PLAIN RADIOGRAPHIC FILMS AND DIGITAL RADIOGRAPHY

In a modern medical imaging department the plain radiograph taken on X-ray film is now almost obsolete. Plain radiographic images are mostly obtained with digital equipment using a much smaller X-ray dose compared to conventional film. The resulting image is displayed on a computer screen where it can be manipulated to alter the image brightness and contrast, magnify areas of interest, and measure the size of abnormalities. Whether obtained with a digital system or on ‘hard copy’ X-ray film, plain radiographs continue to be the commonest imaging modality used in orthopaedics. They are usually the only diagnostic images required for most cases of trauma and the great majority of degenerative disorders. Fracture diagnosis and management is satisfactorily guided by plain films, as is planning joint replacement surgery for arthritic disease in the hip and knee. The use of plain films has diminished in other areas of orthopaedic practice, where for many years they were the standard imaging investigation. An example would be patients with back pain, where it is now recognised that the plain film has little to offer in evaluation and MRI is the investigation of choice.

COMPUTERISED TOMOGRAPHY (CT)

CT is a technique that utilises X-rays, but acquires multiple thin slices axially through the area of interest. With the advances in digital technology modern CT scanners can now acquire this data very rapidly. Current models in use can obtain 16 slices simultaneously, while new machines will be able to acquire 64 and even 256 slices at a time. The result is a stack of very thin radiographic images, usually about 1 mm in thickness. These can be manipulated by the CT computer to generate images in any plane with excellent detail. Typically these images are reformatted into coronal and sagittal planes so that images are similar to looking at AP and lateral radiographs (Fig. 2.1). These images can be acquired in only a few seconds and the reconstructions are also available within a similar short time. Modern scanners can also rapidly reconstruct a three-dimensional model of the scanned area (Fig. 2.2). The commonest role of CT in orthopaedics is assisting in the management of trauma patients with complex fractures. While the fracture is usually easily visualised on plain radiographs, the exact position, number and size of the fracture fragments may be more difficult to identify correctly. This is where CT and its multiplanar capability is of most use in orthopaedic patient management. Typical areas where CT is of most value are in the assessment of fractures in complex anatomical sites, such as the spine, pelvis, tibial plateau and calcaneus. Fracture treatment planning is greatly aided by the multiple views that can be obtained of these complex injuries and in selected cases three-dimensional reconstructions can also add useful information.

Fig. 2.1 Reconstructed CT images acquired in axial plane but reformatted in the coronal and sagittal planes. With current scanners the quality and detail of the images is as good as on the original axial scans. A A sagittal reconstruction of the cervical spine. B A coronal reconstruction of a hip.

Fig. 2.2 CT scan which has been reconstructed to give a three-dimensional image. On computer screen this can be rotated on real time to allow full evaluation of the bony structures in detail.

Non-traumatic conditions where CT can be useful are the evaluation of bone- and cartilage-based tumours. Although MRI is the principal method used for the anatomical staging of these tumours, CT may add further useful information regarding the extent of the ossifying or calcified portions of these tumours which an MR scan can underestimate (Fig. 2.3). Another example is osteoid osteoma, where a CT scan provides the diagnostic finding of a very small central sclerotic nidus (Fig. 2.4). A CT scan is also valuable for the full staging of bone and soft-tissue sarcomas, where it is used routinely to detect the presence of lung metastases (Fig. 2.5).

Fig. 2.3 A MRI scan of a very large mass arising from the pelvis and occupying a large part of the abdomen. B A coronal reformatted CT of the same lesion. This shows the bony origin and the bone content of the tumour which is not well seen on the MRI scan.

Fig. 2.4 An axial CT through an osteoid osteoma. The central sclerotic nidus is identified (arrow) and is best seen on CT.

Fig. 2.5 CT of lungs in a patient with an osteosarcoma of the femur demonstrating pulmonary metastases.

MAGNETIC RESONANCE IMAGING (MRI)

MRI is a new technology now commonly used in the evaluation of orthopaedic patients without the use of ionising radiation and with no known harmful effects. MR scanning can produce images in any plane and the tissue contrast can be altered to highlight particular anatomical structures and areas of pathology.

