Posterolateral Complex Knee Injuries: Magnetic Resonance Imaging with Surgical Correlation

Transcription

1 Acta Radiologica ISSN: (Print) (Online) Journal homepage: Posterolateral Complex Knee Injuries: Magnetic Resonance Imaging with Surgical Correlation D. J. Theodorou, S. J. Theodorou, D. C. Fithian, L. Paxton, D. H. Garelick & D. Resnick To cite this article: D. J. Theodorou, S. J. Theodorou, D. C. Fithian, L. Paxton, D. H. Garelick & D. Resnick (2005) Posterolateral Complex Knee Injuries: Magnetic Resonance Imaging with Surgical Correlation, Acta Radiologica, 46:3, To link to this article: Published online: 09 Jul Submit your article to this journal Article views: 129 View related articles Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at Download by: [ ] Date: 31 December 2017, At: 08:53

2 ORIGINAL ARTICLE ACTA RADIOLOGICA Posterolateral Complex Knee Injuries: Magnetic Resonance Imaging with Surgical Correlation D. J. THEODOROU, S.J.THEODOROU, D.C.FITHIAN, L.PAXTON, D.H.GARELICK &D.RESNICK Department of Radiology, School of Medicine, University of California San Diego, San Diego, Calif., USA; Department of Orthopedic Surgery, Southern California Permanente Medical Group, San Diego, Calif., USA; Midwest Orthopedics, Chicago, Ill., USA Theodorou DJ, Theodorou SJ, Fithian DC, Paxton L, Garelick DH, Resnick D. Posterolateral complex knee injuries: magnetic resonance imaging with surgical correlation. Acta Radiol 2005;46: Purpose: To describe the magnetic resonance imaging (MRI) findings of injuries of the posterolateral aspect of the knee and to evaluate the diagnostic capabilities of MRI in the assessment of these injuries. Material and Methods: The MRI studies of 14 patients (mean age 33 years) with trauma to the posterolateral aspect of the knee were retrospectively reviewed, and the imaging findings were correlated with those of surgery. Results: In all patients, MRI showed an intact iliotibial (ITB) band. MRI showed injury to the biceps tendon in 11 (79%), the gastrocnemius tendon in 1 (7%), the popliteus tendon in 5 (36%), and the lateral collateral ligament (LCL) in 14 (100%) patients. Tear of the anterior cruciate ligament (ACL) was seen in 11 (79%) patients and tear of the posterior cruciate ligament (PCL) in 4 (29%) patients. With routine MRI, visualization of the popliteofibular or fabellofibular ligaments was incomplete. On MRI, the lateral meniscus and the medial meniscus were torn with equal frequency (n54; 29%). Osteochondral defects were seen in 5 (36%) cases and joint effusion in all 14 (100%) cases on MRI. Using surgical findings as the standard for diagnosis, MRI proved 86% accurate in the detection of injury to the ITB band, the biceps tendon (93%),, the gastrocnemius tendon (100%), the popliteus tendon (86%), the LCL (100%), the ACL (79%), the PCL (86%), the lateral meniscus (90%), the medial meniscus (82%), and the osteochondral structures (79%). Surgical correlation confirmed the MRI findings of joint effusion in all cases. Conclusion: MRI is well suited for demonstrating the presence and extent of injuries of the major structures of the posterolateral complex of the knee, allowing characterization of the severity of injury. Key words: Correlative study; knee, anatomy; knee, injuries; knee, ligaments; knee, MR; knee, posterolateral corner; knee, surgery; magnetic resonance imaging Daphne J. Theodorou, M.D., 13 Papadopoulos Street, Ioannina, Greece (fax , . daphne_theodorou@hotmail.com, rjtheodorou@hotmail.com) Accepted for publication 19 November 2004 Injuries to the posterolateral aspect of the knee are complex injuries that may result in posterolateral rotatory instability of the knee, characterized clinically by posterior subluxation and external rotation of the lateral tibial plateau relative to the femur (1, 2, 5, 19 21). Posterolateral instability of the knee is uncommon, and occurs more often in conjunction with injuries to one or both cruciate ligaments (1, 2, 13, 14, 20, 26, 33, 38, 40). For example, in a study of 17 patients with injury to the lateral structures of the knee (37), the anterior cruciate ligament was affected in 56% of cases, and the posterior cruciate ligament in 17% of cases. BAKER and co-workers (1) noted that the anterior cruciate ligament was torn in 65% of patients with posterolateral injury. These investigators (2) also reported a series of 13 consecutive patients with acute combined posterior cruciate ligament injury and posterolateral instability. In another study of 18 patients with posterolateral injury (23), the cruciate ligaments were injured in 89% of the cases. The authors found that the anterior cruciate ligament was injured in 22% of cases, the posterior cruciate ligament in 17% of cases, and both ligaments in 50% DOI / # 2005 Taylor & Francis

