MRI of Acute Meniscal Injury Associated with Tibial Plateau Fractures: Prevalence, Type, and Location

Size: px

Start display at page:

Download "MRI of Acute Meniscal Injury Associated with Tibial Plateau Fractures: Prevalence, Type, and Location"

Imogen Blair
6 years ago
Views:

1 Musculoskeletal Imaging Original Research Mustonen et al. MRI of Meniscal Injury Musculoskeletal Imaging Original Research ntti O. T. Mustonen 1 Mika P. Koivikko 1 Jan Lindahl 2 Seppo K. Koskinen 1 Mustonen OT, Koivikko MP, Lindahl J, Koskinen SK Keywords: arthroscopy, knee, MDCT, meniscal, MRI, tibial plateau fracture DOI: /JR Received February 10, 2008; accepted after revision pril 11, Department of Radiology, Helsinki Medical Imaging Center, Helsinki University Central Hospital, Toolo Trauma Center, Topeliuksenkatu 5, FIN 00029, Helsinki, Finland. ddress correspondence to. O. T. Mustonen (antti.mustonen@helsinki.fi). 2 Department of Orthopedics and Traumatology, Helsinki University Central Hospital, Toolo Trauma Center, Helsinki, Finland. JR 2008; 191: X/08/ merican Roentgen Ray Society MRI of cute Meniscal Injury ssociated with Tibial Plateau Fractures: Prevalence, Type, and Location OJECTIVE. The purpose of this study was to evaluate the prevalence, type, and location of meniscal injuries, particularly to assess the prevalence of unstable meniscal s in acute knee trauma with tibial plateau fractures. MTERILS ND METHODS. total of 78 menisci were evaluated in 39 patients who had undergone knee MDCT and MRI. Meniscal s were classified as horizontal, vertical (subdivided into longitudinal and radial), flap, bucket-handle, or complex. The presence of meniscal contusion was documented. The anterior horn, body, and posterior horn were assessed separately for both menisci. Knee arthroscopy was performed on 28 patients. RESULTS. Of the 39 patients in the study, 24 had detectable abnormal menisci, for a total of 33 abnormal menisci (42%). mong the 33 meniscal abnormalities were 11 longitudinal s (33%), 17 contusions (52%), four flap s (12%), six horizontal s (18%), and six radial s (18%). mong the 16 patients with meniscal s (41% of the 39), 14 patients had an unstable. No significant correlation was found between degree of articular depression and site or morphologic features of the meniscal injury. Correspondingly, no statistical correlation was evident between normal menisci and degree of articular depression, nor was a significant correlation found between differing fracture groups and meniscal findings. CONCLUSION. high percentage of patients (36%) with a tibial plateau fracture had an unstable meniscal. If a meniscal is detected preoperatively, meniscal surgery can be combined with fracture fixation, and reoperation can be avoided. large number of meniscal contusions were found. wareness of this abnormality can help radiologists increase specificity by avoiding false-positive findings of meniscal. T ibial plateau fractures are a complex group of injuries that often manifest with severe ligamentous or meniscal injury challenging to the orthopedic surgeon [1, 2]. Previous studies [1 6] have shown a high number of meniscal s in cases of tibial plateau fractures. ecause meniscal injuries can lead to posttraumatic arthrosis and decreased knee function [7, 8], it is important that meniscal s be accurately repaired or débrided to maintain knee joint stability and congruency and to minimize articular contact pressure [7 9]. Meniscal stability is an important concept in the differentiation of symptomatic and asymptomatic meniscal s. Unstable s are associated with pain, which is an important criterion for many orthopedic surgeons considering meniscal treatment [10]. In addition to radiography and CT, MRI has become an increasingly popular imaging technique for surgical planning [2, 6, 8, 11]. To our knowledge, however, detailed MRI analysis of meniscal injury, such as prevalence of unstable meniscal s and exact location of a, in acute knee trauma with tibial plateau fractures is lacking. The purpose of this retrospective study was to evaluate the prevalence, type, and location of meniscal injury, particularly to assess the prevalence of unstable meniscal s that would justify routine use of MRI in the evaluation of acute knee trauma involving tibial plateau fractures. Materials and Methods This retrospective study was conducted at a hospital that serves almost 1.5 million people and is the leading level 1 trauma center in its country. The study was approved by the hospital ethics committee. Patient Selection Using the hospital PCS, we retrieved all emergency department MDCT requests regarding knee trauma from January 2001 to the end of ugust 1002 JR:191, October 2008

