Acute Pancreatitis Treatment & Management

See the Guidelines section for guidelines recommendations from the American College of Gastroenterology, the American Gastroenterology Association, and the World Society of Emergency Surgery.

Medical management of mild acute pancreatitis is relatively straightforward. The patient is kept NPO (nil per os—that is, nothing by mouth), and intravenous (IV) fluid hydration is provided. Analgesics are administered for pain relief. Antibiotics are generally not indicated.

If ultrasonograms show evidence of gallstones and if the cause of pancreatitis is believed to be biliary, a cholecystectomy should be performed during the same hospital admission. Feeding should be introduced enterally as the patient’s anorexia and pain resolves. Patients can be initiated on a low-fat diet initially and need not invariably start their dietary advancement using a clear liquid diet. Systematic reviews and meta-analyses have shown that administration of enteral nutrition may reduce mortality and infectious complications compared with parenteral nutrition. ^[31] Although the ideal timing to initiate enteral feeding remains undetermined, administration within 48 hours appears to be safe and tolerated. ^[31]

Serum amylase and lipase levels can be elevated in patients with brain injury (eg, cerebrovascular accident or brain trauma). These patients are generally cared for in an intensive care unit (ICU) and require mechanical ventilation. Pancreatic enzyme elevations may rise and fall considerably over many days to weeks. The elevation is believed to result from hyperstimulation of the pancreas via a central mechanism, but no evidence of acute pancreatitis is present on imaging studies.

Patients with severe acute pancreatitis require intensive care. Within hours to days, a number of complications (eg, shock, pulmonary failure, renal failure, gastrointestinal [GI] bleeding, or multiorgan system failure) may develop. The goals of medical management are to provide aggressive supportive care, to decrease inflammation, to limit infection or superinfection, and to identify and treat complications as appropriate.

Autoimmune pancreatitis is a rare condition. Corticosteroids should not be used to treat this condition in the short term in patients who are suspected of having autoimmune pancreatitis and who present with acute pancreatitis.

No evidence-based guidelines specify when a patient should be transferred to a more experienced or skilled medical center. However, if severe acute pancreatitis is suggested either by the Atlanta criteria ^[2] or by a C-reactive protein (CRP) level above 10 mg/dL, Ranson score of 4 or higher, or Acute Physiology and Chronic Health Evaluation (APACHE) II score of 9 or higher, consider transfer to an institution where an intensivist staffs the critical care unit and an interested subspecialist experienced in the diagnosis and treatment of pancreatitis is available.

Further inpatient care depends on whether any of the complications of severe pancreatitis develop and how well patients respond to treatment. This ranges from a few days to several months of intensive care.

Patients can be discharged when their pain is well controlled with oral analgesia, they are able to tolerate an oral diet that maintains their caloric needs, and all complications have been addressed adequately.

Fluid resuscitation

Patients with acute pancreatitis lose a large amount of fluids to third spacing into the retroperitoneum and intra-abdominal areas. Accordingly, they require prompt intravenous (IV) hydration within the first 24 hours. Especially in the early phase of the illness, aggressive fluid resuscitation is critically important. This cannot be overemphasized.

There is no universal consensus definitively favoring one type of fluid over another type; both crystalloids and colloids are used. Resuscitation should be sufficient to maintain hemodynamic stability. This usually involves administration of several liters of fluid as a bolus, followed by continuous infusion at a rate of 250-500 mL/h.

Central venous pressure, pulmonary artery wedge pressure, and urine output (>0.5 mL/kg/h) can be followed up as markers of adequate hydration. Careful attention should be paid to signs of overhydration, such as pulmonary edema causing hypoxia.

