Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

DOI: 10.1148/radiol.2421051369
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kim, S.
Right arrow Articles by Suh, J.-S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, S.
Right arrow Articles by Suh, J.-S.
(Radiology 2007;242:225-235.)
© RSNA, 2007

Musculoskeletal Imaging

Chronic Tibiofibular Syndesmosis Injury of Ankle: Evaluation with Contrast-enhanced Fat-suppressed 3D Fast Spoiled Gradient-recalled Acquisition in the Steady State MR Imaging1

Sungjun Kim, MD2, Yong-Min Huh, MD, Ho-Taek Song, MD, Sung-Ah Lee, MD, Jin-Woo Lee, MD, Jong Eun Lee, PhD, In Hyuk Chung, MD, PhD and Jin-Suck Suh, MD

1 From the Departments of Diagnostic Radiology (S.K., Y.M.H., H.T.S., S.A.L., J.S.S.), Orthopedic Surgery (J.W.L.), and Anatomy (J.E.L., I.H.C.), the Research Institute of Radiological Science of Severance Hospital (J.S.S.), and the Brain Korea 21 Project for Medical Science (J.S.S.), Yonsei University, College of Medicine, 134 Shinchondong, Seodaemun-ku, Seoul 120-752, Korea; and Department of Diagnostic Radiology, Seoul Medical Center, Seoul, Korea (S.A.L.). From the 2004 RSNA Annual Meeting. Received August 15, 2005; revision requested October 18; revision received December 11; accepted January 10, 2006; final version accepted, February 3. Supported in part by Korea Science and Engineering Foundation, Republic of Korea, through National Core Research Center (R15-2004-024-02002-0). Address correspondence to J.S.S. (e-mail: jss{at}yumc.yonsei.ac.kr).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCE IN KNOWLEDGE
 References
 
Purpose: To retrospectively determine the accuracy of coronal contrast material–enhanced fat-suppressed three-dimensional (3D) fast spoiled gradient-recalled acquisition in the steady state (SPGR) magnetic resonance (MR) imaging, as compared with that of routine transverse MR imaging, in the assessment of distal tibiofibular syndesmosis injury, with arthroscopy as the reference standard.

Materials and Methods: The review board of the College of Medicine in Yonsei University approved this study; informed consent was waived. The study group comprised 45 patients (26 men, 19 women; mean age, 32.1 years; range, 18–58 years) with a chronic ankle injury who had undergone MR imaging and arthroscopic surgery. Three independent readers retrospectively reviewed the two sets of MR images (one set of gadolinium-enhanced 3D fast SPGR images and one set of routine T1-, T2-, and intermediate-weighted images). Scores from 1 to 5 in increasing order of the probability of injury were assigned to both sets. Arthroscopy was the reference standard. Syndesmotic recess height was measured on contrast-enhanced images. The two sets of images were compared for diagnostic performance with receiver operating characteristic (ROC) analysis. Dissection and histologic examination of six cadaveric ankles was performed to assess the syndesmotic area and ascertain the enhancing structure at MR imaging.

Results: At arthroscopy, syndesmotic injury was found in 24 ankles but not in 21 ankles. Areas under the ROC curve were significantly higher for the contrast-enhanced images (P < .05). The contrast-enhanced set showed higher accuracy, sensitivity, and specificity compared with the routine set for the assessment of syndesmosis injury. Mean syndesmotic recess height was significantly greater (P < .05) in patients with syndesmotic injury. Dissection and histologic examination revealed a highly vascular synovial fold in the syndesmotic area that is expected to enhance at MR imaging.

Conclusion: In the assessment of chronic syndesmosis injury, coronal gadolinium-enhanced fat-suppressed 3D fast SPGR MR images were more sensitive, specific, and accurate than routine MR images.

© RSNA, 2007


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCE IN KNOWLEDGE
 References
 
The distal tibiofibular syndesmosis is a complex structure stabilized by four separate ligaments: the anterior inferior tibiofibular ligament (AITFL), the posterior inferior tibiofibular ligament, the transverse tibiofibular ligament, and the interosseous ligament (Fig 1). A distal tibiofibular syndesmosis injury may develop as a result of severe ankle trauma, as well as through repeated ankle sprains, with an incidence ranging from 1% to 20% (1–6). External rotation of the foot and internal rotation of the leg are generally accepted as the mechanisms of the injury (3,7). An ankle with syndesmosis injury becomes unstable in its tibiofibulotalar articulation. Improper treatment of such instability results in a longer recovery time and a poor prognosis (3,4,8,9).


Figure 1
View larger version (23K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: Four ligaments stabilize the distal tibiofibular syndesmosis: AITFL, which is about 20 mm wide and 20–30 mm long and often multifascicular, runs obliquely at an approximate 45° angle from anterolateral tubercle of tibia (Tillaus-Chaput tubercle) to the anterolateral fibula. Posterior inferior tibiofibular ligament (PiTFL) covers back of tibiotalar joint. Interosseous ligament connects tibia and fibula from 0.5 to 2 cm above the tibial plafond. It surrounds the syndesmotic recess, which usually extends approximately 1 cm up from the tibiotalar joint. Deep portion of posterior inferior tibiofibular ligament is known as the transverse tibiofibular ligament (TrTFL). It lies anterior to superficial component of ligament and forms most distal aspect of the tibiotalar articulation.

 
Controversy still exists regarding the diagnostic capability of magnetic resonance (MR) imaging for ankle ligament injuries owing to varying levels of sensitivity and accuracy (10–18). The study population (eg, patients with acute vs those with chronic injury), imaging techniques, and imaging planes may affect diagnostic accuracy. For example, owing to the oblique course of the AITFL, it may not be entirely visible on a single transverse image, and a partially visible AITFL might be mistaken as a tear despite careful inspection of contiguous sections (11,19–21) (Fig 2). Signal intensity changes in acutely injured ligaments can appear obvious on nonenhanced MR images, but this may not be the case for chronic injuries. To solve these technical problems and to enable more precise evaluation of articular cartilage and loose bodies, we have included a coronal contrast material–enhanced fat-suppressed three-dimensional (3D) fast spoiled gradient-recalled acquisition in the steady state (SPGR) sequence in our institution's routine MR imaging protocol for ankle ligament sprain evaluation.


