Musculoskeletal Imaging Original Research

Transcription

1 Musculoskeletal Imaging Original Research Use of 3D SPCE Sequences for ssessment of nkle Syndesmosis Injury Musculoskeletal Imaging Original Research Minchul Kim 1 Yun Sun Choi 1 Min Sun Jeong 1 Mira Park 2 Tong Jin Chun 1 Jin Su Kim 3 Ki Won Young 3 Kim M, Choi YS, Jeong MS, et al. Keywords: 3.0 T, 3D isotropic, 3D SPCE, MRI, tibiofibular syndesmosis DOI: /JR Received July 11, 2016; accepted after revision September 11, Department of Radiology, Eulji University Medical Center, 68 Hangeulbiseok-ro, Nowon-gu, Seoul, 01830, Korea. ddress correspondence to Y. S. Choi (cys0128@eulji.ac.kr). 2 Department of Preventive Medicine, School of Medicine, Eulji University, Daejeon, Korea. 3 Department of Orthopedic Surgery, Eulji University Medical Center, Seoul, Korea. This article is available for credit. JR 2017; 208: X/17/ merican Roentgen Ray Society Comprehensive ssessment of nkle Syndesmosis Injury Using 3D Isotropic Turbo Spin-Echo Sequences: Diagnostic Performance Compared With That of Conventional and Oblique 3-T MRI OJECTIVE. The objective of this study is to evaluate the diagnostic performance of 3D sampling perfection with application-optimized contrasts using different flip angle evolution (SPCE) sequences in the evaluation of ankle syndesmosis injuries, compared with that of conventional orthogonal MRI and oblique proton density weighted turbo spin-echo (TSE) sequences performed with 3-T MRI. MTERILS ND METHODS. This retrospective study included 98 patients with suspected ankle syndesmosis injuries who underwent both MRI and surgery. Fifty patients (mean [± SD] age, 34.5 ± 15.3 years) had acute injuries, and 48 patients (mean age, 28.5 ± 9.6 years) had chronic injuries. For both groups, the diagnostic performance of each MRI sequence with regard to syndesmosis injuries was evaluated. rthroscopy findings were used as a reference standard to confirm diagnosis. RESULTS. No statistically significant differences in the sensitivity, specificity, accuracy, and UC values were noted between images of the syndesmosis obtained using proton density weighted TSE sequences and images of the syndesmosis obtained using 3D SPCE sequences (p > 0.05). Interobserver agreement regarding the diagnosis of both acute and chronic syndesmosis injuries was almost perfect for proton density weighted TSE images (κ > 0.80) and was substantial for 3D SPCE images (κ > 0.75). oth methods of obtaining images of the syndesmosis had a diagnostic performance superior to that of conventional orthogonal MRI. CONCLUSION. The performance of 3D SPCE sequences is comparable to that of 2D proton density weighted MR images for the diagnosis of acute and chronic syndesmosis injuries. T he distal tibiofibular syndesmosis is a complex structure that is stabilized by four separate ligaments: the anterior inferior tibiofibular ligament (ITFL), the posterior inferior tibiofibular ligament (PITFL), the transverse tibiofibular ligament, and the interosseous ligament. distal tibiofibular syndesmosis injury may develop as a result of severe ankle trauma as well as repeated ankle sprains; the frequency of such injuries ranges from 1% to 20% [1]. If not accurately diagnosed and properly managed, these injuries may lead to posttraumatic arthritic changes and chondral defects over time [2, 3]. The value of MRI for the visualization of acute and chronic syndesmotic injuries has been reported in several studies [4 8]. The fact that the distal ITFL and PITFL run obliquely to the orthogonal planes is a matter of concern [9] because a scan in the axial plane may lead to a false-positive interpretation of the presence of a syndesmotic injury when MRI results are compared with surgical findings [7, 10]. Studies have shown that, when compared with images acquired in conventional orthogonal planes, images acquired in the oblique plane at an angle of approximately 45 provide a better depiction of the ITFL and PITFL and allow better evaluation of the integrity of these ligaments [11, 12]. Three-dimensional MRI with an isotropic fat-saturated T2-weighted turbo spinecho (TSE) sampling perfection with application-optimized contrasts using variable flip angle evolution (SPCE) sequence can be used to acquire multiplanar reformatted images and can shorten the total acquisition time by obviating the need to acquire the same sequence in different planes. The ankle is one of the most complex human joints, and most of its ligaments and tendons are oriented obliquely; therefore, it may be difficult to evaluate the ankle with sequences oriented in standard planes. Similar techniques include the Cube technique (formally JR:208, pril

