A criticalist's view of canine pancreatitis (Proceedings)

Many canine patients present to the veterinarian with vomiting, diarrhea, anorexia, and abdominal discomfort, with or without fever. Based on these non-specific signs of disease, the veterinarian has to build a differential diagnosis list. This list commonly includes acute gastro-enteritis (parasitic, bacterial, viral), dietary indiscretion/foreign body ingestion, toxicity, renal disease, pancreatitis, hypoadrenocorticism, and cholangiohepatitis, among others. The job of the veterinary clinician is to determine which of these disease processes are involved in the patient's illness and then determine the best course of therapy for the patient. The goals of this lecture are to help the veterinarian differentiate pancreatitis from the other disease processes listed above and then determine what traditional and ancillary therapies are warranted in treating the patient with acute pancreatitis.

Pathophysiology of Acute Pancreatitis

Many different inciting causes have been postulated to cause canine acute pancreatitis. High fat diets and dietary indiscretions are the most common clinically quoted causes of pancreatitis. Yet, there is little scientific evidence for these etiologies. The use of corticosteroids has also been quoted as an underlying cause. Corticosteroid use in dogs causes serum lipase activity to increase. However, several studies have shown no pancreatic lesions in animals receiving corticosteroids; the lipase must be another type of lipase. However, a lot of patients receiving steroid therapy have diseases that predispose them to pancreatitis, so the primary disease process may be at fault, not the steroids administered to treat those diseases. Etiologies that have been shown to cause pancreatitis include certain drugs, most commonly potassium bromide, hypercalcemia, hyperlipidemia, especially in miniature Schnauzers, pancreatic hypoperfusion from general anesthesia or hypovolemia, and trauma. Immune-mediated disease, common in the human population with pancreatitis, may play a role in many of the canine patients with pancreatitis, although this has yet to be studied.

Irrespective of the initiating cause, pancreatitis is generally believed to occur when digestive enzymes are activated prematurely within the pancreas. In the normal pancreas, safeguards are present to ensure that harmful pancreatic enzymes are not activated until they reach the intestinal lumen. Enzymes are stored in zymogen granules within the acinar cell in the presence of pancreatic secretory trypsin inhibition and are released at the apical surface directly into the duct system. They are only activated in the intestine, by trypsin, following the cleavage of trypsinogen by enterokinase. In clinical pancreatitis, it is thought that inappropriate premature activation of trypsin from trypsinogen in the acinar cells initiates a cascade of early activation of zymogens, especially pro-elastase and pro-phospholipase, leading to auto-digestion of the pancreas. Often pancreatic inflammation is a self-limiting process, but in some animals reduced pancreatic blood flow and leukocyte and platelet migration into the inflamed pancreas may cause progression to pancreatic necrosis. Secondary infection may arise by bacterial translocation from the intestine or ascending infection up the pancreatic duct. Release of active pancreatic enzymes and inflammatory mediators from the inflamed pancreas amplifies the severity of pancreatic inflammation, and adversely affects the function of many organs (systemic inflammatory response), and cause derangement in fluid, electrolyte and acid-base balance. It is the development of multi-systemic abnormalities that separates mild from severe, potentially fatal pancreatitis.

Histologically, acute pancreatitis is characterized by findings that range from pancreatic edema to necrosis, variable infiltrates of mononuclear and polymorphonuclear cells, and local changes such as peri-pancreatic fat necrosis and thrombosis. Acute pancreatitis may resolve or persist and can be complicated by secondary infection and pseudocyst or abscess formation. It is tempting to equate mild acute pancreatitis with pancreatic edema, and severe or fatal pancreatitis with pancreatic necrosis, but this relationship has not been critically examined in patients with naturally occurring pancreatitis

Diagnosis of Acute Pancreatitis

Pancreatitis tends to occur in middle-aged, neutered dogs with no gender preference. Dogs with pancreatitis are often overweight. Terrier and non-sporting breeds tend to be overrepresented. As noted earlier, presenting signs are often non-specific and can include vomiting, diarrhea, anorexia, and abdominal pain. Examination findings are also non-specific for many abdominal disorders and can include fever or hypothermia, tachycardia, tachypnea, evidence of decreased perfusion such as pale mucus membranes and hyper- or hypodynamic pulses, evidence of dehydration such as tacky or dry mucus membranes and prolonged skin turgor, absent intestinal borborygmi, abdominal pain, abdominal distention due to ascites, and icterus.