Figure 2
View larger version (154K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a: (a, b) Transverse T2-weighted dual spin-echo MR images (2000/70; matrix, 256 x 256; number of signals acquired, one; field of view, 12 cm) of AITFL. Entire length of AITFL is not always reliably identifiable on transverse scans owing to its oblique course. (c) Drawing shows levels at which a and b were acquired. Transverse section at level of tibial plafond (dotted line in c) shows AITFL has no definite tibial attachment (arrow in b), and transverse section at a certain level of tibiotalar joint (solid line in c) shows just tibial attachment (arrow in a).

 

Figure 2
View larger version (161K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b: (a, b) Transverse T2-weighted dual spin-echo MR images (2000/70; matrix, 256 x 256; number of signals acquired, one; field of view, 12 cm) of AITFL. Entire length of AITFL is not always reliably identifiable on transverse scans owing to its oblique course. (c) Drawing shows levels at which a and b were acquired. Transverse section at level of tibial plafond (dotted line in c) shows AITFL has no definite tibial attachment (arrow in b), and transverse section at a certain level of tibiotalar joint (solid line in c) shows just tibial attachment (arrow in a).

 

Figure 2
View larger version (32K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c: (a, b) Transverse T2-weighted dual spin-echo MR images (2000/70; matrix, 256 x 256; number of signals acquired, one; field of view, 12 cm) of AITFL. Entire length of AITFL is not always reliably identifiable on transverse scans owing to its oblique course. (c) Drawing shows levels at which a and b were acquired. Transverse section at level of tibial plafond (dotted line in c) shows AITFL has no definite tibial attachment (arrow in b), and transverse section at a certain level of tibiotalar joint (solid line in c) shows just tibial attachment (arrow in a).

 
When interpreting enhanced images, it is sometimes difficult to distinguish injured ligaments from the adjacent enhanced synovial tissue because the two are in close proximity. Thus, the purpose of our study was to retrospectively determine the accuracy of coronal contrast-enhanced fat-suppressed 3D fast SPGR MR imaging, as compared with that of routine transverse MR imaging, in the assessment of distal tibiofibular syndesmosis injury, with arthroscopy as the reference standard.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCE IN KNOWLEDGE
 References
 
Patients
We retrospectively reviewed our study group, which comprised 45 ankles in 45 patients (26 men, 19 women; mean age, 32.1 years; range, 18–58 years) who had a history of repeated ankle sprain and who had undergone MR imaging and arthroscopic surgery between January 2001 and August 2004. All patients presented with ankle pain and were referred to our institute because of a lack of response to conservative treatment. None of the patients had a history of rheumatoid arthritis or any other seronegative arthritis. The mean time interval between the major episode causing the sprain and MR imaging was 64.3 weeks, with intervals ranging from 8 to 215 weeks. The interval between MR imaging and subsequent arthroscopic surgery ranged from 3 to 20 days. Arthroscopic ankle surgery was performed to treat lateral ankle injuries, soft-tissue impingement, loose bodies, or osteochondral lesions caused by the ankle sprain. The review board of Yonsei University College of Medicine approved our study; informed consent was waived by the review board.

MR Imaging
All MR imaging examinations were performed with a 1.5-T imager (Signa; GE Medical Systems, Milwaukee, Wis) and a dedicated extremity coil. The patient was laid in a supine position, with the ankle placed in a neutral position.

Our MR imaging study included the performance of a routine protocol set of transverse T1-weighted (repetition time msec/echo time msec, 700/11; matrix, 256 x 256; number of signals acquired, one; field of view, 12 cm; imaging time, 2 minutes 29 seconds) and T2- and intermediate-weighted (2000/20, 70; matrix, 256 x 256; number of signals acquired, one; field of view, 12 cm; imaging time, 6 minutes 48 seconds) dual spin-echo sequences in which the section thickness was 4 mm without an intersection gap. In addition, we performed a coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR sequence (20.9/2.2; flip angle, 15°; matrix, 256 x 192; number of signals acquired, one; field of view, 12 cm). A 9-cm-thick slab that was partitioned into 60 sections was applied for the ankle, resulting in a section thickness of 1.5 mm. Imaging with this sequence was performed before and 30 seconds after bolus intravenous injection of 0.1 mmol gadopentetate dimeglumine (Magnevist; Schering, Berlin, Germany) per kilogram of body weight. This sequence was performed in 4 minutes 22 seconds.

Arthroscopy of Ankles
Arthroscopic procedures were performed by one surgeon (J.W.L., an arthroscopic surgeon with 10 years of experience). Spinal or general anesthesia was administered, and the patients were placed in a supine position with the knee laid over the end of the operating table to allow the ankle to hang down by gravity. First, the ankle joint space was distended with saline to avoid iatrogenic injury to articular cartilage; arthroscopic procedures were then performed through anteromedial and anterolateral portals. Syndesmosis instability was considered present if insertion of the 2-mm-wide probe into the distal tibiofibular joint was possible, with a positive result in the "lateral shift test," which indicates that the fibula is displaced laterally when pushed with the probe. The criterion mentioned above is based on the study by Close (22), who reported that the maximum widening of the intraarticular distal tibiofibular syndesmosis was approximately 1.5 mm in a normal ankle. This criterion was also adopted in a previous report as the "stress test," in which instability was considered to be present with the arthroscopic visualization of more than 2 mm of widening of the distal tibiofibular syndesmosis when the ankle was moved from internal to external rotation (18). Coexisting ankle disorders were also examined.

Image Analysis
Three musculoskeletal radiologists (S.K., H.T.S., and S.A.L.; with 6, 7, and 6 years of experience, respectively) who were not informed of the patients' clinical history and arthroscopic results independently and retrospectively reviewed the two sets (routine and contrast enhanced) of MR images in random order. Analysis of each set was performed at an independent Digital Imaging and Communications in Medicine viewer (Centricity; GE Medical Systems). The interval between the review of the routine set and the review of the contrast-enhanced set was more than 2 weeks for all of the three readers to minimize bias. In addition, cases were reviewed in random order at each session to further minimize bias.

