J Am Acad Orthop Surg, Vol 12, No 1, January/February 2004, 1-5.
© 2004 the American Academy of Orthopaedic Surgeons
Multidetector-Row Computed Tomography
Georges Y. El-Khoury, MD,
D. Lee Bennett, MD and
Gerald J. Ondr, MD
Dr. El-Khoury is Professor, Radiology and Orthopaedics, Department of Radiology, The University of Iowa College of Medicine, Iowa City, IA. Dr. Bennett is Assistant Professor, Department of Radiology, The University of Iowa College of Medicine. Dr. Ondr is Radiologist, Springfield Radiological Group, Springfield, MO.
Reprint requests: Dr. El-Khoury, University of Iowa College of Medicine, 200 Hawkins Drive, 3966 JPP, Iowa City, IA 52242.
Since its introduction in 1973, computed tomography (CT) technology has advanced rapidly, with particular improvement in resolution and speed.1 The recent development of multidetector-row computed tomography (MDCT) is the result of faster computers, slip-ring configuration, enhanced x-ray tube heat capacity, efficient x-ray tube loading, and innovative detector array designs.2,3
MDCT is capable of generating large volumetric data sets from which two- and three-dimensional images of unprecedented quality are created. With this technology, hundreds of images are generated for each patient. The acquired data set is viewed as a volume rather than as individual images.4 Postprocessing workstations are almost a necessity for viewing these data sets, with two-dimensional multiplanar reformations and three-dimensional images being the dominant display modes. Two-dimensional multiplanar reformatted images are reconstructed from the data set in any chosen plane and are not limited to the sagittal and coronal planes.
 |
MDCT Technology
|
|---|
The key technological developments in MDCT are increased x-ray tube rotation speed (from 1 sec/rotation to 0.5 sec/rotation) and increased number of detectors.1–3 In standard CT machines, detectors are aligned along the axial plane of the patient (ie, x-axis); with MDCT, detectors also are stacked along the long axis of the patient (z-axis), enabling collection of 4 to 16 CT slices with each x-ray tube rotation (Fig. 1
). Scanners with 2, 4, 6, 8, and 16 detectors are commercially available; however, because the musculoskeletal system does not have any moving parts, 4-detector scanners are adequate for orthopaedic work. Raw data flow from the detector elements to digital acquisition systems (ie, channels); the terms multi-channel helical CT and multidetector-row CT are used interchangeably. The MDCT design markedly increases the volume covered and reduces scanning time to less than 1 minute in most instances. With conventional CT, the x-ray tube and the detectors are threaded by electrical cables that constrain their rotation to 180°. However, slip-ring technology in MDCT eliminates the need for electrical cables within the gantry. These conductive metal rings allow the x-ray tube and detector array to rotate continuously at high speed without cable entanglement.1–3 Although the gantry is subjected to approximately 13 Gs when the x-ray tube rotates 360° in 0.5 second, the gantry does not vibrate.
View larger version (24K):
[in this window]
[in a new window]
|
Figure 1 Components of the multidetector-row (multichannel helical) CT scanner. Detectors are stacked in both the x- and z-axes. DAS = digital acquisition system. (Courtesy of Michael Vannier, MD, Iowa City, IA.)
|
|
X-ray tubes used in conventional CT machines have limited heat capacity, and scanning thick body parts (eg, pelvis) often has to be interrupted after a few scans for tube cooling. However, the latest x-ray tubes are designed to scan continuously without overheating. They have a significantly higher heat capacity (7.5 million HU) than conventional tubes do and cool at a much faster rate.1 With its narrow-angle, cone-beam configuration rather than a fan beam, the emitted x-ray beam also has a more efficient geometry, thus enabling the use of a substantially greater percentage of the emitted x-rays.3 Also, by simultaneously obtaining multiple CT slices with each rotation, the x-ray tube is used more efficiently, thus eliminating some of the tube heat-loading problems.1
Another significant technological advance is an innovative detector array that allows the acquisition of 0.5-mm–thick slices.1 This breakthrough made isotropic imaging a reality. In isotropic imaging, the dimensions of each voxel (volume element) in the data set are equal4 (Fig. 2). Isotropic data allow reformatting of images in any plane with spatial resolution identical to the original scanning plane.
