MRI is an accurate imaging modality to assess for complete tear of the ACL, with reported sensitivity of 83–95% and specificity of 95–100% [1]. Both primary and secondary signs of tear have been described.

Primary signs of complete ACL tear include ACL discontinuity, abnormally increased intrasubstance ACL signal intensity, enlarged mass-like morphology of the ACL, abnormal orientation of ACL fibers, and non-visualization of the ACL [1]. The most accurate primary signs of ACL tear are discontinuity of the ACL and abnormal orientation of ACL fibers, each with reported positive predictive values of 100% [2]. ACL discontinuity is reported to be visualized best in the sagittal and axial planes, but anecdotally we have found the coronal and axial planes to better demonstrate the continuity or discontinuity of the ACL bundles [3]. Additionally, straight sagittal sequences do not account for the normal oblique orientation of the ACL, which can result in misdiagnosis or lack of recognition of a clinically important tear (Fig. 7.5). While adding a degree of obliquity when prescribing the sagittal sequence has been promoted as a method to accommodate for the orientation of the ACL, it increases the overall scan acquisition time [4].

Discontinuity must be evaluated both within the substance of the ligament and at the origin from the femur. The “empty notch” sign refers to an ACL that has been completely torn from its femoral origin at the lateral intercondylar notch and is best seen on coronal or axial sequences (Fig. 7.5) [5].

Abnormal orientation of ACL fibers can be characterized as abnormal vertical orientation of proximal ACL fibers, abnormal horizontal orientation of distal ACL fibers, or abnormal bowing of the ACL (Fig. 7.6). An angle of greater than 15° between the roof of the intercondylar notch and the proximal ACL and an angle of less than 45° between the tibia and distal ACL are suggested to be highly accurate for the diagnosis of complete ACL tear [6]. In addition, torn distal ACL fibers can sometimes flip anteriorly, a finding that can present clinically with decreased knee extension following ACL tear [7].

Secondary signs tend to have a high specificity, but lower sensitivity for complete ACL tear [3]. They have been shown to have no added benefit in the determination of whether an ACL is completely torn or is intact when compared to primary signs alone [8]. Reported secondary signs of ACL tear include anterior translation of the tibia, uncovering of the lateral meniscus posterior horn, osseous injury, buckling of the posterior cruciate ligament (PCL), reduced PCL angle, posterior PCL line, and the posterior femoral line [1, 3, 6, 9–13].

Anterior translation of the knee is assessed on the midsagittal image of the lateral compartment of the knee. Abnormal anterior translation is present when the posterior aspect of the lateral tibial plateau is subluxated anteriorly 5 mm or more relative to the posterior aspect of the lateral femoral condyle. This finding is reported to have sensitivity of 58% and specificity of 93% for complete ACL tear. With 7 mm or more anterior tibial translation, the specificity increases to 100% [13].

Greater than 3.5 mm of uncovering of the tibial surface of the lateral meniscus posterior horn due to anterior translation of the tibial plateau has reported sensitivity of 44% and specificity of 94% for ACL tear [6].

Bone contusions and/or impaction fractures that result from a pivot shift mechanism ACL tear are typically seen at the anterior to central aspect of the lateral femoral condyle and the posterior aspect of the lateral tibial condyle [10–12]. Often there is additional contusion or fracture of the posterior aspect of the medial tibial condyle, thought to be due to a contrecoup force [9]. Less commonly, osseous injury can be seen along the anterior femoral and tibial condyles from a hyperextension mechanism of ACL injury.

Duration of bone contusions after injury has been studied predominantly in adult patients. In a study of 30 patients with acute knee injury (20 with ACL tears) and mean age 28 years old, all bone contusions were still present at 12–14 weeks [14]. An additional study in adult patients with older mean age of 43.5 years demonstrated a median healing time of 42 weeks [15]. However, in that study concomitant osteoarthritis was shown to nearly double healing time, suggesting that bone contusion healing time may be shorter in pediatric patients.

The presence of subchondral bone depression in association with bone contusion may have important clinical implications. When subchondral bone depression is present in the setting of pivot shift mechanism osseous injuries, there is increased association with meniscal injury in the same compartment and worsened functional outcome at 1 year (Fig. 7.7) [16]. Persistence of subchondral bone depression, as well as articular cartilage thinning, has been observed in patients 2 years after initial ACL injury, even following a successful ACL reconstruction [17].