2 known as fast spin-echo extended echo-train acquisition) designed by GE Healthcare and a volume isotropic TSE acquisition sequence (Vista, Philips Healthcare). Previous studies showed that the performance of 3D isotropic T2-weighted TSE MRI in diagnosing internal derangements of the ankle is comparable to that of conventional 2D MRI sequences [13 15]. However, to our knowledge, no reports have compared the performance of 3D SPCE sequences with that of conventional 2D MRI of the ankle for the diagnostic evaluation of acute and chronic distal syndesmosis ligament injuries. In the present study, we evaluate whether a 3D SPCE sequence formatted into an oblique plane at an angle of 45 has a diagnostic performance similar to that of 2D proton density weighted TSE images taken at an angle of 45 and whether it is superior to conventional orthogonal plane MRI for the evaluation of distal syndesmosis injury. Materials and Methods Case Selection and Clinical Diagnosis The retrospective study was approved by the institutional review board at Eulji University Medical Center. The requirement for informed consent was waived. We evaluated 50 consecutive patients with a suspected acute distal syndesmosis injury of the ankle and 50 consecutive patients with suspected chronic ankle instability. Patients were selected for inclusion in the study on the basis of their history, symptoms, and physical examination findings and on the basis of medical records indicating that they underwent both MRI and surgery between January 2013 and ugust We excluded patients who had a previous history of surgery performed on the affected ankle (thereby excluding one patient in the chronic injury group), a comorbid infection, or an underlying systemic pathologic finding, such as gouty arthritis (thereby excluding one additional patient in the chronic injury group). The final study group therefore included 50 patients with acute injuries and 48 patients with chronic injuries. The acute injury group consisted of 34 male patients and 16 female patients (mean [± SD] age, 34.5 ± 15.3 years; range, years). The chronic injury group consisted of 30 male patients and 18 female patients (mean age, 28.5 ± 9.6 years; range, years). Forty-three fractures were found in the acute injury group. The mean interval between MRI and surgery was 2.8 ± 1.6 days. MRI Techniques MRI studies were performed using a 3-T imaging system (Skyra, Siemens Healthcare). The patients were examined while in a supine position with the ankle in a neutral position, with the use of a phased-array foot-and-ankle coil with 16 channels. sagittal T2-weighted fat-suppression image was obtained using the Dixon technique with a TR/TE of 4100/73, a turbo factor of 19, and a slice thickness of 3 mm with no interslice gap. Sagittal, axial, coronal, and coronal fat-saturated proton density weighted TSE images were obtained with a TR of , an effective TE of 22 33, a turbo factor of 6 or 7, a slice thickness of 3 mm with no interslice gap, and an FOV of cm. matrix of was obtained with one or two excitations. n axial fat-saturated T2-weighted TSE image was obtained with a TR of 3130, an effective TE of 78, a turbo factor of 11, and a matrix of dditional proton density weighted TSE MR views of the syndesmosis obtained in the oblique plane at an angle of 45 (Fig. 1) and 3D SPCE sequences were also obtained. Images acquired using a 3D SPCE sequence were subsequently reformatted into an oblique plane at an angle of 45, with a slice thickness of 0.6 mm, by technicians using commercially available software (Syngo MR, Siemens Healthcare) [16]. The specific parameters used for each sequence are summarized in Table 1. MRI nalysis The diagnostic value of conventional plane 2D images (method 1, which included axial and coronal proton density weighted TSE, axial fat-sat- TLE 1: Imaging Parameters for MRI Sequences Used in the Evaluation of nkle Syndesmosis Injury Imaging Parameter xial Proton Density Weighted TSE xial FS T2-Weighted TSE Method 1 Coronal Proton Density Weighted TSE Fig. 1 Coronal MR image indicating 45 angle oblique image plane for distal tibiofibular syndesmosis. In coronal plane, 45 angle is related to tibial plafond. Coronal FS Proton Density Weighted TSE Method 2 a Method 3 b TR/TE / / / / / /50 Flip angle ( ) Matrix size FOV (cm) Section thickness (mm) Intersection gap (mm) andwidth (Hz/pixel) Echo-train length Mean no. of signals Scan time 2 min 22 s 2 min 2 s 2 min 1 s 2 min 12 s 3 min 38 s 6 min 2 s Note TSE = turbo spin-echo, FS = fat-saturated. a Proton density weighted turbo spin-echo (TSE) image obtained in oblique plane at 45 angle. b Three-dimensional sampling perfection with application-optimized contrasts using different flip angle evolution (SPCE) sequence reformatted into an oblique plane at JR:208, pril 2017

