| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, University of Wisconsin School of Medicine and Public
Health, 600 Highland Ave., E3/311, Madison, WI 53792.
2 Central Illinois Radiological Associates, Peoria, IL.
3 Department of Orthopedics and Rehabilitation Medicine, University of Wisconsin
School of Medicine and Public Health, Madison, WI.
4 Department of Biostatistics, University of Wisconsin School of Medicine and
Public Health, Madison, WI.
Received August 16, 2007;
accepted after revision February 1, 2008.
Address correspondence to A. A. De Smet.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. We reviewed the records of 559 patients who underwent knee MR examinations and arthroscopy. We compared the positive predictive values (PPVs) of an MR diagnosis of a medial meniscal tear for differences in tear location or type, delay between knee injury and MRI, delay between MRI and arthroscopy, and the presence of an anterior cruciate ligament (ACL) tear. We also retrospectively reviewed the MR examinations of 50 longitudinal tears to compare the PPVs of various MRI findings of a longitudinal tear.
RESULTS. There was no association between either the delay between injury and MRI or the delay between MRI and arthroscopy and false-positive diagnoses. The PPV of 64% (32/50) for longitudinal tears was lower than the values of 83% (15/18) to 100% (116/116) for other types of medial meniscal tears. False-positive diagnoses of medial meniscal tears were more common in patients who had a prior episode of acute trauma (p = 0.004) or an ACL tear (p < 0.0001). Review of longitudinal tears revealed a decreased PPV when MRI showed signal contacting only the superior surface (p = 0.016) or when MRI showed signal contacting the surface at the meniscocapsular junction (p = 0.004). Four of the 18 menisci with a false-positive diagnosis of a longitudinal tear had a healed ACL tear noted at arthroscopy.
CONCLUSION. False-positive MR diagnoses of medial meniscal tears are more common for longitudinal tears than other tear types and are also more common with MR abnormalities at either the superior surface or the meniscocapsular junction. Spontaneous healing of longitudinal tears accounts for some false-positive MR diagnoses.
Keywords: anterior cruciate ligament • knee • meniscal tear • meniscus • sports medicine • trauma
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In our clinical practice, we noted that some patients with an apparent longitudinal tear of the medial meniscus on MRI did not have a tear confirmed at arthroscopy. Most of these patients with false-positive MR diagnoses of a meniscal tear had other intraarticular injuries or abnormalities that warranted surgical intervention. However, some patients had medial joint-line pain and the decision to operate was partially based on the MR diagnosis of a tear. We undertook this study to determine if there are clinical variables or certain tear types or locations that are associated with false-positive MR diagnoses of medial meniscal tears. We also wished to determine whether there were MRI findings associated with longitudinal tears that would suggest an apparent tear would not be confirmed at arthroscopy.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Medical Records Review
Using the electronic medical records of 1,963 patients who underwent knee
MR examinations at our institution from October 2001 through June 2004, we
selected all patients who met the following criteria: first, MR examination on
a 1.5-T magnet; second, available medical records with relevant history and
physical examination findings; third, no history of meniscal surgery; and,
fourth, knee arthroscopy after MR examination. Using these selection criteria,
we identified 559 patients as the patient population for this study.
We then reviewed the medical records of these 559 patients to determine the following clinical parameters: whether the patient had a specific knee injury or atraumatic onset of knee pain, the time between the onset of knee symptoms and MR examination, and the time between the MR examination and knee arthroscopy. We reviewed these variables to determine if there might be a healing effect related to the mechanism of injury or increased time for healing to occur.
We also reviewed the original MR inter pretations to determine the type and location of medial meniscal tears that had been diagnosed on MRI. In our department, we diagnose menisci as possibly torn if a sign of a tear is seen on one MR image and as torn if more than one image shows abnormal findings suggestive of a tear using previously published criteria [1, 6]. For this study, all menisci diagnosed as possibly torn or torn were considered as a diagnosis of a torn meniscus because an abnormality on one or more images is the standard MR criterion for the diagnosis of a tear. The MR examinations were originally inter preted by one of seven fellowship-trained muscu loskeletal radiologists who have 2–20 years of experience in knee MRI as faculty in a university medical center.