For both the routine and the contrast-enhanced sets, each observer recorded the presence of syndesmosis injury by using a five-point scoring method in the order of the probability of ligament injury on the basis of alteration of the AITFL: A score of 1 indicated that the AITFL was definitely not injured; a score of 2, that the AITFL was probably not injured; a score of 3, that the AITFL was possibly injured; a score of 4, that the AITFL was probably injured; and a score of 5, that the AITFL was definitely injured. The imaging features suggesting AITFL injury on the routine set were thickening, wavy contour, redundancy, discontinuity, or absence of the ligament (Fig 3). On the contrary, the imaging feature suggestive of AITFL injury on the contrast-enhanced set was linear or nodular enhancement of the AITFL, with or without continuity with the enhancement at the anterolateral gutter. Higher scores were given when the enhancement was separable from the enhancement at the anterolateral gutter and seemed certainly to be due to the AITFL itself. Contrast enhancement of and around the AITFL was regarded as ligamentous inflammation or impinged tissue within the syndesmosis (Fig 4).


Figure 3
View larger version (154K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: (a–d) Imaging findings of injured and noninjured AITFLs on routine transverse intermediate-weighted dual spin-echo MR images (2000/20; matrix, 256 x 256; number of signals acquired, one; field of view, 12 cm) obtained approximately at level of tibial plafond. When AITFL did not demonstrate discontinuity or thickening, it was considered not injured (arrow in a). AITFL was considered injured when it was thickened (arrow in b), redundant with a wavy contour (arrow in c), showing discontinuity (arrow in d), or absent (not shown).

 

Figure 3
View larger version (186K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: (a–d) Imaging findings of injured and noninjured AITFLs on routine transverse intermediate-weighted dual spin-echo MR images (2000/20; matrix, 256 x 256; number of signals acquired, one; field of view, 12 cm) obtained approximately at level of tibial plafond. When AITFL did not demonstrate discontinuity or thickening, it was considered not injured (arrow in a). AITFL was considered injured when it was thickened (arrow in b), redundant with a wavy contour (arrow in c), showing discontinuity (arrow in d), or absent (not shown).

 

Figure 3
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3c: (a–d) Imaging findings of injured and noninjured AITFLs on routine transverse intermediate-weighted dual spin-echo MR images (2000/20; matrix, 256 x 256; number of signals acquired, one; field of view, 12 cm) obtained approximately at level of tibial plafond. When AITFL did not demonstrate discontinuity or thickening, it was considered not injured (arrow in a). AITFL was considered injured when it was thickened (arrow in b), redundant with a wavy contour (arrow in c), showing discontinuity (arrow in d), or absent (not shown).

 

Figure 3
View larger version (181K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3d: (a–d) Imaging findings of injured and noninjured AITFLs on routine transverse intermediate-weighted dual spin-echo MR images (2000/20; matrix, 256 x 256; number of signals acquired, one; field of view, 12 cm) obtained approximately at level of tibial plafond. When AITFL did not demonstrate discontinuity or thickening, it was considered not injured (arrow in a). AITFL was considered injured when it was thickened (arrow in b), redundant with a wavy contour (arrow in c), showing discontinuity (arrow in d), or absent (not shown).

 

Figure 4
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a: (a–d) Imaging findings of injured and noninjured AITFLs on coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°; matrix, 256 x 192; number of signals acquired, one; field of view, 12 cm) obtained at level of AITFL. When AITFL did not demonstrate enhancement, it was considered not injured (arrow in a). AITFL was considered injured when it showed linear (arrows in b and c) or nodular (arrow in d) enhancement. Note enhancing soft tissue, which was considered to be synovial hypertrophy at the anterolateral gutter (arrowheads in b and d).

 

Figure 4
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b: (a–d) Imaging findings of injured and noninjured AITFLs on coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°; matrix, 256 x 192; number of signals acquired, one; field of view, 12 cm) obtained at level of AITFL. When AITFL did not demonstrate enhancement, it was considered not injured (arrow in a). AITFL was considered injured when it showed linear (arrows in b and c) or nodular (arrow in d) enhancement. Note enhancing soft tissue, which was considered to be synovial hypertrophy at the anterolateral gutter (arrowheads in b and d).

 

Figure 4
View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4c: (a–d) Imaging findings of injured and noninjured AITFLs on coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°; matrix, 256 x 192; number of signals acquired, one; field of view, 12 cm) obtained at level of AITFL. When AITFL did not demonstrate enhancement, it was considered not injured (arrow in a). AITFL was considered injured when it showed linear (arrows in b and c) or nodular (arrow in d) enhancement. Note enhancing soft tissue, which was considered to be synovial hypertrophy at the anterolateral gutter (arrowheads in b and d).

 

Figure 4
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4d: (a–d) Imaging findings of injured and noninjured AITFLs on coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°; matrix, 256 x 192; number of signals acquired, one; field of view, 12 cm) obtained at level of AITFL. When AITFL did not demonstrate enhancement, it was considered not injured (arrow in a). AITFL was considered injured when it showed linear (arrows in b and c) or nodular (arrow in d) enhancement. Note enhancing soft tissue, which was considered to be synovial hypertrophy at the anterolateral gutter (arrowheads in b and d).

 
On the contrast-enhanced set, existence of enhancing tissue delineating the syndesmotic recess was observed. If this enhancing tissue was present, its height was measured from the lateral talar dome to its highest point by one observer (S.K.) (Fig 5). The mean height was then compared between the two groups—the one with syndesmosis injury and the one without.


Figure 5
View larger version (112K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5: Measurement of height of enhancing tissue delineating syndesmotic recess on coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR image (20.9/2.2; flip angle, 15°; matrix, 256 x 192; number of signals acquired, one; field of view, 12 cm). Height was measured to nearest millimeter from lateral talar dome (a) to highest point (b). Enhancing tissue is deviated near tibia rather than fibula.

