There are a number of imaging modalities available to the clinician to assist in the evaluation of foot and ankle pathology. An understanding of each technique and its limitations is crucial in providing a rational approach to radiological investigation. The variety of techniques will be described, highlighting the particular advantages and shortcomings of each. Recent advances and variations relating to the individual modalities are discussed together with the normal imaging appearance of ankle and foot structures. The imaging findings for a range of common and important abnormalities are used to emphasize how imaging can be best utilized.
The initial evaluation of many conditions of the foot and ankle is with plain radiographs. A radiograph is produced through variations in the absorption of ionizing radiation by the body’s tissues, resulting in excellent spatial resolution between soft tissues and bone, due to their relative attenuation values. Excepting acute trauma, radiographs should be performed weight bearing, if tolerated by the patient. This provides standardization of views and allows review of subtle malalignments, giving important biomechanical information.
In the ankle and hindfoot, routine radiographs include a lateral, an anteroposterior (AP), and mortise view. The mortise view is acquired with 15 to 20° of internal rotation to allow unobstructed assessment of the talar dome. Variations of the standard radiographs can be used to answer specific queries; this is particularly helpful in the ankle and foot where complex anatomy can obscure important findings. For example, the Harris–Beath view provides an axial assessment of the calcaneum including the sustentaculum tali and is particularly informative in suspected fracture assessment and in imaging suspected coalition. The Harris–Beath view is taken from behind with the patient standing with the sole of the foot on the radiographic plate, and the ankle flexed, a position similar to that adopted for skiing. The X-ray beam is angled 45° downward, toward the midline of the heel.
In the foot the standard views include an AP view and a lateral view. The addition of an external oblique radiograph is essential for fracture assessment and optimizes visualization of the lateral midfoot, it is also useful in diagnosis of calcaneonavicular coalition1 (Figure 3.1).
Radiographs are widely available and relatively inexpensive. Many institutions are now “filmless” allowing radiographs to be acquired, viewed, and stored digitally. The Picture Archive and Communication System (PACS) facilitates rapid comparison with previous imaging studies. Plain radiographs are particularly useful in the diagnosis of bony abnormalities. The presence of a joint effusion or soft tissue swelling can help in cases of radiographically occult injuries2 (Figure 3.2). Soft tissue calcification can point to underlying connective tissue disease, arthropathy, or even tumor (Figure 3.3).
Figure 3.2 Lateral radiograph demonstrates an ankle joint effusion (arrowheads) but no fracture, following trauma.
The acquisition of plain radiographs involves ionizing radiation and while the dose to the extremity is minimal, the potential hazards of radiation should not be ignored. A detailed assessment of the soft tissues is not possible on radiographs as a result of the narrow range of attenuation values of the various soft tissues, even when injured. It is important that the clinician interprets the result of any radiographic examination in the context of the clinical scenario, and that additional imaging is performed if concern persists.
Active or passive stress views may demonstrate indirect evidence of an associated ligamentous injury. The combination of the additional applied force and an underlying ligamentous disruption results in widening of the joint space. In the ankle, stress views can evaluate for disruption of the lateral ligament complex, the medial ligament complex, and the tibiofibular syndesmosis.
Fluoroscopic techniques are typically used in orthopedic surgery and across radiological services to guide fracture reduction or aid interventional procedures. Similar to standard radiography this modality utilizes a related x-ray source but produces real-time dynamic assessment.
Arthrography involves the injection of a radio-opaque contrast agent into a joint. This is typically under fluoroscopic guidance, although ultrasound (US) guidance is also used. Indirect information pertaining to the soft tissues can be deduced from the pattern of distribution of the injected contrast medium. Both diagnostic and therapeutic joint injections are frequently undertaken with arthrographic control to ensure that the agent has been injected into the correct joint. Arthrography also establishes with which joints the injected joint communicates, this is important if the injection is diagnostic.
