Mechanism-based Pattern. Injuries of the Knee Depicted at MR Imaging 1

Transcription

1 SCIENTIFIC EXHIBIT Mechanism-based Pattern Approach to Classification of Complex Injuries of the Knee Depicted at MR Imaging 1 Curtis W. Hayes, MD Monica K. Brigido, MD David A. Jamadar, MB Tim Propeck, MD S121 Complex knee injuries are common, often resulting from multiple forces: varus, valgus, hyperextension, hyperflexion, internal rotation, external rotation, anterior or posterior translation, and axial load. Certain combinations of forces are known to cause specific injury patterns. After a review of the literature, the authors developed a mechanismbased classification system based on patterns of bone marrow edema and ligament injury for complex knee injuries depicted at magnetic resonance imaging. The classification system takes into account knee position and forces and recognition of patterns of bone injury and complementary soft-tissue injury. Ten mechanism-based injury patterns were recognized: (a) pure hyperextension, (b) hyperextension with varus, (c) hyperextension with valgus, (d) pure valgus, (e) pure varus, (f) flexion with valgus and external rotation, (g) flexion with varus and internal rotation, (h) flexion with posterior tibial translation, (i) patellar dislocation (flexion, valgus, and internal rotation of femur on fixed tibia), and (j) direct trauma. Recognition of these patterns may help assess the full extent of knee injury, particularly at the posterolateral and posteromedial corners of the knee. Abbreviations: ACL = anterior cruciate ligament, LCL = lateral collateral ligament, MCL = medial collateral ligament, PCL = posterior cruciate ligament, SE = spin-echo Index terms: Knee, fractures, , , Knee, injuries, Knee, ligaments, menisci, and cartilage, Knee, MR, , RadioGraphics 2000; 20:S121 S134 1 From the Department of Radiology, University of Michigan Health System, Taubman Center, Rm 2910A, 1500 E Medical Center Dr, Ann Arbor, MI Recipient of a Certificate of Merit award for a scientific exhibit at the 1999 RSNA scientific assembly. Received February 10, 2000; revision requested March 16 and received April 25; accepted May 8. Address correspondence to C.W.H. ( hayescw@umich.edu). RSNA, 2000

2 S122 October 2000 RG Volume 20 Special Issue Introduction Complex injuries of the knee are common, resulting from accidents or sports mishaps at all levels of competition. These injuries often occur as a result of multiple forces applied to the knee: varus, valgus, hyperextension, hyperflexion, internal rotation, external rotation, anterior or posterior displacement, and axial load. It is widely known that certain forces predictably produce specific individual or combined patterns of injury. Acquisition of a precise history of the mechanism may be difficult, however, as is performance of an accurate physical examination in the setting of acute injury. Magnetic resonance (MR) imaging is widely used to assess knee injuries more completely. Radiologists are accurate at detecting individual injuries and combinations of injuries. However, little attention has been paid to the use of MR imaging to classify knee injuries into mechanismbased categories. A comprehensive classification system based on mechanisms of injury would be useful because (a) an understanding of the causative mechanism in a given case may improve detection of the complete constellation of injuries and (b) appreciation of the mechanism of injury may help predict both immediate and delayed instability and need for surgery. In this article, we present a mechanism-based classification system for complex knee injuries. With use of MR imaging patterns of ligamentous, meniscal, and capsular injuries in addition to specific patterns of bone marrow edema, or bone bruising, to infer the causative forces, we found that many complex knee injuries can be categorized on the basis of their specific mechanisms of injury. First in this article, the pertinent functional anatomy of the knee is briefly reviewed. Then, forces acting on the joint that are typically responsible for knee injuries are listed. The differences between impaction and avulsion bone marrow edema patterns are demonstrated to help categorize the direction of forces responsible for the injury. The classification system is then presented, followed by examples of each pattern. Functional Anatomy The bones of the knee contribute little to the stability of the joint. Both the static and dynamic stability of the knee are dependent on its supporting soft tissues (1). Menisci, ligaments, tendons, muscles, and fascia all make contributions to knee stability. Dynamically, the supporting structures can be divided by location: anterior, medial, lateral, posterior, and central. The major contributors in each area are demonstrated in Figure 1. Two areas in the knee are critical for stability: the posteromedial and posterolateral corners. The posteromedial corner consists of the attachments of the semimembranosus tendon, the posterior joint capsule, and the posterior oblique ligament. The posterolateral corner is anatomically complex, consisting of the joint capsule, arcuate ligament, fabellofibular ligament, and popliteus muscle and tendon, with support from adjacent structures. Both the posteromedial and posterolateral corners are major resistors of rotational and translational stresses, particularly in extension. The corners may be thought of as anchors of a sling across the back of the knee; traumatic loss of one corner may allow unstable rotation of the knee joint, with a pivoting out around the other corner. Injury at the posterolateral corner, in particular, may lead to severe disability. Therefore, recognition of injuries to these structures is critical in any MR imaging evaluation of the injured knee. Injury-producing Forces and Resisting Structures The major forces acting on the knee joint include translation (anterior and posterior), angulation (varus and valgus), rotation (internal and external), hyperextension, axial load, and direct blow. Most knee injuries are the result of two or more forces exerted across either a flexed or extended joint. However, it is still useful to look at pure forces in terms of the structures that are responsible for resistance. Table 1 lists major forces and their primary and secondary resistors.

