HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cotten, A. Articles by Gougeon, F. Search for Related Content PubMed PubMed Citation Articles by Cotten, A. Articles by Gougeon, F. Social Bookmarking                      What's this?

MR Imaging of the Knee at 0.2 and 1.5 T

Correlation with Surgery

¹ Service de Radiologie Ostéo-Articulaire, Hôpital R. Salengro, Blvd. du Pr. J. Leclercq, 59037 Lille, France.
² Service de Traumatologie, Hôpital R. Salengro, Blvd. du Pr. J. Leclercq, 59037 Lille, France.

Received July 29, 1999; accepted after revision September 3, 1999.

 
Address correspondence to A. Cotten.

Abstract

Top Abstract Introduction Subjects and Methods Results Discussion References

 
OBJECTIVE. The purpose of this study was to compare the diagnostic efficacy of low- and high-field-strength MR imagers in the diagnosis of anterior cruciate ligament tears and meniscus tears.

SUBJECTS AND METHODS. In 219 patients with suspected internal derangement of the knee, MR imaging at 0.2 and 1.5 T was performed with similar sequences. Only patients with surgically confirmed diagnosis (n = 90) were included in the statistical analysis. Radiologists were unaware of diagnosis and field strength. Sensitivity, specificity, diagnostic accuracy, and inter- and intraobserver variability were determined.

RESULTS. There was excellent correlation between the field strengths in accuracy, sensitivity, and specificity for anterior cruciate ligament and meniscus tears. Accuracy for medial meniscus, lateral meniscus, and anterior cruciate ligament tears was 91-93%, 88-90%, and 93-96%, respectively, at 0.2 T and 91-94%, 91-93%, and 97-98%, respectively, at 1.5 T. Inter- and intraobserver variability values showed excellent correlation ( > 0.8).

CONCLUSION. The level of diagnostic accuracy in anterior cruciate ligament tears and meniscus tears is comparable for low- and high-field-strength MR imagers.

Introduction

Top Abstract Introduction Subjects and Methods Results Discussion References

 
The diagnostic value of MR imaging in patients with suspected internal derangement of the knee is well established. However, in some articles investigators have reported different diagnostic performances with low-field-strength MR systems. Parizel et al. [1], Kladny et al. [2], and Kersting-Sommerhoff et al. [3] found no significant difference in diagnostic performance between low- and high-field-strength MR imaging of the knee; Fischer et al. [4] found MR examinations of the medial meniscus at 0.35 T to be less accurate than examinations performed at 1.5 T, and Kinnunen et al. [5] reported a specificity of only 25% for lesions of the lateral meniscus (images obtained at 0.1 T). These different results may, in part, be caused by the different methods used in these studies, technical differences in the MR protocols, and, in some studies, the small patient population. In two recent studies, researchers also found large interobserver variation at low-field-strength MR imaging [6, 7].

We decided to compare the diagnostic efficacy of low-field-strength (permanent magnet, 0.2 T) and high-field-strength (superconducting system, 1.5 T) MR imagers in the diagnosis of anterior cruciate ligament tears and meniscus tears and to incorporate the following features that we believe are missing, at least in part, from earlier studies: direct comparison of low- and high-field-strength images of the knee in the same individuals; assessment of a large number of patients; use of similar sequences and sections at 0.2 and 1.5 T; independent interpretation of images by two radiologists unaware of the diagnosis and field strength; and pathologic confirmation in all patients.

Subjects and Methods

Top Abstract Introduction Subjects and Methods Results Discussion References

 
The study was approved by the committee on human research at our institution and all patients provided informed consent. A total of 221 patients with suspected anterior cruciate ligament tear or meniscal tear of the knee were enrolled in the study. Two patients refused to undergo the second imaging procedure. Thus, 219 patients comprised the study group: 76 were women and 143, men (age range, 18-57 years; mean age, 35 years). Patients underwent imaging at one field strength and within 2 hr underwent imaging at the other field strength in random order. Imaging was performed with 0.2- and 1.5-T imagers (Magnetom Open and Vision, respectively; Siemens, Erlangen, Germany). The imagers have different coil configurations (1.5 T, transmit-receive quadrature coil; 0.2 T, receive quadrature coil), but both have shielded gradients. Both also have digital radiofrequency transmitter and receiver electronics. Gradient strength was from 0 to 25 mT/m at 1.5 T and from 0 to 10 mT/m at 0.2 T, and rise time was 600 msec at 1.5 T and 1200 msec at 0.2 T.

