Teleradiology can vastly improve diagnostic capabilities and patient outcomes, but only if it’s done right. Here’s how to ensure success.
Teleradiology enables veterinarians to consult radiologists about their patients by digitally sharing images over the Internet. Eli Cohen, DVM, DACVR, a clinical assistant professor of radiology at North Carolina State College of Veterinary Medicine, recently hosted a webinar in which he offered tips for getting the most from teleradiology.
Dr. Cohen stressed the value of communication between practitioners and radiologists. “There’s no replacement for having a direct conversation about a case,” he said. However, high caseloads make it impractical to discuss every patient by telephone. In the webinar, Dr. Cohen focused on 3 areas: clinical history, positioning, and technique.
Clinical History Do’s and Don’ts
Context is essential to making clinical decisions, said Dr. Cohen. He explained that image interpretation depends on 2 functions: the ability to pinpoint lesions, which is developed through training, and cognitive decision making, which relies on clinical context. The case history provides context and often determines which conditions remain on a radiologist’s diagnostic differential list. Without clinical context, Dr. Cohen said, a radiologist is simply an image interpreter, not a clinician consultant.
Radiographic findings are often nonspecific, he remarked, noting that a complete case history increases the sensitivity and specificity of his own reading. A referring veterinarian’s own diagnostic differential list is also useful because it means the case has passed through another person’s cognitive filter. In other words, 2 readers are better than 1.
What Not to Do• Don’t use vague or colloquial terms. A good teleradiology relationship depends on clear communication. A consultant should be able to understand your meaning without having to guess. Use standard medical terminology in case descriptions to ensure clarity.
• Don’t use acronyms. Acronyms and abbreviations can also obscure your meaning. Dr. Cohen asks radiology residents not to use acronyms in their own reports.
• Don’t omit pertinent details. Dr. Cohen discussed examples of clinical details that can affect his differential list:
• Fever in a dog with an increased respiratory rate can place pneumonia high on the list even if the lung lesions are not in a typical location for pneumonia.
• In a cat with a bronchial lung pattern, the absence of clinical signs of lower airway disease could put lymphoma higher on the list than asthma.
• Heart rate and respiratory rate are invaluable because they help differentiate between cardiogenic and noncardiogenic causes of pulmonary lesions.
As a specific example, Dr. Cohen discussed a consultation about a dog with a caudodorsal alveolar lung pattern. The referring veterinarian reported lip ulcerations and a history of chewing an electrical cord. These details, in addition to the presence of small curvilinear metal opacities in the stomach, confirmed the diagnosis of electrocution.
What to Do
In addition to the images themselves, Dr. Cohen recommended providing the following information when requesting a teleradiology consult:
• Presenting problem
• Relevant clinical history
• The referring veterinarian’s differential diagnosis
• Specific questions from the referring veterinarian
The dens of the axis should not be visible in a properly positioned lateral cervical radiograph, said Dr. Cohen, because the transverse processes of the atlas should overlie the dens. “If the dens looks great, the rad’s not straight!” For lateral cervical views, Dr. Cohen suggested placing padding under the neck to
keep the cervical vertebrae horizontal and to prevent the neck from bowing toward the table. A vertical curve in the neck creates obliquity, affecting the appearance of the disk spaces and vertebral endplates (Figure 1).
FIGURE 1. (A) In this properly positioned lateral cervical radiograph of a dog, the wings of the atlas overlie the dens of the axis (arrow), and the vertebral endplates are visible (arrowheads). (B) In this improperly positioned radiograph, the dens is visible (thick arrow), but the disk spaces and endplates are superimposed (thin arrow).
Orthogonal Views (taken at 90° angles)
For thoracic radiographs, Dr. Cohen said that 3 or 4 views should be standard: at least 1 ventrodorsal (or dorsoventral) view and 2 lateral views. Lung lesions may be visible on only 1 of the lateral views, so it is best to obtain both. When a veterinary patient is in lateral recumbency, the dependent lung (toward the table) undergoes temporary positional atelectasis. The resulting decrease in gas opacity can obscure soft
tissue lesions that would be visible on the other lateral view (Figure 2). Dr. Cohen also reminded the audience to take thoracic radiographs at maximal inspiration (if possible) and to use dorsoventral, rather than ventrodorsal, projections in unstable patients.
FIGURE 2. These thoracic radiographs of a dog illustrate the importance of obtaining 3 views. A soft tissue lesion in the left lung (arrowheads) is visible when the patient is in right lateral (A) and ventrodorsal (B) recumbency. The lesion is not visible when the patient is in left lateral recumbency and the left lung is dependent (C).
For abdominal radiographs, 3 views should be standard and 2 are necessary, said Dr. Cohen. Dorsoventral views of the abdomen are not usually helpful except for contrast studies and examinations focusing on the stomach, he noted. The first abdominal view taken should be the left lateral. Dr. Cohen cited a 2015 study in which dogs undergoing 3-view abdominal radiography were significantly more likely to have duodenal gas if the first projection was left lateral, followed by ventrodorsal and then right lateral.1 (Duodenal gas is useful as a negative contrast agent to depict mucosal lesions.)
With regard to technique, Dr. Cohen focused on 2 problems: quantum mottle (Figure 3) and saturation artifact. “Quantum mottle is the appearance of graininess on the image and is the result of too few photons reaching the image detector,” he said. He later added by email, “The main determinant of
photon number is milliampere seconds (mAs), but kilovolt peak (kVp) also impacts photon number. Doubling mAs is one method to reduce quantum mottle artifact, but because some systems do not allow changing mA and s (time) independently, it can result in increased time of exposure as time is also increased. This can lead to in-motion blur artifact, particularly in the thorax. Unlike film-screen systems where kVp has a large impact on image contrast, contrast in digital systems is largely the result of processing algorithms and presets (assuming adequate exposure). Therefore, increasing kVp can also be used to reduce quantum mottle and still achieve diagnostic contrast.”
FIGURE 3. Lateral abdominal radiographs of a dog illustrate proper technique (A) and quantum mottle (B).
Saturation artifact is the loss of anatomic information caused by detector overexposure. “This occurs most commonly at thin body parts in an image with variable thickness (often portions of the lung) but can occur in any part of the image,” said Dr. Cohen. “Saturated regions of the image cannot be assessed even if the image is windowed/leveled (adjusting contrast). Depending on the manufacturer of the plate, linear background striations (planking) can also be seen with overexposure. With appropriate technique, digital systems allow diagnostic assessment of the patient even across regions of different thickness.”
Dr. Cohen concluded the seminar by again pointing out the importance of communication between the referring veterinarian and the radiologist. He suggested making a telephone call if the radiology report does not seem to fit the case. “It really needs to be a relationship,” he said.
Dr. Laurie Anne Walden received her doctorate in veterinary medicine from North Carolina State University in 1994. After an internship at Auburn University College of Veterinary Medicine, she returned to North Carolina, where she has been in companion animal general practice for over 20 years. Dr. Walden is also a board-certified editor in the life sciences and owner of Walden Medical Writing.