| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Musculoskeletal Imaging |
1 From the Departments of Radiological Sciences (M.G., F.M., E.S., G.A., S.V., A.B.) and Sport Medicine (D.B., L.M.), University of Messina, Policlinico "G. Martino," Via Consolare Valeria, 98100 Messina, Italy. Received March 1, 2004; revision requested May 11; final revision received August 16; accepted September 8. Address correspondence to F.M. (e-mail: fminutoli@unime.it).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Medical ethics committee approval and informed consent were obtained. A total of 42 patients experiencing tibial pain due to early stress injuries were evaluated. Eight patients had bilateral involvement; thus, 50 tibiae were evaluated. All patients underwent initial radiography that was negative for injury. MR imaging, CT, and bone scintigraphy were performed in all patients within 1 month of onset of symptoms. Ten asymptomatic volunteers served as the control group. Location of stress injuries, types of bone alterations, and presence of periosteal and bone marrow edema were evaluated. Sensitivity, specificity, accuracy, and positive and negative predictive values of MR imaging and CT were assessed, as was sensitivity of bone scintigraphy. McNemar test was used to detect statistically significant differences.
RESULTS: Sensitivity of MR imaging, CT, and bone scintigraphy was 88%, 42%, and 74%, respectively. Specificity, accuracy, and positive and negative predictive values were 100%, 90%, 100%, and 62%, respectively, for MR imaging and 100%, 52%, 100%, and 26%, respectively, for CT. Significant difference in detection of early tibial stress injuries was found between MR imaging and both CT and bone scintigraphy (McNemar test; P < .001 and P = .008, respectively).
CONCLUSION: MR imaging is the single best technique in assessment of patients with suspected tibial stress injuries; in some patients with negative MR imaging findings, CT can depict osteopenia, which is the earliest finding of fatigue cortical bone injury.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Exercise-induced stress reactions and stress fractures involving the tibia are common and may account for up to 75% of exertional leg pain and stress fractures (3–5). Radiologic, radionuclide bone scanning, computed tomographic (CT), and magnetic resonance (MR) imaging features of tibial stress injuries have been described in numerous publications (4–22). These studies usually included patients in whom the symptoms had been present for many weeks or months, however. On the other hand, early detection of tibial stress injuries may be crucial to prevent stress fracture and complications of stress fracture and to lead to early recovery (5,23,24). Thus, the aim of this study was to prospectively compare CT, MR imaging, and bone scintigraphy in athletes with clinically suspected early stress injury of tibia.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
Inclusion was based on fulfillment of all of the following criteria: (a) unilateral or bilateral lower leg pain lasting less than 1 month, (b) no history of trauma, and (c) no evidence of bone abnormalities on conventional radiographic images of the lower leg obtained with orthogonal and oblique views.
A total of 21 patients were excluded from analysis. In 16 of these 21 patients, a different source of lower leg pain was found: Muscular strain and/or tendinopathy was found in seven patients, chronic compartment syndrome was found in three, internal derangement of knee or ankle was found in four, focal myositis (inflammatory pseudotumor) was found in one, and entrapment of the tibial nerve was found in one. Two additional patients experienced claustrophobia and were not able to undergo MR imaging, and two patients refused to take part in the study. In one patient, pregnancy was considered a criterion for exclusion.
Thus, the study group consisted of 42 patients (mean age, 28.2 years; age range, 16–37 years). The group included 16 female patients (age range, 19–37 years; mean age ± standard deviation, 27.8 years ± 5.4; median age, 28.5 years), and 26 male patients (age range, 16–37 years; mean age ± standard deviation, 28.4 years ± 5.9; median age, 29 years). There were no statistically significant differences between the distributions by sex. All patients were involved in competitive or recreational activities: 20 patients participated in running, 13 participated in basketball, seven participated in soccer, one participated in volleyball, and one participated in handball. All but three patients described a recent change in the intensity, volume, or equipment used in their training regimen. Since eight patients experienced bilateral pain, a total of 50 tibiae were evaluated. In addition, 10 asymptomatic volunteers who were involved in sports underwent both CT and MR imaging of one tibia; these volunteers served as healthy controls. The control group consisted of three women and seven men (mean age, 29.6 years; age range, 25–38 years). Five volunteers were involved in soccer, four were involved in running, and one was involved in basketball.