3 RG Volume 20 Special Issue Hayes et al S123 Figure 1. Diagram of functional anatomy of the knee joint, grouped by anterior, medial, posteromedial, posterior, posterolateral, lateral, and central supporting structures. Table 1 Forces Responsible for Knee Injuries and Supporting Structures Responsible for Primary and Secondary Resistance to Those Forces Primary Secondary Force Resistance Resistance Anterior translation (displaces tibia anteriorly) ACL MCL, LCL Posterior translation (displaces tibia posteriorly) PCL None Varus (medial to lateral) LCL ACL, PCL Valgus (lateral to medial) MCL ACL, PCL Internal rotation (femur fixed) LCL ACL, capsule External rotation (femur fixed) MCL PCL, capsule Hyperextension PCL ACL, posterior capsule Impaction versus Avulsion Injury Patterns In general, forces acting on the knee produce an impaction injury at the entry site of the force and a distraction injury, or avulsion, at the opposite, or exit, site of the force. Figure 2 demonstrates a valgus force that produces impaction of the lateral compartment and distraction of the opposite medial compartment. Both types of injuries produce bone marrow edema or frank fractures.

4 S124 October 2000 RG Volume 20 Special Issue However, bone injuries caused by impaction tend to be broad, compared with smaller, more focal areas of bone marrow edema associated with ligament avulsions on the distraction side. Impactions may be further divided into contiguous, kissing, injuries or noncontiguous injuries, which are produced by the abrupt translation of two bones that occurs after ligamentous rupture. The direction and type of force may be inferred from the patterns of bone marrow edema and softtissue injury. Direct blows result in bone contusions and soft-tissue injury at the impaction site only. Mechanism-based Classification System of Knee Injuries On the basis of information from the injury patterns to primary and secondary stabilizing structures, plus bone injury patterns, we have classified complex knee injuries into 10 categories, according to the knee position (flexion, extension), direction of force, and presence or absence of rotation (Figure 3): (a) pure hyperextension; (b) hyperextension with varus; (c) hyperextension with valgus; (d) pure valgus; (e) pure varus; (f) flexion with valgus, external rotation; (g) flexion with varus, internal rotation; (h) flexion with posterior tibial translation; (i) patellar dislocation (flexion, valgus, internal rotation of femur on tibia); and (j) direct trauma. Hyperextension injuries, by virtue of the greater forces exerted on the extended or locked knee, produce more pronounced bone injury patterns, often with frank fractures. Severe distraction injuries on the posterior, exit side of the joint are common with this pattern. These injuries are particularly serious in that they involve the critical posteromedial and/or posterolateral corners of the knee. Flexion injuries, unless combined with significant axial load, tend to show few contiguous impaction bone injuries but have a greater tendency to produce injury due to internal or external rotation. Noncontiguous impaction bone bruises are usually found, as well as smaller avulsion bone bruises. We have observed that flexion injuries with rotation show a greater association with meniscal tears than do hyperextension injuries. Pure valgus and the rare pure varus categories are characterized by a simple coup-contrecoup pattern of impaction bone injuries and oppositesided distraction ligament injuries. Dislocation of the patella