| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Thomas Jefferson University Hospital, 111 S. 11th
St., Philadelphia, PA 19107.
2
Department of Radiology, Wilford Hall Medical Center, San Antonio, TX
78236.
Received August 18, 2000;
accepted after revision October 12, 2000.
Address correspondence to W. B. Morrison.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. At 1.5 T, 141 MR examinations of the ankle (116 clinical examinations of patients with chronic pain, 25 of asymptomatic control patients) were performed using T1-weighted, proton density—weighted fast spin-echo, and T2-weighted fat-suppressed fast spin-echo sequences. Images were retrospectively reviewed by two musculoskeletal radiologists for presence of bone marrow edema occurring in a subcortical location associated with the course of the medial or lateral tendon groups, as well as focal thickening or increased T2 signal within the tendons. These findings were correlated with clinical information regarding symptom location. The association of subtendinous marrow edema with tendinopathy and symptom location was statistically analyzed.
RESULTS. Subtendinous bone marrow edema was present at 26 sites on 24 ankle MR examinations (17%) (at the medial malleolus [n = 17] associated with the posterior tibialis tendon, at the lateral malleolus [n = 6] and the calcaneus [n = 2] associated with the peroneus longus and brevis tendons, and at the cuboid [n = 1] associated with the peroneus longus tendon). These subtendinous bone marrow edema patterns were significantly associated with overlying tendon abnormality medially (p = 0.001) and laterally (p = 0.001), and with symptoms medially (p = 0.0016) but not laterally (p = 0.078).
CONCLUSION. On MR images of the ankle, bone marrow edema localized in a subtendinous location is associated with overlying tendinopathy medially and laterally and with ankle pain medially.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All MR imaging was performed using a 1.5-T scanner (General Electric Medical Systems, Milwaukee, WI) using an extremity coil around the ankle. The MR imaging protocol consisted of coronal, sagittal, and axial T2-weighted fast spin-echo images with fat suppression, axial proton density—weighted fast spin-echo images, and sagittal T1-weighted spin-echo images. T2-weighted fast spin-echo images were acquired with a TR range/TEeff range of 2000-6000/60-90, echo train length of 8, slice thickness of 4 mm with an interslice gap of 1 mm, with 2-3 signal averages, a field of view of 12-18 cm, and a matrix of 256 x 192. T1-weighted spin-echo images were acquired with TR range/TE range of 400-700/10-20, slice thickness of 4 mm with an interslice gap of 1 mm, and 2 signal averages with a field of view of 16-18 cm and a matrix of 256 x 192. Axial proton density—weighted fast spin-echo images were also obtained with TR range/TEeff of 2500-3800/34, echo train length of 4, slice thickness of 4 mm with an interslice gap of 1 mm, and 2 signal averages with a field of view of 14-16 cm, and a matrix of 512 x 256.
Image interpretation was performed retrospectively by two musculoskeletal radiologists in consensus who were unaware of the initial MR findings and clinical data. Criterion for subtendinous bone marrow edema was marrow T2 hyperintensity verified in at least two planes adjacent to the course of a medial or lateral ankle tendon (in absence of artifact or inadequate fat suppression judged to obscure marrow edema). To fulfill the criterion, the marrow edema had to be centered in a subcortical location, away from an articular surface. The specific location and the pattern (localized or diffuse) of subtendinous marrow edema were recorded. T1-signal characteristics in the area of edema were recorded as equal to or less than that of surrounding fatty marrow. The adjacent cortex was also evaluated for bone proliferation, defined as an irregularity of the cortical signal or protrusion of fatty marrow signal from the cortical margin.
Tendinopathy was defined as the presence of tendon intrasubstance T2 signal or focal tendon enlargement, verified on both axial and sagittal images; this definition included tendinosis and tendon tear. The amount of tendon sheath fluid was not considered in the criterion because of the wide variability of fluid previously described in asymptomatic individuals [20]. The tendons were evaluated separately but grouped as medial (posterior tibial tendon, flexor digitorum longus, and flexor hallicus longus) or lateral (peroneus longus and peroneus brevis) for statistical analysis because the clinical information was confined to medial and lateral symptoms.
Clinical information was obtained from multiple sources. The first source was the history and indication given by the health care provider ordering the test. Information in the hospital information system was also used. Additionally, information was obtained from a preprocedural questionnaire administered to all patients referred for ankle MR imaging at our institution. Patients were asked to check one or more descriptive terms on the questionnaire corresponding to the location of their pain: choices included the inner aspect, outer aspect, front, or back of the ankle, as well as the bottom of the foot. Patients were also asked to estimate the duration of their present symptoms.