 
Statistical Analysis
Data were analyzed according to the receiver operating characteristic (ROC) analysis method by using a computer program (MedCalc, version 6.0; MedCalc Software, Mariakerke, Belgium). Each observer's performance in assessing distal tibiofibular syndesmosis injury was evaluated by using the area under the ROC curve (AUC). Differences in AUC values between the two sets were determined by using a z score test. To allow for generalization to the three readers, we used a multireader-multicase ROC approach with the jackknife method by using a computer program (LABMRMC 1.0, beta version 3; Charles E. Metz, University of Chicago, Chicago, Ill) (23). The statistical significance of the mean differences in AUC values between the routine and contrast-enhanced sets was reported at 95% confidence intervals.

Sensitivity, specificity, and accuracy of MR imaging in characterization of syndesmosis injury were calculated by using arthroscopic results as the reference standard. We considered scores of 4 and 5 as positive MR imaging results for both the routine set and the contrast-enhanced set. Each of the false-positive and false-negative results with the routine set was compared with the respective result with the contrast-enhanced set. False-positive and false-negative results in the contrast-enhanced set were evaluated, and comparison with arthroscopy was performed for false-positive cases to explain the possible reasons.

Quadratic weighted {kappa} tests were performed to assess interobserver variability among the three observers in judging syndesmosis injury with both the contrast-enhanced set and the routine set. Because our data were ranked, the weighted {kappa} test was performed to allow for differences in the importance of disagreement between scores (24). Three {kappa} values each were obtained for the contrast-enhanced set and the routine set, and mean {kappa} values for the two sets were obtained. The degrees of agreement were categorized as follows: a {kappa} value of less than 0.00 indicated poor agreement; a {kappa} value of 0.00–0.20, slight agreement; a {kappa} value of 0.21–0.40, fair agreement; a {kappa} value of 0.41–0.60, moderate agreement; a {kappa} value of 0.61–0.80, substantial agreement; and a {kappa} value of 0.81–1.00, almost perfect agreement (24).

The unpaired t test was performed to assess the significance of the difference in the average height of the enhancing tissue delineating the syndesmotic recess between the injured group and the uninjured group.

The {kappa} test and the t test were performed by using a computer program (MedCalc, version 6.0).

Histologic Examination of Syndesmotic Area
Cadaveric histologic examinations were performed to ascertain the existence of a synovial fold and the syndesmotic recess, as well as to compare the results with MR findings. Six fresh human ankles and feet were harvested from three unembalmed cadavers (one woman, two men; age range at death, 52–59 years) from the Department of Anatomy at Yonsei University College of Medicine. The institutional policies of Yonsei University College of Medicine were followed, and informed consent from the donors or their family members was obtained at the time of donation or before death for use of parts or all of the cadaver. Clinical histories before death were not available.

MR imaging and dissection were performed for all six ankles. MR imaging studies were performed with a coronal fat-suppressed fast spin-echo T2-weighted sequence (4000/60; matrix, 256 x 256; number of signals acquired, two; echo train length, 10; field of view, 12 cm) and were evaluated by a radiologist (S.K.) to ascertain the existence of a syndesmotic recess in the cadaveric specimen. Dissection was performed after MR imaging for all six ankles by an anatomist (I.H.C.) and a radiologist (S.K.) in collaboration for comparison with MR images. Three ankles were fully dissected; the syndesmotic ligaments were cut so that whole syndesmotic recesses and synovial folds could be seen.

Three ankles were not fully dissected to preserve them for later evaluation of their histologic features and were examined for their surface morphology only. These three ankles were prepared for histologic section and did not demonstrate abnormality in terms of either gross surface morphology or MR imaging findings. After being fixed, decalcified, and set into a paraffin block, these three specimens were trimmed to leave just the distal tibiofibular syndesmosis. The fixed specimen blocks were sliced by 2 µm for every 4 mm on the true coronal axis. Both a hematoxylin-eosin stain and a trichrome (Masson and Gomori) stain were performed. Prepared specimens were examined with a microscope at low (x1.25) and high (x400) magnification.

Analysis of the histologic features of the distal tibiofibular syndesmotic synovial fold and syndesmotic recess was performed by a histologist (J.E.L., with more than 20 years of experience). The synovial fold and syndesmotic recess were previously investigated by Sabacinski et al (25). In addition, separation of the synovial fold from the AITFL was examined at both dissection and histologic examination to ascertain separability of these two structures on MR images. The height of the syndesmotic recess was measured.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCE IN KNOWLEDGE
 References
 
Arthroscopic and MR Imaging Analysis of Ankles
Arthroscopic results confirmed that 24 of the 45 ankles demonstrated distal tibiofibular syndesmosis instability. For the patients with syndesmotic instability, additional arthroscopic diagnoses were as follows: fibrosed or hypertrophied synovium (n = 21), anterior talofibular ligament injury (n = 22), loose body (n = 4), osteochondral lesion (n = 8), cartilage defect of talus or tibia (n = 9), and/or fracture of lateral malleolus (n = 6). The remaining 21 ankles did not demonstrate syndesmosis injury, but other abnormalities such as fibrosis or hypertrophied synovium (n = 13), anterior talofibular ligament injury (n = 20), loose body (n = 1), osteochondral lesion (n = 5), and/or a cartilage defect of the talus or tibia (n = 8) were present.

For all three readers, the overall accuracy of diagnosing syndesmosis injury at MR imaging (Table 1) was significantly higher (P < .05) with the contrast-enhanced set (AUC values were 0.804 for reader 1, 0.812 for reader 2, and 0.845 for reader 3) than with the routine set (AUC values were 0.569 for reader 1, 0.594 for reader 2, and 0.549 for reader 3). Overall accuracy was also higher with the contrast-enhanced set (AUC = 0.854) than with the routine set (AUC = 0.537) when the results of all readers were combined.