Historically, the diagnosis of acute pancreatitis was made via noted the elevation of amylase and/or lipase on a general chemistry profile. However, amylase and lipase have been shown to be produced by multiple organs and their elimination can be affected by other organ dysfunction, making elevation of these enzymes non-specific for the diagnosis of pancreatitis. Abdominal radiographs may or may not have findings that are suggestive of pancreatitis, even in the presence of known pancreatitis. These findings may include loss of serosal detail, increased opacity in the right cranial quadrant of the abdomen, displacement of the duodenum ventrally and/or to the right, dilated hypomotile duodenum and caudal displacement of the transverse large intestine. Abdominal ultrasound was, and remains, a standard test for diagnosing acute pancreatitis. Abdominal ultrasonography often finds abnormalities that suggest or are consistent with pancreatitis as well as eliminate other potential causes of vomiting and abdominal pain. One may sometimes detect hypoechogenicity of the pancreas surrounded by hyperechoic fat in the region of the pancreas that is due to pancreatitis. At other times, a markedly thickened pancreas may be found. Ultrasonography currently seems to be the fastest test with reasonably "good" sensitivity and specificity for pancreatitis in the dog. However, ultrasonography is not perfect: it is very operator dependent, and false positive and false negative results are possible, even in the hands of an experienced operator. A test for trypsin-like immune-reactivity (TLI) was developed. This test was found to be diagnostic for pancreatic insufficiency, but several issues with the test as a diagnostic for pancreatitis were noted: renal failure can falsely elevated TLI values and a normal TLI value does not exclude acute pancreatitis as a diagnosis.

More recently, several tests to measure pancreatic lipase immuno-reactivity have been developed. These tests evaluate the serum for the specific lipase produced and released by the pancreas. Because of the short half-life of pancreas-specific lipase (90 minutes), continuous secretion of high levels of this enzyme into the circulation is needed for a diagnosis of pancreatitis. A patient-side ELISA-based colormetric canine pancreatic lipase immuno-reactivity (SNAP cPL, Idexx Laboratories) was developed. This test can differentiate abnormally high levels of this enzyme in serum from normal levels. Therefore, if the SNAP cPL is negative, pancreatitis is essentially ruled out. However, if this test is positive, a quantitative assay for canine pancreatic lipase immuno-reactivity (Spec cPL, Idexx Laboratories) should be performed. This test can differentiate those patients that may have pancreatitis (Spec cPL between 200-400ug/L) and those that have pancreatitis (Spec cPL > 400ug/L). Recently, a study did show elevations of cPLI activity without confirmed diagnosis of pancreatitis in a patient population with inflammatory bowel disease. Therefore, a combination of clinical suspicion, elevated Spec CPL, and abnormal abdominal ultrasound should be used to definitively diagnose acute pancreatitis.

Surgery or laparoscopic biopsy for histopathology is the gold standard for diagnosis of canine pancreatitis. There are advantages to obtaining surgical biopsy specimens. The pancreas and peri-pancreatic area can be evaluated for abscesses and necrosis and debridement can be performed. During the procedure, a jejunostomy feeding tube can be placed, thus providing the ability to enterally feed the patient but bypassing the pancreas for digestion. The concern regarding obtaining biopsy samples from the pancreas can worsen the disease process is unfounded. However, just like other diagnostic tests for pancreatitis, biopsy has its drawbacks. The use of general anesthesia can lead to hypoperfusion of the splanchnic circulation, exacerbating the pancreatic inflammatory response. Also, pancreatitis can be a segmental disease and, since typically only 1 or 2 samples are obtained, the biopsy sample may not be representative of the process. Unless there is absolute proof of an infectious complication, which is rare, the benefit of the procedure rarely outweighs the perils of surgical intervention.

Treatment of Acute Pancreatitis

The mainstay of therapy for acute pancreatitis is primarily supportive. Fluid therapy, analgesics, anti-emetics, and antacids constitute traditional treatment. Intravenous fluid therapy is commonly administered with crystalloid and colloidal fluids. The crystalloid fluids are the mainstay of fluid therapy for the pancreatitis patient. Crystalloid fluids consist primarily of water with a sodium or glucose base, plus the addition of other electrolytes and/or buffers. Replacement crystalloid solutions contain dissolved solutes that approximate the solute concentration found in plasma water. These solutions are indicated for the rapid replacement of intravascular volume and electrolytes as seen with severe volume depletion secondary to the losses associated with vomiting and diarrhea. With replacement crystalloid fluids, only 20-25% of the infused volume of fluid remains within the intravascular space 1 hour after infusion. Therefore, large volumes of replacement crystalloids need to be administered initially to replace intravascular volume and are continued to replace the ongoing losses from the vomitus and diarrhea. These fluids are often supplemented with potassium and/or magnesium to replace the intra-vascular and interstitial deficits found in patients affected with severe acute pancreatitis.

Colloidal fluids are high molecular weight compounds that do not readily leave the intravascular space. They exert their effect of expanding intravascular volume by holding and potentially drawing water into the vasculature. Colloid fluid solutions are indicated for the treatment of hypovolemia, sepsis, and other inflammatory conditions where crystalloid fluids may leak from the vasculature, all potentially present in the patient with pancreatitis. They are also used to improve colloid oncotic pressure in patients with low albumin from protein loss secondary to gastro-enteric losses. Common colloidal solutions used in pancreatitis patients include plasma and synthetic compounds such as hydroxyethyl starch and Oxyglobin™.

