Acute Pancreatitis Workup

Once a working diagnosis of acute pancreatitis is reached, laboratory tests are obtained to support the clinical impression. In addition to confirming the diagnosis, laboratory tests are helpful in defining an etiology and looking for complications.

Although diagnostic imaging is unnecessary in most cases of pancreatitis, visualization of inflammatory changes within the pancreas provides morphologic confirmation of the diagnosis. Obtain imaging tests when the diagnosis is in doubt, when severe pancreatitis is present, or when a given imaging study might provide specific information needed to answer a clinical question.

Image-guided aspiration may be useful for differentiating infected from sterile necrosis and for draining fluid collections. Genetic testing for mutations associated with acute pancreatitis may be considered, even if effective treatments for these genetic conditions are lacking.

Amylase and lipase

Serum amylase and lipase levels are typically elevated in persons with acute pancreatitis. However, these elevations may only indicate pancreastasis. In research studies, amylase or lipase levels at least 3 times above the reference range are generally considered diagnostic of acute pancreatitis.

Serum amylase determinations are routinely available, but they are not specific for pancreatitis. Preferably, the amylase P level should be measured, which is somewhat more specific to pancreatic pathology. Elevations can occur in anyone with small intestinal obstruction, mesenteric ischemia, tubo-ovarian disease, renal insufficiency, or macroamylasemia. Rarely, elevations may reflect parotitis. The serum half-life of amylase is short, and elevations generally return to reference ranges within a few days.

Lipase has a slightly longer half-life and abnormalities may support the diagnosis if a delay occurs between the pain episode and the time the patient seeks medical attention. Elevated lipase levels are more specific to the pancreas than elevated amylase levels. Lipase levels remain high for 12 days. In patients with chronic pancreatitis (usually caused by alcohol abuse), lipase levels may be elevated in the presence of a normal serum amylase level.

The level of serum amylase or lipase does not indicate whether the disease is mild, moderate, or severe, and monitoring levels serially during the course of hospitalization does not offer insight into prognosis.

Liver-associated enzymes

Determine alkaline phosphatase, total bilirubin, aspartate aminotransferase (AST), and alanine aminotransferase (ALT) levels to search for evidence of gallstone pancreatitis. An ALT level higher than 150 U/L suggests gallstone pancreatitis and a more fulminant disease course.

Serum electrolytes, BUN, creatinine, glucose, cholesterol, and triglycerides

Obtain measurements for blood urea nitrogen (BUN), creatine, and electrolytes; a great disturbance in the electrolyte balance is usually found, secondary to third spacing of fluids. Measure blood glucose level because it may be elevated from B-cell injury in the pancreas.

Measure calcium, cholesterol, and triglyceride levels to search for an etiology of pancreatitis (eg, hypercalcemia or hyperlipidemia) or complications of pancreatitis (eg, hypocalcemia resulting from saponification of fats in the retroperitoneum). However, be aware that baseline serum triglyceride levels can be falsely lowered during an episode of acute pancreatitis.

Complete blood count and hematocrit

A complete blood count (CBC) demonstrates leukocytosis (white blood cell (WBC) count higher than 12,000/µL) with the differential being shifted toward the segmented polymorphonuclear (PMN) cells. Leukocytosis may represent inflammation or infection.

Hemoconcentration at admission (an admission hematocrit value greater than 47%) has been proposed as a sensitive measure of more severe disease. However, this has subsequently been shown to have value only as a negative predictor—that is, a lack of hemoconcentration effectively rules out severe disease.

If blood transfusion is necessary, as in cases of hemorrhagic pancreatitis, obtain type and cross-match.

C-reactive protein

A C-reactive protein (CRP) value can be obtained 24-48 hours after presentation to provide some indication of prognosis. Higher levels have been shown to correlate with a propensity toward organ failure. A CRP value in double figures (ie, ≥ 10 mg/dL) strongly indicates severe pancreatitis. CRP is an acute-phase reactant that is not specific for pancreatitis.

Other tests

Evaluate arterial blood gases if a patient is dyspneic. Whether tachypnea is due to acute respiratory distress syndrome (ARDS) or diaphragmatic irritation must be determined.

Lactic dehydrogenase (LDH), BUN, and bicarbonate levels should be measured both at admission and at 48 hours in order to help determine the Ranson criteria for survival.

Immunoglobulin G4 (IgG4) levels can be checked to evaluate for autoimmune pancreatitis. However, this test is not specific, because IgG4 levels can be elevated in as many as 10% of patients with acute pancreatitis who do not have autoimmune pancreatitis.