The sample size was appropriate for a statistical power of more than 95% (25).
The protocol was approved by the medical ethics committee of our hospital both for patients and for volunteers. Informed consent was obtained from each subject or from a parent of patients younger than 18 years after the nature of the examinations had been fully explained.
Imaging Technique and Initial Interpretation
All imaging examinations were performed within 1 month of the onset of symptoms (range, 5–28 days; mean, 18 days). Each patient was examined with MR imaging, CT, and bone scintigraphy, respectively. Time interval between examinations was 0–3 days.
All MR imaging examinations were performed with a 1.5-T system (Magnetom Vision; Siemens, Erlangen, Germany) with a phased-array coil.
The MR imaging protocol consisted of the following pulse sequences: (a) a transverse T1-weighted fast spin-echo sequence (repetition time msec/echo time msec, 680/12; echo-train length, 3; section thickness, 3 mm; number of acquisitions, six; image matrix, 210 x 512; acquisition time, 4 minutes 49 seconds), (b) a transverse T2-weighted fast spin-echo sequence (5400/99; echo-train length, 11; section thickness, 3 mm; number of acquisitions, three; image matrix, 220 x 512; acquisition time, 5 minutes 29 seconds), and (c) a transverse fast short inversion time inversion-recovery (STIR) sequence (repetition time msec/echo time msec/inversion time msec, 3600/60/150; echo-train length, 11; section thickness, 3 mm; number of acquisitions, four or six; image matrix, 242 x 256; acquisition time, 2 minutes 56 seconds or 4 minutes 24 seconds).
Transverse MR images were obtained for two reasons. First, they were easier to compare with transverse CT scans. Second, the transverse plane has been demonstrated to be the best in the detection of stress injuries in the tibial shaft (21). On the basis of findings seen on transverse MR images, additional coronal and/or sagittal MR images were obtained in 14 patients to better define craniocaudal extension of the lesion.
All patients and volunteers underwent CT scanning. CT was performed in 15 patients and in four volunteers with a Somatom Plus 4 scanner (Siemens) with use of the following parameters: 2-mm collimation, 15-mm table increment, 120 kVp, and 120 mAs. The other 27 patients and six volunteers underwent CT examination with a Somatom Sensation 16 scanner (Siemens), which became available in our institution in January 2003. Scanning parameters were as follows: 2-mm collimation, 15-mm table increment, 120 kVp, and 120 mAs.
All patients underwent radionuclide bone scanning 3–4 hours after injection of 20 mCi (740 MBq) of technetium-99m methylene diphosphonate. Imaging was performed by using the planar mode on a dual-head gamma camera. Volunteers in the control group did not undergo bone scanning.
All images were stored digitally. At the conclusion of the examinations, the image sets obtained for each patient were displayed on a computer monitor and evaluated by the observers. Each image set was reviewed at the time of acquisition by a senior radiologist or a senior nuclear medicine physician with more than 10 years of experience (G.A. or S.V.). They inspected the images to ensure that they were of diagnostic quality, to make an initial reading, and to give a preliminary interpretation of the findings. Additional coronal and/or sagittal MR images were thought to be useful and were obtained during the initial reading while the patient was still in the imaging room or department. The initial reading (step 1) was performed with knowledge of all available clinical and imaging findings to make a formal report and to optimize patient care.
CT and MR Imaging Analysis
In January 2004, all CT and MR imaging examinations of the 42 patients and 10 volunteers were retrieved and evaluated in consensus by two musculoskeletal radiologists (F.M. and A.B., with 8 and 12 years of experience, respectively) (step 2). These radiologists had not been involved in the execution of the examinations. CT images were displayed with two different window settings to assess bone structures and soft tissues.