In the canine patient with acute pancreatitis, intravenous is the preferred route of fluid therapy administration. Although convenient, the use of fluid therapy via a multiple of "maintenance needs" is inappropriate as most patients do not lose or require fluid therapy in these multiples. In this situation, fluid therapy is administered as an initial bolus followed by a constant rate infusion. Initial bolus fluid therapy is indicated in the severely volume depleted patient (absolute or relative) and the severely dehydrated patient. In the intravascular volume depleted patient, the volume of the bolus is determined based on resolution of clinical signs (slower heart rate and respirations, improved pulse quality, improved mucus membrane color). Therefore, frequent reassessment of the volume depleted patient is required. Fluid rates as high as 90ml/kg/hr of crystalloid fluids may be required. Colloids can also be administered as boluses. Colloids are administered in boluses of 5ml/kg up to 20ml/kg. Again, the patient should be re-assessed after each bolus to determine if the bolus has been effective in resolving the volume depleted state or if additional boluses are needed. Crystalloid fluids are often administered along with colloids to augment their vascular volume expanding effect. Smaller doses of crystalloids than those listed above are necessary with the concomitant use of colloid. In the dehydrated patient, the quantity of a fluid bolus is based on the estimated degree of dehydration. The dehydration deficit can be calculated as body weight (kg) x estimated degree of dehydration (%) x 100 = ml of fluid required. It is recommended that one-half of the dehydration deficit be given as a bolus and the remainder replaced as a constant rate infusion over 12-24 hours.

Once vascular volume and some of the dehydration deficits have been replaced, constant rate fluid administration is initiated. As described above, it is used to replace dehydration deficits. It is also necessary to account for a patient's maintenance fluid requirements, as patients with pancreatitis are not consuming sufficient quantities of fluid on their own. Daily maintenance needs for patients varies on the age and size of the patient. Most of us were taught that the maintenance crystalloid fluid needs of a patient are approximately 54-66ml/kg/day (lower end in large dogs and higher end in small dogs). With the use of colloid fluid therapy, infusions at rates of up to 20ml/kg/day have been used to provide continuous intravascular volume support. With the use of infusions of colloid fluids, lower infusion rates of crystalloid fluids are necessary. If a patient has ongoing losses due to vomiting, diarrhea, or third spacing (ascites, pleural effusion), these losses should be replaced on a 2 milliliter replaced per estimated 1 milliliter lost.

Monitoring for the desired effect of intravenous fluid administration as well as potential adverse effects is necessary for successful fluid therapy. Proper monitoring of the patient receiving fluid therapy is a hands-on endeavor. Much of the information we need is gained through serial examinations. No single parameter evaluated will necessarily provide all the information required to guide fluid therapy. Physical examination parameters that should be evaluated include a patient's weight, mentation, skin turgor, pulse rate and quality, respiratory rate and effort, serial lung auscultation for rales, mucus membrane color, and capillary refill time. Skin turgor allows for gross assessment of hydration status- prolonged skin turgor will be noted with 6-8% dehydration, markedly prolonged skin turgor will be noted with > 8% dehydration. Limitations of this method are that obesity, common in the canine pancreatitis patient, may mask decreases in skin turgor. If an indwelling urinary catheter is in place, serial evaluation of fluid input and urine output can provide significant information regarding whether too little or too much fluid therapy is being administered. Renal chemistry parameters (BUN, creatinine) in conjunction with urine specific gravity measurement provide additional information. Increased renal parameters and/or increasing urine specific gravity may provide clues that a patient is receiving too little fluid therapy. Serum lactate measurement can provide information regarding tissue perfusion. Increased serum lactate concentrations indicate that tissues are not receiving adequate oxygen for metabolism and are instead relying on anaerobic metabolism to provide energy.

The administration of fresh frozen plasma has been advocated for the treatment of severe acute pancreatitis in dogs. Aside from the colloidal effects of the albumin in plasma, albumin is also necessary for the transportation of highly protein bound drugs and helps maintain normal serum calcium levels. It also supplies anti-thrombin III and coagulation factors needed for the therapy of disseminated intravascular coagulation. Most important, although of theoretical benefit, are the α-macroglobulins fresh frozen plasma contain, which include α2-antitrypsin and other α-proteases. Once trypsinogen is activated into trypsin, it goes on to activate a whole cascade of other enzyme activation, including but not limited to kallikrein, chymotrypsin, elastase, carboxypeptidase, phospholipase and inflammatory cells via PAR-2 (trypsin receptor). By administering fresh frozen plasma, this cascade may be ameliorated.

