| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Musculoskeletal Imaging |
1 From the Department of Medical Imaging, Mount Sinai Hospital and the University Health Network, University of Toronto, 600 University Ave, Toronto, Ontario, Canada M5G 1X5 (P.R., L.M.W., D.C.S.); Department of Orthopaedic Surgery, St Michael’s Hospital, University of Toronto, Ontario, Canada (T.R.D.); and Department of Orthopaedic Surgery, Toronto Western Hospital, University of Toronto, Ontario, Canada (D.O.H.). From the 2000 RSNA scientific assembly. Received October 16, 2000; revision requested November 28; revision received February 5, 2001; accepted March 19. Supported in part by the 2000 RSNA Fellowship Trainee Prize. Address correspondence to L.M.W. (e-mail: lwhite@mtsinai.on.ca).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Thirty-two patients with chronic ankle pain prospectively underwent gadolinium-enhanced MR arthrography of the tibiotalar joint. All underwent clinical examination and were included if anterolateral impingement (n = 13) or a control condition (n = 19; suspected osteochondral defect, intraarticular bodies, instability, osteoarthritis) was diagnosed. MR imaging included transverse and coronal T1-weighted and sagittal T2-weighted imaging sequences. Images were prospectively analyzed by two readers blinded to the clinical diagnosis. The anterolateral gutter contour was assessed. MR arthrographic findings were correlated with subsequent arthroscopic appearances.
RESULTS: MR arthrographic assessment of the anterolateral soft tissues had an accuracy of 97%, sensitivity of 96%, specificity of 100%, negative predictive value of 89%, and positive predictive value of 100%. Accuracy was 100% with clinical anterolateral impingement, with an arthroscopically confirmed abnormality in 12 cases and a normal appearance in one. Anterolateral soft-tissue thickening was identified at MR arthrography in 11 control cases, with arthroscopic confirmation in all. The remaining cases had normal appearances, with an arthroscopic soft-tissue abnormality in one case and a normal appearance in seven.
CONCLUSION: MR arthrography of the tibiotalar joint is accurate in assessing the anterolateral recess of the ankle.
Index terms: Ankle, abnormalities, 463.486 • Ankle, injuries, 463.486 • Ankle, MR, 463.1214 • Magnetic resonance (MR), arthrography, 463.12149, 463.12143
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
ALI is largely a clinical diagnosis of exclusion, with pathologic findings confirmed at arthroscopy (5). Retrospective surgical series (3,6) have revealed that anterolateral tenderness, swelling, pain with single leg squatting, and pain with ankle dorsiflexion and eversion are the clinical findings most strongly associated with abnormality at arthroscopy. These features, together with the absence of mechanical instability and tendon subluxation, and the absence of severe degenerative change and fracture complications at conventional radiography, suggest the diagnosis (3).
Most patients with ALI respond to rehabilitative physiotherapy; however, if this is unsuccessful, surgical treatment may be of benefit (1–3,6,7). Previous orthopedic studies (1–3,8) have shown that when synovitis and/or scarring is identified and resected in this patient group, there is a considerable improvement in symptoms and ankle function.
Magnetic resonance (MR) imaging is widely used as a diagnostic tool in the preoperative assessment of patients with chronic ankle pain that does not respond to conservative therapy (9). In a number of retrospective studies (5,6,10–12), investigators have reviewed the use of cross-sectional imaging for the preoperative identification of substantial anterolateral recess abnormalities. One retrospective study (11) of MR examinations performed in patients with a subsequent arthroscopic diagnosis of ALI revealed that conventional MR imaging (without intraarticular contrast material enhancement) is sensitive for anterolateral abnormality. However, findings in three other studies (5,6,10) with similar retrospective methods did not demonstrate a substantial advantage in imaging this group of patients prior to arthroscopy.
The purpose of this study was to prospectively investigate the diagnostic accuracy of MR arthrography in assessing the anterolateral recess of the ankle.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
In all 32 cases, the referring surgeon completed a clinical assessment form that included patient demographic data, mechanism of injury, duration of symptoms, clinical findings, conventional radiographic findings, and previous treatment. Specific clinical findings suggestive of ALI were recorded and included anterolateral tenderness, anterolateral swelling, pain with single leg squatting, and pain with dorsiflexion and eversion (3,6).