To obviate recall bias, CT scans and MR images were analyzed separately in random order. For this reading, the readers were aware that the patients were being evaluated for a possible tibial lesion, but they did not know which images were obtained in patients with a clinical stress injury and which images were obtained in volunteers.
The readers were asked to subjectively evaluate the presence of fracture line, cortical abnormalities, bone marrow edema, and periosteal edema. If cortical abnormality was present, the readers were asked to categorize it as osteopenia, resorption cavity, or striation. These descriptive terms were drawn from conventional radiology (2,12,20,26). Interpretation criteria that were used to assess the presence of lesions on CT scans and MR images are summarized in Table 1. Namely, the presence of cortical osteopenia on CT scans was evaluated as follows: First, the images were analyzed visually and graded in consensus on a scale of 1–4 (grade 1 = definitely absent, grade 2 = probably absent, grade 3 = probably present, and grade 4 = definitely present) for the visual impression of any difference in density of the cortex. Grades 3 and 4 were considered to represent osteopenia. At the same time, attenuation measurements were obtained. A region of interest was placed within the zone of reduced bone attenuation (grades 3 and 4) and in the nearest part of the bone cortex that appeared to be healthy. The area of the region of interest within the bone cortex varied between 0.05 and 0.1 cm2. The size of the region of interest within the area of reduced bone attenuation and the size of the region of interest within the normal bone cortex were identical. Observers found that reduction of attenuation of 10% or more included all the cortical areas with visual grades of 3 or 4; therefore, it was considered to represent osteopenia.
|
|
The readers also reported the number of lesions detected with long-repetition-time sequences (ie, T2-weighted fast spin-echo and fast STIR) versus the number of lesions detected with T1-weighted fast spin-echo sequences.
Bone Scintigraphy Analysis
In January 2004, scintigraphic studies were evaluated by two nuclear medicine physicians with 7 and 9 years of experience (step 2) who were not involved in the performance of the examinations. A consensus opinion was reached in all cases of disagreement between the two readers. For this reading, the readers were aware that the patients were being evaluated for a possible tibial lesion.
Findings of scintigraphy were defined as positive if foci of increased uptake were present in areas unrelated to the normal biodistribution of the radiopharmaceutical.
Standard of Reference
The standard of reference (27,28) consisted of results of a review of clinical findings, physical examination, and detailed history by three experienced sports medicine physicians. Other causes of chronic leg pain were excluded. In all patients, pain was induced by exercise. In some patients, palpation of the tibia produced discomfort. All but three patients indicated a recent increase of activity or a recent change in equipment, especially footwear. There was no history of plantar paresthesia or other symptoms, such as muscle cramping and swelling, which are indicative of other causes of exercise-induced leg pain. All patients underwent an ultrasonography examination to allow evaluation of muscles and tendons. Color Doppler flow imaging was selectively performed to rule out the presence of vascular disease. Electromyography with nerve conduction was selectively used in the evaluation of patients in whom radicular nerve conditions were suspected. In 22 patients, measurement of compartment pressure was performed and permitted to exclude the presence of chronic compartment syndrome, which is the most common cause of exercise-induced leg pain other than tibial stress injury (29).
The results of CT, MR imaging, and bone scintigraphy were not included to avoid bias (30).
Data Analysis and Statistical Evaluation
A radiologist (M.G.) who was not involved in the execution of examinations or the evaluation of images analyzed collected data (step 3). The analysis was performed on the data obtained in the step 2 reading.
All statistical processing was performed with SPSS software, version 11.5 (SPSS, Chicago, Ill).
The sensitivity, specificity, accuracy, and positive and negative predictive values of MR imaging and CT were assessed. With regard to bone scintigraphy, only sensitivity could be calculated, since healthy patients were not examined.
The McNemar test for paired proportions was used to compare the disease detection rates of the three modalities. A P value of less than .05 was considered to indicate a statistically significant difference with a 95% confidence interval.