Analgesia is important in the treatment of acute pancreatitis. The release of proteases and inflammatory mediators into the abdominal cavity causes peritonitis and significant pain. The most commonly used analgesics include parenteral opiate analgesia and transdermal patches. Parenteral analgesia can be supplied as intermittent injections or continuous rate infusions. Transdermal patches may or may not provide adequate analgesia due to variations in perfusion at the site of application and release of drug into the circulation. Epidural analgesia using a pure opioid and local anesthetic can be performed by single lumbar puncture or placement of an epidural catheter.

Canine patients with acute pancreatitis are very nauseous and vomit frequently. The nausea and vomiting often preclude voluntary enteral nutrition and make naso-gastric tube feeding difficult. Ondansetron (Zofran™ 0.1mg/kg BID to TID IV slowly over 2-5 minutes), dolasetron (Anzemet™ 0.6-3.0 mg/kg daily IV) are centrally acting serotoninergic anti-emetics commonly used. Maropitant (Cerenia 1 mg/kg SQ q 24 hrs) is a newer broad spectrum antiemetic that works both centrally and peripherally and is very effective. However, one of the contraindications for the use of maropitant is liver dysfunction. Metoclopramide, which is a dopamine inhibitor, has routinely been used in these patients. However, the effect of dopamine inhibition on perfusion of the pancreas has not been evaluated and may be detrimental.

There is no proven benefit in using antibiotics in dogs with pancreatitis, but they are commonly administered. Dogs very seldom get infected local necroses and it may be that by using early enteral nutrition alone we will reduce the risk of sepsis by reducing bacterial translocation. If antibiotics are indicated based on suspected infection, then selection of a broad spectrum antibiotic effective against gram-positive and gram-negative bacteria given parenterally is recommended, pending culture results.

The traditional therapy of withholding food from patients with acute pancreatitis may not be beneficial and in some ways may be harmful. Enteral nutrition was avoided with the concern that it would stimulate vomiting as well as pancreatic secretions, exacerbating the disease process. However, enteral nutrition is essential to promote enterocyte health. When the gastro-intestinal tract does not receive sufficient nutrients, several pathologic consequences can occur. These include villous atrophy leading to increased mucosal permeability, decreases in gut-associated lymphoid tissue (GALT), and decreased surface area for absorption of nutrients for systemic use. This may lead to gut barrier failure, allowing translocation of bacterial and endotoxin into the portal and, sometimes, systemic circulation, which may lead may lead to a systemic inflammatory response (SIRS). Several human clinical trials have shown that early enteral nutrition is associated with improved outcomes with regard to infectious complications, length of hospital stay, and cost compared to the use of parenteral nutrition and NPO. A jejunostomy tube may be ideal to minimize pancreatic stimulation but there is evidence to support use of the naso-gastric or naso-esophageal feeding tubes without increased complications.

Patients with pancreatitis that are actively vomiting are candidates for micro-enteral nutrition. Micro-enteral nutrition is the delivery of small amounts of water, electrolytes, and readily absorbed nutrients directly to the gastro-intestinal tract to maintain mucosal cell integrity. Microenteral nutrition is commonly delivered by naso-gastric feeding tubes. Naso-gastric tubes are among the easiest and least expensive forms of feeding tubes available for use. Argyle infant feeding tubes are ideal, are soft and pliable, and are well tolerated after placement by most patients. Red rubber feeding tubes are a viable alternative, although they can more reactive with the nasal mucosa. Aside from being able to provide nutrition to the cells lining the gastro-intestinal tract, a naso-gastric feeding tube can allow the removal of residual air. Removal of residual air in the stomach can decrease gastric distention, making the patient more comfortable, decreasing the chance of emesis, and increasing gastric motility. The removal of residual fluid/food will also decrease gastric distention as well as prevent food from stagnating within the stomach if limited motility is present

The use of heparin has gained favor in many critical conditions. In the research setting, heparin has been postulated to increase microvascular circulation, a problem in patients with severe acute pancreatitis. No studies have been performed evaluated the impact of any use of heparin in clinical human patients or veterinary patients.

Dopamine has been proposed as a therapy for pancreatitis. Dopamine in low doses (~ 5 µg/kg/min) may maintain mesenteric blood flow and may also limit increases in microvascular permeability due to beta-adrenergic effects. Research studies have supported attenuation of multiple organ damage by dopamine in rats and diminished severity of pancreatic injury in cats. No evidence exists in the form of clinical studies in people or animals.

Canine acute pancreatitis is a complex disease. Many recent advances have been made regarding the pathophysiology and diagnosis of this disease. Yet, underlying causes for pancreatitis remain elusive and therapy primarily remains supportive. Still, with proper treatment, many canine patients afflicted with pancreatitis can be lead relatively normal lives with only minor changes in lifestyle.

References furnished upon request.