MR Arthrography
After informed consent was obtained, all 32 patients underwent MR arthrography of the affected ankle. A 25-gauge needle was fluoroscopically guided into the tibiotalar joint, and its position was confirmed by using iodinated contrast material (diatrizoate meglumine 60% [Hypaque]; Nycomed Amersham, Princeton, NJ; 0.5–1.0 mL). Gadopentetate dimeglumine solution (Omniscan; Nycomed Amersham; 2 mmol/L, 1 mL in 250 mL normal saline) was then injected into the tibiotalar joint until resistance was felt (volume range, 8–15 mL). The needle was removed, and the patients were transferred to the MR suite without allowing them to ambulate. All imaging was performed with a 1.5-T magnet (Signa; GE Medical Systems, Milwaukee, Wis) by using a quadrature extremity coil (Medical Advances, Milwaukee, Wis), with the ankle in a neutral position. Transverse and coronal T1-weighted fat-suppressed conventional spin-echo sequences (repetition time msec/echo time msec, 583/8) and sagittal T2-weighted fat-suppressed fast spin-echo sequences (3,933/80; echo train length, eight; 3-mm section thickness; no spacing; 256 x 256 matrix; 19.9-cm field of view) were performed through the tibiotalar joint and hindfoot.
Image Analysis
All images were prospectively assessed by two experienced musculoskeletal radiologists (L.M.W., D.C.S.) in consensus. Readers were blinded to the patient data, clinical history, and ratio of cases with clinical ALI of the ankle. In all cases, assessment included evaluation of the ankle and anterolateral soft tissues.
Grading of the anterolateral recess.—The appearance of the soft-tissue interval between the inferior anterior tibiofibular and anterior talofibular ligaments was graded I–IV on the basis of a previously proposed (12) grading scheme. Grade I was a normal appearance (Fig 1); grade II, thickening of the inferior anterior tibiofibular ligament but an otherwise normal appearance (Fig 2); grade III, focal nodular soft-tissue thickening (Fig 3); and grade IV, irregular nodular soft-tissue thickening (Fig 4).
|
|
|
|
Lateral capsular ligaments and articular cartilage.—The anterior tibiofibular, anterior talofibular, and calcaneofibular ligaments were classified as normal, intact but thickened and/or attenuated, or disrupted. Articular cartilage was assessed as normal or abnormal.
Arthroscopy
All 32 patients subsequently underwent arthroscopy after MR arthrographic examination (average time between MR arthrography and surgery, 31 days; median, 43 days; range, 1–184 days). At surgery, the appearances of the anterolateral recess, capsule, ligaments, cartilage, and bones were recorded. The anterolateral recess was considered normal or abnormal (encompassing synovitis and scar tissue).
Statistical Evaluation
The grade of soft-tissue abnormality (with grades III and IV considered abnormal) at MR arthrography was compared with findings at arthroscopy. The Fisher exact test was used to evaluate the relationship between individual imaging findings and anterolateral soft-tissue abnormalities at surgery. Comparisons of the findings in the two groups (the clinical ALI group and the control group) were performed by using the two-tailed P values with the Fisher exact test.
Clinical Follow-up
All 32 patients underwent follow-up clinical examination at 3 (n = 32) and 6 (n = 7) months. The examination included an assessment of ankle pain and overall function in relation to pain and disability.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Grading of the anterolateral recess.—MR arthrographic grading of the anterolateral soft tissues was significantly associated with surgical findings (P < .005). Overall, MR arthrography had an accuracy of 97% (31 of 32 patients), sensitivity of 96% (23 of 24 patients), specificity of 100% (eight of eight patients), negative predictive value of 89% (eight of nine patients), and positive predictive value of 100% (23 of 23 patients) for assessment of the anterolateral soft tissues.
Anterolateral recess distention.—The presence of capsular tissue adherent to the anterior fibula was not significantly associated with abnormality at arthroscopy (P = .14). However, the proportion of cases with adherence of anterolateral soft tissues to the anterior fibula was significantly greater in patients with a clinical diagnosis of ALI (six of 13 patients) compared with that of the control group (one of 19 patients) (P = .012).
Lateral capsular ligaments and articular cartilage.—Abnormality of the other imaging parameters recorded at MR arthrography showed no significant association with surgical findings, as follows: tibiofibular ligament, P = .76; talofibular ligament, P = .69; calcaneofibular ligament, P = .8; articular cartilage, P = .4.
Patients with a Clinical Diagnosis of ALI
Clinical findings in the 13 patients in this group included anterolateral tenderness (n = 13), anterolateral swelling (n = 6), pain with single leg squatting (n = 12), and pain with dorsiflexion and eversion (n = 9).
