| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Diagnostic Radiology, Yale University School of Medicine, 333
Cedar St., New Haven, CT 06520.
2 Present address: Radiology Consultants, PC, Ste. 2B, 40 Temple St., New Haven,
CT 06520.
3 Department of Orthopedic Surgery, Yale University School of Medicine, New
Haven, CT 06520.
4 Connecticut Orthopedic Specialists, PC, 450 Post Rd., Guilford, CT
06437.
Received December 2, 2002;
accepted after revision May 14, 2003.
Address correspondence to T. R. McCauley.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We retrospectively identified 27 patients (mean age, 31 years; range, 18–45 years) with anterior cruciate ligament reconstruction who had undergone MR arthrography followed by arthroscopy within 1 year. Three radiologists independently reviewed the MR arthrograms for the presence or absence of graft tear, localized anterior arthrofibrosis, and impingement.
RESULTS. Graft tears were identified with 100% sensitivity by all
three reviewers with specificities of 100%, 89%, and 94%. Localized anterior
arthrofibrosis was identified with 100% sensitivity by all reviewers, with
specificities of 79%, 71%, and 38%. Impingement was detected with
sensitivities and specificities of 83% and 100%, 83% and 52%, and 33% and 90%
by the three reviewers, respectively. Interobserver agreement was almost
perfect for detection of graft tear (
= 0.83, 0.92, and 0.83), was fair
to moderate for detection of localized anterior arthrofibrosis ( =
0.50, 0.32, and 0.22), and was slight to fair for detection of impingement
( = 0.40, 0.08, and 0.35).
CONCLUSION. MR arthrography can accurately depict the presence of anterior cruciate ligament graft tears. Localized anterior arthrofibrosis and graft impingement were less accurately detected and showed greater observer variability.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the study group, there were 20 men and seven women with a mean age of 31 years (range, 18–45 years). The mean time from MR examination to surgery was 89 days (range, 20–305 days). The reason for MR arthrography was pain in 14 patients, reinjury in nine patients, instability in two patients, stiffness in one patient, and clicking in the knee in one patient. For the initial anterior cruciate ligament reconstruction surgery, the source of the graft was the patellar tendon for 18 patients, the quadriceps tendon for one patient, and unknown for eight patients. The time between the initial anterior cruciate reconstruction and MR arthrography was not known for nine patients. For the remaining 18 patients, the mean time between the graft reconstruction and MR arthrography was 38 months (range, 4–119 months).
In all patients, MR arthrography was performed using fluoroscopic guidance, sterile skin preparation, and a local anesthetic of 1% lidocaine. Injection of contrast material was performed from a lateral approach into the patellofemoral joint with a 20-gauge needle. One of two techniques was used. Either needle localization was confirmed by injection of 1–3 mL of iohexol (Omnipaque, Nycomed, Princeton, NJ) followed by injection of 20 mL of a 1:250 dilution of gadodiamide (Omniscan, Nycomed) in normal saline containing approximately 0.2 mL of 1:1,000 epinephrine (Abbott Laboratories, North Chicago, IL), or a mixture of 5 mL of Omniscan, 5 mL of 1% lidocaine (Abott Laboratories), 2 mL of epinephrine, and 10 mL of a 1:125 dilution of Omniscan in normal saline was injected. Both techniques result in an approximately 1:250 dilution of the gadolinium contrast agent in the knee joint.
MRI was performed on 1.5-T scanners (General Electric Medical Systems, Milwaukee, WI) with a transmit–receive extremity coil (20 cases) or a phased array receive-only extremity coil (seven cases). In all cases, sagittal T1-weighted fat-suppressed images were obtained with the following parameters: TE range/TR range, 14–17/600–800; 14- to 15-cm field of view; 256 x 192 matrix; and 1 excitation. Contiguous 3-mm images were obtained using an interleaved imaging sequence so that the interslice spacing for each interleaved set was 3 mm.