3 MRI Surgical Correlation of Posterolateral Knee Injuries 298 of cases. Failure to recognize and treat injuries of the posterolateral structures, however, can result in anterior cruciate ligament (ACL) failure as well as instability after posterior cruciate ligament (PCL) repair (5, 7, 13, 31). Early diagnosis of injuries to the posterolateral aspect of the knee is critical because surgical repair in the acute period is easier, and is associated with a more favorable outcome for patients (1, 2, 3, 7, 11, 13, 25). Initial evaluation of patients sustaining an injury of the posterolateral complex of the knee is usually made clinically and includes several physical examination maneuvers (6, 9, 14, 19, 21, 22, 25, 41). Because the patient is often in significant pain, findings on clinical evaluation may be inconclusive. On radiographs, an avulsion fracture from the fibular head a finding referred to as the arcuate sign indicates injury to the arcuate complex, which comprises the posterolateral corner of the knee (18, 23, 36). Magnetic resonance imaging (MRI) is useful in the assessment of abnormalities of the knee joint. Our objective was to describe the MRI findings of posterolateral corner injuries in a relatively large series of patients, and to correlate these findings with the results of surgery, which are the gold standard for diagnosis in our study. Material and Methods Fourteen patients with posterolateral complex injuries of the knee were studied. The patients (11 M and 3 F, age range years, mean age 33 years) in most cases had sustained either a posterolaterally directed blow to the anteromedial proximal portion of the tibia or a non-contact hyperextension and varus injury (Table 1). On admission, patients with lateral or posterolateral tenderness underwent plain radiographs of the injured knee, and varus stress radiographs at 30 of knee flexion (20). Varus laxity was confirmed if stress radiographs showed 3 mm or more of asymmetry compared to the contralateral normal knee on varus stress radiographs. In that case, MRI was performed and examination under anesthesia with surgical exploration was scheduled. The interval between injury and surgery ranged from 7 to 200 days (mean 29 days). Eleven (78%) patients also had undergone initial arthroscopy to determine intraarticular abnormalities. In all cases, MR findings were correlated with surgical and arthroscopic findings. MRI examinations were performed on 1.5T MR units (Magnetom Vision; Siemens Medical Systems, Iselin, N.J., USA) (Signa; G.E. Medical Systems, Milwaukee, Wisc., USA) in 8 patients, and on 1.0T MR units (Magnetom Vision; Siemens Medical Systems) in 6 patients. Imaging sequences included T1-weighted (TR /TE 10 20) spinecho images and intermediate-weighted (TR /TE 18 20) and T2-weighted (TR / TE 80 90) spin-echo images in the coronal, axial, and sagittal imaging planes. Fat-suppressed intermediate-weighted (TR 4010/TE 18) fast spin-echo MR images were available in two patients. The field of view varied between 14 cm and 16 cm, and slice thickness varied between 3 and 4 mm, with a 0.5 mm intersection gap. Analysis of the MR images was accomplished by two musculoskeletal radiologists with consensus. The ligamentous structures lateral collateral (LCL), ACL, PCL, and the tendinous structures (biceps femoris, gastrocnemius and popliteus tendons, and the iliotibial (ITB) band were evaluated for injury by analysis of morphology, signal intensity characteristics, integrity of each individual structure, and adjacent bone changes. A full-thickness tear was diagnosed when disruption of the ligament or tendon was identified. A partial-thickness tear was diagnosed when the ligament or tendon appeared thickened with increased intrasubstance signal intensity on T2-weighted MR images. Avulsive injuries were designated by identification of the avulsed structure with or without adjacent cortical disruption or bone marrow changes (or both). Associated injuries of the menisci, osteochondral bone, and joint effusion were also recorded. Surgical correlation was available for all patients. Data Analysis The SPSS statistical software (SPSS Inc., Chicago, Ill., USA) was used to analyze the agreement between MRI and surgical findings. We calculated sensitivity as probability of a positive MRI result, given a positive surgical finding. Specificity was calculated as the probability of a negative MRI result, given a negative surgical finding. The Fisher exact tests were used to evaluate the relationships between injuries of different structures noted at surgery. The statistical significance level was set at Anatomy SEEBACHER (35) simplified the complex anatomy of the posterolateral aspect of the knee, organizing structures into three layers. The first layer, the most superficial, includes the iliotibial band anteriorly, and the superficial component of the biceps