MR arthrography had an accuracy of 100% (13 of 13 patients), sensitivity of 100% (12 of 12 patients), specificity of 100% (one of one patient), negative predictive value of 100% (one of one patient), and positive predictive value of 100% (12 of 12 patients) for the assessment of the anterolateral recess. The anterolateral soft tissues were graded as abnormal in 12 (one, grade III; 11, grade IV) of 13 cases at MR arthrography; these grades corresponded to scarring and/or synovitis at arthroscopy in all 12 cases (Fig 5). In one additional case with a preoperative diagnosis of ALI, the anterolateral recess was graded as normal (grade I) at MR arthrography, and this finding was confirmed at arthroscopy (Fig 6).
|
|
|
|
Subjects with a Control Diagnosis
Clinical findings of the 19 subjects without a clinical diagnosis of ALI included mechanical instability (n = 8), diffuse ankle tenderness (n = 8), diffuse ankle swelling (n = 4), pain and locking (n = 3), anterolateral tenderness (n = 5), pain with single leg squatting (n = 10), and pain with dorsiflexion and eversion (n = 10).
In this subject group, MR arthrography had an accuracy of 95% (18 of 19 subjects), sensitivity of 92% (11 of 12 subjects), specificity of 100% (seven of seven subjects), negative predictive value of 88% (seven of eight subjects), and positive predictive value of 100% (11 of 11 subjects) for the assessment of the anterolateral recess. At MR arthrography, the anterolateral soft tissues were graded as abnormal in 11 (five, grade III; six, grade IV) of 19 cases at MR arthrography; these findings corresponded to scarring and/or synovitis at arthroscopy in all 11 cases.
In two of 19 subjects, the anterolateral recess was assessed as grade II at MR arthrography, and the soft tissues were normal at arthroscopy in both cases. In the remaining six of 19 subjects, the anterolateral recess was graded as normal (grade I) at MR arthrography, and this finding was confirmed at arthroscopy in five cases, but scarring and/or synovitis was found in one case. In one case (one of 19 subjects), the anterolateral tissues were adherent to the anterior surface of the fibula, and abnormal anterolateral soft tissues were present at arthroscopy.
Other arthroscopic findings in this subject group included talar chondral abnormalities in 12 of 19 subjects (eight, grade 1; three, grade 2; one, grade 3). These abnormalities were lateral in three cases (all grade 1) and anteromedial in nine cases (five, grade 1; three, grade 2; one, grade 3). Bone spurs of the talus and tibia were present in six of 19 subjects (anterior, three subjects; lateral, two subjects; medial, one subject). The anterior talofibular ligament appeared abnormal in eight of 19 subjects, with complete rupture in three of eight. Intraarticular bodies were identified in three subjects.
Clinical Follow-up
No immediate or long-term complications occurred as a consequence of arthrography, and the surgeons noted no intraoperative surgical difficulties. Thirty-one of 32 patients showed improvement in pain and function scores after surgery; one subject (a control subject with preoperative instability and intraarticular bodies) developed septic arthritis after surgery.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Previous studies (5,6,10,11) on the value of conventional (nonarthrographic) MR imaging had revealed conflicting results in the examination of patients with ALI. Sensitivity (39%–100%) and specificity (50%–100%) in the detection of anterolateral soft-tissue abnormality varied widely (6, 10,11). One study (5) revealed that MR assessment of the anterolateral recess is accurate only when a substantial joint effusion is present. A single study (12) of CT arthrography revealed a strong association between the appearance of the anterolateral recess on coronal images and surgical findings.
The investigators in each of these prior MR studies (5,10,11) used methods that involved a retrospective imaging review of patients with a clinical diagnosis of ALI and abnormal anterolateral recess at arthroscopy; this method created substantial study selection bias. Thus, patients with a clinical diagnosis of ALI and normal arthroscopic findings were potentially excluded from the investigation. In addition, control cases in previous studies (5,6) were selected on the basis of a normal arthroscopic examination of the anterolateral recess, potentially excluding subjects with anterolateral soft-tissue abnormality without clinical features of ALI.