Avulsion fractures can also occur in the setting of ACL injury and can often be identified on radiographs immediately following the injury. The Segond fracture is an avulsion fracture at the anterolateral proximal tibia that is highly associated with ACL injury [18]. The etiology of the Segond fracture has caused much consternation over the years, but more recent anatomic studies indicate that a distinct ligament, the anterolateral ligament (ALL), is the ligament associated with this avulsion fracture [19]. However, this ligament is unable to be reliably identified as a discrete structure on MRI, and it has been suggested that the term “lateral capsular ligament” be used as a generic term referring to the portion of the lateral knee stabilizers that includes the ALL [18]. An additional, less common avulsion fracture seen in association with cruciate ligament injury occurs at the insertion of the arcuate ligament complex on the fibular head. This fracture is seen radiographically as a linear lucency through the fibular head, a finding known as the “arcuate sign ,” and indicates a posterolateral corner injury (Fig. 7.8) [20].

Tibial eminence fractures can occur via the same mechanisms as ACL tears, but instead of the weak link being the ligament itself, the bone at the distal ACL insertion is avulsed. This type of injury overwhelmingly occurs in skeletally immature patients, typically 8 to 14 years old [21, 22]. Rarely, these fractures can be associated with concomitant injury of the ligament itself (Fig. 7.8). The following classification scheme based on the position of the avulsed fragment has been developed by Meyers and McKeever to assist management: type 1, non-displaced; type 2, posterior hinge; type 3, completely displaced; and type 4, comminuted [23, 24]. Identification of intermeniscal ligament or meniscal entrapment between the fragment and donor bone, typically in type 2 and type 3 fractures, is crucial, as it can preclude closed reduction and necessitate an open procedure [22].

Imaging can play an important role in the diagnosis and characterization of partial ACL tears, as physical examination findings can be quite variable, ranging from ligament stability to complete ACL insufficiency [25]. In addition, accurate characterization of partial tears has important treatment implications, as lower-grade injuries may be successfully treated conservatively, while higher-grade injuries are more likely to progress to ligament insufficiency and require surgical intervention. It is reported that a tear that involves less than 25% of the ACL thickness carries a 12% risk of progression to ACL insufficiency, whereas a tear that involves greater than 75% thickness carries a risk as high as 86% [25] .

In general, the reported sensitivity, specificity, and accuracy of MRI in the diagnosis of a partial ACL tear are lower than that for complete ACL tear. At 1.5 Tesla (T), MRI sensitivities range from 62 to 81% and specificities range from 19 to 97% [26, 27]. In a study of patients with partial ACL tear confirmed at arthroscopy who were imaged at either 1.5 T or 3 T, accuracy rates of just 25–53% for diagnosis of partial tear are reported [28].

Higher-field strength MRI can improve diagnosis of partial ACL tear, with reported sensitivities, specificities, and accuracies at 3 T of 77–87%, 87–97%, and 87–95%, respectively [29, 30]. Axial oblique plane MRI has also been promoted as a method to improve diagnosis of partial ACL tear, but the added benefit of axial oblique plane imaging over standard three-plane imaging has not been shown to be statistically significant [29]. Isotropic three-dimensional (3D) MRI sequences have shown no significant benefit over standard 2D sequences [31].

Primary signs of partial ACL tear include attenuation of the ACL, hyperintense intrasubstance signal with at least some intact fibers, posterior inferior bowing of the ACL, distortion of ACL morphology without obvious ACL discontinuity, and bundle discontinuity/isolated bundle (Fig. 7.3) [27, 28, 30, 32, 33]. Unfortunately, the ability of these findings to differentiate a normal from a partially torn ACL and a partially torn from a completely torn ACL is variable. Given this difficulty, some authors have advocated dividing ACL tears into stable (stable partial ACL tears) and unstable (unstable partial and complete ACL tears) categories rather than the traditional normal, partial tear, and complete tear categories, arguing that this has more practical value in determining management strategy [32, 33]. When discrimination of stable from unstable ACL injury on MRI is considered, studies with surgical correlation have shown that an attenuated ACL, abnormal intrasubstance signal, and elliptical morphology suggest a stable ACL injury, whereas distorted, cloudlike ACL morphology, an isolated intact single bundle, and non-visualization of the ACL suggest an unstable ACL injury, with sensitivities of 77–100% and specificities of 92–96% [32, 33].