Nutritional support

General guidelines for nutritional support of patients with acute pancreatitis include the following:

• In patients with mild uncomplicated pancreatitis, no benefit is observed from nutritional support, and the energy (caloric) intake received with IV dextrose 5% in water (D5W) suffices; oral feedings should be initiated once the patient’s pain and anorexia resolve

• In patients with moderate-to-severe pancreatitis, begin nutritional support early in the course of management, as soon as stabilization of fluid and hemodynamic parameters permits; optimally, nasojejunal feedings with a low-fat formulation should be initiated at admission

• Total parenteral nutrition (TPN) may be required when patients cannot meet their caloric needs with enteral nutrition or when adequate jejunal access cannot be maintained; the TPN solution should include fat emulsions in amounts sufficient to prevent essential fatty acid deficiency

• If surgery is required for diagnosis or complications of the disease, place a feeding jejunostomy at the time of the operation; use a low-fat formula

• Begin oral feedings once abdominal pain has resolved and the patient regains appetite; the diet should be low in fat and protein

Theoretical considerations regarding the ability of the enterocyte to maintain a barrier against bacterial translocation favor nasojejunal feedings—hence the recommendation to attempt initiating nasojejunal feedings at admission in all patients admitted to the intensive care unit (ICU). For patients with mild acute pancreatitis, nasojejunal feedings can be avoided unless patients are unable to tolerate oral intake for over 1 week.

Research has not established whether nasojejunal tubes have an advantage over nasogastric tubes for enteral feeding.

Although TPN, which has been shown to reduce mortality, may be necessary in certain situations, it should generally be reserved for use as a second-line therapy, behind enteral feeding.

A prospective, randomized study showed that initiating oral feeding with a low-fat solid diet was as well tolerated as initiating feeding with a clear liquid diet, but it did not result in a shorter length of hospital stay. ^[32]

In a 2014 randomized, multicenter study of 208 patients with acute pancreatitis, early nasoenteric tube feeding was not superior to an oral diet initiated 72 hours after presentation. ^[33] Tube feeding was provided if the oral diet was not tolerated.

During 6 months of follow-up, major infection or death occurred in 30 of 101 patients (30%) in the early nasoenteric tube feeding group and in 28 of 104 patients (27%) in the oral diet group (risk ratio, 1.07; 95% confidence interval, 0.79 to 1.44; P = 0.76). ^[33] Of the 104 patients in the oral diet group, 72 (69%) did not require tube feeding.

Antibiotics, usually drugs of the imipenem class, should be used in any case of pancreatitis complicated by infected pancreatic necrosis. However, they should not be given routinely for fever, especially early in the disease course, because this symptom is almost universally secondary to the inflammatory response and typically does not reflect an infectious process.

Several controlled trials have evaluated the role of empiric antibiotics in patients with severe acute necrotizing pancreatitis for infectious prophylaxis.

One such trial evaluated the role of imipenem-cilastatin initiated at admission to prevent infected pancreatic necrosis. This drug combination penetrates the pancreatic parenchyma and reduces the risk of intra-abdominal infection. It appeared to offer some benefit in preventing infectious complications. Unfortunately, fungal superinfection tends to develop later in the clinical course, although this risk is probably overemphasized. ^[34]

A randomized trial failed to show any benefit from giving ciprofloxacin and metronidazole to prevent infectious complications. Accordingly, this drug combination is not routinely used for prophylaxis in the setting of acute pancreatitis. ^[35]

The bottom line is that antibiotic prophylaxis in severe pancreatitis is controversial. At this time, the routine use of antibiotics as prophylaxis against infection in severe acute pancreatitis is not recommended.

Although the role that cytokines play in the systemic inflammatory response syndrome (SIRS) appears to be important, a large clinical trial of lexipafant, a platelet-activating factor antagonist, has shown no benefit in patients with severe acute pancreatitis. Because multiple pathways are involved in the inflammatory response, further research is needed in order to define which cytokine or combination of cytokines should be targeted to ameliorate the complications of acute pancreatitis. ^[36]

Therapy directed against tumor necrosis factor-alpha (TNF-α) has been targeted as a potential treatment of acute pancreatitis; however, clinical trials have not yet determined its value in this setting.