3 urated T2 TSE, and coronal fat-saturated proton Use of 3D SPCE Sequences for ssessment of nkle Syndesmosis Injury density weighted TSE images), proton density weighted TSE images obtained in the oblique plane at an angle of 45 (method 2), and 3D SPCE reformatted images reformatted into an oblique plane at an angle of 45 (method 3) were independently evaluated by two radiologists who had 12 years and 3 years of experience in musculoskeletal imaging. First, the radiologists independently graded anterior and posterior syndesmosis with the use of method 1 alone. Neither radiologist had access to clinical information or radiology reports. To prevent recall bias, the radiologists reviewed each method in separate sittings occurring at least 2 weeks apart (i.e., method 2 was viewed 2 weeks after the evaluation of method 1, and method 3 was viewed 2 weeks after the evaluation of method 2). To grade acute syndesmosis injury on a 4-point scale, findings for the distal tibiofibular ligament and information on the presence of a bony avulsion were included in the overall findings, to indicate the presence of anterior or posterior syndesmotic injury. score of 0 denoted normal syndesmosis (i.e., a normal ligament without bony avulsion); 1, periligamentous edema without bony avulsion; 2, partially ruptured syndesmosis presenting as laxity, irregular contour, or partial discontinuity without bony avulsion; and 3, completely ruptured (discontinuous or invisible) syndesmosis (i.e., a ruptured ligament without a bony avulsion or an intact ligament with a bony avulsion) [11, 17]. In terms of appearance, a normal ITFL has a triangular shape in the anteroposterior direction and consists of three bundles that are seen as fascicles that slightly converge in the laterodistal direction. normal PITFL has a quadrilateral shape but also shows fascicles [9, 12]. The imaging features that suggested chronic syndesmosis injury on MRI were thickening, thinning, wavy contour, redundancy, discontinuity, or absence of the ligament [1]. Ligament injury was considered to be present if any one of the aforementioned criteria were fulfilled. For all three methods, each observer recorded the presence of syndesmosis injury by use of a 4-point grading method that indicated the probability of ligament injury. score of 0 indicated that the ligament had a normal appearance; 1, that the ligament probably was not injured; 2, that the ligament was probably injured; and 3, that the ligament was definitely injured. Surgery Two orthopedic surgeons who specialized in foot and ankle surgery performed all surgical procedures by means of arthroscopic evaluation of the distal tibiofibular syndesmosis. To uphold the standard of care, the surgeons were not blinded to the MRI results. Syndesmosis ligament complexes were evaluated using arthroscopy performed with the use of standard anterolateral and C D Fig year-old man with distal tibiofibular fracture. and, Coronal proton density weighted turbo spin-echo (TSE) () and axial 2D fat-saturated T2-weighted TSE () MR images obtained using 2D conventional MRI (method 1) show uncertain continuity of anterior inferior tibiofibular ligament (ITFL) (arrow). T = tibia, F = fibula. C and D, Proton density weighted TSE image (C) obtained in oblique plane at 45 angle (method 2) and 3D sampling perfection with application-optimized contrasts using different flip angle evolution (SPCE) sequence reformatted into oblique plane at 45 (D) (method 3) show mild thinning but intact ITFL (arrow). Patient underwent internal fixation of distal tibiofibular facture. ITFL was intact at surgery. T = tibia, F = fibula. C D Fig year-old woman with chronic ankle injury. and, Coronal () and axial () proton density weighted turbo spin-echo (TSE) MR images obtained by 2D conventional MRI (method 1) show loss of multifascicular pattern in anterior inferior tibiofibula ligament (ITFL) (arrow). F = fibula, T = tibia. C and D, Proton density weighted TSE image obtained in oblique plane at 45 angle (C) (method 2), and 3D sampling perfection with application-optimized contrasts using different flip angle evolution (SPCE) sequence reformatted into oblique plane at 45 (D) (method 3) reveal wavy redundant ITFL (arrow). Surgical diagnosis was syndesmotic instability of ITFL with diastasis. F = fibula, T = tibia. JR:208, pril