Sagittal T1-weighted spin-echo imaging, sagittal and coronal T2^*-weighted gradient-recalled echo imaging, and axial short tau inversion recovery imaging were performed on both imagers. The imaging parameters for the different field strengths are shown in Table 1. The same number of sections was programmed on both imagers to ensure reproducibility between studies. Appropriate window settings for T1- and T2^*-weighted images were used, with technologists ensuring that ligaments and menisci could be analyzed. Short tau inversion recovery images were used to analyze bone bruises. Images were filmed without annotation, and patient details were omitted from the images. Selection of patients for surgery was performed using only the data from the 1.5-T magnet.

To prevent potential reviewer bias, interpretations for this study were performed 3-5 months after acquisition. The reviewers were two musculoskeletal radiologists experienced in MR imaging with more than 5 years of postresidency. To maintain objectivity, reviewers interpreted images without knowledge of the first reviewer's report or the arthroscopic findings. The retrospective findings were then compared with the arthroscopic reports; the average interval between MR imaging and arthroscopy was approximately 57 days (median, 41 days; range, 1-240 days); the arthroscopic descriptions were used as the gold standard. Reviewers were unaware of the patient's name, the diagnosis, and the field strength, and they interpreted the images independently. Images of the same patients presented in a randomized fashion to the reviewers were interpreted four times (0.2-T images interpreted twice, 1.5-T images interpreted twice), with an interval of 1-6 months (mean, 4 months) between the two interpretations (0.2 T-0.2 T, 0.2 T-1.5 T, 1.5 T-0.2 T, 1.5 T-1.5 T; in whatever the order of presentation of the images) to determine intra- and interobserver variability.

Anterior cruciate ligament tears were diagnosed on MR imaging on the basis of the presence of increased signal intensity in the ligament. If ligament margins were intact, the tear was termed partial. If margins were not identified or there was ligament retraction and no identifiable central ligament was present, the tear was termed complete [8,9,10]. Similarly, meniscal injuries were diagnosed by means of abnormal signal intensity reaching the superior or inferior meniscal surface, with secondary features of blunting of the meniscal free edge, meniscal deformity, or absence of a meniscus [4, 8, 11]. Both T1- and T2^*-weighted images were used to make these diagnoses. Arthroscopic or pathologic confirmation of ligamentous or meniscal tear was used as the standard of reference.

Statistical evaluation of the comparisons between the two field strength MR imagers was performed by a biostatistician using kappa measures and confidence interval measures (statistical significance was defined as p < 0.05). Differences in performance between the two observers as well as between the two interpretations of each observer for each magnet were tested for significance by using kappa measures [12]. A kappa value of greater than 0.75 indicated excellent agreement.

Results

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Only surgically confirmed diagnoses (n = 90) were included in the statistical analysis. Of these 90 patients, 29 were women and 61 were men (age range, 18-54 years; mean age, 34 years). Surgery revealed 57 tears of the medial meniscus, 29 tears of the lateral meniscus, and 15 tears of the anterior cruciate ligament (complete, n = 14; partial, n = 1). The accuracy, sensitivity, specificity, positive predictive value, and negative predictive value for medial meniscus, lateral meniscus, and anterior cruciate ligament are indicated in Table 2. Accuracy was 91-93% at 0.2 T and 91-94% at 1.5 T for the medial meniscus, 88-90% at 0.2 T and 91-93% at 1.5 T for the lateral meniscus, and 93-96% at 0.2 T and 97-98% at 1.5 T for the anterior cruciate ligament (Figs. 1A,1B,2A,2B,3A,3B). Kappa coefficient measures (Table 3) showed excellent concordance between the two field strengths with a kappa value of more than 0.8. Confidence interval measures showed no significant difference (p > 0.05) between the interpretations of the two field strengths. No discrepancy in bone bruises was observed between the two field strength results (n = 27). Intra- and interobserver variability values also showed excellent agreement (Tables 4 and 5).

View larger version (117K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1A. —27-year-old man with medial meniscal tear. Sagittal T2*-weighted MR images obtained at 1.5 T (A) and 0.2 T (B) show medial meniscal tear equally well (arrow). Arthroscopy (not shown) confirmed meniscal tear.

View larger version (118K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 1B. —27-year-old man with medial meniscal tear. Sagittal T2*-weighted MR images obtained at 1.5 T (A) and 0.2 T (B) show medial meniscal tear equally well (arrow). Arthroscopy (not shown) confirmed meniscal tear.

View larger version (95K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2A. —31-year-old man with medial meniscal tear. Sagittal T2*-weighted MR image obtained at 1.5 T (A) reveals medial meniscal tear (arrow), whereas sagittal T2*-weighted MR image obtained at 0.2 T (B) does not. Arthroscopy (not shown) confirmed meniscal tear.