Tears were localized on MRI within the meniscus using a previously described classification system in which the meniscus is divided into thirds as the anterior horn, body, and posterior horn [7]. If a tear extended into more than one third of the meniscus, then it was described as involving the anterior horn and body; body and posterior horn; anterior horn and posterior horn; or anterior horn, body, and posterior horn. We added an additional location for tears confined to the root of the meniscus.
For describing the type of tears, we used a commonly used surgical classification that categorizes tears as horizontal, radial, longitudinal, flap, complex, or bucket handle [8]. A horizontal tear had meniscal signal abnormality contacting the surface of the meniscus in a horizontal orientation of less than 30° relative to the adjacent tibial plateau. A radial tear had signal to the meniscal surface in a vertical orientation begin ning in the free edge of the meniscus. A longi tudinal tear had a signal ab normality in a vertical orientation away from the free edge of the meni scus. A flap tear had a signal abnormality in an oblique orientation. A complex tear had signal abnormality contacting the meniscal surface in more than one orientation creating more than one tear fragment. A bucket-handle tear had a displaced fragment within the intercondylar notch.
Finally, we reviewed the operative reports to determine the location and type of meniscal tear and to determine whether an anterior cruciate ligament (ACL) tear or a meniscal tear was found at arthroscopy. The arthroscopic surgeries were performed by one of three orthopedic surgeons who are board-certified in sports medicine with 5–32 years of experience as faculty of a university medical center. One of the surgeons, an author of this study, performed knee arthroscopy in approximately 50% of the patients.
Arthroscopic findings were used as the reference standard for deciding whether an MR diagnosis of a tear was a true-positive or false-positive diagnosis. If the type of tear found at surgery differed from the MR-assigned type, the operative type was used for the final type assignment. If a meniscal tear was noted to be partially healed at arthroscopy, we considered the meniscus to be torn. If a completely healed tear was noted at arthroscopy, we considered the MR diagnosis of a tear to be a false-positive diagnosis.
Retrospective MR Review
One of the authors retrospectively evaluated the knee MR examinations of
all patients who had an MR diagnosis of a longitudinal tear of the medial
meniscus. Fifty longitudinal tears of the medial meniscus were diagnosed on
MRI. There were 31 male and 19 female patients with an average age of 27
years; they ranged in age from 11 to 53 years.
Each of the 559 MR examinations was performed using the same protocol on a 1.5-T magnet (Signa, GE Healthcare) with a 4-channel knee phased-array coil and fast spin-echo imaging. A field of view of 14 cm, slice thickness of 3 mm with 1.5-mm interslice gap, bandwidth of 20 kHz, and matrix of 256 x 192 were used for all four sequences except a matrix of 256 x 224 was used for the coronal T1-weighted images. The parameters for the coronal T1-weighted images were TR range/TE, 600–700/17; 1 signal average; and echo-train length, 3. The parameters for the coronal fat-saturated proton density–weighted images were 1,800–2,000/17, 1 signal average, and echo-train length of 4. The parameters for the sagittal proton density–weighted images were 2,000–2,200/17, 1 signal average, and echo-train length of 4. The parameters for the sagittal fat-saturated T2-weighted sequence were 3,000–3,400/60, 1 signal average, and echo-train length of 6.
The 50 MR examinations with an original diagnosis of longitudinal tear were reviewed with out knowledge of the arthroscopic findings, but the observer did know that a longitudinal tear had been originally diagnosed on the MR exam ination. For this retrospective review, each medial meniscus was evaluated for the following findings of a longitudinal tear: a vertical line of intra meniscal signal contacting the superior sur face, inferior surface, or both on any coronal or sag ittal imaging sequence. In addition, the ob server counted the number of MR images that showed a fluid-signal abnormality contacting the surface on T2-weighted images. The number of images with T2 signal abnormalities were counted because we assumed that fluid extending within the meniscus is a de finitive sign of a tear.