In the coronal plane, a double-echo conventional spin-echo imaging sequence was performed with a TR/first-echo TE, second-echo TE of 2,000/20 and 80. Sections 3 mm thick with a 0.3-mm gap were obtained with a 14-cm field of view, 256 x 192 matrix, and 1 excitation. In 17 patients, axial images were obtained with a double-echo spin-echo pulse sequence with 1,700/20 and 80, 5-mm images with a 2.5-mm gap, 256 x 128 matrix, and 1 excitation. In nine patients, axial images were obtained with a fat-suppressed fast spin-echo proton density–weighted image set with a TR range/effective TE range of 3,500–4,500/16–20, echo train of 10, field of view of 16, 5-mm-thick sections with 0.5-mm intersection gap, matrix of 256 x 256, and excitation of 1.
Images were reviewed by three experienced musculoskeletal radiologists without knowledge of the arthroscopic results. The images were graded for the presence or absence of anterior cruciate ligament graft tear, localized anterior arthrofibrosis, and impingement. The graft was considered intact if it could be followed from the femoral tunnel to the tibial tunnel (Fig. 1A, 1B). A torn graft was diagnosed when the graft fibers could not be identified extending from the femoral tunnel to the tibial tunnel, especially when gadolinium extended through a discontinuity in the graft fibers. The reviewers classified all tendons showing these findings as torn and did not attempt to differentiate partial from complete graft tears. Localized anterior arthrofibrosis was diagnosed when abnormal nonfat soft tissue was seen superior to the tibial plateau anterior to the insertion of the anterior cruciate ligament graft [6, 7]. Impingement was diagnosed when the anterior cruciate ligament graft contained increased signal or was enlarged and was associated with either the tibial tunnel placed anterior to a line drawn along the roof of the femoral notch or the superior surface of the graft deformed by the roof of the femoral notch [8–11].
|
|
Anterior displacement of the tibia relative to the femur in the lateral compartment has been previously described as a secondary sign of anterior cruciate ligament tear [12]. One of the reviewers measured anterior displacement of the tibia relative to the femur by drawing a vertical line from the posterior cortex of the femur and measuring the distance from this line to the posterior cortex of the tibia on the image in the middle of the lateral compartment of the knee that had the greatest anterior displacement of the posterior cortex of tibia relative to the posterior cortex of the femur. Displacement greater than 7 mm was considered indicative of a tear [12]. In addition to the retrospective review, the prospective MR arthrography clinical reports were reviewed to determine the presence or absence of the anterior cruciate ligament graft findings.
The arthroscopic or surgical reports were reviewed to determine the
presence of anterior cruciate ligament graft tear, impingement, and localized
anterior fibrosis. Sensitivity, specificity, accuracy, positive predictive
value, and negative predictive value were determined for each reviewer using
the arthroscopic reports as the standard of reference. Interobserver agreement
was assessed using the kappa statistic. The strength of agreement was
interpreted according to the guidelines suggested by Landis and Koch
[13]: almost perfect (
= 0.81–1.00), substantial ( = 0.61–0.80), moderate (
= 0.41–60), fair ( = 0.21–0.40), slight ( <
0.20), and poor ( < 0.0). The anterior displacement of the tibia
relative to the femur in patients without tears was compared with the anterior
displacement in patients with tears using the unpaired two-tailed Student's
t test.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
Review of the clinical reports showed perfect agreement between MR arthrography interpretations and surgical findings for the presence of anterior cruciate ligament graft tear. Localized anterior arthrofibrosis was not identified in one of the three clinical reports; however, the specificity of the clinical interpretations was higher than that for all of the retrospective reviewers. Impingement was detected in only one of six cases at clinical interpretation.
The mean displacement of the tibia versus the femur in patients with intact grafts was 3.5 mm (range, 0–9 mm), which was less than the mean displacement of 9.8 mm (range, 4–11 mm) found in patients with tears (p < 0.0002).