In current practice, arthrography is often performed in conjunction with MRI scanning and less frequently with CT (Figure 3.4). In the ankle joint both direct and indirect MR arthrography have a role in the evaluation of ligamentous injuries, impingement syndromes, cartilage lesions, loose bodies, osteochondral lesions of the talus, and synovial joint disorders3.
Conventional radiography can be modified to acquire numerous low-dose images of a specific body part resulting in the acquisition of digital tomosynthesis images. This modality was established both for breast imaging and in the evaluation of pulmonary nodules, but has been extended into musculoskeletal imaging4. The radiation dose is greater than for conventional radiography but lower than in CT. Tomosynthesis shows promise in the evaluation of postoperative patients as it can reduce streak artifact. Streak artifact is the dark and bright streaks from metalwork that distorts CT images. Studies have demonstrated the value of tomosynthesis in relation to wrist fractures, but it also has the potential to evaluate the foot and ankle for occult bony injury, where complex anatomy limits evaluation by plain radiography5 (Figure 3.5).
Ultrasound plays a key role in the diagnosis and management of musculoskeletal abnormality. High-frequency sound waves produced by the probe reverberate back from internal structures and the resultant echoes received by the probe are converted into the displayed image. For the evaluation of superficial musculoskeletal structures a high-frequency probe is necessary, typically a linear array probe of at least 7 MHz and ideally 10 MHz or greater. These higher frequency probes offer better spatial resolution, but reduced depth penetration. A small footprint probe is a useful adjunct in the foot and ankle. While intrinsic evaluation of bone is not possible with US, the periosteum is well visualized and occult stress fractures of the ankle or metatarsals can be detected.
Ultrasound is a high-resolution, rapid real-time examination and involves no radiation. Compared to other imaging modalities it offers the advantage of dynamic review with structures being evaluated during active and passive movement. Due to the superficial location of the ankle and foot tendons, US is an ideal tool to evaluate these structures. In the ankle, dynamic US is routinely used to assess for subluxation or dislocation of the peroneal tendons.
Doppler evaluation of vascularity is used in the imaging of joints for synovitis, tendons for neovascularity (Figure 3.6), and assessing blood flow within soft tissue masses. As the US wave is reflected from the moving blood there is a change in the frequency of the wave that is received by the US probe. Power Doppler uses signal amplification to increase sensitivity so that even small, low-flow vessels are recognized. Traditional color Doppler is less sensitive, but can be used to determine the speed and direction of blood flow; important, for example, in deep vein thrombosis imaging. Although both Doppler techniques can be used in musculoskeletal assessment, power Doppler is generally preferred as it is more sensitive to blood flow and the directional information provided by color Doppler is not required.
Figure 3.6 Longitudinal USS of the anterior ankle with severe tendinosis within tibialis anterior including abnormal vascularity on power Doppler. Note normal smooth bony cortex and periosteum of the anterior tibia (arrow) and normal hypoechoic articular cartilage covering the anterior aspect of the talar dome (arrowheads).
Ultrasound is operator dependent and there are a number of intrinsic artifacts that can influence image quality. The most frequently encountered in musculoskeletal US is anisotropy. Anisotropy is an artifact that occurs in muscles and tendons during musculoskeletal US as a result of the linear arrangement of the structures being assessed. When the US beam is perpendicular to a tendon, the normal tendon has a characteristic hyperechoic, fibrillar appearance. If the beam is at an angle to the structure, the tendon becomes hypoechoic, simulating tendon pathology such as tendinosis. Anisotropy may lead to an incorrect diagnosis of tendinosis or tendon tear. This is an important potential pitfall in the imaging of tendons, ligaments, and muscle6.
When beam steering is applied to the transducer array the beam can be electronically tilted by 30 to 40°. This technique, either alone or in conjunction with manual angulation of the probe, allows the operator to acquire images from varying angles and helps reduce, or even eliminate, anisotropic artifact.