In addition, the retrospective observer noted if the signal contacted the surface at the meniscocapsular junction or in the outer, middle, or inner third of the meniscus. A peripheral tear within the meniscal substance was differentiated from a meniscocapsular junction separation by identifying low-signal-intensity meniscal tissue interposed between the tear and the capsule using both the sagittal proton density–weighted and sagittal T2-weighted images. This distinction of a separation from a peripheral tear was based on the MR appearance and not the arthroscopic findings because our orthopedists do not use the term "meniscocapsular separation."
The observer also evaluated the MR studies for T2 hyperintensity in the medial tibial plateau beneath the posterior horn of the medial meniscus. This finding, which is suspected to represent a bone contusion in patients with acute trauma, has been associated with peripheral tears [9].
Statistical Analysis
For the 559 patients, the relationship was examined between the positive
predictive value (PPV) and each of the clinical variables and MR variables.
For the subset of 50 patients with an MR diagnosis of a longitudinal tear, the
relation between the PPV and the MRI findings studied was examined retro
spectively. For the continuous predictors, MR delay and arthroscopy delay,
logistic regressions were per formed for PPV and Fisher's exact tests for
grouping of delays were performed for PPV. For the remaining categoric
variables, Fisher's exact tests were perform ed. A statistically significant
difference was con sidered to be present if the p value was less than
0.05.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
The PPVs for tears in different locations are shown in Table 2. There was no significant difference in the PPVs for tears at differing locations (p = 0.08–1.0).
|
Effect of Clinical Variables and ACL Tear on MR Accuracy of Diagnosis of Medial Meniscal Tear
The PPV for medial meniscal tears was significantly higher at 97% (141/146)
in patients with spontaneous onset of knee pain than at 88% (202/229) for
patients with a specific knee injury (p = 0.004). The PPV for medial
meniscal tears was significantly higher at 96% (270/282) for those without an
ACL tear than at 78% (73/93) for those with an ACL tear (p <
0.0001).
Logistic regression analysis of the delay from the onset of symptoms and MRI and the delay from MRI to arthroscopy did not find an association between the delays and false-positive diagnoses (p = 0.18–0.96). When grouping the delay from onset of knee injury symptoms to MR examination and the delay from MR examination to arthroscopy as less than 30 days, 30–60 days, or more than 60 days, we found no significant difference for true-positive or false-positive diagnoses in either delay (p = 0.18–0.39).
Longitudinal Tears of the Medial Meniscus
Thirty-seven longitudinal tears of the medial meniscus were found at
arthroscopy. All 37 tears were in the posterior horn. Thirty-two (86%) of
these 37 tears were diagnosed on the original MR interpretations. There were
18 false-positive MR diagnoses of a longitudinal tear in the posterior horn of
the medial meniscus with arthroscopic findings of a normal meniscus in 14 of
these patients and a completely healed longitudinal tear in four. Three of the
four tears that showed complete healing at arthroscopy were localized on MRI
as occurring at the meniscocapsular junction and the other as in the
peripheral third of the meniscus.
Clinical Variables and ACL Tear Versus Longitudinal Tear MR Accuracy
Of the 50 patients with an MR diagnosis of a longitudinal tear, 40 suffered
acute sports injuries, nine had other acute knee trauma, and one had chronic
knee pain for years. All 18 patients with false-positive diagnoses and 31 of
the 32 with true-positive diagnoses had a specific episode of knee trauma
before MR examination.
The PPV of an MR diagnosis of a longitudinal meniscal tear was 63% (27/43) for the 43 patients with a complete ACL tear and 71% (5/7) for the seven patients without an ACL tear (p = 1.0).
The PPV of a longitudinal meniscal tear was not significantly different for patients who underwent MRI less than 30 days after injury (19/29), 30–60 days after injury (2/5), and more than 60 days after injury (10/16) (p = 0.51). The PPV of a longitudinal meniscal tear was not significantly different for patients who underwent arthroscopy less than 30 days after MR examination (6/10), 30–60 days after MR examination (6/11), and more than 60 days after MR examination (20/29) (p = 0.66). Nine of the 50 patients with an MR diagnosis of a longitudinal tear had a combined interval between injury and arthroscopy of less than 48 days with two of these seven having false-positive diagnoses. The remaining 41 patients had an interval from injury to arthroscopy ranging from 48 to 329 days with an average of 121 days.
