Diagnosis of canine pancreatitis

Introduction

The synthesis and storage of digestive enzymes in the pancreas carries the risk of self-digestion and subsequent inflammation, i.e., pancreatitis. Strictly speaking, the term pancreatitis refers to inflammation (i.e., infiltration with inflammatory cells) of the exocrine pancreas, but it is also commonly expanded to include diseases of the exocrine pancreas characterized mainly by necrosis (necrotizing pancreatitis) or irreversible structural changes such as fibrosis (chronic pancreatitis), sometimes with only a minimal inflammatory component 1.

The exocrine pancreas has several mechanisms to prevent self-digestion (e.g., enzyme precursors, storage of enzymes in granules separated from lysosomes, localized high pH levels, good blood supply, etc.). Only if all these protective mechanisms are breached at the same time will pancreatitis develop. The disease itself proceeds in two stages. In the first stage, trypsin (activated from trypsinogen) is released, which in turn activates other digestive enzymes, leading to local changes such as edema, hemorrhage, infiltration with inflammatory cells, and necrosis of acinar cells and peripancreatic fat. In the second stage, the actual inflammatory process advances with recruitment of inflammatory cells and release of cytokines, which can lead to systemic organ disorders and ultimately death.

Pancreatitis is classified into acute (AP) and chronic forms (CP), depending on whether permanent histopathologic changes are absent (AP) or present (CP). Histologically, AP consists of neutrophilic inflammation associated with interstitial edema and necrosis of mesenteric fat (Figure 1). In CP, fibrosis is more prominent than inflammatory changes, with cystic degeneration of the tissue gradually increasing as the fibrosis increases.

Etiology

Pancreatitis is the most common disorder of the exocrine pancreas in dogs. Despite this, the pathophysiology is poorly understood, and its etiology remains unknown in most cases. Miniature Schnauzers are more likely to be presented with pancreatitis than other breeds 2, and this is probably due to mutations in the SPINK gene, which determines the pancreatic secretory trypsin inhibitor. Other breeds with a potentially increased risk of pancreatitis include the Cavalier King Charles Spaniel, Cocker Spaniel, Boxer, Border Collie and Yorkshire Terrier 3. Risk factors for pancreatitis include ingestion of fatty meals, trauma, local ischemia, endocrine disorders (hyperadrenocorticism, diabetes mellitus, and hypothyroidism) and the use of various drugs. Calcium, glucocorticoids, L-asparaginase, azathioprine, potassium bromide, zinc and glucantime have been identified as risk factors in the dog, but the causal relationship is not really proven for all of these drugs. Drawing comparisons from the human literature, it is also prudent to consider non-steroidal anti-inflammatory drugs (NSAIDs), thiazide diuretics, furosemide, vinca alkaloids, cholinesterase inhibitors, estrogens and salicylates as potential inducers of pancreatitis. Hyperlipidemia, and especially hypertriglyceridemia, have also been shown to cause pancreatitis in the dog. On the other hand, bacterial and fungal infections are rarely found as triggers, although Babesia canisis known to be a causative agent 4.

Diagnosis

Diagnosing pancreatitis antemortem remains a challenge for the clinician. This is for various reasons, including the undefined etiology, the often mild and nonspecific clinical signs, the poor sensitivity and specificity of most of the clinicopathological and imaging findings, the fact that concomitant disorders frequently occur, and the difficulty in obtaining or interpreting biopsy samples. Histology is still considered the gold standard of pancreatitis diagnostics, even though it is rarely used.

Clinical signs

Pancreatitis presents clinically pronounced differences, ranging from subclinical to multiple organ failure. The clinical picture usually corresponds to the picture of an acute abdomen, with anorexia, vomiting, abdominal pain, and dehydration, with or without diarrhea, commonly noted. Major systemic complications may be seen with severe pancreatitis (e.g., disseminated intravascular coagulation (DIC), pulmonary thromboembolism, cardiovascular shock, and multi-organ failure). Chronic pancreatitis (where the clinical signs are even more non-specific than with the acute version) is less common in dogs than in cats.

Imaging

Abdominal radiographs may reveal a loss of detail in the cranial abdomen and/or a mass effect. However, radiography is both insensitive and non-specific for the diagnosis of pancreatitis, and is mainly recommended to rule out concomitant diseases such as intestinal obstruction and foreign bodies. Abdominal ultrasound is usually considered the imaging method of choice for the diagnosis of pancreatitis, and is also helpful for the diagnosis or exclusion of other diseases that cause similar clinical signs. There are only a limited number of studies that have systematically evaluated the performance of abdominal ultrasonography for the diagnosis of canine pancreatitis, revealing sensitivities of 69% at best, and most of these studies are more than a decade old 5. Since then, there have been significant advances in both the quality of the equipment and the expertise of the radiologists. It is of utmost significance to underline the fact that the performance of abdominal ultrasonography for the diagnosis of pancreatitis is highly dependent on the expertise of the ultrasonographer and the quality of the equipment used. Ultrasonographic findings such as a hypoechoic pancreas, hyperechoic mesentery and abdominal effusion are relatively specific for pancreatitis, although other pancreatic lesions (e.g., neoplasia, hyperplastic nodules) may share a similar appearance with pancreatitis. It is important to recognize that some changes detected during abdominal ultrasonography may be age related, such as pancreatic duct dilation – a finding which was previously thought to be specific to pancreatitis.

Contrast-enhanced computed tomography (CT) is an extremely valuable tool for the evaluation of human patients with suspected pancreatitis. To date, few studies have evaluated the suitability of CT as a diagnostic tool for canine pancreatitis, but a recently published report confirmed that computed tomographic angiography was better than ultrasound at identifying dogs with severe acute pancreatitis and portal vein thrombosis 5. Magnetic resonance imaging (MRI) and magnetic resonance cholangiopancreatography (MRCP) are becoming the imaging modality of choice in humans for the pancreatic and biliary tract, but to date there is only limited experience of these techniques in dogs.

Routine blood work

Hematology and biochemistry tests are not specific for pancreatitis and do not appear to differ significantly between patients with AP and CP. The most common abnormal findings in serum chemistry are elevation of alkaline phosphatase and alanine aminotransferase, azotemia (mostly pre-renal), jaundice (mostly post-hepatic) and hypercholesterolemia; typically between 50-70% of these parameters will be outwith the normal reference range. Serum lipase and amylase activity are not specific to the pancreas and are not particularly sensitive for pancreatitis, but can potentially be used for diagnosis in an emergency if the clinical picture fits. Finally, an elevated serum trypsin-like immunoreactivity (TLI) concentration is quite specific for pancreatitis, but has only a sensitivity of about 30-50%.

DGGR-based lipase assays

Lipase levels can be determined by enzymatic lipase activity or by immunologic assays. Enzymatic assays measure (as the name suggests) the activity of enzymes, whereas immunologic assays detect certain parts of the protein/isoenzyme via antibodies 6 7. Most enzymatic assays utilize a 1,2-diglyceride as a substrate, others triolein, and yet others DGGR (1,2-o-dilauryl-rac-glycero-3-glutaric acid-(6’-methylresorufin) ester).

Recently, some reports have suggested that DGGR-based assays are more specific for the measurement of pancreatic lipase in canine serum than other total lipase activity assays 8 9. However, another study reported that the specificity of a DGGR-based assay for the measurement of serum lipase activity in dogs was lower than that for the more traditional 1,2-diglyceride-based assays 10. If DGGR were a specific substrate for pancreatic lipase, dogs with exocrine pancreatic insufficiency (EPI) should have negligible serum lipase activity when measured with a DGGR-based assay. Indeed, serum lipase activity has been shown to be significantly lower in dogs with EPI than in healthy control dogs 11. However, 33 out of 48 (69%) EPI dogs in this study had serum lipase activities within the reference interval, suggesting that DGGR is not exclusively hydrolyzed by pancreatic lipase, and DGGR-based assays are thus not specific for this enzyme. This would suggest that DGGR also acts as a substrate for non-pancreatic lipases, but which other lipases are detected by the DGGR-based assay remains to be determined.

Given the above, only the use of DGGR as a substrate will most likely not lead to similar results in different DGGR-based assays. However, moderate-to-good sensitivity and specificity of two DGGR-based lipase assays have been reported. A DGGR-based lipase assay¹ has been shown to have high agreement with the best-established test for pancreatic lipase immunoreactivity (Spec cPL^®, Idexx, USA) in dogs with suspected pancreatitis 8, but agreement between ultrasonography and both lipase assays results was only fair. Another DGGR-based lipase assay² has demonstrated excellent precision, reproducibility and linearity and substantial agreement between DGGR lipase and the Spec cPL, with similar sensitivity and specificity for the diagnosis of acute and chronic pancreatitis, even though the study population was very small 9.

¹ Lipase colorimetric for Roche Cobas Integra 800, Roche Diagnostics, Rotkreuz, Switzerland.

Pancreatic lipase immunoreactivity (PLI)

In contrast to lipase activity in serum, PLI only measures lipases synthesized by acinar cells of the exocrine pancreas. The antibodies used in the Spec cPL test² are specific and do not show cross-reactivity with other lipases 6 7 12. Serum PLI is highly specific for exocrine pancreatic function and shows a high sensitivity for moderate-to-severe pancreatitis 13. Furthermore, the Spec cPL test has demonstrated the best overall performance characteristics (sensitivity and specificity) compared to amylase, lipase, and cTLI for diagnosing histopathologic lesions of pancreatitis in dogs 13.

² DiaSys Lipase DC FS, Holzheim, Germany.

In recent years, a test for cage-side use on the patient has also become available (SNAP cPL^®, Idexx, USA). This test is semi-quantitative and should be used to exclude pancreatitis with a negative result. A positive cage-side result should be followed by a determination of PLI concentration for confirmation of the diagnosis and subsequent follow-up measurement. The SNAP cPL test result appears to have good correlation with the Spec cPL test – both tests were developed by the same laboratory and employ the same diagnostic antibodies 14 (Figure 2). Another study has also shown a high correlation between the two tests, suggesting that the cage-side test is the most sensitive single test that can be done in-house 15.

Looking ahead, new commercial immunologic lipase assays are becoming available, but some of them are not yet validated in the literature. For example, in a recent study, a newly released assay for the measurement of canine pancreatic lipase showed significant bias and poor concordance with partially different clinical interpretations when compared with the validated assay 16. Further research is therefore needed before newly released assays can be recommended for clinical use.

As there is currently no real gold standard for antemortem diagnosis of pancreatitis in dogs, the combination of a complete history and physical examination, measurement of pancreatic lipase immunoreactivity, and ultrasonographic examination of the pancreas is the best approach for an accurate non-invasive diagnosis of pancreatitis. The diagnosis should ideally be confirmed by pancreatic cytology and/or pancreatic histopathology (ultrasound-guided, laparoscopy or laparotomy), but this is rarely done. Abdominal ultrasound is useful but requires experience, and normal findings do not exclude pancreatitis.