| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Musculoskeletal Imaging |
1 From the Department of Radiology, Balgrist University Hospital, Forchstrasse 340, CH-8008 Zurich, Switzerland (M.R.S., C.W.P.); and Departments of Diagnostic Radiology (T.S., N.S., S.W., D.W.) and Anatomy (M.M.), University of Zurich, Zurich, Switzerland. Received August 20, 2004; revision requested October 29; revision received December 24; accepted January 21, 2005. Address correspondence to M.R.S. (e-mail: marius.schmid{at}balgrist.ch)
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Cadaver wrists were obtained and used according to institutional guidelines and with informed consent of donors prior to death. Nine cadaver wrists of eight subjects were evaluated. MR images were obtained with a 1.5-T MR unit. Imaging protocol included intermediate-weighted coronal and transverse fast spin-echo and coronal three-dimensional gradient-echo sequences. Multi–detector row CT arthrography was performed after tricompartmental injection of 3–6 mL of contrast material with a concentration of 160 mg per milliliter of iodine. Palmar, dorsal, and central segments of both ligaments were analyzed on transverse and coronal MR images and multiplanar multi–detector row CT reconstructions by two musculoskeletal radiologists working independently. Open inspection of the wrists was the reference standard. Sensitivity, specificity, accuracy, and positive and negative predictive values were calculated from the imaging and gross pathologic readings. Statistical significance was calculated with the McNemar test. Weighted 
values for interobserver agreement were calculated for both imaging modalities.
RESULTS: All ligament segments could be visualized in all cases with both imaging modalities. CT arthrography was more sensitive (100%) than MR imaging (60%) in detection of palmar segment tears (P = .62); specificity of both imaging modalities was 77%. Sensitivity (CT arthrography, 86%; MR imaging, 79%) and specificity (CT arthrography, 50%; MR imaging, 25%) for detection of the central segment tears were determined. Dorsal segment tears were detected only with CT arthrography, while all tears were missed with MR imaging (P = .02). Interobserver agreement was better for multi–detector row CT arthrography ( = 0.37–0.78) than for MR imaging ( = –0.33 to –0.10).
CONCLUSION: Performance in depiction of palmar and central segment tears of SL and LT ligaments is almost equal for multi–detector row CT arthrography and MR imaging, with much higher interobserver reliability for CT arthrography. CT arthrography is significantly superior to MR imaging in the detection of dorsal segment tears of SL and LT ligaments.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
From a clinical point of view, knowledge of the particular anatomy of these ligaments is important for the radiologist, since the site and extent of a disrupted ligament segment may be used to differentiate between a traumatic tear and a degenerative change, which can occur as senescent changes in asymptomatic wrists (2). In addition, the dorsal and palmar segments of both ligaments are more important functionally than is the central segment (3). Dorsal and palmar segments of both ligaments are true ligaments, in contrast with the central segments of both ligaments (4).
Traditionally, conventional wrist arthrography, which—to our knowledge—was first described in 1961 by Kessler and Silberman (5), has been used in the assessment of SL and LT ligament tears. With use of multiple projections, conventional wrist arthrography can enable differentiation of different ligament segments and associated abnormalities (6,7).
In the past, magnetic resonance (MR) imaging with (9–14) and without (15–18) arthrography and computed tomographic (CT) arthrography (19) have been advocated for assessment of SL and LT ligament tears. In almost all of these studies, which were performed to assess the diagnostic performance of an imaging modality with regard to the detection of tears (7,9–11,13,15,17,18), the SL and LT ligaments were considered a single unit, without taking into account the fact that both ligaments are composed of different ligamentous segments. The segmental anatomy of the interosseous ligaments and their abnormalities are potentially important with regard to the distinction between traumatic tears and, most often, centrally located degenerative tears (20). To our knowledge, in no study have investigators directly compared the diagnostic performance of CT arthrography with that of MR imaging with regard to the assessment of SL and LT ligaments, particularly with regard to the assessment of the different segments of these ligaments. Thus, the purpose of our study was to compare the accuracy of multi–detector row CT arthrography and MR imaging in the depiction of tears of dorsal, central, and palmar segments of the SL and LT ligaments in cadaver wrists.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
The cadaver hands were obtained and used according to institutional guidelines. Prior to death, donors consented to the use of their bodies for research purposes.