Clinical Follow-Up
Clinical follow-up was performed for each patient by a sports medicine physician with more than 10 years of experience until disappearance of symptoms.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
|
|
The difference in detection rate between MR imaging and CT and between MR imaging and scintigraphy was statistically significant (McNemar test, P < .001 and P = .008, respectively). A statistically significant difference was also found between scintigraphy and CT (McNemar test, P = .02).
Location
A total of 27 lesions were located in the diaphysis, 12 were located in the proximal tibia, and eight were located in the distal tibia. Diaphyseal lesions consisted of 21 cortical injuries (seven cases associated with periosteal edema, marrow edema, or both) and six cases of periosteal edema.
Cortical abnormalities involved the anterior cortex in 10 tibiae, posterior cortex in three tibiae, and both posterior and anterior cortices in eight tibiae.
Fracture
Two fractures were detected with all MR imaging sequences. In one patient, a fracture in the proximal metaphysis developed very quickly (Fig 2). Neither of these fractures were visible on CT scans, but scintigraphic findings were positive in both cases.
|
|
|
|
|
|
|
|
|
Bone Marrow Edema
MR imaging demonstrated bone marrow edema in 20 tibiae within cancellous bone of proximal or distal tibia (Fig 6) and in five tibiae within central medullar cavity of tibial shaft (Figs 3b, 7). Although edema was always visible on T1-weighted and T2-weighted images, fast STIR images demonstrated marrow edema with more confidence (Fig 7). CT revealed a marrow abnormality in only two tibiae.
|
|
|
|
|
|
In all patients, tibial pain disappeared with rest after 7–25 days.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The incidence of bone stress injuries is increasing among the competitive and recreational athletic population (34). It is likely that stress fractures are underreported (5,35). Highly motivated athletes with tibial pain can refuse to interrupt training in the presence of pain. A strong resistance against interruption of sports activity is common, especially in professional athletes whose income depends on achievement of a good performance. In these circumstances, the imaging demonstration of early stress lesions is a strong argument to persuade athletes to discontinue training. The importance of early diagnosis of a stress lesion has been recently highlighted by Ohta-Fukushima et al (23). These authors have found a statistically significant difference in the return time to the sport between athletes younger than 20 years in whom a stress lesion was diagnosed within 3 weeks of onset and those in whom a stress lesion was diagnosed more than 3 weeks after the onset of symptoms (return time, 10.4 vs 18.4 weeks).
The precise pathogenesis of stress fracture is poorly understood. Stresses related to daily activities stimulate the remodeling process. Increased osteoclastic resorption is the initial response to abnormal stress. If increased stress persists, imbalance between bone resorption and bone replacement leads to weakening of the bone. Weight bearing, muscle actions, and muscle fatigue may play a role in increasing stress on bone. In the tibia, tensile forces are produced along its anterior convex side, while compressive forces occur along its posterior concave margin (2). Accelerated intracortical remodeling causes microscopic cracks, osteopenia, and formation of resorption cavities that may join in larger lesions. Stresses in cancellous bone may initially result in microfractures. If the inciting activity is not decreased, the accumulation of microdamages may result in stress fracture of cortical or trabecular bone (8).
Many authors have described CT and MR imaging findings of stress fractures of the tibia. To our knowledge, however, no previous study in which CT and MR imaging findings were evaluated in patients with early stress injuries exists in the literature (2,5–22).
We have seen that stress-related lesions of the tibial cortex are common. Both CT and MR imaging are able to demonstrate them. Imaging findings correlate well with the spectrum of physiopathologic bone response to stress. We believe that osteopenia, resorption cavity, and striation occur as early lesions preceding the formation of cortical fracture. These prefracture lesions have not been described previously, probably because evaluation of patients with tibial pain with CT and MR imaging is usually delayed for many weeks or months.