Surgical intervention, whether by minimally invasive or conventional open techniques, is indicated when an anatomic complication amenable to a mechanical solution is present (eg, acute necrotizing pancreatitis in which the necrotic phlegmon is excised to limit a potential site of sepsis, or hemorrhagic pancreatitis in which surgical control of bleeding is warranted). Depending on the situation and local expertise, this may require the talents of an interventional radiologist, an interventional endoscopist, or surgeon (individually or in combination).

The images below provide examples of the treatment of severe acute pancreatitis by means of minimally invasive techniques.

Acute pancreatitis. Endoscopic retrograde cholangiopancreatography excluded suppurative cholangitis and established the presence of anular pancreas divisum. The dorsal pancreatogram showed extravasation into the retroperitoneum, and sphincterotomy was performed on the minor papilla. A pigtail nasopancreatic tube was then inserted into the dorsal duct and out into the retroperitoneal fluid collection. The other end of the tube was attached to bulb suction and monitored every shift.

View Media Gallery

Acute pancreatitis. Although percutaneous drains remove loculated fluid collections elsewhere in the abdomen, a nasopancreatic tube drains the retroperitoneal fluid collection. One week later, the retroperitoneal fluid collection was much smaller (the image is reversed in a horizontal direction). By this time, the patient was off pressors and was ready to be extubated.

View Media Gallery

Gallstone pancreatitis

It is optimal for patients admitted with gallstone pancreatitis to undergo cholecystectomy before discharge, rather than being scheduled for a later date as an outpatient. Patients discharged with gallstone pancreatitis without a cholecystectomy are at high risk for recurrent bouts of pancreatitis.

Aboulian et al found that in patients with mild gallstone pancreatitis, performing laparoscopic cholecystectomy within 48 hours of admission—regardless of whether abdominal pain or laboratory abnormalities had resolved—resulted in a shorter hospital stay and had no apparent impact on the technical difficulty of the procedure or the perioperative complication rate. ^[37]

In a retrospective study of data from 316 Italian patients admitted for nonsevere acute gallstone pancreatitis over 14 years, investigators found that less than one third (31%) underwent early laparoscopic cholecystectomy (≤72 hours). ^[38] The most common factors related to surgical delay were the need to (1) stablilize comorbid conditions and (2) preoperatively investigate the common bile duct; other factors were significantly advanced age and an increased incidence of clinical signs indicating the presence of common bile duct stones. Although early laparoscopic cholecystectomy appeared to shorten overall hospitalization, the clinical outcomes were similar between those who underwent early surgery and those who underwent delayed laparoscopic cholecystectomy. ^[38]

If the imaging and laboratory study findings are consistent with severe acute gallstone pancreatitis that is not responding to supportive therapy or with ascending cholangitis with worsening signs and symptoms of obstruction, early endoscopic retrograde cholangiopancreatography (ERCP) with sphincterotomy and stone extraction is indicated.

Pancreatic duct disruption

Damage to the pancreatic ductal system may allow the pancreatic juice to leak from the gland. The sudden development of hypocalcemia or a rapid increase in retroperitoneal fluid on computed tomography (CT) is suggestive of this condition.

When imaging studies provide corroborating evidence, the condition is initially managed by percutaneous placement of a drainage tube into the fluid collection under the guidance of ultrasonography or CT scanning. ^[4] Fluid amylase or lipase levels in the 10,000s strongly suggest the presence of a ductal disruption.

In the appropriate clinical setting, ERCP confirms the diagnosis and provides a treatment option. Transpapillary stent placement or, preferably, placement of a 6 French nasopancreatic tube attached to an external bulb suction device can successfully treat leaks by removing the sphincter tone and changing the dynamics of fluid flow in favor of ductal healing. Occasionally, leaks are associated with downstream stenoses that are also amenable to endoscopic treatment.

Refractory cases may warrant surgery. If a persistent leak is present in the tail of the gland, a distal pancreatectomy is preferred. If the leak is in the head of the gland, a Whipple procedure is the operation of choice.

