Acute pancreatitis is the most common gastrointestinal disease for which patients are acutely hospitalized, and its incidence continues to rise.¹ The majority of patients (approximately 80%) with acute pancreatitis have mild disease, and symptoms usually resolve within 1 week with basic supportive care.^2,3 The other 20% of patients develop a severe form of pancreatitis with organ failure and necrotizing pancreatitis. Necrotizing pancreatitis is now defined as either pancreatic parenchymal necrosis and peripancreatic fat necrosis or peripancreatic fat necrosis alone.^2,4 The clinical course of these patients is often characterized by a persisting systemic inflammatory response syndrome and/or (multiple) organ failure for the first 1 to 2 weeks. Despite maximal supportive care in the intensive care unit, mortality is up to 30% in patients with early persisting organ failure.^5,6 Secondary infection of the necrosis develops in 30% of patients and carries a mortality risk of approximately 15%.^7,8

A better understanding of local and systemic complications, together with better imaging and new interventions in acute pancreatitis, led to the Revised Atlanta Classification.² This revised classification aims to clarify terminology and stimulate the use of uniform definitions and standardized reporting in patients with acute pancreatitis. Based on the absence or presence of local or systemic complications, 3 categories were defined in acute pancreatitis: mild, moderate, and severe (Table 70-1). Acute pancreatitis itself is divided into interstitial edematous pancreatitis and necrotizing pancreatitis, where local complications are linked to 1 of the 2 subtypes. Local complications are peripancreatic fluid collections, pancreatic and peripancreatic necrosis (ie, sterile or infected), and walled-off necrosis (ie, sterile or infected) (Table 70-2).

The Revised Atlanta Classification is definitely a step forward in the classification of patients with acute pancreatitis. However, some practical issues with the classification need to be resolved. The cutoff value of 4 weeks is used for the definition of local complications. In the case of encapsulation of an acute necrotic collection, this is called walled-off necrosis after 4 weeks. The timing of encapsulation of collections, however, differs between patients and should probably be based on imaging instead of time from onset of disease.⁹ In acute pancreatitis, the most commonly used imaging modality is contrast-enhanced computed tomography (CECT). Determining the content of collections (ie, fluid and/or necrosis) can be difficult on CECT.⁹ Magnetic resonance imaging (MRI) is superior in this respect but is impractical in critically ill patients. These limitations of computed tomography (CT) can make implementation of the new classification difficult in some cases.

To identify patients with a high risk for a severe course of their acute pancreatitis, prognostic scoring systems have been developed. There are several scoring systems, which are based on clinical and biochemical parameters (eg, the Ranson, Acute ­Physiology and Chronic Health Evaluation II [APACHE II], Imrie, or modified Glasgow scores). C-reactive protein and blood urea nitrogen are also often used in predicting severity during admission to the hospital. A systematic review on all these parameters showed that they have a high negative predictive value but low positive predictive value.¹⁰ CECT can assess morphologic abnormalities and can be used in scoring systems (eg, the CT severity index). However, CECT can underestimate morphologic abnormalities in acute pancreatitis in the first days of admission and is inferior to clinical scoring systems.^2,11 A head-to-head comparison on the accuracy of scoring systems in predicting persisting organ failure showed that the modified Glasgow score served as best ­predictor.¹² Still, more accurate markers are needed to ­predict the severity of acute pancreatitis.

Persistent organ failure is the key determinant for mortality in acute pancreatitis. Together with the lack of a strong marker to predict severity, the International Association of Pancreatology (IAP)/American Pancreatic Association (APA guidelines recommends that persistent systemic inflammatory response syndrome (>48 hours) be used as a marker to predict the severity of acute pancreatitis.³ Persistent organ failure is also used in the Revised Atlanta Classification to define the severity of acute pancreatitis (Table 70-1).

Pain Management

Abdominal pain is the most dominant feature of acute pancreatitis during admission. Specific pain management in acute pancreatitis is lacking; therefore, the World Health Organization analgesic ladder should be followed during admission.¹³ Of course, proper pain management is crucial for all patients, but increasing evidence also suggests a beneficial impact of pain relief on both physiologic and immunologic response.¹⁴

Fluid Therapy

Early in the onset of acute pancreatitis, there is often accumulation of fluid in the third space than can cause intravascular hypovolemia and might evoke or worsen organ failure. Extensive fluid resuscitation to correct or preferably prevent intravascular hypovolemia and maintain the microcirculation of the pancreas is therefore important in the supportive care during the first days of acute pancreatitis.^15,16 More recent evidence also showed that uncontrolled, aggressive fluid therapy may induce morbidity and even mortality.^17,18 For acute pancreatitis, there is a lack of quality evidence on which type of fluid therapy should be used.¹⁹ Results from a trial in patients with severe sepsis showed that when patients were resuscitated with hydroxyethyl starch (HES) there was an increased risk for death compared to Ringer’s lactate.²⁰ The recent update of the IAP/APA treatment guidelines for acute pancreatitis recommends closely monitoring urine production and vital parameters and using Ringer’s lactate with an infusion rate of 5 to 10 mL/kg/h until resuscitation goals are reached.^3,21