is typically produced by combined flexion, valgus, and internal rota- Figure 2. Illustration of impaction bone bruise pattern (dark gray shading in lateral compartment) and avulsion bone bruise pattern (black shading at margins of medial compartment). This pattern is characteristic of a valgus force (large arrow). tion of the femur on a fixed tibia. In theory, this is very similar to flexion, valgus, and external rotation (tibia on fixed femur), although we have chosen to separate these two patterns on the basis of their different injury patterns. Direct trauma is characterized by broad bone contusions located beneath the site of impaction. Typically, there are no injuries on the opposite side of the knee. Direct anterior trauma, which causes patellar and trochlear groove contusions, and lateral or medial blows are most common. Clinical Application We retrospectively reviewed the MR images in 100 cases of acute knee trauma to determine the ability of our system to help accurate classification of complex knee injuries. In this small sample, 85% of cases could be classified into one of our 10 categories. The pattern of flexion, valgus, and external rotation was most common, accounting for nearly half of all injuries (46%). Next in frequency were the patterns of hyperextension with varus (8%) and flexion with posterior tibial translation (8%), followed by the patterns of pure valgus (6%), patellar dislocation (6%), and direct trauma (5%). As expected, the pattern of pure varus (1%) was rare. Surprisingly, the pattern of flexion, varus, and internal rotation (Segond fracture) was also rare (1%) in this small series. The patterns of pure hyperextension (2%) and hyperextension with valgus (2%) were also

5 RG Volume 20 Special Issue Hayes et al S125 Figure 3. Diagram illustrates the direction of injury-producing forces acting on the knee. a. b. c. Figure 4. Pure hyperextension injury. (a) Axial fast spin-echo (SE), proton-density weighted, fat-suppressed MR image (repetition time msec/echo time msec = 4,000/15, echo train length of eight) shows impaction type anterior tibial bone bruise (*) and popliteus muscle injury with edema (arrows). (b, c) Sagittal SE proton-density weighted MR images (1,000/12) show a posterior cruciate ligament (PCL) tear (arrow in b) and posterior capsule disruption (arrow in c). uncommon. Common reasons for classification system failure included the presence of insufficient injury, massive injury, or bone marrow edema due to preexisting osteoarthritis. Examples of the injury categories are presented in Figures 4 13, with details for each pattern in Table 2.

6 S126 October 2000 RG Volume 20 Special Issue a. b. c. d. e. f. Figure 5. Hyperextension with varus injury. (a, b) Axial fast SE, proton-density weighted, fat-suppressed MR images (4,000/15, echo train length of eight) show anteromedial femoral condyle impaction bone bruise (* in a) and tibial plateau fracture (* in b) with edema over the posterolateral corner (arrows). (c, d) Sagittal fast SE, protondensity weighted MR images (1,000/12) show impaction fracture at the anterior femoral condyle (arrow in c) with popliteus tendon tear (arrow in d). (e) Sagittal fast SE, proton-density weighted, fat-suppressed MR image (3,233/15, echo train length of eight) shows anteromedial bone bruise and posterior capsule tear (arrow). (f) Coronal fast SE, proton-density weighted, fat-suppressed MR image (4,000/17, echo train length of eight) shows a lateral collateral ligament (LCL) tear (arrowhead) with soft-tissue edema laterally.