4 anteromedial portals. The ITFL was well visualized through the anteromedial portal and could be evaluated along its entire intraarticular course. The PITFL was well visualized through both portals and could be reliably evaluated for midsubstance tearing [18]. chronic syndesmotic abnormality was considered to be present when there was distal tibiofibular diastasis of 4 mm or greater; an obturator (2.9-mm diameter) could be inserted arthroscopically between the tibiofibular diastasis [4, 19, 20]. If present, chronic syndesmotic instability was treated by débridement of the hypertrophied syndesmotic tissue, followed by fixation of the tibia and fibula with the use of a suture button or a 5.0-mm cannulated screw placed 1.5 cm above the tibiotalar articular surface. Forty-three of 50 acutely injured patients had associated ankle fractures that were treated by internal fixation surgery and reconstruction of the syndesmosis. Statistical nalysis Statistical analyses were performed using statistical software (SPSS software, version 20.0, SPSS- IM; and MedCalc for Windows, version 16.4, MedCalc Software), for comparison of UC values. Surgically confirmed diagnoses were used as reference standards. To compare the performance of the three methods for diagnosing acute and chronic distal syndesmosis ligament tears, we calculated the sensitivity, specificity, and accuracy for each reader. The differences in sensitivity and specificity were validated using McNemar statistics. The UC value was used to compare the differences between the methods and was interpreted as the mean sensitivity value for all possible values of specificity; an UC value is also a measure of the overall performance of a diagnostic test. n UC value can be any value between 0 and 1, and the higher the value in this range, the better the overall performance of a diagnostic test [21]. ecause only one acute and one chronic PITFL tear were confirmed at surgery, UC values for ITFL only were analyzed. Interreader agreement was evaluated using kappa statistics. For the aforementioned statistical analyses, p < 0.05 was considered to denote statistical significance. The kappa values used to define interreader agreement in the statistical analyses were interpreted as follows: poor (κ 0.1), slight (0.1 < κ 0.2), fair (0.2 < κ 0.4), moderate (0.4 < κ 0.6), substantial (0.6 < κ 0.8), and almost perfect (0.8 < κ 1.0) agreement. Fig year-old man with fracture of posteromedial tubercle of tibia., Proton density weighted turbo spin-echo image obtained in oblique plane at 45 angle (method 2) shows anterior inferior tibiofibula ligament (ITFL) with definite full-thickness tear (arrow) at fibular attachment site. oth readers identified grade 3 tear. F = fibula, T = tibia., Three-dimensional sampling perfection with application-optimized contrasts using different flip angle evolution (SPCE) sequence reformatted into oblique plane at 45 angle (method 3) shows swollen area but some partial continuity of ITFL (arrow). oth readers identified grade 2 tear. Surgical diagnosis was full-thickness tear of ITFL. F = fibula, T = tibia. Results rthroscopic results confirmed that 39 of the 50 patients with acute ankle injuries had ITFL tears (37 complete and two partial tears) and one bony avulsion of the PITFL. mong patients with chronic injury, 25 of 48 ankles had ITFL instability and one PIT- FL was affected. Fibrous or hypertrophied C D Fig year-old man with chronic ankle instability and sense of giving way. and, Coronal proton density weighted turbo spin-echo (TSE) () and axial 2D fat-saturated T2-weighted TSE () MR images, obtained by 2D conventional MRI (method 1) show multifascicular anterior inferior tibiofibula ligament (ITFL) (arrow). oth readers assigned injury score of 0. C, Proton density weighted TSE image obtained in oblique plane at 45 angle (method 2) shows ITFL (arrow) with mild redundancy but preserved continuity. Synovial thickening (arrowhead) is also seen. oth readers assigned injury score of 1. F = fibula, T = tibia. D, Three-dimensional sampling perfection with application-optimized contrasts using different flip angle evolution (SPCE) sequence reformatted into oblique plane at 45 (method 3) shows uncertain continuity and marked thinning of ITFL (arrow). Surgical diagnosis was tear of ITFL. oth readers assigned injury score of 2. F = fibula, T = tibia. 830 JR:208, pril 2017

