Radiographs are often the initial imaging study performed for a patient presenting with knee pain, swelling, or decreased range of motion. Radiographs can also aid in the evaluation of numerous other clinical presentations, including suspected arthritis, osteonecrosis, instability, malalignment, patellofemoral tracking abnormalities, and postoperative pain. Radiographs are usually the initial imaging study performed to assess a suspected bone lesion (e.g., a tumor or infection) and to evaluate fracture healing or hardware complications. Standard imaging protocols vary depending on the clinical presentation but often include an anteroposterior, lateral, and tunnel view of the affected knee. Other supplemental views may be added depending on the clinical presentation ( Table 93-1 ).
|Standing anteroposterior||Alignment/joint space narrowing/arthritis|
|Axial (Merchant view)||Patellofemoral joint|
|Cross-table lateral||Joint effusion often accompanies internal derangement |
Fat-fluid level indicates intracapsular fracture
|Oblique views||Identification of subtle fractures|
|Stress views||Ligamentous injury|
Radiographs are best suited for depicting abnormalities of the osseous structures such as fractures, dislocations, or bone lesions. Most athletic knee injuries, however, involve soft tissue structures rather than the bones, and the presence of subtle and seemingly insignificant osseous abnormalities can sometimes indicate the presence of a more serious soft tissue injury ( Table 93-2 ). A thorough evaluation of the soft tissues should include a search for soft tissue swelling, loss of normal fat planes, soft tissue calcification (e.g., intraarticular body and vascular calcification), and joint effusion (including a fat-fluid level).
|Radiographic Sign||Associated Abnormality|
|Effusion||Internal derangement (nonspecific)|
|Fat-fluid level||Intracapsular fracture|
|Segond fracture||Anterior cruciate ligament tear|
|Deep lateral femoral sulcus||Anterior cruciate ligament tear|
|Arcuate sign (avulsion fracture fibular head)||Posterolateral corner injury|
|Avulsion fracture medial patellar facet||Transient patellar dislocation|
The most current generation of multidetector computed tomography (CT) scanners are capable of providing high-resolution images in all imaging planes and are superb for evaluating the extent and location of intraarticular fractures in the area of the knee. CT imaging can accurately depict the extent and location of articular surface incongruity, particularly as it pertains to the tibial plateau. It is also an excellent method of assessing fracture healing, including evaluation of malunion or nonunion, which is a common complication of fractures in the region of the tibial shaft and the area of the knee. CT imaging is the modality of choice in the evaluation of bone loss and tunnel position after anterior cruciate ligament (ACL) reconstruction. CT imaging can also accurately depict the matrix of a bone lesion, demonstrating osteoid, cartilage, or fibrous matrix to aid in diagnosing focal bone lesions. The administration of intraarticular contrast followed by CT imaging through the knee (CT arthrography) is an accurate means of assessing for internal derangement in patients who have a contraindication to magnetic resonance imaging (MRI).
Ultrasound has limited usefulness with regard to evaluation of the knee. Although a few studies have been performed to assess the accuracy of detecting meniscal pathology with use of ultrasound, it has been shown that only the most peripheral portions of the meniscus can be assessed with ultrasound and that MRI is far more accurate for detection of meniscal tears and other internal derangement of the knee such as the cruciate ligament injury. Ultrasound can accurately assess the status of the collateral ligaments, as well as the quadriceps and patellar tendons, in cases of suspected injury. Ultrasound can also be a useful tool when a targeted evaluation of one of these structures is required. However, ultrasound is not generally accepted as a standard imaging technique for providing a global survey of the knee. The most common use of ultrasound of the knee is to assess for the presence of a Baker cyst and for directing aspiration of a cyst. Ultrasound is also an excellent means of assessing the large vessels of the knee and calf for the presence of arterial disease or deep venous thrombosis.
Magnetic Resonance Imaging
MRI is well accepted as the primary noninvasive imaging modality for global assessment of the knee and accurately depicts abnormalities of the ligaments, tendons, menisci, articular cartilage, and bones. It has a high negative predictive value with regard to the knee. Imaging protocols vary widely depending on the available equipment and the preferences of the imager. However, it is generally accepted that high-field strength magnets provide the highest quality images. Advancements in low-field strength system technology during the past few years have resulted in significant improvements in the overall image quality on the low field systems that are commonly used in the evaluation of the musculoskeletal system.