Imaging
Nine cadaver wrists were examined with MR imaging by using a 1.5-T unit (Signa Horizon; GE Medical Systems, Milwaukee, Wis) with a dedicated quadrature wrist coil. A fat-suppressed intermediate-weighted fast spin-echo sequence (repetition time msec/echo time msec, 1800/24; field of view, 8 x 8 cm; section thickness, 2 mm; intersection gap, 0.5 mm; image matrix, 512 x 256; four signals acquired; echo train length, four) was performed in the coronal and transverse planes. A fat-suppressed coronal three-dimensional gradient-recalled-echo sequence (40–45/12–15; flip angle, 25°; matrix, 512 x 256; field of view, 8 x 8 cm; four signals acquired; section thickness, 1 mm; no intersection gap; image matrix, 512 x 256) and an intermediate-weighted fast spin-echo sequence without fat suppression (1800/24; field of view, 8 x 8 cm; section thickness, 2 mm; intersection gap, 0.5 mm; image matrix, 512 x 256; four signals acquired; echo train length, four) were performed in the transverse plane.
After MR imaging, CT arthrography was performed. All nine specimens were examined with a 16–detector row CT scanner (Sensation 16; Siemens Medical Systems, Forchheim, Germany). Intraarticular injection of contrast material was performed immediately before imaging with the multi–detector row CT unit. With CT guidance, a 1:1 mixture of iodixanol (320 mg of iodine per milliliter) (Visipaque 320; Amersham Health, Buckinghamshire, England) and saline was injected into the distal radioulnar joint. If necessary (ie, no perforation of the triangular fibrocartilage complex or the SL or LT ligaments was present), additional injections into the radiocarpal and midcarpal compartments were performed. All injections were performed by the same musculoskeletal radiologist (M.R.S.). Mean volume of contrast material injected at tricompartmental arthrography was 4 mL (range, 3–6 mL).
Immediately after contrast agent injection, CT arthrography was performed. First, a scout view of the cadaver hand was used to plan multi–detector row CT data acquisition.
For data acquisition, section collimation of 16 x 0.75 mm, section width of 0.75 mm, and reconstruction increment of 0.3 mm were used. Tube current was 120 mAs, and tube voltage was 120 kV.
Image Analysis
Two musculoskeletal radiologists (C.W.P., D.W.) with 10 and 11 years of CT arthrography experience, respectively, who were not involved in image acquisition or gross anatomic inspection interpreted data obtained with both imaging modalities for all nine specimens separately. Consensus was obtained in case of disagreement. Interpretation of images obtained with both modalities was performed separately, with an interval of at least 2 weeks between the reading sessions. For review of MR data, all MR images were available on a workstation (Advantage Windowing Workstation; GE Medical Systems Europe, Buc, France). For CT arthrogram review, the data set was uploaded on a Volume Zoom Wizard workstation (Siemens Medical Solutions); analysis of the data was performed by using multiplanar reformations. All multiplanar reformations were reconstructed on the workstation by the readers themselves. Multiplanar reformation included routine coronal and transverse reformations of the wrist. The window settings for assessment of the multiplanar reformations were adjusted for each patient individually by both radiologists who performed image analysis.