When a patient is placed into the core of a very powerful magnet the protons (hydrogen atoms) of the body align themselves to the magnetic field. If a radiofrequency pulse is then applied the protons will be tipped slightly from the line of the field and for a short time will oscillate (or resonate – hence the name). These resonating protons emit low-power radio waves which can be detected by magnetic coils within the scanner and can then be used to form a three-dimensional image. By altering how the resonance is performed and how the radio waves are captured, it is possible to formulate the images into slices in any plane. It is also possible to alter the characteristics of the image in a large variety of ways. The most common types used in orthopaedic imaging are called T1 and T2 weighted images. Images where signal from fat is suppressed, termed fat saturation, are also useful.

The T1 images show fat as white, bone and tendons as black, with muscle and water appearing grey (Fig. 2.6A). A T2 image of the same structure is most easily recognised because water is seen as white (Fig. 2.6B). This is useful in the detection of fluid structures such as cerebrospinal fluid in the spinal canal and cystic masses in the soft tissues. If the same image then has fat saturation, the fat becomes black leaving even small amounts of fluid as the only white areas on the image. This means that subtle oedema in muscle or bone can be easily identified.

Fig. 2.6 A A sagittal T1 weighted image of the lumbar spine. Note that the CSF in the spinal canal is dark grey. B Sagittal T2 weighted image of same patient. Here the CSF is now white. Fluid appears white on T2 weighted images. The intervertebral discs are healthy as they can be seen to contain high signal fluid within the nucleus pulposus.

Further information can be obtained by using MR contrast agents. These substances are injected intravenously and are most commonly compounds derived from gadolinium, a heavy metal which alters the signal characteristics of various structures, particularly those with a large blood supply. Further scans obtained after injection of this contrast medium may give helpful additional information in selected cases. Typically this is used for the assessment of tumours, infection and post operative patients following spinal surgery (Fig. 2.7).

Fig. 2.7 A Axial T1 weighted scan. This is a patient who had a microdiscectomy at L5/S1 on the right and now has recurrent symptoms. There is a mass of abnormal tissue (arrow) along the right side of the thecal sac. It is not possible to see the right S1 nerve root. A recurrent disc herniation could be the cause of this. B Following administration of intravenous gadolinium there is enhancement of all the abnormal tissue indicating that this is post-operative granulation tissue and not disc material which would not enhance. The S1 root is now clearly visible (arrow) surrounded by the granulation tissue.

MRI is particularly helpful where the abnormality is confined to soft-tissue structures and is most commonly used for evaluation of disease and injury in the spine and knee. In the spine, MRI can identify the intervertebral discs and individual nerve roots, making it the investigation of choice for patients with suspected disc herniation causing nerve root compression (Fig. 2.8). It can also be used to identify spinal stenosis, disc space infection and spinal tumours.

Fig. 2.8 A Sagittal T2 weighted image of lumbar spine. There is a large herniation of the L4/5 disc. B Axial scan through the disc space showing disc material has herniated into the left side of the spinal canal and compressed the left L5 nerve (compare with normal nerve on the right). By convention axial images are viewed as if looking from the patient’s feet. Thus, the patient’s right is on the left of the image and vice versa.

In patients with knee pain, MRI is highly accurate in diagnosing patients with meniscal tears and can distinguish between degenerate oblique tears and the less common bucket handle tears. It is also valuable in detecting injuries of the cruciate and collateral ligaments following trauma to the knee joint.

In the shoulder MR scans are increasingly used to identify degenerative tendinopathy of the rotator cuff muscles and tears of the individual tendons (Fig. 2.9). Around the ankle it has proved useful in the diagnosis of tears and degeneration of the Achilles tendon, as well as for osteochondral fractures of the talar dome, and injuries to the collateral ligaments of the ankle and subtalar joints.

Fig. 2.9 Coronal MRI scan of shoulder. There is fluid seen as an area of bright signal within the supraspinatus tendon indicating a tear of the rotator cuff.