We found that clinical examination was reasonably accurate in the diagnosis of ALI, with only one false-positive case. In all 13 patients, the MR arthrographic appearance of the anterolateral soft tissues agreed with findings at arthroscopy (including the clinically false-positive case). We also found that a substantial number of control subjects (11 of 19 subjects) had anterolateral scarring and/or synovitis at MR arthrography (grade III or IV). This finding was confirmed arthroscopically in all 11 subjects; however, none of these subjects had clinical features consistent with those of ALI. This observation is important because it suggests that the identification of abnormal soft tissue itself does not imply the presence of clinical ALI.
Another finding of this study was that the absence of a recess of arthrographic fluid between the anterolateral soft tissues and the anterior surface of the fibula was always associated with scarring and/or synovitis at MR arthrography and arthroscopy. This absence may be due to the presence of adhesions and scar tissue that prevent fluid from entering the normal recess between the fibula and joint capsule (Fig 5). The frequency of this finding occurred in a significantly greater proportion of patients with ALI (six of 13 patients) compared with that of the control group (one of 19 subjects) (P = .012). Although specific for soft-tissue abnormalities, this finding has limited sensitivity, since it can be absent in patients with ALI.
Potential limitations of this study include possible interobserver variability in the assessment of the anterolateral recess at surgery. To our knowledge, there is no accepted surgical grading system of anterolateral soft-tissue abnormality, and we are therefore dependent on the individual surgeon’s subjective assessment of the area. We tried to reduce this potential bias by limiting our referral base to two experienced surgeons.
ALI itself is not a common cause of chronic ankle pain, and most patients respond to rehabilitative physiotherapy; therefore, prospectively obtaining a large number of cases can be difficult (5,6,10–12). Another limitation of this study is the relatively short clinical follow-up that did not allow the overall benefit of imaging and surgery to be studied. We plan to continue clinical follow-up of existing patients and recruit new patients to further assess the effect of imaging and surgery on ALI.
In conclusion, we found MR arthrography to be an accurate technique for use in assessing the anterolateral recess of the ankle in patients with ALI. However, a significant number of control subjects also had anterolateral synovitis and/or scarring at MR arthrography and arthroscopy. Therefore, image confirmation of anterolateral soft-tissue abnormalities must be considered with the clinical findings. The usefulness of any preoperative imaging modality is dependent on its ability to complement clinical assessment in the determination of patients who will benefit from surgery. Our results support the role of MR arthrography in confirming the presence of soft-tissue scarring and elucidating its extent in patients with ALI prior to arthroscopy.
|
FOOTNOTES |
|---|
Abbreviation: ALI = anterolateral impingement
Author contributions: Guarantor of integrity of entire study, L.M.W.; study concepts, L.M.W., P.R.; study design, L.M.W.; literature research, P.R.; clinical studies, P.R., D.C.S., T.R.D., D.O.H.; data acquisition and analysis/interpretation, P.R., L.M.W., D.C.S.; statistical analysis, P.R., L.M.W.; manuscript preparation and definition of intellectual content, P.R., L.M.W.; manuscript editing, all authors; manuscript revision/review and final version approval, P.R., L.M.W.
This paper received the 2000 Fellow Research Trainee Prize at the 2000 RSNA Scientific Assembly and Annual Meeting
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S Ostlere Imaging the ankle Imaging, September 1, 2007; 19(3): 269 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. W. P. Bearcroft, S. Guy, M. Bradley, and F. Robinson MRI of the ankle: effect on diagnostic confidence and patient management. Am. J. Roentgenol., November 1, 2006; 187(5): 1327 - 1331. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Robinson and S. R. Bollen Posterior ankle impingement in professional soccer players: effectiveness of sonographically guided therapy. Am. J. Roentgenol., July 1, 2006; 187(1): W53 - W58. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S Ostlere Imaging the ankle and foot Imaging, December 1, 2003; 15(4): 242 - 269. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Cerezal, F. Abascal, A. Canga, T. Pereda, R. Garcia-Valtuille, L. Perez-Carro, and A. Cruz MR Imaging of Ankle Impingement Syndromes Am. J. Roentgenol., August 1, 2003; 181(2): 551 - 559. [Full Text] [PDF]
|
|
|
|
 |
|
 P. Robinson and L. M. White Soft-Tissue and Osseous Impingement Syndromes of the Ankle: Role of Imaging in Diagnosis and Management RadioGraphics, November 1, 2002; 22(6): 1457 - 1469. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. S. Steinbach, W. E. Palmer, and M. E. Schweitzer Special Focus Session: MR Arthrography RadioGraphics, September 1, 2002; 22(5): 1223 - 1246. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