In 1984, Daffner (20) described the presence of faint radiolucent striations in the anterior tibial cortex as a sign of stress injury. In 1995, Mulligan (12) reported subtle ill definition of the cortex, the so-called gray cortical sign, as an early sign of stress fracture. These radiographic signs probably correlate with late-stage cortical prefracture abnormalities. It is well known that CT is much more sensitive than radiography in the appreciation of bone loss. Thus, it is not surprising that early cortical stress injuries can be detected more precociously and easily with CT than with radiography. Our data confirm that conventional radiographs are insensitive in the detection of early-stage stress injuries (2,16,18,36).
With regard to cortical injuries, both CT and MR imaging were used to depict more lesions than did bone scintigraphy. However, CT was used to depict more lesions than did MR imaging (Table 4). MR imaging allows depiction of periosteal and endosteal marrow edema, which are believed to be useful ancillary markers for stress injury (10,19,37). Both T2-weighted images and STIR images are reliable in revealing the presence of cortical stress injury. On the other hand, as previously reported by Ahovuo et al (21), we have seen that T1-weighted images are of lesser value.
We want to highlight that another interesting finding is the presence of abnormalities that simultaneously affect the anterior and posterior cortex. The phenomenon can be explained by synchronous action of compressive and tensile forces applied to the tibia. This finding may be useful in the differentiation of cortical stress injury from more aggressive bone disease, such as neoplasms and infections (2,18).
Cancellous bone stress injuries result in nonspecific bone marrow alterations similar to those described in patients with bone marrow edema syndrome or bone bruise (2,38). These injuries appear on T2-weighted or STIR MR images as increased marrow signal intensity. The abnormalities correlate with extensive microfractures that cause edema and hemorrhage in bone marrow spaces. As with traumatic bone marrow contusion, radiography and CT scanning are not able to demonstrate abnormalities. A cancellous bone stress fracture may develop more rapidly than a cortical stress fracture, since cancellous bone is weaker than cortical bone. We have seen a cancellous bone stress fracture that appeared only 5 days after the onset of tibial pain (Fig 2).
While in the past it was assumed that a negative bone scan ruled out a stress lesion unequivocally and bone scintigraphy was consequently considered to be the reference standard in the evaluation of stress injuries (17,39,40), several cases of bone scintigraphy failure in the identification of stress lesions have been described in the literature (41–45). Thus, MR imaging is now considered more sensitive than bone scintigraphy (22,37,46). In our series, bone scintigraphy failed to depict eight cortical stress injuries seen with MR imaging, CT, or both. The absence of significant osteoblastic response in the presence of early stress reactions, namely osteopenia, may explain the lack of increased uptake of the radionuclide. In addition, no cases of positive bone scintigraphic findings revealing a lesion that was not detected with MR imaging or CT were seen. On the basis of these results, we may confirm that bone scanning may be ineffective in the diagnosis of early cortical stress injuries (15,41). Moreover, bone scintigraphy lacks specificity.
There are some weaknesses and limitations to our study. First, the small number of healthy control subjects who underwent MR imaging and CT and the absence of healthy control subjects who underwent bone scintigraphy introduced a selection bias (28). There are, however, certain ethical considerations about performing scintigraphic examinations in asymptomatic individuals.
In athletes, the presence of bone marrow edema is not pathognomonic of stress-related injuries. By using the fast STIR MR sequence, some authors have demonstrated that increased marrow signal resembling stress lesion may be seen in asymptomatic subjects, both those who are runners (47,48) and those who are volunteers, after 2 weeks of altered weight bearing (49). Namely, Lazzarini et al (47) detected bone marrow edema in three of 40 tibiae by studying 20 asymptomatic runners. In our series, we have not seen marrow edema in asymptomatic volunteers. The small size of our control sample may explain the difference. Actually, we feel that imaging findings could precede symptoms, since we have seen cortical abnormalities in patients with pain of a few days duration (Fig 3). It seems improbable that cortical osteopenia may develop in such a short time. A large series of asymptomatic volunteers should be studied to solve this problem.