MRI Findings in False-Positive and True-Positive Diagnoses of Longitudinal Medial Meniscal Tears
Because all of the longitudinal tears diagnosed on MRI involved the
posterior horn, statistical analysis of the MRI findings was performed using
only the sagittal imaging findings. Fifty longitudinal medial meniscal tears
were diagnosed on MRI with 32 true-positive and 18 false-positive diagnoses.
In four of the false-positive diagnoses, a healed tear was found at
arthroscopy corresponding to the location of the tear diagnosed on MRI. In
four of the 32 true-positive diagnoses, partial healing of the tear was noted
at arthroscopy. All 50 longitudinal tears originally diagnosed on MRI were
present in the peripheral third of the meniscus or at the meniscocapsular
junction.
The predictive value of a positive diagnosis was lower on sagittal proton density– weighted images when the meniscal signal contacted only the superior surface (3/10) than when the signal contacted only the inferior surface (14/19) or both surfaces (15/21). True-positive and false-positive diagnoses occurred when the signal abnormality involved either the superior surface alone (Figs. 1A, 1B and 2A, 2B) or both surfaces (Figs. 3A, 3B and 4A, 4B). Fisher's exact tests for the inequality of these values did not reach statistical significance (p = 0.52). However, when comparing superior surface contact with grouped inferior surface and both surfaces contact, the PPV was significantly lower for tears involving only the superior surface (p = 0.016).
|
|
|
|
|
|
|
|
The predictive value of a positive diagnosis was lower on sagittal fat-saturated T2-weighted images when the meniscal signal contacted only the superior surface (5/13) or did not have surface contact (4/8) than when the signal contacted only the inferior surface (8/11) or both surfaces (15/18). Fisher's exact tests for the inequality of these predictive values of each finding separately did not reach statistical significance (p = 0.058). However, comparison of the grouped inferior surface and both surfaces contact with the grouped superior surface contact or no surface contact yielded a significantly lower PPV when signal contacted only the superior surface or did not contact this surface (p = 0.016).
There was no significant difference in the predictive value of a positive diagnosis with an increasing number of MR images showing signal to the surface on T2-weighted images (Table 3), both as ungrouped (p = 0.27) or grouped as 0 or one image compared with two or more images (p = 0.27). Although the PPV appeared to decrease with an increase in the number of positive images from two to four (Table 3), this difference was not statistically significant. The apparent trend was thought to be due to chance related to the small number of cases in each subgroup.
|
The predictive value of a positive diagnosis was considerably lower on T2-weighted images when the tear was at the meniscocapsular junction (7/21) than when the tear was in the outer third but central to the meniscocapsular junction (21/25) (p = 0.0004).
Thirty-two patients had a bone contusion of the tibial plateau underlying the posterior horn of the medial meniscus (Figs. 1B and 4B). Of these 32 patients, 30 had an associated ACL tear. There was no significant difference in the predictive value of a positive diagnosis in menisci with (21/32) or without (11/18) an underlying medial tibial plateau bone bruise (p = 0.77).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We found no significant difference in PPV for tears in different locations (Table 2). However, because 76% (285/375) of the tears diagnosed on MRI were in the posterior horn, significant differences may not have been detected at locations with a lower frequency of tear.
Although investigators have reported errors in the MR diagnosis of meniscal tears in previous studies, no previous study, to our knowledge, has analyzed whether false-positive diagnoses are significantly more common for specific types of tears [1–4]. In one study, researchers noted without statistical analysis that five of seven false-positive MR diagnoses of medial meniscal tears occurred with a longitudinal pattern [2]. Other investigators have noted without statistical analysis that false-positive diagnoses are more common in the posterior horn [1, 3, 4].
A longitudinal tear of the medial meniscus is a distinctive tear that occurs primarily in the posterior horn and is highly associated with acute trauma and an ACL tear [7, 10–12]. In our study, 98% of our 50 patients with an MR diagnosis of a longitudinal meniscal tear had a specific episode of knee trauma, 86% (43/50) had an ACL tear, and all 50 longitudinal tears occurred in the outer third or at the meniscocapsular junction of the meniscus.
