Rupture of the anterior cruciate ligament (ACL) most commonly occurs as a result of a pivot-shift injury, requires substantial force to occur and is rarely a truly isolated event. At least half of ACL tears in young patients are accompanied by other injuries such as meniscal or ligament tears and bony contusions [1–3]. In non-contact injuries, the ACL is at greatest risk of injury when excessive valgus and knee axial rotation loads are combined near full knee extension [4–6], such as when the patient lands on the ground with their centre of gravity moving away from the landing limb. In addition to tearing the ACL, this type of injury places tensile load on the medial collateral ligament (MCL) and posteromedial corner structures, and leads to impaction and shear injuries in the lateral compartment.
The patterns of injury visible at magnetic resonance imaging (MRI) and their relation to specific injury mechanism and clinical outcomes in ACL injury have been well studied in children [3, 7] and adults [4, 6, 8, 9]. Early imaging reviews in the 1990s focused on the detection of ACL tears  or showed few images . The actual ACL injury is generally easily detected [12, 13], particularly with improvements in MRI sequences, but some of the associated injuries remain subtle. Since these injuries may require treatment themselves, may alter the treatment plan when detected, or serve as markers of a poorer prognosis , it is important to identify them accurately. MR images in published studies are generally classic examples of obvious pathology [3, 4, 14], or focus only on the ACL itself and adult patterns . Few studies have focused on ACL tears in paediatric patients [3, 7, 10, 11], a group in which this injury is being seen with increasing frequency . In day-to-day practice, the distinction between normal and pathologic appearance of each structure at the knee can be difficult to make, particularly in adolescents. Therefore, the purpose of this pictorial review is to aid radiologists by demonstrating the spectrum of MR appearances at common sites of associated injury in children and teenagers with ACL tears, with emphasis on patterns specific to this age group.
Young children have weaker physes than ligaments, which can result in epiphyseal avulsion fractures at the anterior tibial spine near the ACL attachment rather than tearing of the ligament itself, for the same pivot-shift injury mechanism (Fig. 1) . This review is limited to patients with ACL tears, but the search for associated injuries in a younger child with a tibial spine avulsion should follow the same pattern. The spectrum of ACL appearance from normal to completely torn is seen in Figs. 2 and 3. The normal ACL is composed of two bundles named for tibial attachment locations, anteromedial (dominant in knee flexion) and posterolateral (most active in extension) . Partial tears can involve one or both bundles. In children the normal ACL can appear quite attenuated (Fig. 2b); in our experience a recently torn ACL is usually visibly thickened and oedematous, and a thin and appropriately oriented ligament without oedema is likely intact. In the last 110 cases at our institution, 7/48 (15 %) read by musculoskeletal radiologists and 26/62 (42 %) read by paediatric radiologists were initially reported as high-grade partial thickness ACL tears, while at later surgery all but one of these tears (99 %) was described as complete (Fig. 2c). Given the time delay between injury and surgery, it is possible that partial tears evolved to complete tears at surgery; however, the distinction between complete and partial tears is well known to be limited in reliability on MRI . One challenge is that the torn ACL may form scar tissue that can mimic a partially intact ligament. A tear that appears to involve most of the ACL fibres is likely to be functionally complete, and we encourage reporting the tear in this fashion.
This region is the site of frequent, generally minor injury, either involving the meniscocapsular junction (Fig. 4b, c) or the peripheral “red” portion of the meniscus (Fig. 4d–i). The capsule posterior to the medial meniscus is usually only slightly higher in signal than adjacent tissue (Fig. 4a). In the posteromedial knee there is a confluence of structures (posterior oblique ligament, oblique popliteal ligament, and fibres of the semimembranosus tendon) that reinforce the posteromedial joint capsule. With valgus stress and knee rotation, capsular tension is transmitted to the meniscocapsular junction which places tensile stress on the periphery of the meniscus which can be subsequently torn . The orientation (peripheral, vertical, and longitudinal) of this medial meniscal tear is typical after sports-related trauma. These tears account for most of the “false-positive” MRI diagnoses of meniscal tears, based on their absence (or interval healing) at later arthroscopy . A likely explanation is that since the periphery of the meniscus is well vascularised, these tears heal well in the interval between MRI and surgery.