Trypsin and its precursor trypsinogen-2 in both the urine and the peritoneal fluid have been evaluated as possible markers for acute pancreatitis (especially post-ERCP pancreatitis) but are not widely used. Trypsinogen activation peptide (TAP) is formed when trypsinogen is cleaved to form trypsin and can be measured commercially in the urine to diagnose acute pancreatitis and to help determine severity.

Although not currently in use clinically, polymorphisms in the chemokine monocyte chemotactic protein 1 (MCP-1) gene may also predict severity. This is the first gene identified that plays a role strictly in predicting the severity of disease.

Abdominal radiographs have a limited role in acute pancreatitis. Kidneys-ureters-bladder (KUB) radiography with the patient in the upright position is primarily performed to detect free air in the abdomen, indicating a perforated viscus, as would be the case in a penetrating, perforated duodenal ulcer. In some cases, the inflammatory process may damage peripancreatic structures, resulting in a colon cut-off sign, a sentinel loop, or an ileus. The presence of calcifications within or around the pancreas may indicate chronic pancreatitis.

Abdominal ultrasonography

Ultrasonography of the abdomen is the most useful initial test in determining the etiology of pancreatitis and is the technique of choice for detecting gallstones. In the setting of acute pancreatitis, sensitivity is reduced to 70-80%. In addition, the ability to identify choledocholithiasis is limited. Although ultrasonography can be used as a screening test, it may not be specific if overlying gas shadows secondary to bowel distention are present. Ultrasonography cannot measure the severity of disease.

Endoscopic ultrasonography

Endoscopic ultrasonography (EUS) is an endoscopic procedure that allows a high-frequency ultrasound transducer to be inserted into the gastrointestinal (GI) tract to visualize the pancreas and the biliary tract. This study allows a more detailed image to be obtained than with transcutaneous ultrasonography because the high-frequency transducer can be introduced directly adjacent to the pancreas.

Its principal role in the evaluation of acute pancreatitis is the detection of microlithiasis and periampullary lesions not easily revealed by other methods. This modality should not be used to help identify chronic pancreatitis until several months after the episode of acute pancreatitis has been completed.

In specialized centers with highly trained medical staff, a secretin-stimulated EUS study may reveal resistance to ductal outflow at the level of the papilla, as evidenced by dilatation of the pancreatic duct to a greater extent and longer duration than in a healthy population. This can be helpful in evaluating patients with recurrent idiopathic pancreatitis.

Abdominal computed tomography (CT) scanning is generally not indicated for patients with mild pancreatitis unless a pancreatic tumor is suspected (usually in elderly patients). It is always indicated in patients with severe acute pancreatitis and is the imaging study of choice for assessing complications. Scans are seldom needed within the first 72 hours after symptom onset unless the diagnosis is uncertain, because inflammatory changes are often not radiographically present until this time.^[9]

Abdominal CT scans also provide prognostic information based on the following grading scale developed by Balthazar and colleagues^{[10, 11, 12]} :

• Grade A - Normal pancreas
• Grade B - Focal or diffuse gland enlargement (see the image below)
• Grade C - Intrinsic gland abnormality recognized by haziness on the scan
• Grade D - Single ill-defined collection or phlegmon
• Grade E - Two or more ill-defined collections or the presence of gas in or nearby the pancreasAbdominal CT scan showing pancreatic enlargement and peripancreatic fat stranding. Gallstones are not visible.

The chances of infection and death are virtually nil in grades A and B but steadily increase in grades C through E. Patients with grade E pancreatitis have a 50% chance of developing an infection and a 15% chance of dying.

Dynamic spiral CT scanning is used to determine the presence and extent of pancreatic necrosis. After the administration of an oral agent to define bowel structures, a study of the upper abdomen is performed twice, before and after administration of an intravenous (IV) bolus of iodine contrast agent.

For a healthy pancreas, density numbers are in the range of 30-40 Hounsfield units on an unenhanced study, increasing to 100-150 Hounsfield units on an enhanced study. When pancreatic necrosis is present, focal or diffuse areas of unenhanced parenchyma on the second study suggest pancreatic necrosis. Pancreatic necrosis for research purposes is defined as loss of enhancement in at least 30% of the pancreatic parenchyma.

Magnetic resonance cholangiopancreatography (MRCP) has an emerging role in the diagnosis of suspected biliary and pancreatic duct obstruction in the setting of pancreatitis. Heavily T2–weighted images provide a noninvasive image of the biliary and pancreatic ducts.^[13]

Although MRCP is not as sensitive as endoscopic retrograde cholangiopancreatography (ERCP), it is safer, noninvasive, and fast, and it provides images useful in guiding clinical care decisions. MRCP should be used if choledocholithiasis is suspected but there is concern that pancreatitis might worsen is ERCP is performed.