Another major weakness was the lack of specimens available for histopathologic analysis, which would be needed to correlate imaging findings with histopathologic analysis findings. This limitation cannot be obviated, since, in the clinical setting of tibial stress injuries, it is not possible to perform bone biopsy. The existence of previous studies on the physiopathology of the stress lesions, however, permits us to hypothesize with good reliability the pathologic abnormalities that may correlate with and explain the imaging findings. For example, the appearance of resorption cavities and striations within the cortex may correlate with osteoclastic proliferation. These abnormalities are similar to those visible in patients with metabolic osteoporosis. High signal intensity depicted on T2-weighted MR images within striations and resorption cavities may be well explained by cell accumulation (osteoclasts, osteoblasts, and fibroblastic cells), increased amount of blood vessels, and production of connective tissue and osteoid matrix, which all occur in osteoporotic bone (8,50). In addition, stress leads to numerous microscopic intracortical microcracks (8) that probably result in hemorrhage and edema.
In conclusion, although evaluation of patients with potential tibial stress injury initially relies on clinical suspicion, imaging plays an essential role in the identification of abnormalities that are worrisome for stress-related disease. Both CT and MR imaging provide early findings of stress injury in patients with activity-related tibial pain. The awareness of these findings is very important, since early diagnosis leads to early recovery. MR imaging should be considered the technique of choice in the assessment of patients with early tibial stress injuries; however, CT is the method of choice in the detection of osteopenia, which is the earliest sign of fatigue damage of the cortical bone.
|
FOOTNOTES |
|---|
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, M.G., E.S.; study concepts, M.G., F.M., E.S., A.B.; study design, M.G.; literature research, M.G., G.A., S.V.; clinical studies, D.B., L.M.; data acquisition, G.A., S.V., D.B., L.M.; data analysis/interpretation, M.G., F.M., D.B., L.H., A.B.; statistical analysis, M.G., F.M., A.B.; manuscript preparation, M.G., F.M., G.A., D.B., A.B.; manuscript definition of intellectual content, M.G., F.M., E.S., S.V., L.M., A.B.; manuscript editing, F.H., G.A., S.V.; manuscript revision/review, M.G., F.M., A.B.; manuscript final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. C. Cabarrus, A. Ambekar, Y. Lu, and T. M. Link MRI and CT of Insufficiency Fractures of the Pelvis and the Proximal Femur Am. J. Roentgenol., October 1, 2008; 191(4): 995 - 1001. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. R. Beck, G. O. Matheson, G. Bergman, T. Norling, M. Fredericson, A. R. Hoffman, and R. Marcus Do Capacitively Coupled Electric Fields Accelerate Tibial Stress Fracture Healing?: A Randomized Controlled Trial Am. J. Sports Med., March 1, 2008; 36(3): 545 - 553. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Fredericson, W. Moore, and S. Biswal Sacral Stress Fractures: Magnetic Resonance Imaging Not Always Definitive for Early Stage Injuries: A Report of 2 Cases Am. J. Sports Med., May 1, 2007; 35(5): 835 - 839. [Full Text] [PDF]
|
|
|
|
 |
|
 G. S. Stacy and L. B. Dixon Pitfalls in MR Image Interpretation Prompting Referrals to an Orthopedic Oncology Clinic RadioGraphics, May 1, 2007; 27(3): 805 - 826. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L Masci, J Pike, F Malara, B Phillips, K Bennell, P Brukner, C Standaert, and L Micheli Use of the one-legged hyperextension test and magnetic resonance imaging in the diagnosis of active spondylolysis * Commentary * Commentary Br. J. Sports Med., November 1, 2006; 40(11): 940 - 946. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Gaeta, F. Minutoli, S. Vinci, I. Salamone, L. D'Andrea, L. Bitto, L. Magaudda, and A. Blandino High-resolution CT grading of tibial stress reactions in distance runners. Am. J. Roentgenol., September 1, 2006; 187(3): 789 - 793. [Abstract] [Full Text] [PDF]
|
|
| ||||||