7 RG Volume 20 Special Issue Hayes et al S127 a. b. c. d. e. f. Figure 6. Hyperextension with valgus injury. (a) Axial fast SE, proton-density weighted, fat-suppressed MR image (4,000/15, echo train length of eight) shows impaction type anterolateral tibial plateau bone bruise (*) with edema over the posteromedial corner (arrowhead). (b e) Serial sagittal fast SE, proton-density weighted, fat-suppressed MR images (4,000/15, echo train length of eight) show anterolateral bone bruise (* in b), capsule disruption (arrow in c), PCL tear (arrow in d), and small avulsion type posteromedial tibial plateau bone bruise (* in e) with meniscotibial ligament tear (arrow in e). (f) Coronal fast SE, proton-density weighted, fat-suppressed MR image (4,000/15, echo train length of eight) shows lateral bone bruises with grade II medial collateral ligament (MCL) injury (arrowhead).

8 S128 October 2000 RG Volume 20 Special Issue 7a. 7b. 8a. 8b. Figures 7, 8. (7) Pure valgus injury. (a) Coronal fast SE, proton-density weighted, fat-suppressed MR image (4,000/15, echo train length of eight) shows impaction type lateral tibial plateau bone bruise (*) and distraction type grade II MCL injury (arrow). (b) Sagittal fast SE, proton-density weighted, fat-suppressed MR image (4,000/15, echo train length of eight) in this case shows the anterior cruciate ligament (ACL) and PCL to be intact, indicating an injury of mild to moderate severity. (8) Pure varus injury. Coronal fast SE proton-density weighted (2,750/30) (a) and T2- weighted (2,750/90) (b) MR images show distraction type iliotibial band injury with irregularity and edema within the adjacent soft tissue (arrow).

9 RG Volume 20 Special Issue Hayes et al S129 a. b. c. d. e. Figure 9. Flexion with valgus and external rotation injury. (a, b) Sagittal fast SE, protondensity weighted, fat-suppressed MR images (4,000/15, echo train length of eight) show the typical combination of noncontiguous posterolateral tibia and lateral femoral condyle bone bruises (* in a) that result from ACL tear (arrow in b) and anterior translation of the tibia. (c) Coronal fast SE, proton-density weighted, fat-suppressed MR image (4,000/15, echo train length of eight) shows tibia bone bruise (*) and grade III MCL tear (arrow). (d, e) In an additional case, fast SE, proton-density weighted, fat-suppressed MR images (4,000/15 17, echo train length of eight) show broad impaction pattern bone bruise in the lateral condyle (large *) and smaller, less specific bruises in the posterolateral and posteromedial tibial plateau and far medial aspect of the medial femoral condyle (small *). Some authors attribute the medial bruises to lesser impactions (contrecoup), whereas others believe they represent avulsion injuries at the MCL and meniscofemoral and meniscotibial ligament attachments. Note the incomplete MCL injury (arrow in d).

10 S130 October 2000 RG Volume 20 Special Issue a. b. Figure 10. Flexion with varus and internal rotation injury. (a, b) Sagittal fast SE, proton-density weighted, fat-suppressed MR images (4,000/15, echo train length of eight) show typical posterolateral tibial plateau and noncontiguous lateral femoral condyle bone bruises (* in a) with ACL tear (solid arrow in b) and posterior capsule disruption (open arrow in b). (c) Coronal fast SE, proton-density weighted, fat-suppressed MR image (4,000/15, echo train length of eight) shows small cortical avulsion fracture at the lateral tibial rim (Segond fracture, arrow) with a small amount of focal subcortical edema (*). c.

11 RG Volume 20 Special Issue Hayes et al S131 11a. 11b. 11c. Figures 11, 12. (11) Hyperflexion injury with posterior tibial translation. (a, b) Fast SE, proton-density weighted, fat-suppressed MR images (4,000/15, echo train length of eight) show an isolated PCL tear (arrow in a) with no bone bruise or capsule disruption produced by moderate force. In b, the rule indicates centimeters. (c) Sagittal T2-weighted fast SE (3,270/90) image shows transient posterior dislocation with PCL avulsion (arrow) and additional injuries produced by severe force. (12) Patellar dislocation (flexion and internal rotation of the femur on a fixed tibia). Axial fast SE, proton-density weighted, fat-suppressed MR image (4,000/15, echo train length of eight) shows typical medial patella and lateral femoral condyle bone bruises (*) 12. a. b. Figure 13. Direct trauma. Axial (a) and coronal (b) fast SE, proton-density weighted, fat-suppressed MR images (4,000/15, echo train length of eight) show localized bone contusion (*) in the medial femoral condyle, which is away from the articular surface. Note the absence of other injuries.