2 MRI of Meniscal Injury 2007 (718 patients). Indications for MDCT were ruling out a fracture or assessing the morphologic features of a fracture. MDCT in the acute phase (within 1 week of injury; mean, within 24 hours) showed 554 patients had acute knee injury and tibial plateau fracture. When clinically indicated, preoperative MRI was performed on all patients with suspected soft-tissue injuries. total of 46 patients underwent MRI within 3 weeks (mean, 3 days) of injury. Patient charts allowed verification of clinical history, and patients who had previously undergone knee surgery were excluded (three patients). To reduce the number of menisci with possible degenerative changes, patients with any sign of osteoarthritis also were excluded (four patients). This exclusion would help to differentiate meniscal findings due to contusion from those due to degeneration. bsence of osteoarthritis was determined from MDCT and MR images. The final study group comprised 39 patients (21 men, 18 women; mean age, 37 years; range, years) and 78 menisci. Only the injured knee was studied. The injury mechanisms were traffic accident in 19 cases, simple fall in nine cases, sports injury in six cases, and twisting injury in five cases. rthroscopy was considered the reference standard in the cases of 28 patients. Eight of the 28 had normal MRI findings, and arthroscopy was performed with clinical indications. MRI was considered the reference standard in the cases of 11 patients. These patients had neither MRI findings of abnormal menisci nor clinical symptoms of meniscal, so arthroscopy was deemed ethically unacceptable. Image nalysis The images were evaluated on clinical PCS workstations (Impax DS3000, version 4.5, gfa- Gevaert) by two radiologists with 2 and 8 years of subspecialty experience in musculoskeletal radiology. ware of the preoperative clinical history, they reviewed MDCT and MR images and reached a consensus while blinded to initial interpretations and surgical findings. MDCT Protocol ll patients underwent knee MDCT with a 4-MDCT scanner (LightSpeed QX/i, GE Healthcare). The routine protocol was collimation, mm; interval, 0.62 mm; gantry rotation time, 1.0 second; pitch 3; table feed, 3.75 mm; 120 kv; 150 m; approximate total exposure time, seconds. MDCT acquisition ranged from the upper pole of the patella to caudal to the fibular head. Routine multiplanar reconstructions were made in standard sagittal and coronal planes with a slice thickness of 2.0 mm and reconstruction increment of 2.0 mm. MRI Protocol The MR images were obtained with a 1.5-T unit (Signa MRI Echospeed, GE Healthcare) with a dedicated transmit receive quadrature lower extremity coil. The standard clinical sequences were coronal T2-weighted fast spin-echo with fat saturation, sagittal proton-density spin-echo, sagittal T2-weighted fast spin echo, and axial proton-density fast spin-echo with fat saturation (Table 1). rthroscopy Protocol rthroscopy was performed in the acute phase of injury by three orthopedic surgeons specialized in trauma surgery and knee arthroscopy with more than 10 years of experience each. ll but one patient with MRI findings of underwent arthroscopy. The exception was patient 24 (Table 2), who had a radial in the posterior horn of the medial meniscus (Fig. 1). The knee was stable, and the nondisplaced 2 fracture was managed conservatively. In this case, repairing the menisci and arthroscopy were deemed unnecessary. Classification of Tibial Plateau Fractures Tibial plateau fractures were assessed according to the rbeitsgemeinschaft für Osteosynthesefragen (O)-Orthopaedic Trauma ssociation classification [12]. ll fractures were classified type or C and group 1, 2, or 3 (Fig. 2). The fractures were not further divided into subgroups because that practice is not routinely used at our institution. vulsion fractures of the anterior or posterior cruciate ligament (1.3) (Fig. 2) from its tibial insertion were included when the tibial plateau also was fractured (e.g., fractures) (Fig. 3). The maximal depression of the tibial articular surface was measured. Classification of Meniscal Injury at MRI The menisci were assessed as normal or torn. Tears were classified as horizontal (Fig. 4), vertical (subdivided into longitudinal [Fig. 3] and radial), TLE 1: MRI Parameters Parameter T2-Weighted Fat-Suppressed Coronal Fast Spin Echo flap (Fig. 5), bucket-handle (Fig. 6), or complex [7]. complex consisted of two or more configurations (Fig. 5). The presence of any meniscal contusion [13] was documented. The anterior horn, body, and posterior horn were assessed separately for both menisci. Possible meniscocapsular separation was documented. MRI Criteria meniscus was considered torn when an area of increased internal signal intensity was in unequivocal contact with a meniscal articular surface on one or more images. Meniscal contusion (Fig. 7) was defined as an area of increased internal signal intensity in the meniscus in contact with the articular surface but less discrete and less well defined than the signal intensity associated with a or with intrasubstance degeneration [13]. Except for discoid and buckled (flounce) menisci, menisci with abnormal morphologic features (e.g., radial s) were considered torn [7]. Horizontal s were considered stable, and longitudinal, radial, flap, and complex s unstable [10]. Ligamentous Injuries nterior cruciate ligament, posterior cruciate ligament, medial collateral ligament, and lateral collateral ligament injuries were documented at MRI. Management of ligamentous injuries was verified from patient charts. Classification of Meniscal Injury at rthroscopy The surgeons at our institution divide menisci into three parts (anterior horn, body, and posterior horn) and classify meniscal s as longitudinal, flap, radial, horizontal, root, or bucket handle. Data nalysis In cases with arthroscopic verification, values determined were sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of MRI in the detection of meniscal. Proton Density Sagittal Spin Echo T2-Weighted Sagittal Fast Spin Echo Proton-Density Fat-Suppressed xial Fast Spin Echo TR (ms) 4,700 1,800 4,000 3,000 Effective TE (ms) Echo train length 8 N 16 6 Matrix size Number of signals acquired Field of view (cm 2 ) Slice thickness/space (mm) 4/1 3/1 3/1 3/1 Note N = not applicable. JR:191, October