View this table:
[in this window]
[in a new window]

  		Table 1. Mean AUC Values for Accuracy in Characterization of Syndesmosis for Each Reader and Each Technique

 
Assessment of sensitivity, specificity, and accuracy with both the routine set and the contrast-enhanced set with a cutoff score of 4 (Table 2) revealed that the contrast-enhanced set had greater sensitivity, specificity, and accuracy than the routine set in the diagnosis of syndesmosis injury according to all three readers.


View this table:
[in this window]
[in a new window]

  		Table 2. Diagnostic Performance of Each Technique in Assessment of Syndesmosis Injury

 
We could not detect syndesmosis injury with the routine set (reader 1 had nine, reader 2 had 13, and reader 3 had 11 false-negative results out of 24). These false-negative results were interpreted correctly with the contrast-enhanced set (seven of nine results for reader 1, 12 of 13 results for reader 2, and 10 of 11 results for reader 3) (Fig 6). On the other hand, there were false-positive results with the routine set (readers 1 and 2 had eight and reader 3 had 10 false-positive results out of 21). These false-positive results were interpreted correctly with the contrast-enhanced set (six of eight results for reader 1, five of eight results for reader 2, and eight of 10 results for reader 3) (Fig 7).


Figure 6
View larger version (184K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6a: (a, b) Transverse MR images at level of plafond in 20-year-old man with 3-year history of repeated ankle sprains. (a) T1-weighted (700/11) and (b) T2-weighted (2000/70) images show normal continuity of AITFL (arrow). (c, d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show nodular enhancement (arrow) at AITFL. Arthroscopy showed syndesmotic disruption. Routine set showed a false-negative while contrast-enhanced set showed a true-positive result. Focal enhancing lesion at lateral talar dome (arrowheads in c and d) was confirmed to be osteochondral fracture at arthroscopy.

 

Figure 6
View larger version (179K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6b: (a, b) Transverse MR images at level of plafond in 20-year-old man with 3-year history of repeated ankle sprains. (a) T1-weighted (700/11) and (b) T2-weighted (2000/70) images show normal continuity of AITFL (arrow). (c, d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show nodular enhancement (arrow) at AITFL. Arthroscopy showed syndesmotic disruption. Routine set showed a false-negative while contrast-enhanced set showed a true-positive result. Focal enhancing lesion at lateral talar dome (arrowheads in c and d) was confirmed to be osteochondral fracture at arthroscopy.

 

Figure 6
View larger version (183K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6c: (a, b) Transverse MR images at level of plafond in 20-year-old man with 3-year history of repeated ankle sprains. (a) T1-weighted (700/11) and (b) T2-weighted (2000/70) images show normal continuity of AITFL (arrow). (c, d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show nodular enhancement (arrow) at AITFL. Arthroscopy showed syndesmotic disruption. Routine set showed a false-negative while contrast-enhanced set showed a true-positive result. Focal enhancing lesion at lateral talar dome (arrowheads in c and d) was confirmed to be osteochondral fracture at arthroscopy.

 

Figure 6
View larger version (179K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6d: (a, b) Transverse MR images at level of plafond in 20-year-old man with 3-year history of repeated ankle sprains. (a) T1-weighted (700/11) and (b) T2-weighted (2000/70) images show normal continuity of AITFL (arrow). (c, d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show nodular enhancement (arrow) at AITFL. Arthroscopy showed syndesmotic disruption. Routine set showed a false-negative while contrast-enhanced set showed a true-positive result. Focal enhancing lesion at lateral talar dome (arrowheads in c and d) was confirmed to be osteochondral fracture at arthroscopy.

 

Figure 7
View larger version (181K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7a: (a–d) MR images in 40-year-old man with 8-month history of ankle sprains and pain. (a) T1-weighted (700/11) and (b) T2-weighted (2000/70) transverse MR images at level of plafond show wavy configuration and discontinuity of AITFL (arrow). (c, d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show no enhancement at AITFL (arrow). Arthroscopy showed no syndesmotic disruption. Routine set yielded a false-positive and contrast-enhanced set a true-negative result. The patient had a contusion at the medial talar dome (arrowhead).

 

Figure 7
View larger version (175K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7b: (a–d) MR images in 40-year-old man with 8-month history of ankle sprains and pain. (a) T1-weighted (700/11) and (b) T2-weighted (2000/70) transverse MR images at level of plafond show wavy configuration and discontinuity of AITFL (arrow). (c, d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show no enhancement at AITFL (arrow). Arthroscopy showed no syndesmotic disruption. Routine set yielded a false-positive and contrast-enhanced set a true-negative result. The patient had a contusion at the medial talar dome (arrowhead).

 

Figure 7
View larger version (181K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7c: (a–d) MR images in 40-year-old man with 8-month history of ankle sprains and pain. (a) T1-weighted (700/11) and (b) T2-weighted (2000/70) transverse MR images at level of plafond show wavy configuration and discontinuity of AITFL (arrow). (c, d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show no enhancement at AITFL (arrow). Arthroscopy showed no syndesmotic disruption. Routine set yielded a false-positive and contrast-enhanced set a true-negative result. The patient had a contusion at the medial talar dome (arrowhead).

 

Figure 7
View larger version (180K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7d: (a–d) MR images in 40-year-old man with 8-month history of ankle sprains and pain. (a) T1-weighted (700/11) and (b) T2-weighted (2000/70) transverse MR images at level of plafond show wavy configuration and discontinuity of AITFL (arrow). (c, d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show no enhancement at AITFL (arrow). Arthroscopy showed no syndesmotic disruption. Routine set yielded a false-positive and contrast-enhanced set a true-negative result. The patient had a contusion at the medial talar dome (arrowhead).

 
With the contrast-enhanced set, there were small numbers of false-negative results (two of 24 for reader 1 and one of 24 for readers 2 and 3). However, there were a considerable number of false-positive results (six of 21 for readers 1 and 3 and five of 21 for reader 2). All false-positive results were in patients with synovial hypertrophy in the anterolateral gutter.