4 299 D. J. Theodorou et al. Table 1. Mechanism of Injury and Physical Examination Findings in 14 Patients with Posterolateral Complex Injuries of the Knee Case Cause of Injury Age (yrs) Days from Injury to Surgery Physical Examination Findings 1 Fall from a height Lachman, + Posterior sag, Asymmetric varus opening 2 Hyperextension 57 9 Asymmetric varus opening 3 Motorcross injury Dial, Asymmetric varus opening, + Lachman 4 Anteromedial blow Asymmetric varus opening 5 MVA Knee dislocation Lachman, + Posterior sag, Asymmetric varus opening 6 Hyperextension varus Dial, Asymmetric varus opening, + Hyperextension/recurvation, + Lachman 7 Hyperextension varus Pivot shift, Asymmetric varus opening 8 Anteromedial blow Asymmetric varus opening, + Lachman 9 Anteromedial blow Asymmetric varus opening 10 Pedestrian vs. MVA, Asymmetric varus opening, Open tibial fx + Dial, + Lachman 11 Hyperextension Asymmetric varus opening, + Pivot shift, + Dial, + Hyperextension/recurvation, + Reverse pivot shift 12 Anteromedial blow 15 8 Asymmetric varus opening 13 N.A Asymmetric varus opening at 0 0 and 30 0, + Lachman, + Dial 14 Anteromedial blow Lachman, + Pivot shift, Asymmetric varus opening Abbreviations: N.A., Non-available; MVA, motor-vehicle accident; fx, fracture. posterolaterally. Layer two, the middle layer, is represented by the quadriceps retinaculum anteriorly. Posteriorly, there are two ligamentous thickenings which originate from the lateral patella. The most proximal structure joins the terminal fibers of the lateral intermuscular septum, and the distal ligament terminates at the femoral insertion of the posterolateral capsule and the lateral head of the gastrocnemius tendon. Layer three, the deepest, forms the lateral part of the capsule, which contains several important thickenings that function as discrete structures. One such thickening, the fibular collateral ligament (also called the lateral collateral ligament), originates from the lateral femoral condyle and inserts on the fibular head. A second important thickening in the posterior joint capsule is the arcuate ligament, which originates from the styloid process of the fibular head, interdigitates with the popliteus, and inserts into the posterior capsule near the oblique popliteal ligament (Fig. 1). Results The clinical examination findings in our patients are given in Table 1, and the agreement between the MRI and surgical results in Table 2. With MRI, the ITB band appeared intact in all 14 cases. At surgical exploration, the ITB band was intact in 12 (86%) patients; in one knee there was a complete rupture and a second knee showed partial disruption of the ITB band. Although MRI confirmed the 12 intact ITB bands, it failed to detect ITB band injuries in two cases. In 11 (79%) of the 14 patients, MRI revealed avulsive injuries of the biceps tendon, which included avulsion of the tendon from the fibular head (n57), avulsion of the tendon from the femoral attachment (n52), and avulsion fracture of the styloid process of the fibula (n52) (Figs. 2A and 3). Surgical findings confirmed MRI findings in 10 of these patients. MRI showed an avulsive injury of the biceps tendon from its femoral attachment, which was not confirmed at surgery. In one (7%) patient, MRI demonstrated rupture of the lateral head of the gastrocnemius tendon at its origin on the lateral femoral condyle, which was confirmed at surgical exploration. On MR images, abnormalities in the popliteus tendon were noted in five (36%) patients (Fig. 2C). Surgical findings confirmed the MRI findings in four of the five patients. In one patient, MRI demonstrated an intra-substance tear of the popliteus tendon, which was not confirmed at surgery. In another patient, surgical findings revealed partial detachment of the popliteus tendon from its femoral origin, which was not depicted on MR images. Operative findings showed that injuries occurred proximally at the origin of the popliteus in four