Initially, all ankle MR imaging examinations performed from August 1995 to May 1999 were included in the study; however, some examinations were excluded from the final study group on the basis of MR imaging findings or clinical information. Exclusion criteria based on the MR imaging examination included substantial artifacts or incomplete fat suppression interfering with evaluation of tendon signal abnormality or subcortical marrow edema. Examinations with osteochondral defect of the talus were also excluded from the study group (defined for exclusionary purposes as focal high T2 signal at the talar dome), because this entity is common and can be a substantial source of pain and marrow edema that could confound the results [7, 13].
Some patients were excluded from the study group on the basis of available clinical information. Clinical exclusion criteria included a suspected or confirmed inflammatory arthropathy or infection at the ankle, neoplasm in the ankle region, or symptom duration of less than 1 month. These criteria were used in an attempt to exclude conventional causes of marrow edema. Patients who did not report medial or lateral pain were also excluded. This was done because only the medial and lateral tendon groups lie immediately adjacent to the bones of the ankle as they course toward their insertions. Patients who did not complete the questionnaire were also excluded.
After clinical and MR imaging exclusion criteria were applied, 116 ankles in 116 patients comprised the study group. Twenty-five ankles of 14 individuals without ankle pain or history of significant ankle injury on the side or sides imaged served as a control group, yielding a total of 141 ankles imaged.
On the basis of clinical information and MR imaging findings, the symptoms, subtendinous bone marrow edema, and tendinopathy were categorized as located at the medial aspect of the ankle, at the lateral aspect of the ankle, at both the medial and lateral aspects of the ankle, or at neither medial nor lateral (control group). Statistical analysis was performed to determine if there is an association between symptom location and subtendinous bone marrow edema or tendinopathy on the ipsilateral or contralateral aspect of the ankle. Analysis was performed using an odds ratio with 95% confidence interval. Significance of the association was tested using the Fisher's exact test with 2 x 2 contingency tables. Association of ipsilateral tendinopathy and subtendinous bone marrow edema was tested for significance using the chi square test, also with 2 x 2 contingency tables. The duration of symptoms for patients with subtendinous bone marrow edema versus those without was analyzed using the Wilcoxon's rank sum test (assuming a nonnormal distribution).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Symptoms were reported medially in 74 ankles (52%), laterally in 79 (56%), both medially and laterally in 37 (26%), and neither medially nor laterally (asymptomatic controls) in 25 (18%) (Table 1). Medial or lateral tendon abnormalities were identified on 37 MR examinations (26%) (both medial and lateral in four): 16 met criteria for tendinopathy of the medial tendons (13/16 increased signal, 12/16 thickened), versus 25 laterally (12/25 increased signal, 23/25 thickened). The peroneus longus tendon was most commonly abnormal (23 examinations), followed by the posterior tibialis tendon (14 examinations), the peroneus brevis tendon (11 examinations), and the flexor digitorum longus tendon (one examination). No abnormalities were seen within the flexor hallicus longus tendon.
|
Of the 16 MR imaging examinations revealing medial tendinopathy, medial symptoms were reported in 15, yielding an odds ratio of 16.78 and a p value of 0.0003 (Table 2). In contrast, in this group, lateral symptoms were reported in six, yielding an odds ratio of 0.427 and a p value of 0.180. The association of lateral symptoms and tendinopathy was not as strong: of the 25 MR examinations showing tendinopathy of the lateral group, lateral symptoms were reported in 18, yielding an odds ratio of 2.32 and a p value of 0.119. Similarly, in this group, medial symptoms were reported in 11, yielding an odds ratio of 0.66 and a p value of 0.384.
|
Subtendinous bone marrow edema was present on 24 (21%) of the 116 MR imaging examinations. On two examinations, more than one bone was involved (medial malleolus and lateral malleolus; lateral malleolus and calcaneus). Subtendinous marrow edema was present medially on 17 examinations (15%), laterally on eight examinations (7%), and both medially and laterally on one examination (1%). Medially, the edema was present in the medial malleolus on all 17 examinations and was only associated with the course of the posterior tibialis tendon (Fig. 1A,1B,1C). No subtendinous bone marrow edema was seen adjacent to the flexor digitorum longus tendon or the flexor hallicus longus tendon. Laterally, the edema was present in the lateral malleolus on six examinations, in the calcaneus on two examinations (Fig. 2A,2B,2C,2D), and in the cuboid on one examination. Edema in the lateral malleolus and in the calcaneus was associated with the course of both the peroneus longus and brevis tendons, but the edema in the cuboid was only associated with the course of the peroneus longus tendon. Bone marrow edema was not diffuse, but was confined to the subcortical bone on all examinations. T1 signal intensity was normal (equal to the surrounding fatty marrow) on all examinations in the region of subtendinous bone marrow edema; no fracture lines or other abnormalities were visualized. Bone proliferation was evident in the region of subtendinous bone marrow edema on the MR images of two patients, one at the medial malleolus and the other at the lateral calcaneus (Fig. 2A,2B,2C,2D). Finally, no subtendinous marrow edema was observed on MR examinations of the asymptomatic controls.