The mean weighted {kappa} value representing the degree of interobserver agreement for the three readers with the contrast-enhanced set was 0.845 (almost perfect agreement). (Specifically, {kappa} values were 0.895 for reader 1 vs reader 2, 0.823 for reader 1 vs reader 3, and 0.818 for reader 2 vs reader 3.) The mean weighted {kappa} value for the three readers with the routine set was 0.417 (moderate agreement). (Specifically, {kappa} values were 0.557 for reader 1 vs reader 2, 0.362 for reader 1 vs reader 3, and 0.331 for reader 2 vs reader 3.) The degree of interobserver agreement was also higher with the contrast-enhanced set than with the routine set.

A thin, linear enhancing structure that was seen closer to the tibia than to the fibula within the syndesmosis was depicted on the entire contrast-enhanced set. The mean height of enhancing tissue delineating the syndesmotic recess in the contrast-enhanced set was 16.2 mm ± 3.3 (range, 12–24 mm) for the group with syndesmosis injury and 12.6 mm ± 5.1 (range, 6–19 mm) for the group without syndesmosis injury. There was a statistically significant difference (P < .05) in mean height between the two groups, although the standard deviation and the range of the heights were relatively high (Fig 8).


Figure 8
View larger version (154K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8a: (a–d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show enhancement of syndesmotic region in (a, b) noninjured and (c, d) injured syndesmoses. Relatively thin linear enhancement (arrowhead) near tibia is well visualized in a and b; on the basis of histologic results, this was presumed to be enhancing tissue delineating the syndesmotic recess. In the injured syndesmoses, enhancement (arrow) appears thick and nodular and extends further upward.

 

Figure 8
View larger version (145K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8b: (a–d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show enhancement of syndesmotic region in (a, b) noninjured and (c, d) injured syndesmoses. Relatively thin linear enhancement (arrowhead) near tibia is well visualized in a and b; on the basis of histologic results, this was presumed to be enhancing tissue delineating the syndesmotic recess. In the injured syndesmoses, enhancement (arrow) appears thick and nodular and extends further upward.

 

Figure 8
View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8c: (a–d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show enhancement of syndesmotic region in (a, b) noninjured and (c, d) injured syndesmoses. Relatively thin linear enhancement (arrowhead) near tibia is well visualized in a and b; on the basis of histologic results, this was presumed to be enhancing tissue delineating the syndesmotic recess. In the injured syndesmoses, enhancement (arrow) appears thick and nodular and extends further upward.

 

Figure 8
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8d: (a–d) Coronal contrast-enhanced frequency selective fat-suppressed 3D fast SPGR MR images (20.9/2.2; flip angle, 15°) show enhancement of syndesmotic region in (a, b) noninjured and (c, d) injured syndesmoses. Relatively thin linear enhancement (arrowhead) near tibia is well visualized in a and b; on the basis of histologic results, this was presumed to be enhancing tissue delineating the syndesmotic recess. In the injured syndesmoses, enhancement (arrow) appears thick and nodular and extends further upward.

 
Histologic Examination of Synovial Fold and Syndesmotic Recess
Grossly, the synovial fold and the syndesmotic recess were identified clearly in relation to the distal tibiofibular syndesmosis at dissection (Fig 9). MR imaging showed that the syndesmotic recess contained fluid with a high signal intensity that coursed close to the distal tibia (Fig 10a). Histologically, the syndesmotic recess was delineated from the synovial membrane by a single-cell layer (Fig 10). In all specimens, the syndesmotic recess was bordered medially by the distal tibia with a small amount of connective tissue and laterally by the synovial fold. The synovial fold was lined with a single-cell–lined synovial membrane and composed of loose connective tissue with collagenous fibers and adipose tissue. An abundance of vascular channels was present throughout the loose connective tissue around the syndesmotic recess (Fig 10c, 10d), which is expected to be enhanced on contrast-enhanced MR images. The synovial fold did not show any direct gross attachment with the AITFL at the joint level at both the gross and the histologic examinations. The heights of the syndesmotic recesses in these specimens were measured as between 0.5 and 1.0 cm.


Figure 9
View larger version (151K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 9a: Synovial fold and syndesmotic recess in dissected right ankle. (a) Inferior, (b) anterior, and (c) anteroinferior views show "teardrop " appearance of synovial fold (thick arrow). Syndesmotic recess (thin arrows) is seen between synovial fold and lateral aspect of distal tibia. F = fibula, T = tibia, Tl = talus, black arrowhead = cut AITFL during dissection, white arrowhead = cut posterior inferior tibiofibular ligament during dissection.

 

Figure 9
View larger version (165K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 9b: Synovial fold and syndesmotic recess in dissected right ankle. (a) Inferior, (b) anterior, and (c) anteroinferior views show "teardrop " appearance of synovial fold (thick arrow). Syndesmotic recess (thin arrows) is seen between synovial fold and lateral aspect of distal tibia. F = fibula, T = tibia, Tl = talus, black arrowhead = cut AITFL during dissection, white arrowhead = cut posterior inferior tibiofibular ligament during dissection.

 

Figure 9
View larger version (160K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 9c: Synovial fold and syndesmotic recess in dissected right ankle. (a) Inferior, (b) anterior, and (c) anteroinferior views show "teardrop " appearance of synovial fold (thick arrow). Syndesmotic recess (thin arrows) is seen between synovial fold and lateral aspect of distal tibia. F = fibula, T = tibia, Tl = talus, black arrowhead = cut AITFL during dissection, white arrowhead = cut posterior inferior tibiofibular ligament during dissection.

 

Figure 10
View larger version (95K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10a: (a) Coronal fat-saturated fast spin-echo T2-weighted MR image (4000/60; matrix, 256 x 256; number of signals acquired, two; echo train length, 10; field of view, 12 cm) and (b–d) photomicrographs of distal tibiofibular syndesmosis of cadaveric ankle specimen. MR image shows the syndesmotic recess contains high-signal-intensity fluid that courses close to the distal tibia. In b, the syndesmotic recess (SR) (arrowheads) separates the light-red synovial fold (SF) from the tibia and can be traced up to the interosseous ligament (open arrow). (Hematoxylin-eosin stain; original magnification, x1.25.) In c and d, the syndesmotic recess is also clearly visible and is lined by a single-celled synovial membrane (arrowheads). Abundant subsynovial vascular channels (arrows) in loose connective tissue of synovial fold are also seen. (Hematoxylin-eosin stain in c, Gomori trichrome stain in d; original magnification, x400.)