View larger version (13K):
[in this window]
[in a new window]
|
Figure 2 Isotropic and anisotropic data sets.Avoxel is isotropic when all of its sides have equal dimensions (ie, x = y = z). If the sides do not have equal dimensions, the voxel is anisotropic. (Reproduced with permission from
Buckwalter KA, Rydberg J, Kopecky KK, Crow K, Yang EL: Musculoskeletal imaging with multislice CT. AJR Am J Roentgenol 2001;176:979–986.[Free Full Text])
|
|
|
Advantages of MDCT
|
|---|
MDCT may be the only imaging procedure needed for a patient with a complex musculoskeletal problem (Fig. 3). Its advantages are unprecedented speed, the capability to cover large volumes, isotropic imaging, reduced hardware artifacts, soft-tissue imaging, and ease of image interpretation.
View larger version (55K):
[in this window]
[in a new window]
|
Figure 3 An 88-year-old woman presented to the emergency department with neck pain after a car accident. A, Lateral radiograph of the cervical spine shows soft-tissue fullness anterior to C1 and C2 (arrowhead). The tip of the dens (arrow) is high in relation to C1. B, Coronal reformatted CT image shows occipitalization of the lateral masses of C1 (arrows). C, Axial CT section showing a three-part Jefferson fracture (arrows) and calcium pyrophosphate dihydrate deposits in the soft tissues (arrowheads) posterior to the dens. The latter finding and the occipitalization of C1 are unrelated to the trauma.
|
|
Increased speed has resulted in less motion artifact (ie, improved temporal resolution) and, therefore, less need for sedation. A single-pass whole-body protocol is faster than the conventional segmental approach; this aspect is particularly important for patients with multiple trauma. The extended coverage is especially useful for the detection of noncontiguous spinal fractures, evaluation of diffuse congenital vertebral anomalies, and study of extensive long bone abnormalities. Whole-body scanning from the head to below the hips with a 16-detector scanner can be done in less than 1 minute. Imaging speed depends on slice thickness, pitch, and number of detectors.1–3 Pitch is the ratio between slice thickness and table speed. The pitch is high when the table moves more distance in relation to the slice thickness. With high pitch, the turns of the spiral are spaced farther apart than with low pitch, in which the turns of the spiral are closer together (Fig. 1
). High pitch settings result in faster scanning time and a lower dose to the patient but generate less detailed images. Low pitch settings result in slower scanning time, a higher dose to the patient, and better image quality.1 Data are not acquired as individual slices but rather as a volume, which is sectioned into slices during postprocessing.
Although single-slice helical CT, the predecessor of MDCT, consistently produced high-quality axial images, sagittal and coronal reformatted images showed stair-step artifacts. True isotropic imaging becomes possible with submillimeter slice thickness and when the acquired volume consists of tiny isotropic voxels.5,6 Isotropic imaging allows the creation of two-dimensional multiplanar reformatted images in any arbitrary plane as well as high-quality three-dimensional images.5,6 However, isotropic imaging is not always desirable because the images tend to be noisy due to photon starvation, especially in large patients in whom streak artifacts degrade the images. The x-ray tube produces a given amount of photons for a particular setting. When thin slices are acquired, as with isotropic imaging, fewer photons are available for each slice, resulting in mottle (ie, noise) in the images. Some manufacturers have developed image-processing methods to control noise; others reduce noise by increasing the radiation dose to increase the number of photons in each slice. Near-isotropic imaging (eg, slice thickness of 1 to 2 mm with images reconstructed at 0.5-mm intervals) is used in musculoskeletal imaging. It produces images with less noise and eliminates the need for scanning in two planes, which was often required with older CT machines when evaluating the wrist or foot. Near-isotropic imaging reduces radiation dose and increases speed without compromising image quality.
Soft-tissue imaging has been done predominantly with magnetic resonance imaging (MRI) and ultrasound. However, MDCT can be a good substitute when MRI is unavailable or contraindicated because of the presence of metallic hardware, a pacemaker, or shrapnel in the orbit. MDCT is particularly helpful in evaluating abnormalities of the intervertebral disk, articular cartilage, and tendons.7 In a pilot study at our institution, we found that ankle arthrography followed by MDCT is more accurate than MRI alone in assessing joint cartilage thickness.4 Tendon imaging also has been effectively accomplished with MRI and sometimes, ultrasound. For most orthopaedic surgeons, ultrasonographic images are more difficult to interpret than MRIs. Neither MRI nor ultrasound can be used in patients with metal hardware, surgical wounds, or open lacerations over the area of interest. Independent workstations can quickly display three-dimensional volume-rendered images7 (Fig. 4).