MRI is particularly sensitive for imaging the bone marrow because of its high fat content, so that any disease associated with abnormal bone marrow will be identified. These include malignant primary bone tumours, as well as metastases and myeloma, all of which are marrow-based disorders (Fig. 2.10). Infection of bone, avascular necrosis and stress fractures, only cause changes visible on plain radiographs late in the disease process, but MR scans will identify all of these conditions at an early stage.

Fig. 2.10 A Plain film of pelvis does not show any abnormality. B MRI shows extensive abnormal signal in right proximal femur. This is due to marrow replacement by tumour. Patient subsequently shown to have metastatic disease.

ULTRASOUND

Ultrasound has a useful role in orthopaedic imaging. Ultrasound machines are readily available in almost every hospital. The scan only takes a short time, but its value and reliability is very operator dependent. Like MRI, there are no known dangerous side effects associated with the use of ultrasound. While little or no information is obtained from bony structures, ultrasound can visualise the soft tissues in many areas. Particular structures which are well seen on ultrasound are tendon and muscles. Thus suspected abnormalities of the patellar tendon, Achilles tendon and rotator cuff of the shoulder can all be diagnosed by this technique (Fig. 2.11). Muscle tears are easily identified as are their associated haematomas. In skilled hands evaluation of tendon lesions around the hand and wrist can be assessed and some operators are also able to identify ligamentous injuries around the elbow and ankle. Ultrasound is also of use in the evaluation of soft-tissue masses (Fig. 2.12), and fluid-filled structures such as Baker’s cyst of the knee (Fig. 2.13), which can be diagnosed reliably by this technique alone.

Fig. 2.11 Ultrasound of shoulder. The humeral head and overlying supraspinatus muscle and tendon are labelled. The arrow points to an area of altered reflectivity within the tendon. This indicates a tear at this site.

Fig. 2.12 (left) Ultrasound of a small soft-tissue lump in the mid thigh. It contains multiple echoes seen as white speckles. This indicated a solid mass rather than a fluid lesion and is clearly delineated from surrounding muscle. The appearances suggest a nerve sheath tumour and a schwannoma was confirmed at surgery.

Fig. 2.13 (right) Ultrasound of a clinical mass in the popliteal fossa. On this occasion there is an oblong structure which does not have any echoes within it. This indicated the presence of fluid (compare with Fig. 2.12) and is a popliteal or Baker’s cyst.

Because of the lack of ossification in the femoral head at birth, ultrasound has been advocated as a method for screening the neonatal hip joint for developmental dysplasia (DDH). Its value has not been proven for universal population screening, though it is used extensively in many European countries. However, in Britain and North America, the technique is only used for selective screening in babies showing clinical evidence of hip instability or dislocation.

ISOTOPE SCANNING

Isotope bone scanning requires the intravenous injection of a radioactive tracer, usually technetium (^99mTc), linked to a molecule such as disphosphonate which will become incorporated into areas of high bone turnover where there is increased osteoblastic and osteoclastic activity. The bone-seeking radioisotope is injected intravenously and the emitted rays are charted with a gamma camera. After injection of the isotope there is an immediate diffusion of isotope into the bone matrix from the blood: thus an increased uptake of isotope locally in the skeleton denotes hyperaemia of the bone. This is seen on a scan obtained a few minutes after injection (early phase). It will only be seen in conditions with markedly increased blood flow to the area, typically in acute infection. Within 2–4 hours (late phase) the isotope is taken up by active bone-forming cells, increased uptake thus denoting increased osteogenic activity. However, this is non specific as osteoclastic activity is increased by many disease processes, including the site of healing fractures as well as tumours and bone infection. Nevertheless, taken in conjunction with the history and clinical findings, areas of increased uptake can be identified and the likeliest cause suggested. The commonest use for isotope scanning is in identifying the site of bony metastases (Fig. 2.14). Bone scanning will provide images of the complete bony skeleton and this is its major advantage. It is often necessary to supplement the scan with plain radiographs of the areas showing abnormal increased isotope uptake. An example where this is required is when the plain films may be able to identify benign causes of increased isotope uptake and distinguish these from tumour (Fig. 2.15). Isotope bone scanning may also have a role in the diagnosis of infections of bone or joint, primary bone tumours, osteoid osteoma, and stress fractures.