5 MRI Surgical Correlation of Posterolateral Knee Injuries 300 A Fig. 1. A, Schematic drawing of the posterior aspect of the knee. FE5femur, FI5fibula, TI5tibia, f5fabella, 15fabellofibular ligament, 25fibular collateral ligament, 35oblique popliteal ligament, 45popliteus tendon, 55lateral head of gastrocnemius muscle, 65medial head of gastrocnemius muscle, 75semimembanosus tendon, 85arcuate ligament, 95medial collateral ligament. B, Schematic drawing of the lateral aspect of the knee. FE5femur, FI5fibula, TI5tibia, PA5patella, f5fabella, ITT5iliotibial band, P5popliteus muscle, LG5lateral head of gastrocnemius muscle, 15popliteus tendon, 25fibular collateral ligament. Asterisk indicates the arcuate complex. Table 2. MRI Findings Compared with Surgical Findings in 14 Patients with Lateral Ligament Injuries N True (+) Sensitivity True (2) Specificity Agreement (%) Accuracy Iliotibial band 14 0% 100% 86% Biceps tendon % 75% 93% Popliteus tendon 14 80% 89% 86% Lateral collateral ligament % NA 100% Lateral meniscus 10 75% 100% 90% Medial meniscus % 78% 82% Anterior cruciate ligament % 50% 79% Posterior cruciate ligament % 83% 86% Gastrocnemius tendon % 100% 100% Osteochondral defect 14 75% 80% 79% Popliteofibular ligament 14 0% NA 0% B NA5Variable is constant. patients, and in one patient the zone of injury occurred at the insertion of the popliteus on the posterior surface of the tibia, superior to the soleal line. The MR sequences in the coronal, axial, and sagittal imaging planes used in this study did not include any oblique images that would allow optimal visualization of the popliteofibular, fabellofibular, or arcuate ligaments, and thus accurate assessment of injuries to these structures (29, 43). At surgery, the popliteofibular ligament was found to be injured in three (21%) of the patients. In all cases, MRI displayed abnormalities of the LCL, which included avulsion of the ligament from the fibular head (n510; 71%), detachment from the lateral femoral condyle (n53; 21%), and intrasubstance tear (n51; 7%) (Figs. 2B and 4). Surgical exploration confirmed these findings in all 14 (100%) cases. MRI demonstrated partial tear, or complete rupture of the ACL in 11 (79%) patients. At surgery, the ACL was torn in eight of these patients. Indeed, MRI showed partial tears of the ACL in three patients with no evidence of abnormality on surgical exploration of the joint. MRI

6 301 D. J. Theodorou et al. A B C Fig. 2 (Patient 4). A, Coronal T2-weighted (TR 1900/TE 80) spin-echo MR image of the knee shows avulsion of the biceps femoris tendon off the fibular head (short arrow). Edema infiltrating soft tissues adjacent to the lateral femoral condyle is evident (long arrow). B, In this same patient, consecutive T2-weighted (TR 1900/TE 80) spin-echo MR image of the knee 4mm more anteriorly, shows avulsion of the fibular collateral ligament of the fibular head (short arrow). Again, soft tissue edema adjacent to the lateral femoral condyle is seen (long arrow). C, Coronal T2-weighted (TR 1900/TE 80) spin echo MR image of the knee in the same patient as before, demonstrates rupture of the musculotendinous junction of the popliteus with retraction of the proximal end of the popliteus tendon (arrow) and considerable muscle edema (thick arrow). depicted 4 (29%) injuries to the PCL including 2 partial tears and 2 complete ruptures. Although surgical findings confirmed the MRI findings of complete PCL ruptures (50%), partial tears were not identified at surgery (50%). Associated injuries of the lateral meniscus were depicted in 4 (29%) cases. Surgical findings corresponded closely with the MRI findings in three of the four cases. All negative lateral meniscus findings identified by surgical exploration (n56) were also noted as negative in MRI results. MRI demonstrated tears of the medial meniscus in four (29%) cases. Surgical correlation confirmed the MRI findings in 2 (50%) of these cases. MR images showed 5 (36%) osteochondral defects, which were surgically confirmed in 3 (60%) of these patients. MRI failed to identify an osteochondral defect proved at surgery. In 8 (57%) patients, MRI showed variable patterns of bone contusion in the femoral and tibial condyles, which were not appreciated at surgery. Joint effusion was seen in all 14 cases on MR images. Five (36%) of the patients demonstrated the signal intensity characteristics of hemarthrosis, with (n51) or without (n54) a fluid level. The signal intensity of hemorrhagic effusion was similar to that of synovial fluid, or hemarthrosis had high signal intensity on both T1-weighted and T2-weighted spin-echo MR images. Surgery confirmed the MR findings in each patient. Three (21%) patients had intra-articular bodies on MRI, which were detected at surgical exploration of the joint. In addition, we examined the relationships between the different types of injuries noted at surgery. At the time of surgery, 5 (36%) of 14 knees demonstrated isolated injuries to the posterolateral complex without concomitant cruciate ligament injury. In the remaining 9 (64%) patients, 7 injuries

