limiting factor. Several radiological grad- J Orthop Sports Phys Ther Volume 35 Number 10 October 2005 671

TABLE 2. Grading of stress fractures on the basis of radiological findings. From Arendt and Griffiths. 1 Grade Radiographic Finding Bone Scan Finding MRI Finding 1 Normal Poorly defined area of increased activity Increased activity on short inversion time inversion recovery (STIR) image 2 Normal More intense, but still poorly defined Positive STIR and T2-weighted images 3 Discrete line; discrete periosteal reaction Sharply marginated area of increased activity focal or fusiform 4 Fracture or periosteal reaction More intense transcortical localized uptake Positive T1- and T2-weighted images; no focal cortical break Fracture line on T1- and T2-weighted images ing systems for stress fractures have been proposed. One grading system used to diagnose patients with stress fractures based on radiological findings can be found in Table 2. 1 In the 2 case reports presented here, the decision to add MRI in light of positive signs of stress-related injury on radiographs was based on circumstances unique to the medical treatment facility and not because the images were necessary to make treatment decisions for these 2 patients. At the time of these cases, the West Point medical facility had MRI capability, but not bone scintigraphy. Use of MRI at this facility created no additional cost for either the patients or the facility and was readily available in both cases. In addition, the facility also has a teaching mission for orthopedic surgeons and physical therapists. Though the radiographic findings together with the history and physical examination could have guided treatment alone, adding the MRI allowed residents the chance to correlate MRI findings with the signs and symptoms of their patients. Practitioners in other settings should consider the cost of the MRI with any additional benefit of its use. Though rehabilitation for a femoral stress injury is relatively straightforward, with pain being the primary indicator for progressing and limiting activity, the individual s response to activity progression should be closely monitored. Crutches, with toe touch or partial weight bearing, are indicated when gait is antalgic. Full weight bearing can begin as soon as pain subsides, usually within 1 to 6 weeks. It is important to educate patients on the need for proper rest and nutrition to maximize their healing process. Gentle hip, knee, and ankle range of motion can begin immediately to prevent soft tissue restrictions from developing during the time of limited weight bearing. Nonimpact aerobic conditioning may begin as well, such as swimming, low-resistance stationary biking, and upper body ergometry. Upper body resistance training may be initiated early in rehabilitation, as long as the stress-related symptoms are not provoked. Progressive, low-impact aerobic activity, such as walking, pool running, the elliptical trainer, and stationary biking with increased resistance may begin at approximately 6 to 10 weeks, as long as there is no leg pain and signs of callous formation are present on plain radiographs. Return to running or sports is dependent upon a negative clinical exam (negative fulcrum, no pain with single-leg hopping, and normal strength) and the absence of pain during impact activities. The 2 cadets presented here had negative fulcrum tests, no pain with single-leg hopping, and normal lower extremity strength at 10 weeks. Most athletes are able to return to full activity in 8 to 14 weeks. 7 Delayed union or nonunion of femoral shaft stress fractures is rare following conservative therapy. 4 Physical therapists working in direct-access roles must be skilled in differential diagnosis. While the tibia, femoral neck, and pubic ramus are more common sites for stress fractures, stress injuries to the proximal femoral shaft do occur, primarily in distance runners and military recruits. These 2 case reports emphasize the importance of including the femoral shaft in the clinical differential diagnosis and imaging work-up. REFERENCES 1. Arendt EA, Griffiths HJ. The use of MR imaging in the assessment and clinical management of stress reactions of bone in high-performance athletes. Clin Sports Med. 1997;16:291-306. 2. Bennell KL, Malcolm SA, Thomas SA, et al. Risk factors for stress fractures in track and field athletes. A twelvemonth prospective study. Am J Sports Med. 1996;24:810-818. 2. Bennell KL, Matheson G, Meeuwisse W, Brukner P. Risk factors for stress fractures. Sports Med. 1999;28:91-122. 4. Boden BP, Speer KP. Femoral stress fractures. Clin Sports Med. 1997;16:307-317. 5. Clement DB, Ammann W, Taunton JE, et al. Exerciseinduced stress injuries to the femur. Int J Sports Med. 1993;14:347-352. 6. Deutsch AL, Coel MN, Mink JH. Imaging of stress injuries to bone. Radiography, scintigraphy, and MR imaging. Clin Sports Med. 1997;16:275-290. 672 J Orthop Sports Phys Ther Volume 35 Number 10 October 2005

7. Hershman EB, Lombardo J, Bergfeld JA. Femoral shaft stress fractures in athletes. Clin Sports Med. 1990;9:111-119. 8. Johnson AW, Weiss CB, Jr., Wheeler DL. Stress fractures of the femoral shaft in athletes more common than expected. A new clinical test. Am J Sports Med. 1994;22:248-256. 9. Matheson GO, Clement DB, McKenzie DC, Taunton JE, Lloyd-Smith DR, MacIntyre JG. Stress fractures in athletes. A study of 320 cases. Am J Sports Med. 1987;15:46-58. 10. Milgrom C, Finestone A, Shlamkovitch N, et al. Youth is a risk factor for stress fracture. A study of 783 infantry recruits. J Bone Joint Surg Br. 1994;76:20-22. 11. Milgrom C, Giladi M, Stein M, et al. Stress fractures in military recruits. A prospective study showing an unusually high incidence. J Bone Joint Surg Br. 1985;67:732-735. 12. Monteleone GP, Jr. Stress fractures in the athlete. Orthop Clin North Am. 1995;26:423-432. 13. Moore JH, Ernst GP. Therapeutic exercise. In: O Connor FG, Wilder RP, Nirschl R, eds. Textbook of Running Medicine. New York, NY: McGraw-Hill; 2001. 14. Orava S. Stress fractures. Br J Sports Med. 1980;14:40-44. 15. Shin AY, Morin WD, Gorman JD, Jones SB, Lapinsky AS. The superiority of magnetic resonance imaging in differentiating the cause of hip pain in endurance athletes. Am J Sports Med. 1996;24:168-176. J Orthop Sports Phys Ther Volume 35 Number 10 October 2005 673