Fig. 2.14 A Normal isotope bone scan showing normal distribution of isotope through the bony skeleton. B Isotope scan showing multiple areas of increased isotope uptake. This pattern is seen in patients with multiple bone metastases.

Fig. 2.15 A Isotope scan in a patient with known breast carcinoma and left shoulder pain. There is marked increased uptake in the region of the left humeral head. In view of the history a metastatic deposit was thought likely. B Plain radiograph of the shoulder shows that there is an old humeral neck fracture which accounts for the isotope uptake.

A less common application of radioisotope imaging is for the detection of focal bone or joint infection where this is not demonstrable clinically. The technique takes 24–48 hours, uses the patient’s own white cells, which are labelled with indium-111 and then reinjected intravenously to localise in areas of focal inflammation.

POSITRON EMISSION TOMOGRAPHY (PET CT)

Positron emission tomography is the newest addition to imaging, now usually combined with a CT scan. In essence it is an isotope imaging technique, but one which uses a more complex isotope which emits positrons. The isotope can be linked to various molecules but the one most commonly used is fluorodeoxyglucose (FDG), which accumulates in areas of altered glucose metabolism. The localisation of the site of this uptake is improved by fusing the isotope image with that of a CT scan acquired at the same time. The most common application is in the identification of active tumours and although its role in orthopaedics is currently limited, it may have a future role in the evaluation of bone and soft-tissue tumours. The technique may also be useful in the evaluation of infection around joint replacements but as yet this is not well defined. The technology for PET CT is currently limited to a few specialist centres and is not generally available.

ARTHROGRAPHY

Arthrography is a technique where an X-ray contrast is injected directly into a joint to enable visualisation of the joint surfaces and structures within the joint which would otherwise not be visible on an X-ray. This is being used much less frequently than in the past, as MRI and ultrasound can now often visualise the structures of interest without the need for any contrast. Where it is still used, it is usually in conjunction with MRI and occasionally CT. The resultant MR arthrogram can give even greater detail about intra-articular structures than plain MRI. The commonest indications are; in the shoulder to demonstrate disruption of the glenoid labrum in patients with recurrent dislocation or instability (Fig. 2.16), and in the hip when looking for evidence of impingement and labral tears.

Fig. 2.16 MR arthrogram of the shoulder of a patient with recurrent dislocation. The scan demonstrates that the anterior cartilaginous glenoid labrum (arrow) is detached from the underlying bony rim. This is the so-called Bankart lesion.

ANGIOGRAPHY (ARTERIOGRAPHY)

This is a specialised technique which is now used only infrequently in orthopaedics. The two major indications are trauma and tumour. Following trauma with a severely displaced bone fracture, angiography may be indicated to evaluate associated vascular injury, such as may occur in the pelvis and humerus. Note that angiography can now be carried out using CT and MRI, as well as by conventional intra-arterial catheterisation. In the evaluation of tumours, arteriography may be useful to delineate the extent of the vascular supply, or the relationship of vascular structures to the tumour, to enable safe surgical planning. In both circumstances the use of conventional catheterisation for angiography has the advantage of being able to carry out therapeutic procedures. In life-threatening haemorrhage, embolisation or stenting can be undertaken at the site of vascular damage. In cases of highly vascular tumours, preoperative embolisation of the tumour may reduce intraoperative haemorrhage. This is most commonly used in patients with metastasic renal tumour as these are generally highly vascular and prone to bleed profusely at surgery without prior embolisation (Fig. 2.17).

Fig. 2.17 A Angiogram showing a large tumour blush with vessels feeding a metastasis from renal carcinoma. B Following embolisation the blood flow to the lesion has been occluded allowing surgery to take place with much less blood loss and an easier operative field.