3 Mustonen et al. TLE 2: MRI Findings on bnormal Menisci (n = 24) Patient No. Lateral Meniscus Statistical nalysis The appropriate procedures in the SPSS program (version 15.0, SPSS) were used for the statistical computation. Correlation of articular depression and meniscal findings between different groups (normal menisci, meniscal contusions, and torn menisci) was analyzed with one-way analysis of variance (Tukey test). Correlation of articular depression and meniscal findings between normal menisci and abnormal Medial Meniscus nterior Horn ody Posterior Horn nterior Horn ody Posterior Horn menisci ( or contusion) was analyzed with the Mann-Whitney U test (nonparametric data). The statistical significance level was set at p < Class a rticular Depression (mm) 1 Normal Contusion Contusion Normal Normal Contusion C2 4 2 Longitudinal b Longitudinal Longitudinal b Normal Normal Normal 3 11 b 3 Normal Normal Normal Contusion Radial Normal 1 0 (false-positive) 4 Normal Normal Normal Normal Contusion Normal Contusion Normal Normal Normal Normal Normal Longitudinal b Longitudinal Longitudinal b Normal Normal Longitudinal, 1 0 b horizontal, flap 7 Contusion Normal Normal Normal Normal Normal Normal Normal Longitudinal, Normal Normal Longitudinal 2 3 Flap (false-positive) 9 Normal Normal Normal Normal Normal Longitudinal, flap Normal Normal Normal Normal Longitudinal Longitudinal, flap Contusion Normal Normal Normal Normal Longitudinal Normal Normal Normal Normal Horizontal Horizontal Contusion Longitudinal Contusion, horizontal Normal Horizontal Horizontal Contusion Normal Normal Normal Normal Horizontal Normal Normal Contusion Normal Contusion, radial Contusion Normal Normal Radial Normal Contusion Radial (falsepositive) (false-positive) 17 c Normal Normal Normal Normal Normal Contusion c Contusion Normal Normal Normal Normal Normal Normal Normal Contusion Normal Normal Normal (false negative) 20 c Normal Normal Normal Normal Normal Contusion Normal Normal Longitudinal Normal Normal Normal Normal Normal Normal Normal Normal Longitudinal Normal Normal Radial Normal Normal Normal c Normal Contusion Normal Normal Horizontal Radial 2 2 a rbeitsgemeinschaft für Osteosynthesefragen Orthopaedic Trauma ssociation classification. b ucket-handle. c No arthroscopy. Results MDCT Findings The distribution of tibial plateau fractures on MDCT according to the O Orthopaedic Trauma ssociation classification is shown in Table 3. ll types of tibial plateau fractures except C1 were visible. The average articular depression of the tibial plateau was 3.2 mm (range, 0 25 mm). MRI Findings Of the 78 menisci studied, 33 abnormal menisci (42%) (17 medial, 16 lateral) were detected in 24 patients (Table 2). Sixteen the 39 patients (41%) had a meniscal, 14 of them having an unstable. n unstable was evident in JR:191, October 2008