 

Figure 10
View larger version (166K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10b: (a) Coronal fat-saturated fast spin-echo T2-weighted MR image (4000/60; matrix, 256 x 256; number of signals acquired, two; echo train length, 10; field of view, 12 cm) and (b–d) photomicrographs of distal tibiofibular syndesmosis of cadaveric ankle specimen. MR image shows the syndesmotic recess contains high-signal-intensity fluid that courses close to the distal tibia. In b, the syndesmotic recess (SR) (arrowheads) separates the light-red synovial fold (SF) from the tibia and can be traced up to the interosseous ligament (open arrow). (Hematoxylin-eosin stain; original magnification, x1.25.) In c and d, the syndesmotic recess is also clearly visible and is lined by a single-celled synovial membrane (arrowheads). Abundant subsynovial vascular channels (arrows) in loose connective tissue of synovial fold are also seen. (Hematoxylin-eosin stain in c, Gomori trichrome stain in d; original magnification, x400.)

 

Figure 10
View larger version (78K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10c: (a) Coronal fat-saturated fast spin-echo T2-weighted MR image (4000/60; matrix, 256 x 256; number of signals acquired, two; echo train length, 10; field of view, 12 cm) and (b–d) photomicrographs of distal tibiofibular syndesmosis of cadaveric ankle specimen. MR image shows the syndesmotic recess contains high-signal-intensity fluid that courses close to the distal tibia. In b, the syndesmotic recess (SR) (arrowheads) separates the light-red synovial fold (SF) from the tibia and can be traced up to the interosseous ligament (open arrow). (Hematoxylin-eosin stain; original magnification, x1.25.) In c and d, the syndesmotic recess is also clearly visible and is lined by a single-celled synovial membrane (arrowheads). Abundant subsynovial vascular channels (arrows) in loose connective tissue of synovial fold are also seen. (Hematoxylin-eosin stain in c, Gomori trichrome stain in d; original magnification, x400.)

 

Figure 10
View larger version (82K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10d: (a) Coronal fat-saturated fast spin-echo T2-weighted MR image (4000/60; matrix, 256 x 256; number of signals acquired, two; echo train length, 10; field of view, 12 cm) and (b–d) photomicrographs of distal tibiofibular syndesmosis of cadaveric ankle specimen. MR image shows the syndesmotic recess contains high-signal-intensity fluid that courses close to the distal tibia. In b, the syndesmotic recess (SR) (arrowheads) separates the light-red synovial fold (SF) from the tibia and can be traced up to the interosseous ligament (open arrow). (Hematoxylin-eosin stain; original magnification, x1.25.) In c and d, the syndesmotic recess is also clearly visible and is lined by a single-celled synovial membrane (arrowheads). Abundant subsynovial vascular channels (arrows) in loose connective tissue of synovial fold are also seen. (Hematoxylin-eosin stain in c, Gomori trichrome stain in d; original magnification, x400.)

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCE IN KNOWLEDGE
 References
 
Distal tibiofibular syndesmosis injury is fixed with a syndesmotic cortical screw to limit instability of the ankle joint in patients with instability of the tibiotalar and/or tibiofibular joints. However, there exists some controversy regarding the treatment method and the merit of screw fixation (4,6,9,26–35). Distal tibiofibular syndesmosis injury has most commonly been thought to be associated with high fibular fractures such as the Weber C fracture or certain types of Weber B fractures (26–28,36,37). However, distal tibiofibular syndesmosis injury can also result from severe or high ankle sprains that result in a purely ligamentous injury without an obvious fracture (3,4,6,9). These isolated ligamentous injuries do not often result in gross disruption of the syndesmosis (30).

Disruption of the AITFL is thought to be a reliable indicator of syndesmosis injury at MR imaging because the AITFL is the syndesmotic stabilizer most vulnerable to injury (11,38). Rupture of the AITFL can be diagnosed at routine transverse MR imaging by using features such as discontinuity, waviness, an irregular or curved configuration, and nonvisualization of the ligament (10,12,18,38). Vogl et al (11) reported that the injured ligament could be enhanced on a contrast-enhanced T1-weighted MR image.

As compared with results in previous reports (10,18), our results with the routine set revealed diagnostic performance in terms of sensitivity, specificity, and accuracy that was lower than expected. The oblique course and/or multifasciculation of the AITFL may lower diagnostic performance in general (10,13,39), but relatively poor results with the routine set in our study may have been caused by the chronicity of the injury in our patients. In chronic injuries, the injured ligament can show variable configurations while healing, and its signal alteration can return to normal and therefore render it indistinguishable from uninjured ligaments. It has been shown that at 2 or more months after an injury, the features of the injured ligaments can be thickened, completely absent, or completely normal (20,40,41). Moreover, interpretation of the thickening of a ligament could be subjective. In our study, there was a mean time interval of 64.3 weeks between the onset of symptoms and MR imaging. In contrast, patients with acute syndesmosis injury were included in the studies of Uys and Rijke (12), who performed MR imaging within 1 week, and Vogl et al (11), who performed MR imaging within 3 days. Other articles have not described the time interval between symptom onset and MR imaging (10,18,38).