Using the criterion of anterior displacement of the tibia versus the femur in the lateral compartment greater than 7 mm as indicative of anterior cruciate ligament graft tear resulted in one of nine cases of graft tear not being detected (sensitivity, 89%); however, it resulted in two of 18 intact grafts being classified as torn (specificity, 89%; accuracy, 89%).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Only one partial anterior cruciate ligament graft tear in this study was categorized as torn by all reviewers. We did not attempt to differentiate partial from complete anterior cruciate graft tears because prior studies have shown that MRI is not accurate for differentiation of partial from complete tears when examining the native anterior cruciate ligament [16, 17]. We did not compare MR arthrography with conventional MRI; thus, we do not know if MR arthrography is more accurate than conventional MRI for detection of graft tears. We also did not assess the different imaging planes separately to determine if sagittal or coronal images are more useful for assessment of anterior cruciate graft integrity.
We found that measurement of anterior displacement of the tibia identified all but one of the anterior cruciate ligament graft tears. We do not know if this measurement would be equally sensitive for conventional MRI or if distention of the joint with fluid may accentuate this finding. The reviewers' overall assessments resulted in higher accuracies; thus, measurement of tibial displacement is helpful but should be combined with other primary findings to assess anterior cruciate graft integrity.
The reviewers did not diagnose localized anterior arthrofibrosis with high accuracy. This result is in contrast to the previous study using unenhanced MRI, in which localized anterior arthrofibrosis was diagnosed with 85% accuracy [6]. The reason for the lower accuracy in our study is uncertain. We had only a small number of patients with localized anterior arthrofibrosis in this study; that may have lead to statistical variation accounting for the lower accuracy than shown previously. The use of fat suppression for the sagittal T1-weighted images can obscure the differentiation between the adjacent fat and focal arthrofibrosis that could lead to an inaccurate assessment. Non-fat-suppressed images could have resulted in higher accuracy. Injection of contrast material can cause underestimation or overestimation of abnormal tissue adjacent to the anterior cruciate ligament graft. The amount of tissue that is required for a diagnosis of localized anterior arthrofibrosis at arthroscopy is a subjective assessment, which could contribute to the overdiagnosis of localized anterior arthrofibrosis on the retrospective reviews, causing the low specificity. This possibility is supported by the clinical reports that had a much higher specificity very likely because when reviewing clinical reports, we usually only comment on localized anterior arthrofibrosis when it is believed to be large enough to cause symptoms. The clinical reports in our study showed an accuracy of 89%, which is similar to the prior reported accuracy of MRI performed without contrast material [6].
Impingement was diagnosed with variable accuracy and with low interobserver agreement. We do not know of any prior studies that have evaluated the accuracy of MRI findings for detection of impingement. The high variability in interpretation found in our study very likely is due to the subjectivity of the findings evaluated for diagnosis of impingement both at MR arthrography and at arthroscopy. Future studies would be needed to determine if the criteria for diagnosis of impingement can be more clearly defined to improve reliability of MR arthrography for this diagnosis.
There are limitations to this study. The number of patients with anterior cruciate ligament graft abnormalities was small. Unfortunately, patients rarely undergo MRI after graft reconstruction with subsequent arthroscopic follow-up. We did not compare MR arthrography with conventional MRI; thus, we do not know if MR arthrography results in different accuracy than would be found for conventional MRI. The surgeons had the clinical MR arthrography reports at the time of arthroscopy; thus, their diagnoses at arthroscopy could be biased by these reports, which very likely lead to higher accuracies than would be found if the surgeons did not have access to the clinical MR arthrography reports.
In conclusion, MR arthrography provides accurate assessment of anterior cruciate ligament graft integrity. Localized anterior arthrofibrosis and graft impingement were less accurately detected and showed greater observer variability.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  B. R. Runyan, L. W. Bancroft, J. J. Peterson, M. J. Kransdorf, T. H. Berquist, and C. J. Ortiguera Cyclops Lesions That Occur in the Absence of Prior Anterior Ligament Reconstruction RadioGraphics, November 1, 2007; 27(6): e26 - e26. [Abstract] [Full Text]
|
|
|
|
 |
|
 T. R. McCauley MR Imaging Evaluation of the Postoperative Knee Radiology, January 1, 2005; 234(1): 53 - 61. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