$MRI of Meniscal Injury E Fig. 1 76-year-old man with MDCT findings of 2 tibial plateau fracture after simple fall.$ $2 rbeitsgemeinschaft für Osteosynthesefragen Orthopaedic Trauma ssociation classification of tibial plateau fractures. 1.3 fractures were included in study if they existed with type or C fractures.$ $Type 1 () is pure slit fracture, 2 (C) is pure depression fracture, and 3 (D) is split depression fracture. E G, Drawings show type C fractures involve both condyles.$ stable was found in seven medial (18%) and 10 lateral (25%) menisci.

stable was found in seven medial (18%) and 10 lateral (25%) menisci.

The distribution of s and contusions in different meniscal parts is shown in Table 4. rthroscopic Findings rthroscopy was performed on the knees of 28 patients (72%), and 56 menisci were assessed.

$In the 1, 2, and 3 fracture groups, a total of 12 meniscal s with arthroscopic verification were detectable.$ MRI Correlation with rthroscopy total of two false-negative findings (a small radial and a small isolated flap in the posterior horn of the medial meniscus, which were slightly trimmed at

MRI Correlation with rthroscopy total of two false-negative findings (a small radial and a small isolated flap in the posterior horn of the medial meniscus, which were slightly trimmed at

In these cases, the menisci were intact, and no degenerative changes were evident at arthroscopy.

4 MRI of Meniscal Injury E Fig year-old man with MDCT findings of 2 tibial plateau fracture after simple fall. Coronal T2-weighted fatsuppressed MR image (TR/TE, 4,740/40) shows radial (arrow) in posterior horn of medial meniscus. Findings were not verified with arthroscopy. Fig. 2 rbeitsgemeinschaft für Osteosynthesefragen Orthopaedic Trauma ssociation classification of tibial plateau fractures. 1.3 fractures were included in study if they existed with type or C fractures., Drawing shows avulsion fracture (1.3) of anterior or posterior eminence. D, Drawings show type fractures (not specific to medial or lateral plateau) involve one condyle. Type 1 () is pure slit fracture, 2 (C) is pure depression fracture, and 3 (D) is split depression fracture. E G, Drawings show type C fractures involve both condyles. Type C1 (E) is metaphyseal simple fracture, C2 (F) is metaphyseal multifragment fracture, and C3 (G) is multifragment fracture. C F D G medial (25%) and seven lateral (18%) menisci. stable was found in seven medial (18%) and 10 lateral (25%) menisci. Complex s occurred in nine menisci (12%) and bucket-handle s in two (3%), but neither isolated flap s nor meniscal root injuries were found. The distribution of s and contusions in different meniscal parts is shown in Table 4. rthroscopic Findings rthroscopy was performed on the knees of 28 patients (72%), and 56 menisci were assessed. That group comprised 20 patients with a detected on MRI; eight patients with normal MRI findings had symptoms. In the 1, 2, and 3 fracture groups, a total of 12 meniscal s with arthroscopic verification were detectable. Six of those 12 s were at the site of the fracture, four were ipsilateral, and in two cases both menisci were torn. Two meniscocapsular separations, also evident on MRI, were sutured. MRI Correlation with rthroscopy total of two false-negative findings (a small radial and a small isolated flap in the posterior horn of the medial meniscus, which were slightly trimmed at arthroscopy) and four false-positive interpretations (three posterior horn of the medial meniscus and one posterior horn of the lateral meniscus) were made. In these cases, the menisci were intact, and no degenerative changes were evident at arthroscopy. For meniscal, the diagnostic values of MRI were sensitivity, 88%; specificity, 90%; accuracy, 89%; positive predictive value, 78%; and negative predictive value, 95%. Ligamentous Injuries total of 19 anterior cruciate ligament, 12 posterior cruciate ligament, 13 medial collateral ligament, and eight lateral collateral ligament injuries were detected. Of those injuries, 15 anterior cruciate ligament, nine posterior cruciate ligament, two medial collateral ligament, and three lateral collateral ligament were managed surgically. Statistical Findings No significant correlation emerged between site and morphologic features of meniscal injury and degree of articular depression. Correspondingly, no significant correlation was apparent between normal meniscus and degree of articular depression, nor was there a significant association between fracture group and meniscal findings. JR:191, October