7 MRI Surgical Correlation of Posterolateral Knee Injuries 302 A B Fig. 3 (Patient 10). A, Coronal fat-suppressed intermediate-weighted (TR 4010/TE 18) fast spin-echo MR image of the knee shows avulsion injury of the biceps femoris tendon from its fibular attachment (long arrow). The musculotendinous junction demonstrates the changes of edema secondary to the injury (short arrow). Marrow edema is apparent in the fibular head (asterisk). Note extensive soft tissue edema (thick arrow). B, Coronal T1-weighted (TR 766/TE 14) spin-echo MR image of the knee shows avulsion of the biceps femoris tendon from its fibular attachment (arrow). A Fig. 4 (Patient 5). A, Coronal intermediate-weighted (TR 2000/TE 20) spin-echo MR image shows complete rupture of the lateral collateral ligament close to its femoral attachment (arrow). B, Corresponding T2-weighted (TR 2000/TE 80) spin-echo MR image again shows rupture of the lateral collateral ligament (long arrow) and associated edema of adjacent soft tissues (short arrow). to the posterolateral complex were found in association with an injury to the ACL, one injury was associated with a PCL tear, and another injury was seen with a combined injury to both the ACL and PCL. In the two (14%) cases with an ITB band injury, the biceps tendon was injured; and, every popliteal tendon injury was associated with an injury to the ACL in our patients. Tears of the ITB band occurred only in the presence of biceps tendon tears in this study. The Fisher exact test did not indicate a significant relationship between biceps tendon and ITB band injuries, likely owing to the low incidence of ITB band injuries (Pw0.05), but it did indicate a strong relationship between popliteus injuries and ACL tears (Pv0.01) in this series. No significant relationship was noted B between PCL injuries and LCL or biceps tendon injuries (Pw0.05). Discussion Injuries of the posterolateral aspect of the knee are severe injuries that may result in the worst functional instability of any knee injury (1, 2, 5, 19, 20, 25). Failure to diagnose and manage these injuries can lead to development of abnormal gait patterns with symptomatic knee hyperextension thrusting (7, 30). Not infrequently, patients with injuries to the posterolateral aspect of the knee present to the orthopedic surgeon in the chronic stage, which is associated with suboptimal reconstruction (1, 2, 7, 11, 13, 20, 25). Several physical

8 303 D. J. Theodorou et al. examination maneuvers and tests (6, 8, 9, 21, 22, 25, 41) are used for evaluating posterolateral complex injury. However, these tests may be non-diagnostic. For example, fibrous scar tissue can cause falsenegative posterolateral drawer test results; and, hamstring spasm and tightness, meniscal tear, and an intact ACL may prevent positive external rotation recurvatum test results (6, 21). A positive varus stress test at 30 of knee flexion is suggestive of instability, but is not diagnostic (1). Instrumentation for measuring posterolateral rotatory instability has also been developed (3). Stress radiography is helpful in documenting varus and valgus rotational knee instability (17). With the exception of asymmetric varus opening of the knee noted in all our patients, no other finding was consistently identified on physical examination maneuvers. MRI allows direct visualization of the important structures of the posterolateral aspect of the knee, as well as assessment of the integrity of these structures (16, 18, 23 25, 29, 33, 34, 39, 43). Therefore, MRI may aid in the preoperative evaluation of patients to determine the presence of injury to the posterolateral structures, even when diagnosis is missed on physical examination. MILLER and co-workers (28) reported 30 cases of posterolateral ligament injury determined on MR images, where injury was clinically suspected in only 3 of these 30 cases. TWADDLE and co-workers (39) reported that MRI correctly identified 82% of the fibular collateral ligament injuries in 17 patients