MSD Manual

Please confirm that you are a health care professional

honeypot link
Professional Version

Computed Tomography in Animals


Jimmy C. Lattimer

, DVM, DACVR, Department of Veterinary Medicine and Surgery, College of Veterinary Medicine, University of Missouri

Reviewed/Revised Nov 2019 | Modified Nov 2022
Topic Resources

In computed tomography (CT), an x-ray tube moves around the body and continuously projects a thin fan of x-rays through the body. Electronic detectors opposite the tube continuously monitor the number of x-rays passing through the body and the angle at which the beam is being projected. The number of x-rays reaching the detector changes as the beam passes through different tissues because of the tube movement. A computer mathematically evaluates the data and determines the most probable density of any point within the volume of tissue scanned. The density is generally measured in terms of Hounsfield units, which divide the entire spectrum of possible densities into 4,000 levels extending from a value of –1,000 to +3,000. Air is considered to have a value of –1,000, water is at 0, and lead or some other heavy metal is at +3,000.

Because these units represent the degree to which the x-ray beam is attenuated, the image generated is really an attenuation map, which is displayed on the monitor and stored as a digital file by the computer. Together, all of the attenuation points make an image of the cross-section of the portion of the body through which the beam is passed. This tomographic image is usually referred to as a slice, and each of the individual attenuation points in the image is referred to as a voxel (volume element).

The animal is then moved a specified number of millimeters, and the process is repeated. By sequentially scanning a body area, the entire volume of interest can be imaged without any superimposition of structures. CT also has much better contrast discrimination than standard radiographs, so structures such as individual parts of the brain or individual muscle bellies are seen as separate and distinct on the CT scan. The same contrast media used to enhance structures on planar x-ray images is frequently used to further enhance the contrast between structures and help characterize lesions.

Modern multi-slice CT scanners can acquire up to 620 cross-sectional images at once; each rotation may be as short as ¼ second. These systems are capable of continuous rotation (helical or spiral scanning), in which motion of the patient through the scanner occurs in concert with the rotation of the scanner. These systems can perform a complete scan of the abdomen or thorax in ~10 seconds, and with appropriate ECG-based control can provide CT images of a beating heart during all portions of the cardiac cycle. The image reconstruction time is short, and the entire study can be completed in less time than was required to acquire a single image 25 years ago. With this greater speed and flexibility has come a proliferation in the number of images produced in a single study. For a CT scan on the skull of a small dog in the mid 1980s, more than an hour was required to collect a dozen images of low resolution. Today, a 10–15 minute examination using a 64-slice scanner often generates in excess of 5,000 images.

Many veterinary referral practices and some general practices now have 8- to 64-slice scanners installed, and the actual scan takes less time than it takes to position the animal on the scanning table. Even with such extraordinarily fast systems, veterinary patients must still be anesthetized and immobilized to perform most studies, but the period of anesthesia is short and the value and volume of the information derived is great. Although studies have been published in which fully awake or minimally sedated animals were imaged, scans performed in this manner lack controlled positioning and reflect movement and are therefore difficult to interpret. This type of study should be reserved for instances in which the animal is severely obtunded by its medical condition or for the rare instances in which sedation would significantly alter the status of the tissues (such as an attempt to evaluate the effect of drug intervention in an asthmatic cat).

CT systems are now being installed in private and academic equine practices to help evaluate diseases of the skull. These are standing systems in which a sedated horse is placed in a restraining device and, instead of the patient being moved, the CT gantry moves. Although this is a limited and expensive use for CT scanning, some practices believe that the value of the information derived is worth it. Most CT scans on horses are still done on anesthetized patients, but systems are under development that would allow CT scans to be performed on all anatomic areas in large animals with the patients standing. Such systems may revolutionize imaging of the equine limb when they become available. CT images of the equine limb are far superior to radiographs in their ability to detect and localize both bony and soft-tissue lesions. Unfortunately, as of yet there are no scanners capable of scanning the abdomen or thorax of adult horses. Were such systems available, it is likely that CT would rapidly become the imaging modality of choice for these areas in horses.

As with conventional radiography, the use of nonionic contrast agents can markedly improve the diagnostic accuracy and sensitivity of CT. For vascular studies, the contrast dosage is also substantially less than that used in conventional radiographic angiography. The utility of contrast enhancement on CT imaging is so important that nearly all veterinary CT studies of soft-tissue structures such as the brain or abdominal organs are contrast enhanced. This is unlike human medicine, in which many CT examinations are performed without contrast because the scan is being performed to answer specific questions that do not require the use of contrast. Obtaining a contrast-enhanced image after a nonenhanced study allows comparison of the two images to give insights into the hemodynamic and physiologic changes present.

Modern reconstruction algorithms provide reformatting options to generate images in planes orthogonal to the original axial (transverse) image as well as oblique reformatted images in virtually any plane, as well as three-dimensional reconstruction of structures with a given density. Bones can be depicted without the overlying soft tissues, and vascular structures that have been contrast enhanced can be depicted without any overlying tissues. Newer scanners can produce images of vessels that rival those obtained by conventional contrast angiography. These images can then be displayed as a rotating image that can be examined from any angle. Even if the CT scanner itself does not have such software, many image viewing systems optimized to evaluate CT scans also routinely include software to generate these images.

In addition to some of the imaging procedures unique to CT, this modality is replacing conventional radiography for evaluation of some structures and diseases traditionally assessed by radiography. CT scans of the skull in any species are far more informative and diagnostic than any conventional radiograph because the complex anatomy of the skull, which results in a pattern of overlying structures on a radiograph, is vastly simplified on a CT scan, making the diagnosis much more specific and accurate and improving treatment results. The major exception to this is dental radiography, for which the spatial and contrast resolution of radiographic images is unrivaled.

CT scanning has nearly replaced myelography for evaluation of spinal cord disease because of its greater safety and speed as well as being able to directly image the discs and vertebrae. However, this use of CT imaging is itself being challenged by magnetic resonance imaging. CT imaging is also receiving much attention as a screening procedure to evaluate the lungs and other anatomic areas for evidence of distant metastasis in cancer patients. Metastatic lesions in the lung are far more evident at a smaller size on CT scans than on conventional radiographs. If a CT scan is performed after detection of metastatic lesions on radiographs, the CT scan will almost always find more nodules than are evident on the radiographic images.

CT is also used to guide the acquisition of biopsy and aspirate of samples from many areas of the body, including the lungs, spine, and brain, that cannot be approached using ultrasound or other imaging modalities. The same type of approach may be used to perform image-guided therapy of some diseases such as radiofrequency ablation of tumor nodules in the liver. Such image-guided treatment obviates the need for a surgical procedure, reducing the impact of treatment on the patient.

CT scans are used to detect structural changes deep within the body, including tumors, abscesses, vascular abnormalities, occult fractures, and hematomas. Modern, high-speed scanners are also used to evaluate dynamic physiologic processes such as blood flow, changes in respiratory volume, cardiac function, and intestinal dynamics. The radiologist must have a firm knowledge of anatomy and physiology to ascertain the identity of structures in any plane through the body and evaluate changes in its anatomy and/or physiology. Knowledge of artifacts is also important in evaluating CT scans. Extensive study, experience, and training are required to become adept at performing CT procedures and deriving the most information from the images. Technologists operating CT scanners should be specifically trained and experienced in their operation to provide the best quality studies possible for a given machine.

For More Information

quiz link

Test your knowledge

Take a Quiz!