12 S132 October 2000 RG Volume 20 Special Issue Table 2 Injury Categories Fre- Bone Bruise Ligament Mechanism quency or Fracture Injuries Comments Pure hyperextension 2% Anterior central tibia, anterior femoral PCL, posterior capsule, ACL Associated with anterior translation of tibia condyles (impactions) (ACL tear) or posterior translation of tibia (PCL tear) Hyperextension with varus 8% Anteromedial tibia, femoral condyle Posterolateral corner, ACL, popliteal Unstable posterolateral corner injury (impactions); posterolateral corner, tendon, posterior capsule proximal fibula (avulsions) Hyperextension with valgus 2% Anterolateral tibia, femoral condyle MCL, posteromedial corner, posterior Contiguous ( kissing ) bone marrow edema (impactions); posteromedial tibia capsule, PCL pattern aids in distinguishing lateral im- (avulsion) paction from typical noncontiguous ACL injury pattern (flexion, valgus, and external rotation) Pure valgus 6% Lateral tibia, lateral femoral condyle MCL, ACL, PCL (depending on Pure pattern is uncommon (impactions) severity of force) Pure varus 1% Medial tibia, femoral condyle (impactions) Iliotibial band, LCL Rarely seen pattern, as varus is usually associated with flexed position and internal rotation Flexion valgus, 46% Lateral femoral condyle, posterolateral MCL, ACL Medial and lateral menisci at risk external rotation tibia (impactions); posteromedial tibia, femoral condyle (avulsions vs contrecoup) Flexion varus, internal 1% Lateral femoral condyle, posterolateral ACL, posterolateral corner Lateral and medial menisci at risk rotation tibia (ACL tear related impactions), posterolateral tibia (Segond avulsion), fibular head (avulsion) Flexion with posterior tibial 8% None, unless severe force or associated Isolated PCL, posterior dislocation Most common mechanism for isolated PCL translation with axial load with severe force tear Patellar dislocation (flexion 6% Medial patella, lateral condyle Medial patellar retinaculum, MCL, Search for chondral defect, often associated and internal rotation of (impactions) ACL (with sufficient force) with predisposing conditions (eg, patella femur on fixed tibia) alta) Direct trauma 5% Directly beneath site of trauma None May have superficial soft-tissue injury adjacent to bone contusion

13 RG Volume 20 Special Issue Hayes et al S133 Discussion Distinctive MR imaging patterns in certain knee injuries have been recognized by many investigators (2 7). The association between acute ACL tears and bone bruises involving the lateral femoral condyle and posterior lateral tibial plateau is well known. A less frequent association between ACL tears and medial-sided bone bruises is also described, attributed by some authors to avulsion at the semimembranosus attachment (8,9) and by others as a contrecoup impaction due to a secondary rebound rotation (10). Distinctly differing bone marrow edema patterns also accompany the flexion, varus, and internal rotation mechanism, which produces the Segond fracture (11,12). While also associated with ACL tear, this latter pattern can be differentiated from the more common injury mechanism of flexion, valgus, and external rotation that causes the O Donoghue triad. Other previously described distinctive MR imaging patterns include patellar dislocation and complete knee dislocation (13). We attempted to consolidate known MR imaging patterns of complex knee injury into a unifying mechanism-based classification system. Owing to the complexity of many knee injuries, this classification system is necessarily incomplete. For example, injuries to the extensor mechanism, including patellar dislocations, are significantly influenced by anatomic variation, which is not addressed in our classification system. Complex injuries characterized by predominantly axial loading forces have also been excluded. These injuries produce such pronounced bone injuries that individual patterns of bone marrow edema become obscured. We believe our classification system is particularly useful in the distinction between injuries that occur in extension versus flexion. Hyperextension injuries are characterized by broad areas of contiguous bone bruising at the anterior aspect of the knee. Forces required to produce these injuries are substantial, and the degree of soft-tissue injury at the opposite, posterior aspect of the knee can be appreciated on the basis of the extensive edema depicted at MR imaging. These hyperextension injuries involve the critical posteromedial and posterolateral corners of the knee, and a thorough description of the extent of soft-tissue injury is essential. In comparison, injuries that occur in flexion are characterized by relatively less extensive bone bruising. With the knee in flexion, shearing and rotational forces dominate over impaction forces. Many bone bruises in flexion injuries involve noncontiguous bone surfaces because they are due to secondary rotational impactions after ligamentous rupture. We believe that secondary rotation in flexion injuries may cause a trap-and-twist mechanism, which explains our observation of more frequent meniscal tears in this group of injuries. The ability to distinguish avulsion from impaction bone marrow edema patterns, as described by Palmer et al (14), is fundamental to our classification system. With increased use of highly sensitive fast SE sequences with fat suppression, the frequency of injury-associated bone marrow edema will be fully appreciated. For example, small areas of bone marrow edema at the margins of the tibia and femur at the sites of meniscal and capsular attachments are now commonly seen. We believe these small bone marrow edema foci usually indicate adjacent avulsion injury secondary to distraction or rotation. In conclusion, this mechanism-based classification system for complex knee injuries is based on MR imaging patterns of bone and soft-tissue injuries. The position of the knee, direction of force, and presence or absence of rotation combine to produce consistent injury patterns in many but not all cases. A thorough understanding of knee injury patterns and their mechanisms may help achieve more accurate assessment of these complicated injuries.