Panoramic scanning or EFOV US can be used to demonstrate an abnormality that is greater than the width of the US probe by reconstructing several images to form a composite view of the structure under review (Figure 3.7). While the diagnostic quality of the US is not improved, this technique produces a continuous image, which can be a useful overview for the referring clinician.
Figure 3.7 An extended field of view (EFOV) sonogram of the calf demonstrates normal medial gastrocnemius and soleus. Note the organized linear appearance of the muscle fascicles running obliquely.
Standard sonography can be combined with the intravenous administration of specific microbubble contrast agents. These are markedly echogenic and can be used to assess microcirculation. At present, for musculoskeletal pathologies, these techniques lie firmly within the research forum, but CEUS is emerging as a promising adjunct in musculoskeletal US imaging, particularly in rheumatological conditions7.
Traditional or B-mode US relies on morphological changes within a structure to denote an underlying pathological process. Modification of the technique and equipment with elastography can provide a measure of tissue stiffness by gentle manual compression of the tissues under evaluation. The benefit of sonoelastography over conventional techniques in musculoskeletal US has not yet been fully evaluated8. Nevertheless, tissue softening has been identified as an early indicator of several pathological processes, including tendon degeneration. Thus elastography can be used as an adjunct to B-mode US in, for example, evaluation of the tendo Achillis (Figure 3.8).
Multiple parallel images are produced through an array of x-ray detectors that move circumferentially around a patient, while the patient is moved through the CT scanner. The spatial resolution of calcified structures on CT renders it an ideal modality to evaluate bone and soft tissue mineralization. While acquired axially, images can subsequently be reconstructed in multiple planes, typically coronal and sagittal. For surgical planning a 3D surface rendered image can be generated from the 2D data.
Intravenous contrast is not routinely utilized in musculoskeletal CT, although it may be administered to evaluate peripheral vascularity in cases of trauma with suspected vascular compromise or in the further evaluation of a soft tissue mass.
The process of acquiring a CT takes seconds and is well tolerated by patients. Cross-sectional imaging with CT can help detect loose bodies, osseous coalitions, and help in the preoperative planning of complex fracture fixation.
As with conventional radiography, CT involves ionizing radiation – but at a higher dose. The soft tissue structures have similar attenuation values, consequently the predominant limitation of musculoskeletal CT is poor soft tissue evaluation. Even when abnormal the soft tissues cannot be differentiated from the adjacent normal soft tissues. While much more problematic with MRI, metallic artifact from a surgical prosthesis can obscure diagnostic detail on CT.
In dual-energy CT two x-ray tubes at different kilovoltages simultaneously acquire two data sets of the desired region. A comparison between different materials’ attenuation values at these two acquisitions allows differentiation between uric acid and calcium, and allows imaging of uric acid crystals in tophaceous gout (Figure 3.9)9. Dual-energy CT also has potential in the evaluation of traumatic bony injuries and detecting acute bone marrow edema. It may provide an alternative assessment technique for bone bruise and stress response in the foot and ankle. Dose-reduction techniques both in standard and dual-energy CT are being utilized without compromising the diagnostic ability of the study10–11.
Magnetic resonance imaging has revolutionized musculoskeletal imaging and offers excellent spatial and contrast resolution. Magnetic resonance images are produced by the effect of a strong homogeneous magnetic field on the body’s hydrogen nuclei in water molecules, hence avoiding ionizing radiation and its associated risks.
Magnetic resonance technology is rapidly advancing, in particular functional MRI techniques. The complexity of MRI is not helped by the vast variety of sequences available, with inconsistency in the terminology between different manufacturers for similar sequences.
In routine musculoskeletal imaging, three general groups of MR sequences are commonly used, although there are many variations.
Fat and fluid both return a high signal in T2‒weighted sequences, and to increase fluid conspicuity sequences are often performed with fat suppression to reduce the fat signal. These sequences are denoted as T2fs. A frequently used alternative to fat saturation is the short tau inversion recovery (STIR) sequence.