Use of a dedicated surface coil is mandatory to provide high-quality images regardless of the field strength of the magnetic resonance (MR) system. Images are performed in the sagittal, axial, and coronal imaging planes. The T2-weighted images are often referred to as the pathology images and best depict injuries of the muscles, tendons, ligaments, and bone. Fat-saturation techniques are commonly applied to the T2-weighted sequences to improve conspicuity of the fluid, which is particularly helpful in the detection of bone marrow abnormalities such as bone contusions. T1-weighted and proton density images have the highest signal-to-noise ratio and are often described as the anatomy images . These images are best suited for detecting meniscal pathology and for depicting the anatomy of the knee.
MR arthrography (MRA) can be performed with either direct injection of gadolinium into the joint (direct MRA) or after injection of intravenous gadolinium (indirect MRA). MRA has been shown to more accurately assess the postoperative meniscus for persistent or recurrent tears compared with conventional MRI. It can also be helpful in the detection of intraarticular bodies and in the grading of osteochondral lesions. Studies have shown that direct and indirect MRA are equivalent in the evaluation of the postoperative meniscus except during the first postoperative year after meniscal repair, when indirect MRA can result in enhancement of the granulation tissue at the site of repair and mimic a recurrent tear, resulting in a false-positive study. Compared with direct MRA, indirect MRA has the advantage of being less invasive and does not necessarily require the presence of an on-site physician for administration of the contrast medium.
The presence of meniscal pathology is the most common reason for knee arthroscopy, and MRI is well accepted as an accurate, noninvasive, preoperative screening method for meniscal evaluation. The reported accuracy of MRI ranges between 92% and 95% for detection of meniscal tears, and it has been shown that the experience of the person who reads the images is an important factor with regard to overall accuracy. A thorough knowledge of the normal meniscal MR appearance and primary and secondary MRI signs of meniscal tear, along with an understanding of the common pitfalls and common sources of diagnostic error, are mandatory for achieving the highest level of accuracy with regard to the detection of meniscal pathology.
Normal MRI Appearance
The menisci are composed of fibrocartilage and therefore appear dark on all MR pulse sequences. The normal meniscus can demonstrate intrasubstance signal, but signal extending to the superior or inferior articular surfaces of the meniscus is usually indicative of pathology. Intrameniscal signal in a child or adolescent usually represents normal meniscal vascularity, whereas in an adult it most commonly indicates the presence of myxoid degeneration or meniscal contusion.
The sagittal and coronal low echo time pulse sequences (T1 and proton density) are the primary sequences used to detect meniscal pathology, and these images provide the highest sensitivity with regard to the detection of meniscal tears. The axial imaging plane can be complementary, particularly in the detection of a radial tear or a medial meniscal root injury. The T2-weighted images lack sensitivity with regard to detecting meniscal tears but provide increased specificity when compared with T1/proton density imaging.
Direct MRI Signs of Meniscal Tear
Unequivocal surfacing signal seen on two contiguous images in the same imaging plane or in different imaging planes is nearly diagnostic of a meniscal tear ( Fig. 93-1 ). The presence of surfacing signal on only a single image has no more than a 50% chance of representing a tear.
Abnormal meniscal morphology in the absence of prior meniscal surgery, including missing or displaced meniscal tissue, is also diagnostic of a tear.
A double posterior cruciate ligament (PCL) sign ( Fig. 93-2 ) results from a displaced bucket handle fragment of the medial meniscus located within the intercondylar notch just anterior to the PCL.
Meniscocapsular separation is a sign of a meniscal tear.
The presence of intrameniscal fluid signal on T2-weighted imaging is strongly indicative of a tear even in the absence of a surfacing signal.
Indirect MRI Signs of Meniscal Tear
Although most meniscal tears are easily identified with use of direct MRI signs, occasionally equivocal findings make detection of a meniscal tear difficult. In these instances, the presence of secondary or indirect signs can increase the confidence level of detecting a tear. Indirect imaging signs in and of themselves are not diagnostic of a meniscal tear but add significance to the presence of equivocal MR findings.
The presence of a parameniscal cyst is usually associated with a horizontal or complex tear of the adjacent meniscus. A parameniscal cyst appears as a lobulated fluid-signal intensity mass on T2-weighted images and is usually located in direct contact with the adjacent meniscus ( Fig. 93-3 ). Other cystic lesions about the knee, such as a pericruciate ganglion or fluid-filled bursae or ganglia, can mimic a parameniscal cyst on MRI.