14 S134 October 2000 RG Volume 20 Special Issue References 1. Hayes CW, Conway WF. Normal anatomy and magnetic resonance appearance of the knee. Top Magn Reson Imaging 1993; 5: Gentili A, Seeger LL, Yao L, Do HM. Anterior cruciate ligament tear: indirect signs at MR imaging. Radiology 1994; 193: Kaplan PA, Walker CW, Kilcoyne RF, et al. Occult fracture patterns of the knee associated with anterior cruciate ligament tears: assessment with MR imaging. Radiology 1992; 183: Lynch TCP, Crues JV, Morgan FW, et al. Bone abnormalities of the knee: prevalence and significance at MR imaging. Radiology 1989; 171: Murphy BJ, Smith RL, Uribe JW, et al. Bone signal abnormalities in the posterolateral tibia and lateral femoral condyle in complete tears of the anterior cruciate ligament: a specific sign? Radiology 1992; 182: Robertson PL, Schweitzer ME, Bartolozzi AR, Ugoni A. Anterior cruciate ligament tears: evaluation of multiple signs with MR imaging. Radiology 1994; 193: Yao L, Lee JK. Occult intraosseous fracture: detection with MR imaging. Radiology 1988; 167: Chan KK, Resnick D, Goodwin D, Seeger LL. Posteromedial tibial plateau injury including avulsion fracture of the semimembranosus tendon insertion site: ancillary sign of anterior cruciate ligament tear at MR imaging. Radiology 1999; 211: Yao L, Lee JK. Avulsion of the posteromedial tibial plateau by the semimembranosus tendon: diagnosis with MR imaging. Radiology 1989; 172: Kaplan PA, Gehl RH, Dussault RG, Anderson MW, Diduch DR. Bone contusions of the posterior lip of the medial tibial plateau (contrecoup injury) and associated internal derangements of the knee at MR imaging. Radiology 1999: 211: Dietz GW, Wilcox DM, Montgomery JB. Segond tibial condyle fracture: lateral capsular ligament avulsion. Radiology 1986; 159: Weber WN, Neumann CH, Barakos JA, et al. Lateral tibial rim (Segond) fractures: MR imaging characteristics. Radiology 1991; 180: Yu JS, Goodwin D, Salonen D, et al. Complete dislocation of the knee: spectrum of associated soft-tissue injuries depicted by MR imaging. AJR Am J Roentgenol 1995; 164: Palmer WE, Levine SM, Dupuy DE. Knee and shoulder fractures: association of fracture detection and marrow edema on MR images with mechanism of injury. Radiology 1997; 204: