Pancreatitis has traditionally been a very challenging diagnosis from both clinical and imaging standpoints, as signs are frequently vague and exhibit great overlap with other disorders. Although serum testing and advanced imaging developments in recent years have improved our ability to diagnose pancreatitis, definitive diagnosis remains elusive.
The pancreas is a primarily glandular organ with both exocrine and endocrine functions. Anatomically, the pancreas is typically divided into the right limb, body, and left limb. The right limb of the pancreas is located within the mesoduodenum, medial to the proximal descending duodenum and coursing parallel to the long axis of the intestinal segment. The body is located adjacent to the pyloric junction, and the left limb extends along the omentum just caudal to the greater curvature of the stomach. The exocrine function of the pancreas is the secretion of bicarbonate and enzymes responsible for the breakdown of proteins, carbohydrates, and lipids. The endocrine function of the pancreas is primarily related to control of blood glucose, through the secretion of insulin and glucagon. The primary pancreatic disorders for which imaging plays an important role in veterinary medicine are exocrine pancreatic inflammation (pancreatitis) and pancreatic neoplasia (especially adenocarcinoma and insulinoma).
Image 1 – Ventrodorsal radiographic projection of the abdomen of a dog with pancreatitis. Note the loss of serosal detail in the right cranial abdomen (dotted yellow oval) and the widening of the pyloro-duodenal angle (white arrowheads).
Pancreatitis is the most common clinically encountered pancreatic disorder, and an important disease in both dogs and cats. The spectrum of outcome varies widely, from mild self-limiting signs to fatal systemic shock and DIC. While abdominal ultrasound has historically been considered the best test for pancreatitis, in truth the sensitivity and specificity of ultrasound as a test for pancreatitis varies greatly, and it is currently believed that pancreatic lipase immunoassays (Spec cPL, SNAP cPL, Spec fPL, SNAP fPL) are generally more reliable. Even so, the literature is somewhat mixed on the actual sensitivity and specificity of these tests, owing at least in part to the lack of a reliable gold standard for diagnosis of pancreatitis (even histopathology is imperfect, as a biopsy may miss an area of focal inflammation). The range of reported sensitivities in studies of quantitative cPLI is 21-78% while the range of specificities is 81-100%. The SNAP cPL reportedly has a higher sensitivity (91-94%) but lower specificity (71-78%), thus making the snap test better as a screening tool and the quantitative better as a confirmatory test. In cats, the reported range of sensitivities for Spec fPL is 67-79% with specificities from 67-100%. The SNAP fPL studies have yet to be performed, although the manufacturer claims 82-92% agreement with the Spec fPL test.
Image 2 – Normal pancreas in a Bernese Mountain Dog. Note the similarity in echogenicity between the pancreas (white arrowheads) and surrounding fat.
The most commonly performed imaging studies for pancreatitis are abdominal radiographs, ultrasound, and CT. Abdominal radiographs are generally considered to be non-useful in the diagnosis of pancreatitis, although they are still recommended as an initial imaging study due to the great degree of overlap in clinical signs between pancreatitis and other diseases such as gastrointestinal disease, hepatic disease, and neoplasia. Classic radiographic signs of pancreatitis include focal loss of serosal detail in the right cranial quadrant, widening of the pyloro-duodenal angle, a soft tissue mass effect in the right cranial quadrant, and/or gas in the duodenum (Image 1). In one study of dogs with fatal pancreatitis, fewer than a quarter of the cases had radiographic signs consistent with pancreatitis, underscoring the lack of usefulness of radiographs as a test for pancreatitis. Abdominal ultrasound is also an imperfect test for pancreatitis, due in part to the difficulty in visualizing the entire pancreas. Much of the pancreas is sandwiched between the stomach and colon, two organs which are typically gas-filled and obscure part or all of the pancreas. Ultrasound evaluation of the pancreas is also highly dependent on the skill of the sonographer as well as the quality of the ultrasound machine. In the 1980s and 1990s, during the rise in popularity of ultrasound as a medical imaging modality, it was thought that a normal pancreas was not visible on ultrasound. With the improvement of the equipment in the past two decades, a normal pancreas should always be visible, but its similarity in appearance to the surrounding fat still makes a normal pancreas very difficult to see (Image 2). Typical ultrasound findings in pancreatitis include thickening, a hypoechoic parenchyma, and hyperechoic peripancreatic fat (Image 3). Ultrasound sensitivity for pancreatitis in dogs is most commonly reported as 68%, although this is from a 1998 study, and ultrasound machines have advanced considerably since then. Ultrasound sensitivity for pancreatitis in cats ranges from 11-67%, with a recent study in cats reporting a sensitivity of 84% and specificity of 75% when using serum fPLI as gold standard. It is worth repeating that operator skill and equipment quality are paramount, and these studies are generally performed at academic institutions with top-of-the-line equipment and authored by imaging experts who by definition have a professional research interest in abdominal ultrasound. In the average clinical setting, one or both of these conditions may not be true, meaning the sensitivity and specificity can be expected to be lower than the published numbers (unlike serum PLI tests, which are not dependent on operator skill). In humans, abdominal CT is the preferred imaging test for pancreatitis. Previous studies in cats and dogs have shown mixed results with CT, although a recent pilot study of 3 phase angiographic CT in ten dogs with pancreatitis showed promising results.
Image 3 – Severe pancreatitis in a cat. Note the moderately hypoechoic pancreas (white arrowheads) and hyperechoic peripancreatic fat.
In clinical practice, the workup for suspected pancreatitis is even more confusing than the literature would suggest. While on paper the most accurate and reliable tests would seem to be the PLI, the reported sensitivities and specificities for all tests (including ultrasound) should be interpreted with caution due to the lack of a gold standard. It should be stressed that a diagnosis of pancreatitis should never be made on the basis of a single test result alone. At Angell, most internal medicine clinicians prefer abdominal ultrasound, due to both the presumed high specificity for diagnosing pancreatic abnormalities and the ability to assess the remainder of the abdomen for other potential causes of inappetence, nausea, vomiting, lethargy, and/or pain. The most important criteria are signalment, history, and physical exam findings compatible with pancreatitis. The diagnostic workup strategy then becomes one of ruling out other conditions rather than ruling in pancreatitis.