with knee dislocation. In their study, only 71% of the injuries were suspected on clinical examination. In addition, MRI affords characterization of the extent of injury to the posterolateral corner of the knee. In our study, MRI was helpful in the determination of location and extent of injury to the posterolateral structures, serving as a road-map for surgical treatment. Patients with posterolateral injuries commonly have associated cruciate ligament injuries (5, 13, 23, 33). Combined ligamentous injuries of the knee may compromise function and stability to a greater extent than an isolated injury (1, 5, 15). In those patients with posterolateral ligament injury combined with cruciate ligament injury, failure to reconstruct the cruciate ligament with the posterolateral corner will most likely lead to failure of the posterolateral reconstruction (5). In our patients, MRI allowed us to evaluate associated injuries in the knee and improve characterization of the complexity of injury. In each individual case, identification of the injured structures prompted repair or reconstruction as appropriate. Our results indicate that MRI proved superior to radiography in affording anatomic detail, lesion identification, localization, and characterization. In our patients, MRI was 100% accurate in determining injuries to the LCL and gastrocnemius tendon, and proved equally less accurate (86%) in the assessment of the ITB band and the popliteus tendon. Using surgical findings as the standard for reference, MRI overestimated biceps tendon, cruciate ligament, meniscal, and osteochondral injuries. LAPRADE and co-workers (24) in a series of normal knees and knees with injuries to the posterolateral structures reported that MRI was 95% accurate in diagnosing injuries to the LCL and the ITB band, and 90% accurate in diagnosing injury to the popliteus tendon. HUANG and co-workers (18) reported a series of 13 patients with posterolateral injury where MRI was 100% accurate in diagnosing injuries to the LCL, PCL, the popliteus tendon, and the menisci. In a series of 15 patients with complex knee injuries, TWADDLE and co-workers (39) reported 100% accuracy of MRI in the diagnosis of cruciate ligament injury. In that series, MRI was 82% accurate in the diagnosis of injury to the LCL. We believe overestimation of partial tears of the ACL in our patients was secondary to a partialvolume effect of the ligament with the lateral femoral condyle, producing a pseudotear. Likewise, overestimation of partial tears of the PCL was related to the magic angle phenomenon owing to oblique ligament orientation (12). Because of joint effusion and knee laxity in our patients with ligamentous injuries, we overcalled meniscal tears on the MR images (4, 10). In our study, the magic angle phenomenon was another potential cause of linear intermediate signal intensity in the meniscus (32). Finally, abnormal deepening of the condylopatellar sulcus of the lateral femoral condyle was incorrectly thought to represent an osteochondral fracture. With the prescribed standard imaging planes used in our study, visualization of the popliteofibular ligament and other fine, obliquely oriented structures of the posterolateral aspect of the knee, including the fabellofibular and arcuate ligaments, was incomplete. There is considerable debate about the prevalence in the population of the popliteofibular (93 100%), fabellofibular (24 80%), and the arcuate (24 100%) ligaments, however (27, 35, 42). HUANG and co-workers (18) also reported inadequate depiction of the fabellofibular popliteofibular and arcuate ligaments in their series. LAPRADE and co-workers (24) reported that MRI was 68% accurate in the diagnosis of injuries to the

9 MRI Surgical Correlation of Posterolateral Knee Injuries 304 popliteofibular ligament, and 86% accurate in the diagnosis of injuries to the fabellofibular ligament. Using the oblique coronal plane, YU and coworkers (43) improved visualization of the small posterolateral structures. In a cadaveric study using MRI in standard and coronal oblique planes, MUNSHI and co-workers (29) identified the popliteofibular ligament and arcuate ligament in 57% and 71% of specimens, respectively. The fabellofibular ligament was not identified on MR images in any of the specimens. MRI provided us with additional information regarding the pathomechanics of injuries to the lateral aspect of the knee. In our series of 14 patients, isolated injuries to the posterolateral complex occurred in 36% of cases. Of the remaining 64% of