View larger version (97K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 2B. —31-year-old man with medial meniscal tear. Sagittal T2*-weighted MR image obtained at 1.5 T (A) reveals medial meniscal tear (arrow), whereas sagittal T2*-weighted MR image obtained at 0.2 T (B) does not. Arthroscopy (not shown) confirmed meniscal tear.

View larger version (143K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3A. —29-year-old man with intact anterior cruciate ligament. Sagittal T1-weighted MR image obtained at 1.5 T reveals intact anterior cruciate ligament.

View larger version (165K): [in this window] [in a new window] [as a PowerPoint slide]   Fig. 3B. —29-year-old man with intact anterior cruciate ligament. Sagittal T1-weighted MR image obtained at 0.2 T was interpreted as tear of anterior cruciate ligament. Clinically and arthroscopically (not shown) intact anterior cruciate ligament was confirmed.

The 129 of 219 patients without surgical evaluation were not included in the statistical evaluation because surgical descriptions were used as the gold standard. However, we obtained the following results: no meniscal tear was seen on both magnets in 125 patients; a meniscal tear was revealed on both magnets in one patient (this patient did not undergo arthroscopy because pain spontaneously disappeared); and discrepancies in the meniscal appearance were seen in three patients (three tears seen at 0.2 T were not described at 1.5 T). Normal appearance of the anterior cruciate ligament was reported in 122 patients; anterior cruciate ligament tear was seen on both magnets in six patients with clinically confirmed anterior cruciate ligament tear; and a discrepancy between the findings on two field strength MR imagers was observed in one patient with clinically intact anterior cruciate ligament (anterior cruciate ligament tear seen at 0.2 T but not detected at 1.5 T).

The radiologists reviewed the cases (patients with and without surgical evaluation) with discrepancies between the two field strengths and attributed the different results to the lower image definition of the images obtained at 0.2 T (Fig. 2A,2B).

Discussion

Top Abstract Introduction Subjects and Methods Results Discussion References

 
Although cheaper and potentially more cost-effective, low-field-strength MR systems have traditionally suffered from several technical drawbacks, including long acquisition times, poor signal-to-noise ratio, inability to obtain thin slices, and poor spatial resolution [1]. However, recent improvements including better magnet homogeneity, improved receiving coil technology, and dedicated pulse sequences with low receiver bandwidth have generated a renewed interest in low-field-strength MR imaging [1].

To the best of our knowledge, this study represents the largest study providing a direct comparison of low- and high-field-strength images of the knee in the same individuals. Our results show that the level of diagnostic accuracy in anterior cruciate ligament tears and in meniscus tears is statistically identical for low- and high-field-strength MR imaging. For diagnosis of anterior cruciate ligament tear, our accuracy (93-98%) falls within the range of accuracies reported in the radiology literature for the diagnosis of anterior cruciate ligament tears (82-100%) [13,14,15]. Our accuracy in diagnosing meniscal tears (medial meniscus tears, 91-94%; lateral meniscus tears, 88-93%) also is well within the range reported in the literature (medial meniscus tears, 88-92%; lateral meniscus tears, 87-93%) [4, 10, 15, 16]. Our results agree with those of Parizel et al. [1] who used the same 0.2-T magnet as that used in our study and with those of Kladny et al. [2] and Kersting-Sommerhoff et al. [3] who used a dedicated 0.2-T system. Indeed, these three groups of researchers found no significant difference between the two field strength MR imagers in 10, 22, and 20 subjects, respectively. Franklin et al. [17], Shellock et al. [18], and Riel et al. [19] also reported that compared with arthroscopy, a dedicated-extremity low-field-strength system (0.2 T) was reliable when diagnosing meniscal tears and anterior cruciate ligament ruptures. However, a significant difference in diagnostic performance between low- and high-field-strength MR imaging of the knee has also been reported. Fischer et al. [4] found MR examinations of the knee at 0.35 T to be less accurate than examinations performed at 1.5 T for the medial meniscus, but the technique, interpreting radiologists, and magnets were not controlled; Kinnunen et al. [5] reported a specificity of only 25% for lesions of the lateral meniscus for images obtained at 0.1 T. Penrod et al. [20] reported a sensitivity of 17% and 25% for tears of the anterior horn of the medial meniscus and posterior horn of the lateral meniscus, respectively, imaged with a dedicated-extremity MR unit at 0.2 T.