We found that PPV was decreased when patients had an acute knee injury rather than chronic pain and when patients had an ACL tear. We hypothesize that these apparent false-positive tears are tears that had healed. This hypothesis is supported by the clinical observation that meniscal tears can heal spontaneously, especially with the hemorrhage associated with an ACL tear [13].
Despite our initial hypothesis that spontaneous tear healing would increase with the delay between the initial injury and surgery, we found that PPV was not lower with longer MRI or operative delay. We had expected that with longer delays, a tear would be more likely to heal spontaneously. Spontaneous healing of meniscal tears is documented in the surgical literature. Weiss et al. [14] followed up eight patients in whom peripheral longitudinal tears were identified at initial arthroscopy but were left untreated. At repeat arthroscopy an average of 26 months later, four of the eight peripheral longitudinal tears had healed [14]. In another study, seven longitudinal tears were noted at arthroscopy in the posterior horn of the medial meniscus in patients with ACL tears. These tears were treated conservatively, and on repeat arthroscopy 3 months later, six of the seven tears had healed [15].
It is likely that we did not find a decreased PPV with increasing MRI or operative delay for our patients with longitudinal tears because a high percentage (89%) of the patients with false-positive diagnoses had an interval of more than 6 weeks from injury to arthroscopy. Apparently, this 6-week interval is sufficient for healing to occur so an increasing time for healing is not an important factor in determining the potential for spontaneous healing. In the two patients with a false-positive MR diagnosis of a longitudinal tear and whose interval from injury to arthroscopy was less than 30 days, the tears may have healed rapidly, the apparent tear on MRI might have been an old injury, the tear may have been missed at arthroscopy, or these cases may have been an incorrect MR diagnosis of a tear.
We identified several MRI findings that suggest an MR diagnosis of a longitudinal tear may be a false-positive diagnosis. A false-positive diagnosis was more likely when the signal on either sagittal proton density–weighted or sagittal T2-weighted images contacted only the superior surface, with PPVs of such signal of 30% and 38%, respectively. The PPV was also lower when the apparent tear was at the meniscocapsular junction, with a PPV of only 33%.
In a previous study, 40% of patients with bone contusions in the tibial plateau underlying the posterior horn of the medial meniscus were found to have peripheral tears of the overlying meniscus [9]. All of the patients in that prior study had ACL tears while 30 of our 32 patients with such a contusion had an ACL tear. Although we identified this bone contusion pattern in 64% of our patients with an MR diagnosis of a longitudinal meniscal tear, we did not find any significant difference in the frequency of false-positive MR diagnoses in patients with or in those without a tibial bone contusion.
Our MR study also confirmed our clinical observation that spontaneous healing of meniscal tears accounts for some cases of false-positive MR diagnoses of a meniscal tear. In our study, four of 18 patients with a false-positive MR diagnosis of a longitudinal tear were found to have a healed tear at arthroscopy. In addition, four patients with a true-positive MR diagnosis of a longitudinal tear had partially healed tears noted at arthroscopy. Based on the similar MR appearances of the confirmed tears, the healed tears, and the partially healed tears, we hypothesize that some of our false-positive diagnoses are due to tears that spontaneously healed without residual evidence of the tear at arthroscopy.
There have been two previous studies on knee MRI in which healed meniscal tears have been identified at arthroscopy as a cause of a false-positive diagnosis of a meniscal tear [16, 17]. In a study of MRI with correlative arthroscopy in 1,014 patients, a healed tear was found at arthroscopy in seven of 71 false-positive diagnoses of a medial meniscal tear [16]. In a study comparing MRI findings and arthroscopic findings in 91 patients, a healed tear with scarring was noted at arthroscopy in one of seven false-positive diagnoses of a medial meniscal tear. These two studies did not indicate the location or type of tear that had healed.
Our study is limited by its retrospective nature. Because we did not perform a prospective study with criteria for diagnosing a longitudinal tear accepted by both radiologists and orthopedic surgeons, we cannot be certain that we identified all menisci with a longitudinal pattern of tear. However, this pattern of tear is distinctive and it is likely that most of these tears were correctly categorized. On retrospective review, all of the tears originally diagnosed on MRI as a longitudinal tear had a vertically oriented line of signal abnormality contacting the meniscal surface.