Palmar, central, and dorsal segments of the SL and LT ligaments were evaluated separately in the coronal (Fig 1) and transverse (Fig 2) planes by using multiplanar reformation CT images and MR images. Visibility of the three segments of both ligaments was rated as visible or not visible, and each ligament segment was judged to be intact or torn. For multi–detector row CT arthrography, ligament tear was defined as communication of intraarticular contrast material through any segment of the ligaments. For MR imaging, a tear was diagnosed if there was discontinuity of any part of the ligament or presence of fluid-type signal intensity within the substance of the SL or LT ligaments. Each reader assessed the ligament status separately, with subsequent consensus reading in cases of disagreement.
|
|
Statistical Analysis
Sensitivity, specificity, accuracy, positive predictive value, and negative predictive value were calculated from the imaging findings and gross pathologic readings. Statistically significant differences were calculated with the McNemar test. Statistical significance was set at a P value of less than .05. Both ligaments were analyzed together. This combination of segments of SL and LT ligaments seems acceptable, since SL and LT ligaments are identical in anatomic construction, with thick palmar and dorsal segments composed of oriented collagen fascicles and the weaker central segment composed of fibrocartilage without collagen orientation (4).
Weighted values for interobserver agreement were calculated for both imaging modalities with statistical software (SPSS, version 11.0; SPSS, Chicago, Ill). According to Landis and Koch (21), a value of 0.00–0.20 indicated poor agreement, 0.21–0.40 indicated fair agreement, 0.41–0.60 indicated moderate agreement, 0.61–0.80 indicated good agreement, and 0.81–1.00 indicated excellent agreement.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
|
|
|
|
|
|
values of 0.38, 0.67, and 0.78 for the central, palmar, and dorsal segments, respectively. Poor interobserver agreement was found for MR imaging in all three segments (central segment, = –0.33; palmar segment, = –0.22; dorsal segment, = –0.10). |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
As proposed in earlier studies (2,20), degenerative tears of the SL and LT ligaments are usually located in the central segment. Exact examination and analysis of all three segments of both ligaments are important because dorsal and palmar segment tears indicate a potential traumatic cause of a ligament tear.
SL and LT ligaments can be depicted with several imaging modalities, including conventional arthrography, MR imaging (with and without intraarticular contrast agent administration), and CT arthrography. The role of these imaging modalities, particularly MR imaging and CT, in the visualization and assessment of segmental anatomy of the SL and LT ligaments is not well investigated in the literature. In conventional arthrography, these three ligament segments can be differentiated by an experienced radiologist using oblique views of the wrist (6) in half and full flexion in the supinated position and half and full extension in the pronated position. To our knowledge, there are only two studies in which the ability of MR imaging to depict the segmental anatomy of the interosseous SL ligament has been investigated (12,16). In the study of Scheck et al (12), three-compartment MR arthrography was performed in 41 wrists of patients with wrist pain. In their investigation, arthroscopy served as the reference standard. Scheck et al (12) demonstrated that the three segments of the SL ligament can be depicted by using MR arthrography with coronal thin-section gradient-recalled-echo imaging. Scheck et al (12) used this technique to clearly delineate dorsal, central, and palmar segments of the SL ligament in 95% of the cases. In the study of Totterman and Miller (16), normal appearance of dorsal, central, and palmar segments of the SL ligament in 14 asymptomatic volunteers and five cadavers with an intact SL ligament was depicted with a coronal three-dimensional gradient-recalled-echo sequence. To our knowledge, the ability of CT or CT arthrography to depict different segments of the interosseous ligaments has not yet been demonstrated.
In accordance with previously published findings for MR imaging (16) and MR arthrography (12), we have shown that all three segments of the SL and LT ligaments may be demonstrated with either MR imaging or CT arthrography to good advantage. Our results showed almost equal detectability of palmar and central segment tears of both ligaments with CT arthrography and MR imaging. No statistically significant differences between imaging modalities were found. When compared with gross pathologic analysis, MR imaging did not depict four dorsal segment tears in our study group. This difference was statistically significant. MR arthrography has proved superior to standard MR imaging in the detection of ligament tears in the wrist (12); therefore, it also could be superior to MR imaging in the detection of dorsal segment tears.