|
|
|
|
|
|
|
On the 17 MR imaging examinations with medial subtendinous bone marrow edema, medial symptoms were reported in 15, yielding an odds ratio of 8.26 (p = 0.0016) (Table 3). In this group, lateral symptoms were reported in 12 examinations, yielding and odds ratio of 2.04 (p = 0.297). The association of lateral symptoms and lateral subtendinous bone marrow edema was not as pronounced as the association medially. On the eight MR examinations with lateral subtendinous bone marrow edema, lateral symptoms were reported in seven, resulting in an odds ratio of 5.93 (p = 0.078). In this group, medial symptoms were reported in four, yielding an odds ratio of 0.90 (p = 1.00).
|
Chi square analysis showed a significant association between tendinopathy and subtendinous bone marrow edema medially (p = 0.001) and laterally (p = 0.001). The association of subtendinous bone marrow edema and reported symptom duration was not clear. In patients with subtendinous bone marrow edema, the average symptom duration was 15.9 months, whereas in patients without subtendinous bone marrow edema the average symptom duration was 18.9 months (Wilcoxon's rank sum, p = 0.051).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found that subtendinous marrow edema patterns are present on MR images in a relatively high proportion of patients who present with chronic ankle pain (21%). Subtendinous marrow edema is most common at the medial malleolus, associated with the course of the posterior tibialis tendon, and at the lateral malleolus related to the course of the peroneal tendons. However, edema is also observed at the calcaneus associated with the peroneal tendons and at the cuboid associated with the peroneus longus tendon. Subtendinous bone marrow edema is significantly associated with morphologic or T2-signal abnormality of the overlying tendon at the medial and lateral aspects of the ankle. There is also a significant association of subtendinous marrow edema and medial symptoms, but laterally the association with symptoms does not reach statistical significance; this finding may be related to difficulty in defining morphologic or signal abnormalities in the peroneal tendons. Alternatively, the finding may be related to the difficulty in separating lateral ligament pain from peroneal tendon pain. This is supported by the finding that the association of lateral symptoms and peroneal tendinopathy does not reach statistical significance.
The cause of these marrow edema patterns remains unclear—it may be related to marrow hyperemia and related to overlying tendon friction or it may be a result of hyperemia of the overlying tendon sheath. We excluded patients who had symptoms for less than 1 month, so this process is not likely to represent edema from an acute traumatic event, nor is it likely to represent edema-type signal caused by trabecular reorganization in response to stress, as reported for other locations in the foot and ankle [4, 5]. Subtendinous bone marrow edema has a high association with location of symptoms and overlying tendon abnormality, and when present, may be a reliable indicator of true disease. Bone proliferation was observed on MR images of two ankles in the region of subtendinous bone marrow edema, supporting the hypothesis proposed in previous studies that nonarticular spurring may be associated with tendinopathy.
This study contains limitations that should be recognized. First, there is no reference standard to verify the MR imaging finding: the diagnosis of tendinopathy is presumed on the basis of the imaging appearance. For this reason, no attempt was made to separate tendinosis from tendon tear that could not be surgically documented. Also, although we postulate that tendon sheath hyperemia may be the source of subtendinous bone marrow edema, we did not perform dynamic contrast-enhanced sequences, which could assess areas with increased blood flow. A selection bias may also exist because this study is based mainly on clinical cases that may not be representative of the population as a whole. Bias may also have been introduced because a truly blinded evaluation of tendon abnormalities and bone marrow edema could not be performed by the reviewers as a result of the proximity of the tendons and subjacent bone on the images. Another potential limitation is that the location of symptoms is based primarily on self-reporting from a questionnaire: clinical and follow-up information is limited. Another confounding issue is that patients could report both medial and lateral pain; although pain may primarily originate at one site, it may secondarily radiate from either the primary process or may be generated by altered gait (which commonly occurs with chronic foot and ankle pain). We did not assess the degree of pain or attempt to define the main source of pain in patients with multifocal or diffuse symptoms. Finally, in this study cause-effect relationships between tendon abnormalities, subtendinous bone marrow edema, and pain cannot be established; we merely describe associations.
Bone marrow edema is a common finding on MR images of the ankle. The differential diagnosis includes a variety of conditions from the following broad categories: traumatic, developmental (e.g., coalition), neoplastic, biomechanical (e.g., stress response), inflammatory (infectious or non-infectious), vascular, neuropathic, metabolic, and degenerative. Tendinopathy is an additional factor associated with bone marrow edema, which occurs in a characteristic subtendinous pattern; this edema could potentially be an MR imaging marker for symptom location in patients with chronic ankle pain.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Karchevsky, M. E. Schweitzer, W. B. Morrison, and J. A. Parellada MRI Findings of Septic Arthritis and Associated Osteomyelitis in Adults Am. J. Roentgenol., January 1, 2004; 182(1): 119 - 122. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