$Mustonen et al. Fig. 3 28-year-old man with MDCT findings of 2 and 1.3 tibial plateau fractures after traffic accident. Findings were verified with arthroscopy.$ $4 33-year-old man with MDCT findings of 2 tibial plateau fracture after traffic accident. Findings were verified with arthroscopy.$ $one marrow edema is present in proximal tibia (arrowheads, ). Discussion cute knee injury with tibial plateau fracture is commonly encountered in emergency departments.$ $Radiography and MDCT have been widely used for preoperative evaluation of these injuries [14, 15], but high-energy tibial plateau fractures are often associated with severe soft-tissue injury and$ dequate physical examination of the knee in the acute phase of injury is demanding, and clinical detection of meniscal injury is particularly challenging or even impossible because of pain and

dequate physical examination of the knee in the acute phase of injury is demanding, and clinical detection of meniscal injury is particularly challenging or even impossible because of pain and

$The association between meniscal s and tibial plateau fractures has been well documented [1 6, 9, 11]. In an arthroscopic study, Vangsness et al.$ $[4] found meniscal injury necessitating surgery in 47% of knees with tibial plateau fractures.$

5 Mustonen et al. Fig year-old man with MDCT findings of 2 and 1.3 tibial plateau fractures after traffic accident. Findings were verified with arthroscopy., Coronal T2-weighted fat-suppressed MR image (TR/TE, 3,800/40) shows fracture line (arrows)., Sagittal proton density MR image (1,700/20) shows longitudinal (arrow) in posterior horn of medial meniscus. Fig year-old man with MDCT findings of 2 tibial plateau fracture after traffic accident. Findings were verified with arthroscopy. and, Coronal T2-weighted fat-suppressed (TR/TE, 4,740/40) () and sagittal proton-density (1,800/20) () MR images show horizontal (arrow) in posterior horn of medial meniscus. one marrow edema is present in proximal tibia (arrowheads, ). Discussion cute knee injury with tibial plateau fracture is commonly encountered in emergency departments. Radiography and MDCT have been widely used for preoperative evaluation of these injuries [14, 15], but high-energy tibial plateau fractures are often associated with severe soft-tissue injury and meniscal s [2, 3, 5, 6, 8, 11]. dequate physical examination of the knee in the acute phase of injury is demanding, and clinical detection of meniscal injury is particularly challenging or even impossible because of pain and swelling [6, 9]. MRI is painless, and the presence of possible joint effusion does not interfere with its accuracy in the detection of meniscal injury [16, 17]. MRI has therefore become an increasingly popular imaging technique before surgery [2, 3, 5, 6, 8, 11]. The association between meniscal s and tibial plateau fractures has been well documented [1 6, 9, 11]. In an arthroscopic study, Vangsness et al. [4] found meniscal injury necessitating surgery in 47% of knees with tibial plateau fractures. To our knowledge, however, no MRI studies with a more detailed analysis of meniscal injury (such as prevalence of unstable meniscal s and exact location of s) in the setting of acute knee trauma with tibial plateau fracture have been conducted. In clinical practice, debate has arisen regarding symptomatic and asymptomatic meniscal s, pain being thought to relate to meniscal stability [10]. Pain has been an important criterion for many orthopedic surgeons considering meniscal treatment [10]. t palpation with a probe at arthroscopy, Fig year-old woman after simple fall. Findings were verified with arthroscopy., Coronal MDCT image shows 2 tibial plateau fracture of lateral condyle with 3-mm articular depression (arrows)., Coronal T2-weighted fat-suppressed MR image (TR/TE, 3,800/40) shows complex (longitudinal and flap) (arrows) in posterior horn of lateral meniscus. No isolated flap is visible. one marrow edema is present in proximal tibia (arrowheads). C, Sagittal proton-density MR image (1,800/20) shows part of meniscus (white arrow) is flipped behind and below posterior horn. rticular depression (black arrow) is visible. C 1006 JR:191, October 2008