The presence of focal subchondral marrow edema underlying the meniscus or pericapsular soft tissue edema in cases of nonspecific findings is often seen in association with meniscal tears.
Greater than 3 mm of extrusion of the meniscus along the medial joint line is often associated with a radial tear or meniscal root injury of the affected meniscus.
Pitfalls and Sources of Diagnostic Error with Regard to Detection of Meniscal Pathology
Although the overall accuracy of MRI in detecting meniscal tears is often quoted as being between 92% and 95%, to achieve this level of accuracy, the person reading the images must have a thorough knowledge and understanding of the common pitfalls and sources of diagnostic errors. These sources of errors can be broadly grouped into four categories:
MR artifact: One of the most common sources of diagnostic error is the presence of patient motion artifact, which can result in an artifactual surfacing signal ( Fig. 93-4 ). Great care should be taken when attempting to diagnose a meniscal tear in the presence of a patient motion artifact. Repeat imaging without patient motion is necessary to confirm the presence of true surfacing signal.
Normal anatomic structures located in close anatomic position to the meniscus: The presence of an anatomic structure in close proximity to the meniscus can occasionally mimic grade III surfacing signal. The most common sources of errors include:
Anterior transverse meniscal ligament mimicking a tear of the anterior horn lateral meniscus ( Fig. 93-5 )
Meniscofemoral ligament mimicking a tear of the posterior horn lateral meniscus ( Fig. 93-6 )
Popliteus tendon sheath mimicking a tear of the posterior horn lateral meniscus
Miscellaneous conditions of the meniscus
Discoid meniscus ( Fig. 93-7 ) indicates a congenitally enlarged meniscus, involves the lateral meniscus in more than 90% of cases, and is associated with an increased incidence of both medial and lateral meniscal tears.
Meniscal ossicles represent an intrameniscal ossification, which most often occurs in the posterior horn of the medial meniscus near the meniscal root and demonstrates MR signal characteristics similar to bone (increased T1-weighted signal) within the substance of the meniscus.
The normal postoperative meniscus may demonstrate abnormal morphology or a persistent surfacing T1 signal. However, the presence of a surfacing fluid signal (seen on T2-weighted imaging), a displaced meniscal fragment, or a meniscal abnormality remote from the site of previous meniscal surgery are signs of persistent or recurrent meniscal tears.
Having knowledge of prior meniscal surgery is critical to avoid this pitfall. MRI scans often contain clues that indicate prior arthroscopic surgery, such as linear arthrofibrosis in Hoffa’s fat pad that correlates with the arthroscopic surgical tract.
Postoperative changes that can mimic a tear include a persistent surfacing signal on low echo time sequences, abnormal meniscal morphology, or missing meniscal tissue.
Anterior Cruciate Ligament
The ACL is composed of two separate bundles, the anteromedial and posterolateral bundles, so named for their tibial attachments. The ligament originates along the intercondylar notch portion of the lateral femoral condyle and inserts distally on the tibia near the anterior tibial spine.
High-quality MRI is considered approximately 95% accurate in the detection of an acute ACL tear (similar to that of physical examination) and is slightly less accurate in the detection of chronic ACL disruption. The role of MRI is to confirm the clinical suspicion of ACL injury and to differentiate a partial-thickness tear from a full-thickness tear. MRI can also delineate the location of a tear (proximal, midsubstance, or distal) and detect the presence of an osseous avulsion injury. In addition, MRI plays an important role in the detection of associated meniscal, cartilage, and concomitant ligament injury.
Normal MRI Appearance of the Anterior Cruciate Ligament
Sagittal T2-weighted images provide the primary evaluation of the ACL ( Fig. 93-8 ), with coronal and axial T2-weighted images being complementary. The axial images are particularly useful in the evaluation of the proximal attachment of the ACL. The two bundles are closely opposed proximally but fan out distally with broadening of the ligament distally with interspersed fibro-fatty tissue that results in intermediate signal striations within the substance of the ACL fibers distally. The striated appearance of the distal ACL should not be misinterpreted as an ACL injury. The ACL fibers should appear continuous, and on the sagittal images, the ACL should parallel but not touch the roof of the intercondylar notch.