More serious injury to the medial meniscus is significantly more common later post ACL injury [1, 21, 22]. One possible explanation is that displaced flap tears or bucket handle tears of the medial meniscus may develop due to ongoing joint instability rather than occurring directly at time of initial injury. Bucket handle tears are recognised by a combination of findings (Fig. 5).
The posterolateral corner is an often overlooked site of soft tissue injury in patients with ACL tears. Injuries most commonly involve the popliteomeniscal fascicles (Fig. 6a, b), and the adjacent lateral meniscus (Fig. 6c) . The popliteomeniscal fascicles are meniscocapsular extensions that blend with the popliteus and are considered functionally important stabilisers of the lateral meniscus, working in conjunction with the popliteus to prevent excessive lateral meniscal movement . In adults, posterolateral corner injury implies a higher risk of ACL graft failure, likely due to the associated increased rotational instability . This is likely also true in children but is not as well documented. Unfortunately, particularly in children these tiny structures are poorly seen, especially with motion artefact or a relatively large MRI field of view. The combination of regional oedema and non-visualisation of these tiny structures suggests, but does not confirm, a posterolateral corner injury. An abnormal superior popliteomeniscal fascicle serves as a useful clue for the detection of subtle lateral meniscal abnormalities and has been shown to be highly associated with a tear of the lateral meniscus, with reported positive predictive value 79 % and specificity 96 % .
The meniscofemoral ligament of Wrisberg and/or Humphrey, and rarely the oblique menisco-meniscal ligament, merge with the posterior lateral meniscus, complicating the assessment for tears. This has been well studied in adults [23, 27]. Practically speaking, if the ligament-to-meniscus attachment appears irregular or appears to extend to a slice where the fibula is also visible, a tear is likely present (Fig. 7).
This portion of meniscus is difficult to assess at MRI, because the meniscus deviates abruptly anteriorly to reach its bony attachment and loses its usual triangular shape [28, 29]. In practice, the attachment is more anterior than one might expect, and if no low-signal structure can be seen directly attaching to bone, a tear is likely present (Fig. 8). A meniscal detachment in this location is important to detect because it renders the entire lateral meniscus unstable and substantially increases tibial loading, which could predispose to early osteoarthritis .
Valgus positioning in most ACL injuries results in tensile stress on the medial collateral ligament, which can lead to an MCL tear (Fig. 9). For MR diagnosis, in our experience the axial sequence is more sensitive and specific than coronal sequences, where apparent ligament thickening is often due to volume averaging. The coronal plane is useful to determine location and extent of tear once diagnosed in the axial plane. The MCL was sprained (low or high grade) in 22 % of patients in a paediatric series . In the last 110 cases of ACL tears at our institution, there were seven high grade MCL tears, four proximal at femoral attachment, one at midsubstance and two distal avulsions at tibial attachment. Although our numbers are small, this distribution differs from an adult series in which combined ACL and MCL ruptures more often completely tore the MCL tibial attachment (53 %) than the femoral attachment (26 %) . The variation might be spurious, or due to differences in relative tension of ACL, MCL and joint capsule compared with adults. The MCL is a complex structure with two main layers visible at MRI, the superficial tibiofemoral layer and deep meniscofemoral and meniscotibial fibres firmly adherent to the medial meniscus . Isolated tears of deep MCL fibres are diagnosed when the main superficial fibres are intact but no low-signal struts are visible joining the MCL to the medial meniscus. These tears can occur with ACL injury (Fig. 9b) and can also extend into the attached medial meniscus. Most MCL tears, whether involving deep and/or superficial fibres, are treated non-operatively, but it is important to recognise the location of injury within the MCL because proximal tears may adhere to the capsule and limit motion, requiring aggressive physiotherapy, while distal avulsions may require surgical intervention since the pes anserinus tendons can become entrapped between ligament and bone .