$, Coronal MDCT image shows type 3 tibial plateau fracture with 11-mm articular depression (arrow).$ and C, Coronal T2-weighted fat-suppressed (TR/TE, 4,200/40) () and T1-weighted (500/22) (C) MR images show bucket-handle. Part of torn meniscus body is flipped medially (arrow).

and C, Coronal T2-weighted fat-suppressed (TR/TE, 4,200/40) () and T1-weighted (500/22) (C) MR images show bucket-handle. Part of torn meniscus body is flipped medially (arrow).

$, Coronal MDCT image shows type 1 tibial plateau fracture (arrow) of lateral condyle.$ meniscal s can been classified as stable or unstable [18, 19]. functional study [10] showed that longitudinal, radial, and complex meniscal s are usually unstable but that horizontal s are stable.

meniscal s can been classified as stable or unstable [18, 19]. functional study [10] showed that longitudinal, radial, and complex meniscal s are usually unstable but that horizontal s are stable.

$However, our study differed markedly from that of Cothran et al. Our patient population had tibial plateau fractures with a notable average articular depression of more than 3 mm.$

6 MRI of Meniscal Injury Fig year-old man after traffic accident. Findings were verified with arthroscopy., Coronal MDCT image shows type 3 tibial plateau fracture with 11-mm articular depression (arrow). and C, Coronal T2-weighted fat-suppressed (TR/TE, 4,200/40) () and T1-weighted (500/22) (C) MR images show bucket-handle. Part of torn meniscus body is flipped medially (arrow). Rest of body is in normal position (arrowhead). Fig year-old woman after simple fall. Findings were verified with arthroscopy., Coronal MDCT image shows type 1 tibial plateau fracture (arrow) of lateral condyle., Sagittal proton-density MR image (TR/TE, 1,800/20) shows contusion (arrow) in posterior horn of lateral meniscus as area of increased signal intensity. sterisk indicates proximal fibula. meniscal s can been classified as stable or unstable [18, 19]. functional study [10] showed that longitudinal, radial, and complex meniscal s are usually unstable but that horizontal s are stable. Unstable s are associated with a considerable amount of pain. We found a large percentage of unstable s in both menisci: 25% of medial menisci and 18% of lateral menisci. The term meniscal contusion was adopted in 2001 by Cothran et al. [13]. In our study, we found a large number of meniscal contusions (22 menisci, 28%). However, our study differed markedly from that of Cothran et al. Our patient population had tibial plateau fractures with a notable average articular depression of more than 3 mm. These lesions can be considered high-energy fractures, whereas the knees evaluated by Cothran et al. had bone contusions only, no tibial plateau fractures. We can hypothesize that the trauma energy in our cases was greater, which may explain the large number of contusions. Shearing and compressive forces by the femoral condyles against the tibial plateau are involved in tibial plateau fractures [5], so the menisci may undergo concurrent compressive injury, eventually leading not to frank but to meniscal contusion. arrow et al. [11] found a number of menisci with increased intrameniscal signal intensity thought to represent intrameniscal injury. s was ours, that study was conducted with patients with tibial plateau fractures. The definitive cause of the abnormal intrameniscal signal intensity remains unknown. Results of animal experiments have suggested fibrovascular scarring, but to our knowledge, this finding has not been made in humans [13]. In the study by Cothran et al. [13], follow-up imaging showed the abnormal signal intensity had disappeared in two patients, no change was visible in one patient, and the abnormal signal intensity had decreased slightly in one patient. The authors view was that it seemed that a transient injury had occurred that did not have long-term effects on the meniscus. Vangsness et al. [4], in an arthroscopic study, found no correlation between fracture pattern and meniscal injury. We also found no significant difference between degree of TLE 3: Distribution of Tibial Plateau Fractures ccording to O Orthopaedic Trauma ssociation Class (n = 39) Meniscal Finding No. of Fractures C2 C3 Mean rticular Depression (mm) Normal Contusion Tear Total Note Dash ( ) indicates no fracture in class. ll fracture types except C1 were visible. O = rbeitsgemeinschaft für Osteosynthesefragen. C JR:191, October