Another limitation was our division of peripheral tears on MRI into meniscocapsular junction versus outer one third. Our surgeons performing arthroscopy seldom use the term "meniscocapsular separation" because distinguishing a partial or a healed meniscocapsular junction injury from a deep sulcus can be difficult. Therefore, the T2-weighted images were used to make this distinction. The retrospective reviewer found this distinction occasionally difficult because edema related to the injury resulted in the posterior margin of the meniscus not always being clearly defined. The difficulty in diagnosing meniscocapsular junction tears has been previously described [18].
A third limitation of our study is the use of fast spin-echo imaging for the diagnosis of meniscal tears because one study found that fast spin-echo imaging has a lower sensitivity for meniscal tears [19]. However, a recent study found no significant change in sensitivity or specificity with fast spin-echo imaging compared with conventional spin-echo imaging [6]. Because specificity has not been reported to be affected by pulse sequence, the incidence of false-positive examinations, which was the focus of our study, should not have been affected by pulse sequence.
A fourth limitation of this study was the decision to consider original MR diagnoses of a possible tear based on the presence of a meniscal abnormality on only one image as a tear. We elected to use this criterion because most published studies on MR accuracy for the diagnosis of a meniscal tear consider MRI findings to be positive for a tear if one or more images show abnormal findings. If we had considered these MR diagnoses of a possible tear as not torn, we would have had fewer false-positive MR studies. As noted in a review on the MR diagnosis of meniscal tears, a tear can be diagnosed when internal signal unequivocally contacts the surface of a meniscus on one image but specificity increases when this criterion is met on two or more images [20].
A final limitation of the study is the use of arthroscopy as a reference standard. Although it is unlikely that the arthroscopies in this series were 100% accurate in diagnosing all meniscal abnormalities, care was taken to have as accurate an arthroscopic diagnosis as possible. In all knees, the meniscus was visualized from the standard anterior portals and was probed to expose under and upper surface tears and to evaluate the mobility of the meniscus. When visualization proved difficult or when MRI suggested a peripheral posterior meniscal tear, the posterior compartment was inspected with the arthroscope placed through the contralateral portal and passed through the intercondylar notch to look down on the meniscal root. The surgeons were aware of the MRI findings in each case; in fact, the MR examinations were often reviewed in the operating room before or during surgery. Thus, when a tear was suspected on imaging, particular attention was paid to the involved meniscus in an attempt to minimize the risk of missing a torn meniscus.
We undertook this study to determine if clinical variables or MRI findings that might suggest which medial menisci with apparent longitudinal tears on MRI would be intact at arthroscopy. The clinical decision of one of the coauthors to operate on patients in this study was based on the following criteria: For patients with ACL tears, the primary indication for surgery was the instability caused by the ACL tear. A peripheral repairable meniscus tear noted on MRI was also considered as a factor that increased the risk of nonsurgical treatment. For those patients without an ACL tear, the decision to operate was based on the clinical picture including history, physical examination, and MRI. If the diagnosis of a tear was solely supported by MRI, patients were observed for the development of corroborative signs or symptoms.
Based on our study, the criteria to operate may be expanded to include our findings that false-positive diagnoses are significantly more common after acute trauma and when there is an associated ACL tear. In addition, false-positive MR diagnosis of a longitudinal tear is more common when the tear is at the meniscocapsular junction and when the intrameniscal signal contacts only the superior surface.
Our study also confirmed our clinical impression that false-positive MR diagnoses of a medial meniscal tear are most common with longitudinal tears and that these tears can spontaneously heal.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Nishikawa, Y. Imanaka, M. Sekimoto, K. Hayashida, and H. Ikai Influence of Verification Bias on the Assessment of MRI in the Diagnosis of Meniscal Tear Am. J. Roentgenol., December 1, 2009; 193(6): 1596 - 1602. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. W. Grossman, A. A. De Smet, and K. Shinki Comparison of the Accuracy Rates of 3-T and 1.5-T MRI of the Knee in the Diagnosis of Meniscal Tear Am. J. Roentgenol., August 1, 2009; 193(2): 509 - 514. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