If these different results in assessment of meniscal lesions may be explained in part by the different performances of the low-field-strength MR magnets, differences may also be caused by the different parameters used for the sequences. In our study, we used similar sequences and sections at 0.2 and 1.5 T. However, because of the inherent physical differences between the two field strengths (different relaxation times, different receiver bandwidth, different coil technology, and different field orientation), obviously sequence parameters for both systems were different. In fact, we decided to compare sequences for the two different MR systems with parameter settings as used in our optimal clinical routine on each unit. Because of the lower intrinsic signal-to-noise ratio at 0.2 T, more excitations were required, resulting in longer measurement times. The measurement time required for MR examination of the knee was approximately 15 min longer at 0.2 T than at 1.5 T.

Assessment of inter- and intraobserver variability values was also an aim of our study, because these features may affect the performance of the magnets [21]. These values, to our knowledge, have not been extensively reported. In fact, in two articles, researchers reported large interobserver variation at low-field-strength MR studies [6, 7]. In our study, inter- and intraobserver variability values showed excellent agreement at both field strengths.

Our study has several limitations. First, we did not use spin-echo or turbo spin-echo T2-weighted sequences in our study, because we use gradient-echo T2-weighted sequences in our clinical routine. Our accuracy in diagnosing anterior cruciate ligament or meniscal tears is indeed well within the range of accuracies reported in the literature. However, it is well known that at low field strength, the signal-to-noise ratio is better with gradient-echo than with spin-echo sequences, provided the field strength is homogeneous. Results might have been different if spin-echo T2-weighted sequences had been used at both field strengths. Second, a bias was inherent to the study because 129 of the 219 patients did not have surgical evaluation. Obviously, there is no way of knowing whether any of them were false-negatives except if a long-term follow-up is performed. However, we would like to stress that correlation between the results of the two field strengths was also excellent in these patients. Third, although the observers were unaware of the field strength used, they were able to identify the field strength used because of the difference in the appearance of the 0.2-T images (Figs. 1A,1B and 2A,2B). However, no difference in the diagnostic accuracy between high and low field strengths was found. Finally, the two reviewers were musculoskeletal radiologists and this also may have affected our results because we did not assess the quality of images and the ease of interpretation.

In conclusion, we found no significant difference in diagnostic performance between low- and high-field-strength MR imaging of the knee in the same patients. MR examination of the knee for meniscoligamentous lesions with a low-field-strength system may represent a low-cost alternative in terms of increasing health costs. Our results show that modern low-field-strength MR equipment successfully counters the intrinsically lower signal-to-noise ratio at the cost of somewhat longer imaging times.

References

Top Abstract Introduction Subjects and Methods Results Discussion References

 