7 Mustonen et al. TLE 4: natomic Distribution of bnormal MRI Findings mong 78 Menisci (39 Lateral, 39 Medial) Meniscal Part Contusion Longitudinal Tear Radial Tear Flap Tear Horizontal Tear Lateral Medial Lateral Medial Lateral Medial Lateral Medial Lateral Medial nterior horn ody Posterior horn Total Note Dash ( ) indicates finding was not made. Values are number of cases. Total number of s (n = 34) does not equal total number of menisci with s (n = 20) because of multiple s in some cases. articular depression and meniscal findings. lthough the small number of patients in our study might have influenced this finding, it can be hypothesized that even nondisplaced tibial plateau fractures can be associated with severe meniscal injuries and that clearly displaced tibial plateau fractures can have normal menisci. Meniscal s were most often seen in our patients with split-type fractures (1), indicating the harmful effect of axial load with shearing force combined, eventually leading to meniscal. In contrast, most contusions in our study occurred in patients with type 2 fractures (Table 1), in which the articular depression (axial compressive force) is the characteristic finding. Despite the high-energy fractures, somewhat surprisingly, neither MRI nor arthroscopy depicted meniscal root injuries. When designing the study, we had assumed that we would detect such injuries. large number of ligamentous injuries, most of which necessitated surgery, were evident in our study. This finding is also in agreement with previous findings [3, 5, 6, 11]. The lack of arthroscopic or surgical verification of meniscal s in 11 patients with normal MRI findings might have been a limitation of this study, but a fact well documented [7, 16, 17, 20, 21] is that MRI can depict meniscal s with excellent sensitivity, specificity, and accuracy. The results of our study also confirmed this finding. Reicher et al. [17] found that the presence of joint effusion, commonly associated with tibial plateau fractures, does not affect accurate MRI detection of meniscal lesions. Thus the high negative predictive value in our study strongly suggests that the 11 patients had true-negative findings, and those patients were included in the study. Clinical tests revealed no signs of meniscal. rthroscopy after normal findings at MRI and without symptoms would be ethically unacceptable. In nine menisci, contusion was an isolated finding, and at arthroscopy, the menisci were intact. We can thus hypothesize that if it is an isolated finding, contusion should be managed conservatively. However, lack of follow-up MRI represents a limitation of this hypothesis, and the natural behavior of this increased contusion signal intensity in the menisci over a long period remains unknown. Further studies with follow-up MRI are necessary. The two false-negative interpretations (a small radial and a small flap in the posterior horn of the medial meniscus at arthroscopy) were invisible even at repeated retrospective evaluation. In four false-positive findings (three posterior horns of the medial meniscus and one posterior horn of the lateral meniscus), the menisci also had abnormal signal intensity at repeated retrospective evaluation, and meniscal contusion may explain the findings. t arthroscopy, menisci were intact with no degenerative changes evident. Degenerative changes in the menisci ( or intrasubstance degeneration) are known to be strongly associated with osteoarthritis [22]. The results of our study, in which the patients had a mean age of 37 years and no signs of osteoarthritis at MDCT or MRI, strongly suggest that increased signal intensity in the menisci would be due to contusion. Thus in our study group, degenerative changes in the menisci instead of contusion would be unlikely. The arthroscopic findings supported this hypothesis. high percentage of patients (33%) with tibial plateau fractures had an unstable meniscal. Pain and poor outcome have been related to unstable s, usually necessitating surgical intervention. If a meniscal is Total detected preoperatively, meniscal surgery (arthroscopic or open) can be combined with fracture fixation, avoiding another operation. Moreover, because a large number of meniscal contusions were found, awareness of this abnormality may help radiologists increase specificity by avoiding the false-positive finding of meniscal. We therefore suggest that knee MRI be considered a complementary study after MDCT in the care of patients who have sustained high-energy tibial plateau fractures. References 1. Gardner MJ, Yacoubian S, Geller D, et al. The incidence of soft tissue injury in operative tibial plateau fractures: a magnetic resonance imaging analysis of 103 patients. J Orthop Trauma 2005; 19: Yacoubian SV, Nevins RT, Sallis JG, Potter HG, Lorich DG. Impact of MRI on treatment plan and fracture classification of tibial plateau fractures. J Orthop Trauma 2002; 16: rophy DP, O Malley M, Lui D, Denison, Eustace S. MR imaging of tibial plateau fractures. Clin Radiol 1996; 51: Vangsness CT Jr, Ghaderi, Hohl M, Moore TM. rthroscopy of meniscal injuries with tibial plateau fractures. J one Joint Surg r 1994; 76: Colletti P, Greenberg H, Terk MR. MR findings in patients with acute tibial plateau fractures. Comput Med Imaging Graph 1996; 20: Kode L, Lieberman JM, Motta O, Wilber JH, Vasen, Yagan R. Evaluation of tibial plateau fractures: efficacy of MR imaging compared with CT. JR 1994; 163: Fox MG. MR imaging of the meniscus: review, current trends, and clinical implications. Magn Reson Imaging Clin N m 2007; 15: Weigel DP, Marsh JL. High-energy fractures of the tibial plateau: knee function after longer follow-up. J one Joint Surg m 2002; 84: JR:191, October 2008