Pseudocysts

Peripancreatic fluid collections persisting for more than 4 weeks are referred to as acute pseudocysts. Pseudocysts lack an epithelial layer and thus are not considered true cysts. They also differ from true cysts in that they are usually filled with necrotic debris rather than fluid. Accordingly, pseudocysts may be better described by the term organized necrosis.

Most pseudocysts can be followed clinically. However, when they are symptomatic (ie, associated with pain, bleeding, or infection) or are larger than 7 cm and are rapidly expanding in an acutely ill patient, intervention is indicated. Several different therapeutic approaches may be implemented, depending on the anatomic relations and on the duration of the natural history of the complication.

In selected patients with very large fluid collections, percutaneous aspiration of pancreatic pseudocysts is a reasonable approach. Even though treatment failures are common when the pseudocyst communicates with the pancreatic ductal system, percutaneous drainage serves as a temporizing measure that may later lead to successful endoscopic or surgical intervention. Often, an infected pseudocyst (which by definition is regarded as a pancreatic abscess) can be successfully managed by means of percutaneous drainage.

Pseudocysts may also be managed endoscopically with transpapillary or transmural techniques. Transpapillary drainage requires the main pancreatic duct to communicate with the pseudocyst cavity, ideally in the head or body of the gland. The proximal end of the stent (which should be smaller than the diameter of the pancreatic duct) is placed into the cavity. The technical success rate is 83%, the complication rate 12%. Generally, however, pancreatic stents are difficult to monitor, are prone to obstruction, and carry an increased risk of infection and ductal injury.

Some noncommunicating pseudocysts may be amenable to transmural enterocystostomy. Technical success requires a mature cyst that bulges into the foregut, and the distance from the lumen to the cyst cavity should be less than 1 cm. The success rate is 85%, the complication rate 17%. The transduodenal approach is associated with fewer complications and recurrences than the transgastric approach.

On the basis of prospective studies in the 1970s, surgery was recommended for persistent large (> 7 cm) pancreatic pseudocysts because complications developed in 41% of patients, 13% of whom died. Internal pseudocyst-enteric anastomosis became the standard of care, with an operative mortality of 3-5%. This dogma was subsequently challenged by 2 retrospective studies in which patients with smaller (ie, < 5 cm) asymptomatic pseudocysts rarely (< 10%) developed complications.

Infected pancreatic necrosis

The clinician cannot rely on clinical findings alone to differentiate infected and sterile pancreatic necrosis. When clinical signs of infection or SIRS are present in the setting of necrotizing pancreatitis, CT-guided needle aspiration is indicated.

Surgery is recommended when large areas of the pancreas are necrotic and percutaneous CT-guided aspiration demonstrates infection on the basis of a positive Gram stain. Antibiotic therapy alone is not sufficient to achieve a cure. Aggressive surgical debridement and drainage are necessary to remove dead tissue and to clear the infection.

A study of patients with necrotizing pancreatitis and infected necrotic tissue determined that a step-up approach to treatment (consisting of percutaneous drainage followed, if necessary, by minimally invasive retroperitoneal necrosectomy) yielded better results than standard care with open necrosectomy. ^[39] Patients who received step-up treatment had a lower rate of major complications (new-onset multiorgan failure, multiple systemic complications, perforation of a visceral organ, enterocutaneous fistula, or bleeding) and death.

Pancreatic abscess

Pancreatic abscesses generally occur late in the course of pancreatitis. Many of these respond to percutaneous catheter drainage and antibiotics. Those that do not respond require surgical debridement and drainage.

When the cause of pancreatitis can be determined, prevention depends on stopping the etiologic agent from causing subsequent episodes.

In patients with documented gallstone pancreatitis—and probably in those with idiopathic recurrent pancreatitis as well—cholecystectomy is required. In patients who abuse alcohol, a dedicated person (eg, physician, psychologist, addiction counselor) who can help the patient overcome the addiction to alcohol is required. When an uncommon cause of pancreatitis is identified, the path of prevention is specific to the etiology.