With the exception of reports published by Scheck et al (12) and Totterman and Miller (16), all other reports on the diagnostic performance of MR imaging in the detection of interosseous ligament tears of the wrist did not differentiate this segmental anatomy of the SL and LT ligaments. Several reports have been published on the diagnostic performance of MR imaging with regard to the detection of interosseous ligament tears, irrespective of the segmental anatomy; therefore, a comparison of their results with ours is difficult. Manton et al (17) used standard MR imaging and indirect MR arthrography and found sensitivity and specificity values for detection of SL ligament tears of 42%–69% and 32%–81%, respectively; in the same study, these researchers found sensitivity and specificity values for detection of LT ligament tears of 25%–36% and 68%–84%, respectively. In the study of Manton et al (17), secondary signs, such as focal osseous offset, arc disruption, and focal osteoarthritis, increased specificity of ligament tear detection. High sensitivity (SL ligament, 93%; LT ligament, 56%) and specificity (SL ligament, 89%; LT ligament, 90%) in ligament tear detection were also found in the study of Zlatkin et al (23), in which the results of standard MR examinations of 43 patients with 24 tears of the SL ligament and seven tears of the LT ligament were analyzed.
In the literature, we found only one study (19) in which CT arthrography was used to detect intrinsic ligament injury of the wrist. Theumann et al (19) evaluated 36 wrists with conventional arthrography and CT arthrography for the presence of SL and LT ligament tears. In their study (19), sensitivity and specificity values for conventional arthrography were 85% and 100% for SL ligament tears and 80% and 100% for LT ligament tears. With CT arthrography, sensitivity and specificity values were both 100% for SL ligament tears and 85% and 100%, respectively, for LT ligament tears (19). Although mentioned in the report of Theumann et al (19), the three individual segments of the two ligaments were not discussed further in their article.
A limitation of our study was the small number of wrists that were evaluated. Thus, we performed a combined evaluation of both ligaments. This combination seems appropriate because both ligaments are very similar anatomically, with thicker dorsal and palmar segments composed of transversely oriented collagen fascicles (4) and a central segment composed of fibrocartilage, with no collagen orientation (4). Another limitation of our study was the fact that MR images were obtained without intraarticular contrast material administration contrary to CT, in which CT arthrography was performed. We used this study setup because the time delay for both examinations would have resulted in a substantial diffusion of intraarticular contrast material within the ligaments. Diffusion of contrast material within the ligament would have hampered analysis of the images, which would have been performed subsequent to intraarticular administration of contrast material. Moreover, in clinical practice, MR imaging of the wrist is performed without intraarticular contrast agent administration in most institutions. Finally, this study was performed in cadaver wrists; thus, no clinical data about potential wrist pain and its duration was available.
In conclusion, we have shown that, in cadaver wrists, both standard MR imaging and CT arthrography are able to demonstrate the three different segments of the SL and LT ligaments. A precise segmental evaluation of the two ligaments is important because dorsal and palmar segments of both ligaments are functionally more important than the central segment (3,4). In all three segments, interobserver reliability for tear detection with CT arthrography is much higher than interobserver reliability for tear detection with MR imaging. CT arthrography seems to be superior to standard MR imaging in the detection of tears involving the dorsal segment of the SL and LT ligaments.
|
FOOTNOTES |
|---|
Abbreviations: LT = lunotriquetral • SL = scapholunate
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, M.R.S., C.W.P., D.W.; 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, M.R.S., M.M., S.W.; experimental studies, T.S., N.S.; statistical analysis, C.W.P.; and manuscript editing, M.R.S., T.S., C.W.P., N.S., S.W., D.W.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Magee Comparison of 3-T MRI and Arthroscopy of Intrinsic Wrist Ligament and TFCC Tears Am. J. Roentgenol., January 1, 2009; 192(1): 80 - 85. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Moser, J.-C. Dosch, A. Moussaoui, X. Buy, A. Gangi, and J.-L. Dietemann Multidetector CT Arthrography of the Wrist Joint: How to Do It RadioGraphics, May 1, 2008; 28(3): 787 - 800. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