cases, ACL tear was the most common associated injury, representing 89% of cases with associated injuries. Injury to the popliteus tendon and muscle, which internally rotates the tibia on the femur as the knee is flexed from full extension, was only seen in combination with an ACL injury in our series. However, ACL deficiency was not always associated with injury to the popliteus. In our patients, we observed a weak relationship between hyperextension/varus injury mechanisms and popliteus tendon injuries. Based on our results, it appears that identification of popliteal injury should alert the physician to carefully evaluate the knee for concomitant ACL deficiency. Similarly, our observations indicate that diagnosis of an injury to the ITB band would be suggestive of injury to the biceps femoris tendon due to high correlation of these injuries in this study. We recognize some limitations in our study. Inclusion of only patients who underwent knee surgery may have caused a selection bias; this was necessitated by availability. Injuries to the posterolateral aspect of the knee are relatively rare. Because of the small series of patients, we are not allowed to determine the incidence of this injury in the general population. In addition, chronicity of injury varied in our patients who sustained acute, subacute, and chronic injuries. Finally, we did not employ special imaging planes (24, 29, 43) to visualize any obliquely oriented and delicate structures of the posterolateral complex. In our series, however, routine MRI of the knee was found to be an important and helpful diagnostic tool for evaluating the presence and extent of injuries of the posterolateral complex. In conclusion, MRI provides useful information about the anatomy and abnormalities of the posterolateral structures. The oblique course of fine, lateral capsular structures, however, may require specialized imaging planes. Our study results show that MRI allows identification and localization of injury of the posterolateral structures, as well as characterization of the severity of injury, which is important in surgical planning. We suggest the use of MRI as part of the routine preoperative work-up in all patients with either suspected injury to the posterolateral ligamentous structures or varus knee laxity after trauma. References 1. Baker C, Norwood L, Hughston J. Acute posterolateral rotatory instability of the knee. J Bone Joint Surg 1983;65A: Baker C, Norwood L, Hughston J. Acute combined posterior cruciate and posterolateral instability of the knee. Am J Sports Med 1984;12: Bleday RM, Fanelli GC, Giannotti BF, Edson CJ, Barrett TA. Instrumented measurement of the posterolateral corner. Arthroscopy 1998;14: Chew FS. Medial meniscal flounce: demonstration on MRI of the knee. Am J Roentgenol 1990;155: Clancy W, Shepard M, Cain Lyle E. Posterior lateral corner reconstruction. Am J Orthop 2003;32: Cooper DE. Tests for posterolateral instability of the knee in normal subjects. Results of examination under anesthesia. J Bone Joint Surg 1991;73: Covey DC. Injuries of the posterolateral corner of the knee. J Bone Joint Surg (Am) 2001;83A: Daniel DM, Malcom LL, Losse G, Stone ML, Sachs R, Burks R. Instrumented measurement of anterior laxity of the knee. J Bone Joint Surg (Am) 1985;67: Daniel DM, Stone ML, Barnett P, Sachs R. Use of the quadriceps active test to diagnose posterior cruciate ligament disruption and measure posterior laxity of the knee. J Bone Joint Surg (Am) 1988;70: Davis SJ, Teresi LM, Bradley WG, Burke J. The notch sign: meniscal contour deformities as indicators of tear in MRI of the knee. J Comput Assist Tomogr 1990;14: Delee JC, Riley MB, Rockwood CA, Jr. Acute posterolateral rotatory instability of the knee. Am J Sports Med 1983;11: Erickson SJ, Cox IH, Hyde JS, Carrera G, Strandt J, Estkowski L. Effect of tendon orientation on MRI signal intensity: a manifestation of the magic angle phenomenon. Radiology 1991;181: Fleming R, Blatz D, McCarroll J. Posterior problems in the knee: posterior cruciate insufficiency and posterolateral rotatory insufficiency. Am J Sports Med 1981;9: Fowler P. The classification and early diagnosis of knee joint instability. Clin Orthop 1980;147: Gollehon D, Torzilli P, Warren R. The role of the posterolateral and cruciate ligaments in the stability of the human knee: a biomechanical study. J Bone Joint Surg (Am) 1987;69: Haims AH, Medvecky MJ, Pavlovich R Jr, Katz LD. MRI of the anatomy of and injuries to the lateral and