1. Parizel PM, Dijkstra HAJ, Geenen PJ, et al. Low-field versus high-field MR imaging of the knee: a comparison of signal behaviour and diagnostic performance. Eur J Radiol 1995;19: 132 -138[Medline]
2. Kladny B, Gluckert K, Swoboda B, Beyer W, Weseloh G. Comparison of low-field (0.2 Tesla) and high-field (1.5 Tesla) magnetic resonance imaging of the knee joint. Arch Orthop Trauma Surg 1995;114: 281 -286
3. Kersting-Sommerhoff B, Gerhardt P, Golder W, et al. MRI of the knee joint: first results of a comparison of 0.2-T specialized system and 1.5-T high field strength magnet. Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 1995;162: 390 —395[Medline]
4. Fischer SP, Fox JM, DelPizzo W, Friedman MJ, Snyder SJ, Ferkel RD. Accuracy of diagnoses from magnetic resonance imaging of the knee. J Bone Joint Surg Am 1991;73-A: 2 -10[Abstract/Free Full Text]
5. Kinnunen J, Bondestam S, Kivioja A, et al. Diagnostic performance of low field MRI in acute knee injuries. Magn Reson Imaging 1994;12: 1155 -1160[Medline]
6. Drape J, Coevoet V, Sarazin LR, Pessis EG, Minoui A, Chevrot A. Accuracy of low-field MR imaging in detection and staging meniscal lesions of the knee (abstr). Radiology 1997;205(P): 186
7. Hof N, Kersting-Sommerhoff BA, Steinbom M, Sittek H, Wacker F, Bolze X. Low field MR imaging of the knee: a multicenter trial (abstr). Radiology 1997;205(P): 492
8. Dale Vellet A, Lee DH, Munk PL, et al. Anterior cruciate ligament tear: prospective evaluation of diagnostic accuracy of middle- and high-field-strength MR imaging at 1.5- and 0.5-T. Radiology 1995;197: 826 -830[Abstract/Free Full Text]
9. Mink JH, Levy BA, Crues JV III. Tears of the anterior cruciate ligament and menisci of the knee: MR imaging evaluation. Radiology 1988;167: 769 -774[Abstract/Free Full Text]
10. Barnett MJ. MR diagnosis of internal derangements of the knee: effect of field strength on efficacy. AJR 1993;161: 115 -118[Abstract/Free Full Text]
11. Crues JV III, Mink J, Levy LT, Ltysch M, Stoller DW. Meniscal tears of the knee: accuracy of MR imaging. Radiology 1987;164: 445 -448[Abstract/Free Full Text]
12. Cohen J. A coefficient of agreement for nominal scales. Educ Psych Meas 1960;20: 37 -46
13. Vahey TN, Broome DR, Kayes KJ, Shelbourne KD. Acute and chronic tears of the anterior cruciate ligament: differential features at MR imaging. Radiology 1991;181: 251 -253[Abstract/Free Full Text]
14. Lee JK, Yao L, Phelps CT, Wirth CR, Czajka J, Lozman J. Anterior cruciate ligament tears: MR imaging compared with arthroscopy and clinical tests. Radiology 1988;166: 861 -864[Abstract/Free Full Text]
15. Quinn SF, Brown TR, Szumowski J. Menisci of the knee: radial MR imaging correlated with arthroscopy in 259 patients. Radiology 1992;185: 577 -580[Abstract/Free Full Text]
16. Rubin DA. MR imaging of the knee menisci. Radiol Clin North Am 1997;35: 21 -44[Medline]
17. Franklin PD, Lemon RA, Barden HS. Accuracy of imaging the menisci on an in-office, magnetic resonance imaging extremity system. Am J Sports Med 1997;25: 382 -388[Abstract/Free Full Text]
18. Shellock FG, Stone K, Crues JV III. Assessment of the knee using an extremity MRI system: diagnostic findings compared to arthroscopy (abstr). Radiology 1998;209(P): 192
19. Riel KA, Reinisch M, Kersting-Sommerhoff B, Hof N, Merl T. 0.2-Tesla magnetic resonance imaging of internal lesions of the knee joint: a prospective arthroscopically controlled clinical study. Knee Surg Sports Traumatol Arthrosc 1999;7: 37 -41[Medline]
20. Penrod BJ, Simmons GE, Traino PJ, Maynard PR, Wennogle CF, Chang PS. Diagnostic accuracy of a low field strength extremity-only MR scanner in internal derangement of the knee: correlation with arthroscopy and 1.5 Tesla MRI (abstr). Radiology 1998;209(P): 192
21. De Smet AA, Norris MA, Yandow DR, Graf BK, Keene JS. Diagnosis of meniscal tears of the knee with MR imaging: effect of observer variation and sample size on sensitivity and specificity. AJR 1993;160: 555 -559[Abstract/Free Full Text]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				S. S. Boks, D. Vroegindeweij, B. W. Koes, M. G. M. Hunink, and S. M. A. Bierma-Zeinstra Follow-up of Posttraumatic Ligamentous and Meniscal Knee Lesions Detected at MR Imaging: Systematic Review Radiology, March 1, 2006; 238(3): 863 - 871. [Abstract] [Full Text] [PDF]

				E. H. G. Oei, J. J. Nikken, A. C. M. Verstijnen, A. Z. Ginai, and M. G. Myriam Hunink MR Imaging of the Menisci and Cruciate Ligaments: A Systematic Review Radiology, March 1, 2003; 226(3): 837 - 848. [Abstract] [Full Text] [PDF]

				F. Dubrulle, J. Delomez, A. Kiaei, P. Berger, C. Vincent, F.-M. M. Vaneecloo, and L. Lemaitre Mass Screening for Retrocochlear Disorders: Low-Field-Strength (0.2-T) versus High-Field-Strength (1.5-T) MR Imaging AJNR Am. J. Neuroradiol., June 1, 2002; 23(6): 918 - 923. [Abstract] [Full Text] [PDF]

				M. A. Bredella, C. Losasso, S. C. Moelleken, R. W. Huegli, H. K. Genant, and P. F. J. Tirman Three-Point Dixon Chemical-Shift Imaging for Evaluating Articular Cartilage Defects in the Knee Joint on a Low-Field-Strength Open Magnet Am. J. Roentgenol., December 1, 2001; 177(6): 1371 - 1375. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cotten, A. Articles by Gougeon, F. Search for Related Content PubMed PubMed Citation Articles by Cotten, A. Articles by Gougeon, F. Social Bookmarking                      What's this?