5 Use of 3D SPCE Sequences for ssessment of nkle Syndesmosis Injury TLE 2: Sensitivity, Specificity, and ccuracy of Each Method Used for the Diagnosis of cute Syndesmosis Ligament Injuries Sensitivity Specificity ccuracy Method and Ligament Reader 1 Reader 2 Reader 1 Reader 2 Reader 1 Reader 2 Method 1 ITFL 89.7 (35/39) 97.4 (38/39) 27.3 (3/11) 54.5 (6/11) 76.0 (38/50) 88.0 (44/50) PITFL 100 (1/1) 100 (1/1) 79.6 (39/49) 89.8 (44/49) 80.0 (40/50) 90.0 (45/50) Method 2 ITFL 100 (39/39) 100 (39/39) 100 (11/11) 90.9 (10/11) 100 (50/50) 98.0 (49/50) PITFL 100 (1/1) 100 (1/1) 100 (49/49) 98.0 (48/49) 100 (50/50) 98.0 (49/50) Method 3 ITFL 100 (39/39) 100 (39/39) 81.8 (9/11) 90.9 (10/11) 96.0 (48/50) 98.0 (49/50) PITFL 100 (1/1) 100 (1/1) 100 (49/49) 100 (49/49) 100 (50/50) 100 (50/50) Note Data are the percentage of injured ligament, with number of injured ligaments/total number of ligaments shown in parentheses. Normal and edematous ligaments (grade 0 and 1, respectively) and partially and completely ruptured ligaments (grade 2 and 3, respectively) are grouped together. Method 1 = 2D conventional MRI, ITFL = anterior inferior tibiofibular ligament, PITFL = posterior inferior tibiofibular ligament, Method 2 = proton density weighted turbo spin-echo image obtained in oblique plane at 45 angle, Method 3 = 3D sampling perfection with application-optimized contrasts using different flip angle evolution (SPCE) image reformatted into an oblique plane at a 45 angle. synovium was confirmed in 14 patients with chronic ankle injuries. For method 1, interobserver agreement was moderate (κ = 0.59) for patients with acute injuries and substantial for patients with chronic injuries (κ = 0.61). When images obtained using method 2 were considered, interobserver agreement between the readers was near perfect (κ > 0.80). Substantial interobserver agreement (κ > 0.75) was noted for images obtained using method 3. oth readers found that methods 2 and 3 had greater sensitivity, specificity, and accuracy for the diagnosis of acute injury than did method 1 (Table 2). No statistically significant differences were noted between methods 2 and 3 (p = , for both readers). For the diagnosis of acute PITFL injury, method 1 had statistically significantly lower accuracy (p = , for both readers) than did methods 2 and 3. The fact that method 1 generated false-positive results that were correctly interpreted using methods 2 and 3 suggested that method 1 had low specificity (Fig. 2). ssessment of chronic syndesmosis injury with a cutoff score of 2 (Table 3) revealed that images in the oblique plane at an angle of 45 (i.e., images obtained using methods 2 and 3) had greater sensitivity, specificity, and accuracy for the diagnosis of ITFL injury than did the routine set of images (i.e., those obtained using method 1), and the difference was found to be statistically significant (p = , for both readers). However, no statistically significant difference was noted between methods 2 and 3 (p = , for both readers) (Fig. 3). The overall diagnostic accuracy was lower than that in the acute injury group, with no method showing accuracy of more than 80%. Table 4 summarizes the estimated UC values for acute and chronic ITFL injuries. For both readers, the overall accuracy of diagnosing acute syndesmosis injury with the use TLE 3: Sensitivity, Specificity, and ccuracy of Each Method Used for the Diagnosis of Chronic Syndesmosis Ligament Injuries Method and Ligament Sensitivity Specificity ccuracy Reader 1 Reader 2 Reader 1 Reader 2 Reader 1 Reader 2 Method 1 ITFL 64.0 (16/25) 60.0 (15/25) 69.6 (16/23) 47.8 (11/23) 66.7 (32/48) 54.2 (26/48) PITFL 100 (1/1) 100 (1/1) 100 (47/47) 100 (47/47) 100 (48/48) 100 (48/48) Method 2 ITFL 92.0 (23/25) 92.0 (23/25) 56.5 (13/23) 34.8 (8/23) 75.0 (36/48) 64.6 (31/48) PITFL 100 (1/1) 100 (1/1) 100 (47/47) 100 (47/47) 100 (48/48) 100 (48/48) Method 3 ITFL 92.0 (23/25) 88.0 (22/25) 65.2 (15/23) 47.8 (11/23) 79.2 (38/48) 68.8 (33/48) PITFL 100 (1/1) 100 (1/1) 100 (47/47) 100 (47/47) 100 (48/48) 100 (48/48) Note Data are the percentage of injured ligaments, with number of injured ligaments/total number of ligaments shown in parentheses. Normal and edematous ligaments (grade 0 and 1, respectively) and partially and completely ruptured ligaments (grade 2 and 3, respectively) are grouped together. Method 1 = 2D conventional MRI, ITFL = anterior inferior tibiofibular ligament, PITFL = posterior inferior tibiofibular ligament, Method 2 = proton density weighted turbo spin-echo image obtained in oblique plane at 45 angle, Method 3 = 3D sampling perfection with application-optimized contrasts using different flip angle evolution (SPCE) image reformatted into an oblique plane at a 45 angle. JR:208, pril

6 TLE 4: Mean UC Values for the ccuracy of nterior Inferior Tibiofibular Ligament Evaluation for Each Reader and Each Method Reader and Method of MRI was higher for images obtained in the oblique plane at an angle of 45 (i.e., images obtained using methods 2 and 3, for which UC values were ) than for the routine set of images (for which UC values were 0.710, for reader 1, and 0.866, for reader 2). For chronic ITFL injury, images obtained in the oblique plane at an angle of 45 (i.e., images obtained using methods 2 and 3, for which UC values were ) also had a higher UC value than did images obtained using method 1. For the acute injury group, method 2 had a better diagnostic performance than did method 3 (Fig. 4), whereas for the chronic injury group, method 3 had a better diagnostic performance (Fig. 5), although the difference was not statistically significant (p = , for both readers). For both readers and for each method, the diagnostic accuracy was greater for the acute injury group than for the chronic injury group. cute Injury Group Chronic Injury Group UC Value 95% CI p UC Value 95% CI p Reader 1 Method Method < < Method < < Difference a a b a b Reader 2 Method Method < Method < Difference a a b a b a Data are mean differences in UC values between method 2 and method 3. b For the comparison of UC values for method 2 and method 3, as calculated using the DeLong method. Discussion Three-dimensional MRI techniques allow delineation of anatomic details within an image volume and may be useful for 3D reformation in arbitrary orientations [22]. Highly resolved 3D TSE sequences are suitable for depicting complex ankle anatomy. study by Notohamiprodjo et al. [15] reported that the use of MRI with a 3D SPCE sequence is a comprehensive approach that helps to visualize complex ankle anatomy, including distal syndesmosis, at a sufficient signal-tonoise ratio and carrier-to-noise ratio. For acute and chronic syndesmosis injuries, the diagnostic performance (i.e., sensitivity, specificity, and accuracy) of our methods was similar to that noted in previous reports [1, 10, 12]. For other 3D TSE MR images, including those obtained using a volume isotropic TSE acquisition sequence (Vista, Philips Healthcare) and fast spinecho extended echo-train acquisition sequences, increased image blurring, relatively poor image sharpness, and indistinctness of structural edges have been reported as weaknesses [23, 24]. In the present study, the margin sharpness for 3D SPCE sequences was not as good as that for 2D TSE sequences. However, because the diagnosis of syndesmosis injuries mainly depends on ligament continuity, contour, and homogeneous signal intensity, poor image sharpness did not seem to significantly influence the diagnostic performance of 3D SPCE, especially in the diagnosis of acute ligament injury. Several factors contributed to diagnostic discordance between different MRI methods. It is known that obliquely oriented ankle ligaments and tendons usually show partial volume artifacts in images obtained using conventional 2D TSE sequences; therefore, highly resolved 3D sequences may depict complex ankle anatomy more accurately [15]. However, for detection of injury (and not for evaluation of anatomy), good visualization of small syndesmosis ligaments may result in less confident diagnoses of full-thickness tears (Fig. 4). Instances existed where a case was classified as a grade 3 tear by use of method 2 but as a grade 2 tear by use of method 3. Unlike acute injuries, ankle joint effusion, which distends the ankle joint space and is associated with changes on arthrography, was rarely noted in chronically injured ankles. Changes in signal intensity may not appear