When the contrast-enhanced set and the routine set analyses were compared for the same patient group, the contrast-enhanced set had better diagnostic performance at ROC analysis and also had better sensitivity, specificity, and accuracy for the identification of syndesmosis injuries. These results suggest that the coronal gadolinium-enhanced fat-suppressed 3D fast SPGR sequence is more sensitive, specific, and accurate than routine transverse T1-, T2-, and intermediate-weighted sequences in the assessment of chronic syndesmosis injury. However, the contrast-enhanced set had the drawback of a relatively high false-positive rate. At arthroscopy, patients had synovial hypertrophy (21 of 24 patients in the injured group and 13 of 21 patients in the noninjured group), and all false-positive results were in patients with synovial hypertrophy in the anterolateral gutter. We speculate that this inflamed synovial tissue could be enhanced and mistaken for the injured AITFL. Thus, we believe that careful distinction of the enhanced injured AITFL from the enhanced synovial tissue in the anterolateral gutter could increase accuracy in the assessment of AITFL injury. Because the synovial fold is not attached to the AITFL at the joint level (25), enhancement at the AITFL injury should be separable from synovial tissue enhancement. Enhancement of the injured capsule adjacent to the AITFL or joint fluid enhancement caused by contrast agent diffusion may also be mistaken for enhancement of the AITFL injury, although we minimized the imaging time (4 minutes 22 seconds) to lessen the diffusion effect (42).

A thin, linear enhancing structure within the syndesmosis was depicted on the entire contrast-enhanced set and was seen closer to the tibia than to the fibula. The synovial fold and syndesmotic recess have been investigated and discussed in some histology and radiology articles (25,38,39). However, to the best of our knowledge, no previous report stresses the linear enhancing structure delineating the syndesmotic recess at MR imaging that we addressed here. On the basis of our histologic correlation with cadaveric specimens and results of previous studies (25,39), we speculate that this enhancing tissue corresponds to the subsynovial vascularized tissue along the syndesmotic recess. The syndesmotic recess (distal tibiofibular recess) is the fluid-filled space extending upward from the ankle joint; it is covered laterally by a single-cell–lined synovium of the synovial fold and is closely attached to the tibia medially.

As expected, the mean height of the enhancing tissue delineating the syndesmotic recess was significantly greater in the injured group (16.2 mm ± 3.3) than in the noninjured group (12.6 mm ± 5.1). However, the height of the enhancing tissue was greater than the height of the syndesmotic recess noted in a previous report (25). This discrepancy could be due to the fact that enhancing tissue delineating the syndesmotic recess is inherently longer than the fluid in the syndesmotic recess. Another possible reason for the discrepancy is the method of measurement. We measured the height from the lateral talar dome to the highest extent of the enhancing tissue delineating the syndesmotic recess for objective measurement, whereas histologic measurement was performed within just the distal tibiofibular syndesmosis complex (25). We expect that an abnormal upward extension of this enhancing tissue could be an ancillary sign of syndesmotic instability caused by syndesmotic disruption, although further investigation is needed to analyze the pattern and height of this enhancing tissue delineating the syndesmotic recess.

There were some limitations of our study. First, we adopted injury of the AITFL as a marker of syndesmosis injury. We did not use the MR imaging appearance of altered syndesmosis and other stabilizing ligaments. But, as we noted previously, the AITFL could be a marker of syndesmosis injury because it is the most vulnerable ligament and is most reliably and consistently visualized at MR imaging, and the syndesmosis itself is not a clearly defined structure in MR imaging. Second, we used an arthroscopic criterion that defines syndesmosis injury as when a 2-mm probe can be placed into the distal tibiofibular syndesmosis and yield a positive lateral fibular shift test result. This could be a source of error because, in some cases, we did not see the syndesmosis and the AITFL disruption directly.

Third, we did not compare conventional coronal T2-weighted contrast-enhanced images with our contrast-enhanced images because the former kind of image was not included in our routine protocol. Fourth, selection bias could have been present. The study patients were referred to our tertiary hospital because of a lack of response to conservative treatment. That could be not only the reason for the high prevalence of syndesmosis injury in our patient group but also the reason our patient group showed chronicity of syndesmosis injury. Fifth, we compared the contrast-enhanced MR images with cadaveric specimens indirectly. This indirect comparison was inevitable because contrast-enhanced MR imaging of cadaveric specimens is inherently impossible. Last, we did not include a control group because a strict control group was difficult to create in that such a group should consist of individuals who have never had an ankle injury. Ankle injuries are very common and can be forgotten by patients.

In conclusion, a coronal gadolinium-enhanced fat-suppressed 3D fast SPGR sequence was more sensitive, specific, and accurate than routine transverse T1-, T2-, and intermediate-weighted sequences in the assessment of chronic syndesmosis injury. Also, enhancing tissue that abnormally extends upward in the syndesmosis may be an ancillary sign of syndesmotic disruption.


   ADVANCE IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCE IN KNOWLEDGE
References
 


   ACKNOWLEDGMENTS
 
The authors gratefully thank Joon-Seok Lim, MD, in the Department of Diagnostic Radiology of Yonsei University, College of Medicine, and Seong Ho Park, MD, in the Department of Radiology, University of Ulsan College of Medicine, Asan Medical Center, Seoul, Korea, for perceptive statistical comments and help. The authors also thank Hwa-Seon Kim, MD, in the Department of Diagnostic Radiology of Yonsei University, College of Medicine, for editing the manuscript.


   FOOTNOTES
 

Abbreviations: AITFL = anterior inferior tibiofibular ligament • AUC = area under the ROC curve • ROC = receiver operating characteristic • SPGR = spoiled gradient-recalled acquisition in the steady state • 3D = three-dimensional

2 Current address: Department of Diagnostic Radiology, Hanyang University, College of Medicine, Kuri City, Kyeonggi-do, Korea Back

Authors stated no financial relationship to disclose.

Author contributions: Guarantors of integrity of entire study, S.K., J.S.S.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, all authors; clinical studies, S.K., Y.M.H., H.T.S., S.A.L., J.W.L., J.S.S.; experimental studies, S.K., J.E.L., I.H.C., J.S.S.; statistical analysis, S.K.; and manuscript editing, S.K., J.S.S.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCE IN KNOWLEDGE
 References
 

  1. Cedell CA. Ankle lesions. Acta Orthop Scand 1975;46:425–445.[Medline]
  2. Gerber JP, Williams GN, Scoville CR, Arciero RA, Taylor DC. Persistent disability associated with ankle sprains: a prospective examination of an athletic population. Foot Ankle Int 1998;19:653–660.[Medline]