Another way to quantify severity of a partial thickness ACL tear is to first determine if only one or both bundles are involved. If one bundle is intact, but one is completely torn (e.g., with signs of discontinuity or abnormal bowing), the tear is considered to involve 50% of the total ligament. If one bundle is completely torn, and the other bundle is attenuated and has increased intrasubstance signal intensity on fluid-sensitive sequences suggesting a partial tear, then this injury approximates to 75% of the total ligament. If there is only partial tear of a single bundle manifested by mild increased intrasubstance signal intensity, attenuation of the bundle, or fiber discontinuity in only part of a single bundle, then the tear involves less than or equal to 25% of the ligament (Fig. 7.3). Although these estimates are unlikely to be completely accurate, they can be a useful guide to therapy, as the severity of tear has been shown to impact clinical management and prognosis [25].

Secondary signs on MRI used to diagnose complete ACL tears also have been evaluated in the diagnosis of partial ACL tears. In one study there was improved sensitivity and specificity for the diagnosis of partial ACL tear from 62% sensitivity and 19% specificity with primary signs alone to 81% and 51%, respectively, when both primary and secondary signs are considered [26]. However, when evaluating whether an ACL injury is a stable partial tear or an unstable partial or complete tear, secondary signs can be seen in both types of tears and may actually decrease diagnostic ability of MRI relative to use of primary signs alone [28, 33]. The most specific secondary signs for unstable tear are anterior tibial translation, uncovering of the lateral meniscus posterior horn, and abnormal PCL buckling with specificities of 100%, but these signs had low sensitivity of 23%. Bone contusions are nonspecific and can be seen in both stable and unstable tears [33].

An in-depth consideration of all concomitant injuries that may be found in the setting ACL injury is beyond the scope of this chapter. However, we will review several important findings that accompany ACL injury, namely, posterolateral corner injury, medial collateral ligament (MCL) tears that may require acute surgical repair, and meniscal tears.

Posterolateral corner injury can have a negative impact on the longevity of an ACL graft if not initially recognized (Fig. 7.9). Specifically, untreated clinical grade 3 (most severe) posterolateral corner injuries are associated with higher forces on the graft and can contribute to ultimate graft failure [34]. On MRI, complete tears of two or more structures at the posterolateral corner have been proposed to correlate to clinical grade 3 injuries [35]. Specific structures at the posterolateral corner that are important to evaluate on MRI include the fibular collateral ligament, the popliteus musculotendinous unit including the popliteofibular ligament, and the posterolateral joint capsule. Thickening and intermediate signal intensity on T2-weighted fat-suppressed images within a ligament is indicative of partial thickness tear, while complete tear is diagnosed when discontinuity is present. Secondary signs of posterolateral corner injury include bone marrow edema pattern or fracture at the fibular styloid, bone contusions of the anterior medial femoral condyle, and lack of a substantial joint effusion presumed due to tear of the posterolateral capsule with subsequent leakage of fluid [35].

Most MCL tears heal without consequence, although may require acute repair or reconstruction. Specifically, osseous avulsion of the MCL from the medial epicondyle and tear of the distal MCL from the tibia, either with interposition beneath the medial meniscus or retraction superficial to the pes anserine tendons that results in a Stener-like lesion, can necessitate surgery (Fig. 7.9) [36, 37].

Although many different meniscal tear patterns can be seen with ACL injury, two in particular deserve mention, namely, peripheral tears of the posterior horn lateral meniscus and root tears or complete radial tears.

Tears at the posterior horn of the lateral meniscus are commonly missed in the context of ACL injury likely due, at least in part, to the known “pseudotear” that can occur at the meniscal origin of the meniscofemoral ligament [38, 39]. Tear should be suggested when peripheral abnormal signal intensity is present at the posterior horn of the meniscus on four or more 3-mm-thick slices located lateral to the posterior cruciate ligament (PCL) [39].

Meniscus root tears, which involve the attachment of the meniscus to the tibia, and complete radial tears result in disruption of the normal hoop stress function of the meniscus, and can lead to increased axial compressive forces transmitted to articular cartilage and to eventual osteoarthritis [40]. Root tears in particular are important to identify because the posterior joint space is not routinely evaluated at arthroscopy and injury in this area may be missed [40]. This type of tear is becoming increasingly recognized even in young patients, and lateral root tears in particular have been associated with ACL tear (Fig. 7.9) [33].

While the imaging findings of partial or complete ACL tear in the pediatric population are similar to those seen in adults, the normal expected postoperative appearance following reconstruction can differ depending on the degree of skeletal maturity and reconstruction technique used. In this section, we will introduce the various reconstruction techniques utilized in pediatric patients across the spectrum of skeletal maturity and discuss their normal postoperative imaging appearances. We also will highlight major differences between children and adults with respect to the normal postoperative appearance of various ACL reconstruction techniques and will review the MRI findings of postoperative complications.