obvious on MRI, therefore possibly reducing the overall diagnostic performance of all three MRI methods evaluated. In addition, variable appearances of chronic ligament injury may result in incorrect diagnoses. The poor image sharpness of 3D SPCE sequences accentuated thinning features, resulting in higher grading of injury scores. Several chronic syndesmosis injuries presented as marked thickening of the affected ligament. This finding may have contributed to the low sensitivity of conventional orthogonal MRI methods. ecause thickened ligaments do not appear as discontinuity or high signal intensity, other features (e.g., ligament contour) that are well depicted on images of the syndesmosis obtained in the oblique plane at an angle of 45 played an important role in the grading of chronic ligament injury (Fig. 3). The use of 3D SPCE imaging has several advantages, including reduced scan acquisition times. In the present study, the mean scan time for 3D SPCE source images was approximately 6 minutes 2 seconds. This does not include the time required for reformatting, because it takes only a few seconds to acquire images reformatted into the oblique plane, and because it is difficult to standardize. Half-Fourier imaging and elliptical scanning shortened the acquisition time for isotropic 3D datasets, which was only 3 minutes longer than the acquisition time for a single 2D dataset. The total scan time for conventional MRI of the ankle was 14 minutes 43 seconds. n additional 3 minutes 38 seconds 832 JR:208, pril 2017

7 was required to acquire 2D Use TSE of images 3D SPCE in Conclusion Sequences for ssessment of nkle Syndesmosis 2011; 40:75 83 Injury the oblique plane. Previously published articles The performance of 3D SPCE sequences reported that the diagnostic performance of 3D images is comparable to that of 2D TSE orthogonal images in the evaluation of ankle ligament injuries [13 15]. Thus, if we replace orthogonal MR images with 3D SPCE reformatted images, the scan time is reduced by more than 67%. Hence, this sequence can reformatted into an oblique plane at an angle of 45 is comparable to that of 2D oblique proton density weighted TSE MR images for the diagnosis of acute and chronic distal syndesmosis injuries, and it has the additional advantage of reduced scan acquisition times. be recommended for certain patients who can tolerate short acquisition times only, such as those with pain, pediatric patients, and patients with claustrophobia, and it is a more cost-effective procedure. The present study has some limitations. First, selection bias may have occurred. The study included surgically confirmed cases of severe trauma or long-standing chronic ankle instability only, and it excluded patients with mild injuries. Second, this was a retrospective study that involved the collection of MRI and arthroscopic data on the ankle, and surgeons were not blinded to MRI findings at the time of arthroscopy. linding may have affected the standard of care for these patients and was not implemented. This inherent limitation may have influenced the arthroscopy results, so the findings of the current study must be interpreted accordingly. Third, the researchers knowledge of whether the sequence was a 3D image or a 2D image may have resulted in bias during evaluation of the ligaments. Last, the transverse tibiofibular ligament and the interosseous ligament were not evaluated in the present study. ecause it is accepted that the ITFL can be a marker of syndesmosis injury and is also the most vulnerable ligament of the ankle [1], the surgical findings regarding the interosseous ligament and the transverse tibiofibular ligament were not thoroughly recorded for all patients in this retrospective study. The basic concept behind obtaining an image of the syndesmosis in the oblique plane at an angle of 45 is the belief that the best depiction of a ligament occurs when the scan plane runs parallel to the direction of that ligament. In the present study, we did not find any statistically significant differences between the diagnostic performance of oblique 3D SPCE and oblique 2D proton density weighted TSE imaging, including distinction between the presence and absence of a tear and discrimination of the four grades of distal syndesmosis ligament injuries. Thus, 3D SPCE images in the oblique plane can replace 2D TSE images in the oblique plane for the diagnosis of syndesmosis injuries. References 1. Kim S, Huh Y-M, Song H-T, et al. Chronic tibiofibular syndesmosis injury of ankle: evaluation with contrast-enhanced fat-suppressed 3D fast spoiled gradient-recalled acquisition in the steady state MR imaging. Radiology 2007; 242: Rammelt S, Zwipp H, Grass R. Injuries to the distal tibiofibular syndesmosis: an evidence-based approach to acute and chronic lesions. Foot nkle Clin 2008; 13: Ramsey PL, Hamilton W. Changes in tibiotalar area of contact caused by lateral talar shift. J one Joint Surg m 1976; 58: rown KW, Morrison W, Schweitzer ME, Parellada J, Nothnagel H. MRI findings associated with distal tibiofibular syndesmosis injury. JR 2004; 182: Muhle C, Frank LR, Rand T, et al. Tibiofibular syndesmosis: high-resolution MRI using a local gradient coil. J Comput ssist Tomogr 1998; 22: Schneck CD, Mesgarzadeh M, onakdarpour. MR imaging of the most commonly injured ankle ligaments. Part II. Ligament injuries. Radiology 1992; 184: Takao M, Ochi M, Oae K, Naito K, Uchio Y. Diagnosis of a tear of the tibiofibular syndesmosis the role of arthroscopy of ankle. J one Joint Surg r 2003; 85: Vogl TJ, Hochmuth K, Diebold T, et al. Magnetic resonance imaging in the diagnosis of acute injured distal tibiofibular syndesmosis. Invest Radiol 1997; 32: artoníček J. natomy of the tibiofibular syndesmosis and its clinical relevance. Surg Radiol nat 2003; 25: Oae K, Takao M, Naito K, et al. Injury of the tibiofibular syndesmosis: value of MR imaging for diagnosis. Radiology 2003; 227: Hermans JJ, eumer, Hop WC, Moonen F, Ginai Z. Tibiofibular syndesmosis in acute ankle fractures: additional value of an oblique MR image plane. Skeletal Radiol 2012; 41: Hermans JJ, Ginai Z, Wentink N, Hop WC, eumer. The additional value of an oblique image plane for MRI of the anterior and posterior distal tibiofibular syndesmosis. Skeletal Radiol 19: Yi J, Cha JG, Lee YK, Lee R, Jeon CH. MRI of the anterior talofibular ligament, talar cartilage and os subfibulare: comparison of isotropic resolution 3D and conventional 2D T2-weighted fast spin-echo sequences at 3.0 T. Skeletal Radiol 2016; 45: Park HJ, Lee SY, Park NH, et al. Three-dimensional isotropic T2-weighted fast spin-echo ( VIST) ankle MRI versus two-dimensional fast spin-echo T2-weighted sequences for the evaluation of anterior talofibular ligament injury. Clin Radiol 2016; 71: Notohamiprodjo M, Kuschel, Horng, et al. 3D-MRI of the ankle with optimized 3D-SPCE. Invest Radiol 2012; 47: Chun K-Y, Choi YS, Lee SH, et al. Deltoid ligament and tibiofibular syndesmosis injury in chronic lateral ankle instability: magnetic resonance imaging evaluation at 3T and comparison with arthroscopy. Korean J Radiol 2015; 16: Jeong MS, Choi YS, Kim YJ, Kim JS, Young KW, Jung YY. Deltoid ligament in acute ankle injury: MR imaging analysis. Skeletal Radiol 2014; 43: Clanton TO, Ho CP, Williams T, et al. Magnetic resonance imaging characterization of individual ankle syndesmosis structures in asymptomatic and surgically treated cohorts. Knee Surg Sports Traumatol rthrosc 2016; 24: Choi WJ, Lee JW, Han SH, Kim S, Lee SK. Chronic lateral ankle instability the effect of intra-articular lesions on clinical outcome. m J Sports Med 2008; 36: Han SH, Lee JW, Kim S, Suh J-S, Choi YR. Chronic tibiofibular syndesmosis injury: the diagnostic efficiency of magnetic resonance imaging and comparative analysis of operative treatment. Foot nkle Int 2007; 28: Park SH, Goo JM, Jo CH. Receiver operating characteristic (ROC) curve: practical review for radiologists. Korean J Radiol 2004; 5: Kwon JW, Yoon YC, Choi SH. Three-dimensional isotropic T2-weighted cervical MRI at 3T: comparison with two-dimensional T2-weighted sequences. Clin Radiol 2012; 67: Park HJ, Lee SY, Park NH, et al. Three-dimensional isotropic T2-weighted fast spin-echo ( VIST) knee MRI at 3.0 T in the evaluation of the anterior cruciate ligament injury with additional views: comparison with two-dimensional fast spin-echo T2-weighted sequences. cta Radiol 2015; 57: Ristow O, Steinbach L, Sabo G, et al. Isotropic 3D fast spin-echo imaging versus standard 2D imaging at 3.0 T of the knee: image quality and diagnostic performance. Eur Radiol 2009; FOR YOUR INFORMTION This article is available for CME and Self-ssessment (S-CME) credit that satisfies Part II requirements for maintenance of certification (MOC). To access the examination for this article, follow the prompts associated with the online version of the article. JR:208, pril