There is a normal shallow (less than 1.5 mm) cortical undulation in the lateral femoral condyle known as the lateral condylopatellar sulcus . In the setting of ACL tear, abnormal subchondral signal (oedema acutely, and eventually sclerosis) at or near this sulcus represents contusion, while a focal angular depression together with abnormal signal indicates a fracture, at site of impaction on the lateral tibial plateau (Fig. 10). In our experience, on careful examination a focal abnormality of contour and/or signal is present in this location in most children with ACL tears, comparable to the 83 % rate reported in adults . The overlying cartilage is necessarily injured to some degree, even if this is not obvious at arthroscopy or routine MRI sequences. One recent study in adults used T2 relaxation mapping to demonstrate a cartilage injury in every imaged patient with a torn ACL . We expect this to also hold true in children, with unknown long-term implications.
There is often either contusion or an undisplaced fracture in subchondral bone at the lateral tibial plateau; less frequently, also at the medial tibial plateau. In teenagers this can be difficult to distinguish from a partially closed physis, especially since the site of the last portion of the physis to close is very near the most common fracture site (Fig. 11). A fracture is likely if the physis is separately visible and/or there is associated oedema. Like the femoral condyle impaction fracture, this fracture is not treated clinically, but suggests a high force impact mechanism.
The Segond fracture is a subtle avulsion fracture from the lateral tibial plateau at the combined attachment site of the iliotibial tract and the anterior oblique band. The anterior oblique band is a capsular thickening that joins the fibular collateral ligament to the rim of the lateral tibial plateau where it merges with posterior fibres of the iliotibial tract . Nearly all patients with Segond fractures have ACL tears  or tibial spine avulsion which is mechanically equivalent in younger children (Fig. 1). The fracture is usually produced by internal rotation and varus stress at the knee , which differs from the usual axial rotation/valgus mechanism of ACL tears, likely accounting for its relative rarity. It is subtle on both MRI and radiographs.
A discoid lateral meniscus is particularly prone to tear in association with an ACL tear (Fig. 12). As for the medial meniscus, it is our experience that bucket handle tears of the lateral meniscus are unusual in the early post injury period and more common months or years post untreated ACL tear. Lateral collateral ligament sprains are not expected with valgus ACL injury mechanism. Posterior cruciate ligament sprains in this setting generally indicate higher impact forces than associated with sports injury, such as motor vehicle collisions. Patellar dislocation or vascular injury are also rarely associated with ACL tears. As noted at the lateral femoral condyle, cartilage-specific sequences in adults show cartilage injury to be essentially always present in patients with ACL tears, even if often not obvious on routinely obtained MRI sequences . We are not aware of a study confirming this in children.
This review demonstrates the spectrum of associated findings expected in paediatric ACL injury. Paediatric-specific issues for the radiologist to be aware of include: tibial spine avulsion instead of ACL rupture; a thin but intact ACL; normal poor visualisation of posterolateral corner structures; differentiation between incompletely closed physis and impaction fracture. The posterior lateral meniscus is particularly complex and should be studied in three regions (posterolateral corner, meniscofemoral ligament attachment and bony attachment). If an ACL tear is detected, the associated findings described here should be specifically sought and, conversely, observing one of these findings should increase the suspicion for an ACL tear.
It is important for the interpreting physician to remember that ACL tears are not isolated events, but only the most obvious sign of a complex knee injury involving multiple structures. A child with an ACL tear has a lifetime of loading ahead, and for optimal planning of surgical treatment and prediction of long-term clinical outcome, the radiologist should report the ACL tear with an integrated description of associated findings implying the likely injury mechanism.
The authors declare no conflict of interest. No funding was received for this work. Institutional Review Board approval was obtained.
Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.
De Maesenner M, Van Roy F, Lenchik L, Barbaix E, De Ridder F, Osteaux M (2000) Three layers of the medial capsular and supporting structures of the knee: MR imaging-anatomic correlation. Radiographics 20(Spec):S83–S89 CrossRef
- Spectrum of injuries associated with paediatric ACL tears: an MRI pictorial review
Jacob L. Jaremko
Zachary D. Guenther
Lennart B. O. Jans
Peter J. MacMahon
- Springer Berlin Heidelberg
Neu im Fachgebiet Radiologie
Meistgelesene Bücher aus der Radiologie
e.Med Kampagnen-Visual, Mail Icon II