Axial CT cut through a lumbar vertebra showing the difference between dense compact cortical bone and the trabecular bone. Note the Y-shaped basivertebral venous plexus within the vertebral body.
Each CT data value represents the arithmetic mean of all attenuation values measured in an individual volume element. The grayscale display of a scanned object alone provides some information on the relative density (radiodensity) of a structure on the image. Upon comparison with the surrounding tissue, the structure may be described as isodense (same density), hypodense (low density), or hyperdense (high density).
The attenuation values for image reconstruction, ranging from -1,000 HU to +1,000 HU, are conventionally depicted in numerous corresponding shades of gray. However, the human eye can distinguish only about 15 to 20 of these shades. If the entire density scale of 2,000 HU were to be displayed in a single image, the evaluator would be able to distinguish only a single shade of gray in the diagnostically important soft-tissue range. He or she could not visualize all densitometric nuances measurable by the computer, and important diagnostic information would be lost.
The image window was therefore developed as a means of producing vivid contrasts of even slight densitometric differences. The concept of the window makes it possible to expand the gray scale (window width) according to an arbitrarily set density range (25 to 1,000 HU). Attenuation values above the upper window limit appear white, and those below the lower limit are black on the image. The window level (center of the density scale) determines which attenuation values, and therefore which organ structures, are represented in the medium shades of gray.
The window adjustments must be set in accordance with the structures to be diagnosed. Narrow window widths provide high-contrast images; however, there is a danger that structures outside that window range may be inadequately demonstrated or overlooked. With broad window settings, minor density differences appear homogeneous and are thus masked. Resolution is thereby reduced.
Traditional CT study of the spine consists of multiple slices traversing the spinal elements in axial plains, perpendicular to the long axis of the spinal column, along its different curvatures—the cervical and lumbar lordosis and the dorsal and sacral kyphosis.
The slice thickness can be chosen to fit the desired resolution. It can go down to less than 1 mm in width, especially when multipla-nar reconstructions (MPRs) are desired or needed. A smaller slice width will usually achieve better resolution, down to a certain limit. Each axial image through a vertebral body will demonstrate its shape, usually cylindrical. It contains cancellous medullary bone with trabeculae and marrow (Figure 2-10) covered by a thin layer of cortical bone delineating the vertebral body. Cortical bone delineating the posterior elements is thicker. At a disc level, the density of the cartilage and fibrous tissues will be lower than that of the bone yet higher—hyperdense—to that of the perivertebral muscles and of normal intraspinal components (Figure 2-11A). A normal lumbar intervertebral disc is slightly concave posteriorly in shape, except at L5-S1, where it appears rounded. The intraspinal normal epidural fat has a very low density on CT images as compared with all other spinal components. The epidural fat is more abundant at the lumbosacral level, where it is located mainly behind the dural sac, filling the triangular shape formed by the laminae, and in the lateral aspects of the spinal canal, medial to the intervertebral foramen (Figure 2-11B). The dural sac at the lumbosacral level is measured at about 0 HU and above in direct proportion to the relative amount of CSF. At the level of the dorsal spine the spinal cord appears as an oval-shaped structure of higher density surrounded by CSF. Remember that the cord tends to "take the side" of the concave aspect of the spine, or the "short way." So at the dorsal segment it will be anteriorly located and at the cervical segments it will be situated posterior within the spinal canal.
Modern CT systems can produce volumetric images so that reconstructions can be produced through the whole volume at any desirable plane with reasonable and good-contrast resolution. True sagittal and coronal images through the spine can be produced and enable evaluation of the alignment of the vertebrae and comparison of their shape and size, as well as evaluation of the width and shape of the spinal canal (Figures 2-12A and 2-12B).
The following are the major uses of CT in spinal imaging:
• Evaluation of acute spinal trauma
• Diagnosis of bony spinal stenosis
• Detection of degenerative spinal diseases
• Detection of calcification
• When MR is contraindicated or unavailable
Was this article helpful?