10 305 D. J. Theodorou et al. posterolateral aspects of the knee. Am J Roentgenol 2003;180: Harilainen A, Myllynen P, Rauste J, Silvennoinen E. Diagnosis of acute knee ligament injuries: the value of stress radiography compared with clinical examination, stability under anaesthesia and arthroscopic or operative findings. Ann Chir Gynaecol 1986;75: Huang GS, Yu J, Munshi M, Chan W, Lee CH, Chen CY, Resnick D. Avulsion fracture of the head of the fibula (the arcuate sign): MRI findings predictive of injuries to the posterolateral ligaments and posterior cruciate ligament. Am J Roentgenol 2003;180: Hughston JC, Andrews JR, Cross MJ, Moschi A. Classification of knee ligament instabilities. Part II. The lateral compartment. J Bone Joint Surg (Am) 1976;58: Hughston J, Jacobson K. Chronic posterolateral rotatory instability of the knee. J Bone Joint Surg (Am) 1985;67A: Hughston JC, Norwood LA. The posterolateral drawer test and external rotational recurvatum test for posterolateral rotatory instability of the knee. Clin Orthop 1980;147: Jakob RP, Hassler H, Staeubli HU. Observations on rotatory instability of the lateral compartment of the knee. Experimental studies on the functional anatomy and the pathomechanism of the true and the reversed pivot shift sign. Acta Orthop Scand 1981;191Suppl: Juhng SK, Lee JK, Choi SS, Yoon KH, Roh BS, Won JJ. MR evaluation of the arcuate sign of posterolateral knee instability. Am J Roentgenol 2002;178: Laprade RF, Gilbert TJ, Bollom TS, Wentorf F, Chaljub G. The magnetic resonance imaging appearance of individual structures of the posterolateral knee. A prospective study of normal knees and knees with surgically verified grade III injuries. Am J Sports Med 2000;28: Laprade RF, Wentorf F. Diagnosis and treatment of posterolateral knee injuries. Clin Orthop 2002;402: Larson R. Combined instabilities of the knee. Clin Orthop 1980;147: Maynard MJ, Deng X, Wickiewicz TL, Warren RF. The popliteofibular ligament. Rediscovery of a key element in posterolateral stability. Am J Sports Med 1996;24: Miller T, Gladden P, Staron R, Henry J, Feldman F. Posterolateral stabilizers of the knee: anatomy and injuries assessed with MRI. Am J Roentgenol 1997;169: Munshi M, Pretterklieber ML, Kwak S, Antonio GE, Trudell D, Resnick D. MRI, MR arthrography, and specimen correlation of the posterolateral corner of the knee: an anatomic study. Am J Roentgenol 2003; 180: Noyes FR, Dunworth LA, Andriacchi TP, Andrews M, Hewett TE. Knee hyperextension gait abnormalities in unstable knees. Recognition and preoperative gait retraining. Am J Sports Med 1996;24: O Brien S, Warren R, Pavlov H, Panariello R, Wickiewicz T. Reconstruction of the chronically insufficient anterior cruciate ligament with the central third of the patellar tendon. J Bone Joint Surg (Am) 1991;73A: Peterfy C, Janzen D, Tirman P, Van Dijke C, Pollack M, Genant H. Magic-angle phenomenon: a cause of increased signal in the normal lateral meniscus on short-te MR images of the knee. Am J Roentgenol 1994;163: Recondo JA, Salvador E, Villanua JA, Barrera MC, Gervas C, Alustiza JM. Lateral stabilizing structures of the knee: functional anatomy and injuries assessed with MRI. RadioGraphics 2000;20:S Ross G, Chapman A, Newberg A, Scheller A Jr. Magnetic resonance imaging for the evaluation of acute posterolateral complex injuries of the knee. Am J Sports Med 1997;25: Seebacher JR, Inglis AE, Marshall JL, Warren RF. The structure of the posterolateral aspect of the knee. J Bone Joint Surg (Am) 1982;64: Shindell R, Walsh W, Connolly J. Avulsion fracture of the fibula: the arcuate sign of posterolateral knee instability. Nebra Med J 1984;69: Towne L, Blazina M, Marmor L, Lawrence J. Lateral compartment syndrome of the knee. Clin Orthop 1971;76: Trickey E. Injuries to the posterior cruciate ligament: diagnosis and treatment of early injuries and reconstruction of late instability. Clin Orthop 1980;147: Twaddle B, Hunter J, Chapman J, Simonian P, Escobedo E. MRI in acute knee dislocation. J Bone Joint Surg (Am) 1996;78B: Veltri DM, Warren RF. Posterolateral instability of the knee. J Bone Joint Surg (Am) 1994;76A: Veltri DM, Warren RF. Anatomy, biomechanics, and physical findings in posterolateral knee instability. Clin Sports Med 1994;13: Watanabe Y, Moriya H, Takahashi K, Yamagata M, Sonoda M, Shimada Y, et al. Functional anatomy of the posterolateral structures of the knee. Arthroscopy 1993;9: Yu JS, Salonen DC, Hodler J, Haghighi P, Trudell D, Resnick D. Posterolateral aspect of the knee: improved MRI with a coronal oblique technique. Radiology 1996;198:199.