8 MRI of Meniscal Injury 9. Shepherd L, bdollahi K, Lee J, Vangsness CT Jr. 14. Mustonen O, Koivikko MP, Haapamaki VV, 1990; 72: The prevalence of soft tissue injuries in nonopera Kiuru MJ, Lamminen E, Koskinen SK. Multi 19. Newman P, Daniels U, urks RT. Principles tive tibial plateau fractures as determined by mag detector computed tomography in acute knee in and decision making in meniscal surgery. rthros- netic resonance imaging. J Orthop Trauma 2002; juries: assessment of cruciate ligaments with copy 1993; 9: : magnetic resonance imaging correlation. cta 20. Schafer FK, Schafer PJ, rossmann J, et al. Value 10. oxheimer L, Lutz M, Zanetti M, et al. Charac Radiol 2007; 48: of fat-suppressed proton-density-weighted turbo teristics of displaceable and nondisplaceable meniscal s at kinematic MR imaging of the knee. Radiology 2006; 238: arrow, Fajman W, Parker LM, lbert MJ, Drvaric DM, Hudson TM. Tibial plateau fractures: evaluation with MR imaging. RadioGraphics 1994; 14: Muller ME, Nazarian S, Koch P, Schatzker J. The comprehensive classification of fractures of long bones. New York: Springer, Cothran RL Jr, Major NM, Helms C, Higgins LD. MR imaging of meniscal contusion in the knee. JR 2001; 177: Mustonen O, Koskinen SK, Kiuru MJ. cute knee trauma: analysis of multidetector computed tomography findings and comparison with conventional radiography. cta Radiol 2005; 46: Crues JV 3rd, Mink J, Levy TL, Lotysch M, Stoller DW. Meniscal s of the knee: accuracy of MR imaging. Radiology 1987; 164: Reicher M, Hartzman S, Duckwiler GR, assett LW, nderson LJ, Gold RH. Meniscal injuries: detection using MR imaging. Radiology 1986; 159: Dandy DJ. The arthroscopic anatomy of symptomatic meniscal lesions. J one Joint Surg r spin-echo sequences in detecting meniscal lesions: comparison with arthroscopy. cta Radiol 2006; 47: Zanetti M, Pfirrmann CW, Schmid MR, Romero J, Seifert, Hodler J. Patients with suspected meniscal s: prevalence of abnormalities seen on MRI of 100 symptomatic and 100 contralateral asymptomatic knees. JR 2003; 181: Link TM, Steinbach LS, Ghosh S, et al. Osteoarthritis: MR imaging findings in different stages of disease and correlation with clinical findings. Radiology 2003; 226: JR:191, October

Meniscal Tears: Role of Axial MRI Alone and in Combination with Other Imaging Planes

Meniscal Tears: Role of Axial MRI Alone and in Combination with Other Imaging Planes Nefise Cagla Tarhan 1,2 Christine. Chung 1 urea Valeria Rosa Mohana-orges 1 Tudor Hughes 1 Donald Resnick 1 Received September 30, 2003; accepted after revision February 2, 2004. 1 Department of Radiology,