The most effective and soundly based treatment plan for any disorder is one aimed at the mechanism responsible for the development of the disorder. With that axiom in mind, consultations must be obtained with an eye to addressing the underlying cause of pancreatitis.

Treatment of patients with alcohol-induced pancreatitis should go beyond the physical manifestations of this disease and address the underlying psychological addiction to alcohol. Simply telling patients they must stop drinking alcohol is not satisfactory. Successful treatment often requires the involvement and expertise of a chemical dependency counselor. The author favors in-hospital consultation for all patients admitted with alcoholic pancreatitis.

Patients with hypertriglyceridemia- or hypercalcemia-induced pancreatitis require consultation with an endocrinologist. Rarely, such patients require surgical intervention for treatment of hyperparathyroidism or control of hyperlipidemia refractory to medical therapy.

Patients with gallstones or microlithiasis revealed on imaging studies should have a surgical consultation for gallbladder removal. Because microlithiasis is the most common cause of idiopathic pancreatitis, a patient with recurrent idiopathic pancreatitis should undergo cholecystectomy before procedures associated with a higher risk of complications (eg, ERCP) are performed.

Patients with medication-induced acute pancreatitis may benefit from clinical pharmacology consultation during their hospitalization to optimize their therapeutic regimen.

Once the patient is stable enough to be discharged from the hospital, routine clinical follow-up care (typically including physical examination and amylase and lipase assays) is needed to monitor for potential complications of pancreatitis, especially pseudocysts.

No evidence-based guidelines have been established for outpatient follow-up care. Generally, a reasonable time to see the patient is within 7-10 days from the date of hospital discharge to evaluate how he or she is doing and to check for signs or symptoms of complications. If the pancreatitis was moderate to severe and was associated with peripancreatic fluid collections, subsequent imaging studies are indicated to determine if a pseudocyst has developed.

Recurrent acute pancreatitis

Recurrent acute pancreatitis (see the image below) can be a challenging clinical problem. First, seek to determine the etiology using modalities that subject the patient to the least risk while simultaneously assessing for treatable causes.

Acute pancreatitis. A normal-appearing ventral pancreas is seen in a patient with recurrent acute pancreatitis. Dorsal pancreas (not pictured) showed evidence of chronic pancreatitis.

View Media Gallery

Abdominal computed tomography (CT) scanning is a reasonable first approach. If neoplasia or chronic pancreatitis is found, it must be addressed and treated accordingly.

If the CT scan is not diagnostic, order a magnetic resonance cholangiopancreatography (MRCP). If MRCP shows developmental abnormalities, strictures, or evidence of chronic pancreatitis, remember that endoscopic or surgical treatment may be of benefit in a subset of patients.

If the MRCP findings are normal, further evaluation with endoscopic ultrasonography (EUS) is indicated. EUS has better sensitivity for detecting biliary sludge and microlithiasis, which are probably the most common causes of recurrent idiopathic pancreatitis. It may also help detect periampullary lesions missed by abdominal CT scanning or MRCP.

If previous studies showed evidence of microlithiasis or biliary sludge, cholecystectomy is indicated. If the patient has already undergone cholecystectomy or if pancreatitis recurs, further evaluation by ERCP is indicated. This may reveal papillary stenosis, in which case a pancreatic and, possibly, biliary sphincterotomy is indicated.

The role of genetic testing is emerging, and whether testing for cationic trypsinogen mutations, SPINK1 mutations, or CFTR mutations should be performed remains unclear, because no effective treatment currently exists for these diseases.

If recurrent pancreatitis continues, ERCP with sphincter of Oddi manometry is indicated. This procedure is placed last in the evaluation because patients with suspected sphincter of Oddi dysfunction (especially that resulting from dyskinesia of the sphincter) have a very high rate of post-ERCP pancreatitis, and the possibility that ERCP may create iatrogenic complications rather than cure the recurrent pancreatitis is a concern.