Prevention of Infection

Prophylactic strategies should be focused on infections that occur early in acute pancreatitis and have a significant impact on mortality, especially bacteremia.²²

It was hypothesized that administration of probiotics would be such a strategy. Before the PROPATRIA trial was performed, 2 smaller randomized, controlled, single-institution trials were conducted with probiotic prophylaxis in patients with severe acute pancreatitis. The first trial showed that probiotics reduce pancreatic sepsis and the need for surgical intervention²³; the second study suggested that early nasojejunal feeding with synbiotics may prevent organ dysfunction in the late phase of severe acute pancreatitis.²⁴ In the larger randomized controlled multicenter trial (PROPATRIA), probiotics were compared with placebo in patients with predicted severe pancreatitis. No effect was found on infectious complications, and increased rates of bowel ischemia (9% vs 0%) and mortality (16% vs. 6%) were present when compared to the placebo group.²⁵ Although the mechanism of this adverse effect remains unknown, use of probiotics in acute pancreatitis is not recommended.

Studies in intensive care units showed that selective decontamination of the intestinal tract reduces mortality in ­general.²⁶ Selective decontamination of the intestinal tract has also been performed in severe acute pancreatitis with beneficial results. However, no effect on mortality was found, and the design of the study was not optimal, with a lack of blinding and absence of a placebo.²⁷ Potential benefits of selective decontamination of the intestinal tract should be weighed against an increase in antibiotic resistance and fungal colonization.²⁸

Several systematic reviews of randomized trials have shown that prophylactic administration of intravenous antibiotics does not prevent infected necrosis.^29-31 Therefore, antibiotics in acute pancreatitis are only indicated when infections are proven or there is a high suspicion of infected necrosis.³

Nutrition

When oral nutrition is not tolerated, enteral or parenteral nutrition should be administered. In severe pancreatitis, enteral nutrition is considered a therapeutic measure. Enteral nutrition through a nasoenteric feeding tube is superior to parenteral nutrition in terms of reducing organ failure, infected necrosis, and even mortality.³² These complications are thought to be caused by gut permeability, bacterial overgrowth, and bacterial translocation. Enteral nutrition is believed to stimulate intestinal motility and, with that, reduce bacterial overgrowth, which helps to conserve the gut mucosa.^33,34 In the PYTHON trial, patients with predicted severe pancreatitis were randomized between early nasoenteric tube feeding within 24 hours of presentation to the emergency department or an oral diet initiated 72 hours after presentation, with tube feeding only provided if the oral diet was not tolerated. The primary endpoint was a composite of major infection (infected pancreatic necrosis, bacteremia, or pneumonia) or death during the 6 months of follow-up.³⁵ The main findings of the PYTHON trial were that routine early nasoenteral nutrition did not reduce the composite endpoint of infections or mortality. In the on-demand group, 69% of patients tolerated an oral diet and did not require tube feeding. Therefore, enteral nutrition is only recommended when an oral diet is not tolerated during the first 3 to 5 days of acute pancreatitis.

Endoscopic Retrograde Cholangiopancreatography

In the case of biliary pancreatitis and cholangitis, an urgent endoscopic retrograde cholangiopancreatography (ERCP) with endoscopic sphincterotomy is indicated. There is no evidence that supports the use of ERCP in mild biliary pancreatitis in the absence of cholangitis or symptomatic common bile duct stones.^3,36 It is hypothesized that performing an ERCP with endoscopic sphincterotomy in patients with predicted severe pancreatitis and without cholangitis may decrease the severity of the disease course. However, meta-analyses of randomized trials show conflicting results.^3,36,37 The Dutch Pancreatitis Study Group is currently enrolling patients in a randomized multicenter trial to compare early ERCP with endoscopic sphincterotomy in patients with predicted severe pancreatitis without cholangitis (ISRCTN Registry No. ISRCTN97372133).

When patients fail to show clinical improvement from their systemic inflammatory response syndrome and/or organ failure or when they deteriorate during the course of the disease, infected necrosis should be suspected.

A CECT in these patients may show acute necrotic collection(s) or walled-off necrosis.^2,3 Routine fine-needle aspiration (FNA) of peripancreatic collections to detect bacteria is not indicated because clinical signs and imaging signs are accurate predictors of infected necrosis in the majority of patients.³ Pathognomonic for infected necrosis are gas bubbles in collections, which are caused by gas-forming bacteria or fistulas between the collection and the intestinal tract.⁹ FNA of the collection may be required when the diagnosis of infected necrosis is unclear. Still, one has to bear in mind that false-negative FNA results have been reported in 12% to 25% of patients and prospective studies with a proper design are lacking on this subject.^38-40