View larger version (104K):
[in this window]
[in a new window]
|
Figure 4 A three-dimensional volume-rendered CT image showing lateral dislocation of the peroneal tendons (arrowheads) in a 12-year-old boy with tarsal coalition.
|
|
|
Postprocessing of the Data Set
|
|---|
Three-dimensional surface rendering and volume rendering are used to display anatomic relationships clearly. Three-dimensional surface rendering is similar to taking a photograph of the surface of a structure; the CT data are converted to show the outline or outside shell of the structure. Volume rendering is a three-dimensional transparent representation of the imaged structure, and each voxel contributes to the image. Volume rendering has replaced surface rendering for musculoskeletal imaging. Because variations in transparency and lighting are represented on the three-dimensional volume-rendered images, such images are best viewed in color.4,8
Curved structures or structures not aligned along the axial plane used to be partially displayed in a series of images. With the availability of two-dimensional multi-planar reformatted images, a multitrauma patient can lie comfortably in any position on the CT table and high-quality two-dimensional multiplanar reformatted images can be acquired, allowing viewing of structures in any plane and comparison between sides on the same image. Certain anatomic relationships rely on the proper image plane; for example, a fracture line is best demonstrated when two-dimensional multiplanar reformatted images are reconstructed perpendicular to the plane of the fracture.
Curved structures (eg, scoliotic spine, clavicle, rib, sternum) are difficult to study on axial images because they can be displayed only in bits and pieces. Curved planar reformation is an ideal postprocessing technique for showing uninterrupted sections of these structures (Fig. 5). Such views are possible because the plane of each reconstructed section lies along the curved structure.
View larger version (116K):
[in this window]
[in a new window]
|
Figure 5 MDCT image of a trauma patient to rule out a sternal fracture. A, Sagittal image showing the line (a) drawn along the marks (x) placed to obtain a curved planar reconstruction of the sternum. B, Curved planar reformatted image showing the entire sternum with no abnormalities.
|
|
Two- and three-dimensional region-of-interest editing is used to remove obscuring structures and especially to improve visualization of intra-articular fractures of the acetabulum, tibial plateau, tibial plafond, and distal radius. Editing out the intact opposing articular surface provides an unobstructed three-dimensional view of the fractured surface8 (Fig. 6).
View larger version (68K):
[in this window]
[in a new window]
|
Figure 6 Three-dimensional image of the left acetabulum looking from the left side of the patient up into the acetabular roof. The femoral head has been edited out to provide a clear view of the acetabular roof. A transverse fracture is clearly shown (arrows).
|
|
|
Indications for MDCT
|
|---|
MDCT has revolutionized musculoskeletal imaging. CT protocols have been rewritten and image viewing radically changed. A typical MDCT study of the cervical spine, for example, is 300 to 600 images. When hard-copy images are required, every third or fourth image, or selected representative images, can be printed. Usually, however, the surgeon reviews the studies and plans the surgery at the work station. For difficult cases, such as acetabular fractures or tumors, rapid prototyping (ie, wax/plaster models) can be obtained.
MDCT is used for all of the indications that apply to conventional CT and single-slice helical CT; however, because MDCT is capable of largely suppressing metal artifacts, indications for MDCT have expanded to include a variety of conditions in which it is necessary to evaluate the integrity of metal fixation devices and implants. MDCT has more indications for soft-tissue imaging than conventional CT does—especially tendons, menisci, and articular cartilage. Menisci and cartilage are studied with arthrography, followed by MDCT (Fig. 7). MDCT is used extensively in the emergency department to study a variety of complex fractures in the spine, pelvis, acetabulum, and extremities. It has replaced plane radiography in high-risk (ie, very ill, polytrauma) patients and is the standard of care in most emergency departments. CT examinations are routinely used in the preoperative evaluation and follow-up of comminuted intra-articular fractures of the knee, ankle, foot, shoulder, elbow, and distal radius.6,8
View larger version (63K):
[in this window]
[in a new window]
|
Figure 7 Double-contrast MDCT arthrogram of the articular cartilage. This coronal reformatted image of the ankle shows normal articular cartilage. Adistraction device was applied to the ankle joint before imaging to separate the articular surfaces.