For simplicity, proton density (PD) sequences can be considered fluid-sensitive intermediate-weighted sequences, particularly useful for hyaline cartilage assessment. Proton density sequences are often combined with fat suppression, denoted PDfs.
Magnetic resonance imaging provides excellent spatial and contrast resolution of the ankle and foot without any associated ionizing radiation. Consequently it is widely utilized in the evaluation of bony and soft tissue abnormalities of the foot and ankle.
As a result of the strong magnetic field, patients with many implantable devices, including pacemakers, are currently unsuitable for MRI, although this is an area of rapid advancement. The development of wide, short-bore MRI scanners has increased compliance in claustrophobic patients, and some centers use open-bore scanners to further help in scanning the claustrophobic patient12. Each sequence of a diagnostic study requires the patient to remain completely still, as even minor movement during image acquisition can cause significant artifact and loss of diagnostic detail. This can be problematic as sequences take several minutes to acquire with a much longer overall scanning time compared to CT. Although there are exceptions, the majority of routinely used sequences can only be viewed in the form in which they are obtained, as it is not possible to manipulate images into alternative planes, in contrast to CT.
Ferrous materials create artifacts, which impact on image quality as a result of image distortion and signal voids. The newer generation orthopedic titanium and non-ferrous prostheses are less of a problem; however, imaging of prostheses is an ongoing challenge with development of metal artifact reduction sequences (MARS) to maximize the diagnostic information achievable in these patients13.
While MRI is highly sensitive, it is not always specific and study findings must be interpreted in the context of the clinical scenario and the appearance on other imaging modalities. For example, an MRI examination will detect increased fluid or edema, but cannot always differentiate between the various etiologies, which include trauma, infection, or malignancy.
Magnetic resonance arthrography is widely used in the evaluation of the labrum of the hip and glenohumeral joints. Both direct and indirect arthrography have a role in the evaluation of a variety of ankle joint pathologies, including ligamentous injuries, evaluation of loose bodies, and osteochondral lesions14; however, with the advent of more advanced MR sequences and higher Tesla scanners, many units no longer routinely use arthrography.
Magnetic resonance imaging provides an excellent non-invasive evaluation of articular cartilage, which as the treatment of chondral damage evolves has led to an increased focus on the development of accurate cartilage-specific sequences. These sequences focus both on the biochemical alterations and on the morphological changes including fissuring, thinning, and cartilage loss. Morphological changes are excellently evaluated with fluid-sensitive fast-spin echo and 3D T1 weighted spoiled gradient recalled echo sequences. Alterations in the water or sodium content of cartilage, or in the proteoglycan composition or distribution, can predict cartilage damage. Evaluation with T2 mapping or sequences such as delayed gadolinium-enhanced MR imaging cartilage (dGEMRIC) can identify an irregularity of chondral make-up, which precedes any morphological abnormality15–16.
A standard bone scintigram is the most common nuclear medicine technique used for the evaluation of musculoskeletal disorders and is particularly useful in the evaluation of acute stress fractures of the metatarsals. A radioactive substance, typically technetium-99m labeled methylene diphosphonate (99mTcMDP), is injected into the patient. As this undergoes radioactive decay gamma rays are emitted and detected by a gamma camera, with a resolution of 5 to 8 mm. Three-phase imaging to include an arterial phase, blood pool phase, and bone scan image can be performed to improve differentiation between bone and soft tissue. The tracer detects increased osteoblastic activity and hence areas of increased bone turnover.
Bone scintigraphy is widely available and allows evaluation of bone metabolism of the entire skeleton in a single study. It is highly sensitive for a range of osseous conditions that result in increased bone turnover.
While bone scintigraphy is highly sensitive, it has a low specificity. A wide range of bone disorders including trauma, degeneration, malignancy, and infection result in increased bone turnover. Processes without associated osteoblastic activity, such as multiple myeloma or lytic metastases, may be occult on scintigraphy. The spatial resolution is inferior to other imaging modalities and as a result the evaluation of complex anatomical areas can be limited. The radiation dose involved is greater than standard radiography.