|
|
Because of its ability to scan hardware, MDCT is a powerful tool for postoperative follow-up of complex surgical procedures, fracture nonunion, surgical fusion, and hardware failure.6 Metal artifacts are caused by photo-penic holes in the projection and are displayed on CT images as sunburst streaks. Several factors influence MDCT artifact reduction, including the makeup of the detector elements, scanning parameters, and image reconstruction software. Another important factor is hardware composition; cobalt-chrome alloys cause notable artifacts, whereas titanium causes minimal artifacts. Scanning along the long axis of a hardware appliance produces notable artifacts.
|
Radiation Dose
|
|---|
MDCT delivers a radiation dose to the patient several times larger than that used with plane radiography. In their study comparing helical CT to plane radiography of the cervical spine, Rybicki et al9 reported that the CT dose was >14 times higher than radiography. Another drawback of MDCT is the complexity of the technology and the constant risk of delivering an inappropriately high dose. Dose level is particularly important in pediatric patients.10 Development of online x-ray tube current modulation to correspond with the shape and size of the body part being scanned has resulted in dose reductions of 30% to 50%.11 More recent machines also provide default imaging techniques for each structure being scanned. This capability safeguards against scanning the wrist, for example, with technique settings that are used for the pelvis. The 16-slice scanners have better dose efficiency than the 4-slice scanners do and, therefore, can provide a lower dose of radiation.
|
Summary
|
|---|
MDCT is a powerful imaging tool whose capabilities continue to expand with increasing numbers of detectors and improved postprocessing techniques. Its speed and extensive volume coverage make it ideal for imaging multitrauma patients. Unprecedented image quality, even in the presence of metallic implants or fixation devices, makes MDCT ideal for studying a multitude of skeletal, soft-tissue, and cartilage abnormalities.
|
Footnotes
|
|---|
None of the following authors or the departments with which they are affiliated has received anything of value from or owns stock in a commercial company or institution related directly or indirectly to the subject of this article: Dr. El-Khoury, Dr. Bennett, and Dr. Ondr.
|
References
|
|---|
- Mahesh M: Search for isotropic resolution in CT from conventional through multiple-row detector. Radiographics 2002;22: 949–962.[Abstract/Free Full Text]
- Hu H, He HD, Foley WD, Fox SH: Four multidetector-row helical CT: Image quality and volume coverage speed. Radiology 2000;215:55–62.[Abstract/Free Full Text]
- Dawson P, Lees WR: Multi-slice technology in computed tomography. Clin Radiol 2001;56:302–309.[ISI][Medline]
- Cody DD: AAPM/RSNA physics tutorial for residents: Topics in CT. Image processing in CT. Radiographics 2002;22:1255–1268.[Abstract/Free Full Text]
- Kalender WA: Thin-section three-dimensional spiral CT: Is isotropic imaging possible? Radiology 1995;197:578–580.[Free Full Text]
- Buckwalter KA, Rydberg J, Kopecky KK, Crow K, Yang EL: Musculoskeletal imaging with multislice CT. AJR Am J Roentgenol 2001;176:979–986.
- Pelc JS, Beaulieu CF: Volume rendering of tendon-bone relationships using unenhanced CT. AJR Am J Roentgenol 2001;176: 973–977.[Abstract/Free Full Text]
- Kuszyk BS, Heath DG, Bliss DF, Fishman EK: Skeletal 3-D CT: Advantages of volume rendering over surface rendering. Skeletal Radiol 1996;25:207–214.[ISI][Medline]
- Rybicki F, Nawfel RD, Judy PF, et al: Skin and thyroid dosimetry in cervical spine screening: Two methods for evaluation and a comparison between a helical CT and radiographic trauma series. AJR Am J Roentgenol 2002;179:933–937.[Abstract/Free Full Text]
- Brenner DJ, Elliston CD, Hall EJ, Berdon WE: Response to the statement by The Society for Pediatric Radiology on radiation risks from pediatric CT scans. Pediatr Radiol 2001;31: 389–391.
- Kalender WA, Wolf H, Suess C: Dose reduction in CT by automatically adapted tube current modulation: II. Phantom measurements. Med Phys 1999;26:2248–2253.[ISI][Medline]
Top
MDCT Technology
Advantages of MDCT