ERCP is an endoscopic procedure used to evaluate the biliary and pancreatic ductal system and is indicated in a subset of patients with acute pancreatitis (see the image below). However, ERCP should be used with extreme caution in patients with acute pancreatitis and should never be used as a first-line diagnostic tool in this disease.^[13] It should be performed only in the following situations:

• The patient has severe acute pancreatitis that is believed, and is seen on other radiographic studies, to be secondary to choledocholithiasis
• The patient has biliary pancreatitis and is experiencing worsening jaundice and clinical deterioration despite maximal supportive therapy This patient with acute gallstone pancreatitis underwent endoscopic retrograde cholangiopancreatography. Cholangiogram shows no stones in common bile duct and multiple small stones in gallbladder. Pancreatogram shows narrowing of pancreatic duct in area of genu, resulting from extrinsic compression of ductal system by inflammatory changes in pancreas.

When combined with sphincterotomy and stone extraction, ERCP may reduce the length of hospital stay, the complication rate, and, possibly, mortality.

In the case of biliary pancreatitis where a dilated obstructed common bile duct is diagnosed on the basis of CT or any other imaging modality and an elevated plasma bilirubin level (> 5 mg/dL), ERCP with sphincterotomy is warranted within the first 72 hours. Biliary pancreatitis should always be treated eventually with a cholecystectomy after the process has subsided.

Occasionally, ERCP with sphincter of Oddi manometry reveals sphincter of Oddi dysfunction as the cause of recurrent idiopathic pancreatitis. However, in the setting of suspected sphincter of Oddi dysfunction SOD, the risk that the procedure will induce pancreatitis is at least 30%.

CT-guided needle aspiration is used to differentiate infected necrosis from sterile necrosis in patients with severe necrotizing pancreatitis. The needle is placed into an area of low attenuation in or around the pancreas of a patient with fever and tachycardia or other signs of a systemic inflammatory response syndrome, generally following the first week of severe pancreatitis. The procedure may be repeated weekly if clinically indicated.

The specimen should be delivered to the laboratory within an hour and interpreted promptly. The specimen should always be evaluated for Gram stain, culture, and sensitivity. If the Gram stain shows bacteria or fungi, surgical debridement of the infected necrosis is generally indicated. An exception would be if the patient could not tolerate surgery; in this case, CT-guided catheter drainage may be more effective.

EUS-guided needle aspiration can often be used to differentiate infected necrosis from sterile necrosis in the same manner as CT-guided needle aspiration. EUS can also be used to guide drainage pancreatic and peripancreatic fluid collections that have complicated an episode of acute pancreatitis. These procedures should only be carried out once the pseudocyst has had the opportunity to mature.

A complete description of these procedures is beyond the scope of this article.

With the advent of molecular medicine, various genetic abnormalities have been associated with pancreatitis. Although effective treatments for these conditions have yet to be discovered, it is sometimes reasonable to begin testing for these mutations in patients with otherwise idiopathic pancreatitis, rather than subjecting the entire group to the risks of endoscopic sphincterotomy or stent placement for presumptive sphincter of Oddi dysfunction.

As more is learned about molecular mechanisms and therapy, logical novel treatments can eventually be offered to these patients as part of clinical trials. Of course, as with any type of genetic testing, expert genetic counseling that addresses social, familial, insurance, and financial issues is essential for all persons before testing. In fact, it is the author’s policy not to perform any genetic testing unless patients are first counseled by a qualified genetic counselor.

Hereditary pancreatitis has been associated with a mutation of cationic trypsinogen (PRSS1). At least 4 mutations have thus far been identified. These mutations appear to render the protein resistant to second-line proteolytic defense mechanisms. Patients with PRSS1 mutations typically develop their first episode of pancreatitis by the time they are in their mid teens. Most often, there is a strong family history of pancreatic disease (eg, acute or chronic pancreatitis, pancreatic malignancy). It should be noted that PRSS1 -induced pancreatitis is very rare.

Some patients with idiopathic pancreatitis have atypical mutations in the CFTR gene, which follows an autosomal recessive pattern of inheritance. This is a fascinating advance in understanding the spectrum of cystic fibrosis, in which phenotypic expression depends on the degree to which the mutation affects the activity of the CFTR enzyme’s function. Relatively minor mutations that do not affect pulmonary function may influence chloride transport in the pancreas enough to predispose individuals to recurrent idiopathic pancreatitis.

Mutations in the SPINK1 gene can predispose patients to acute pancreatitis. The SPINK1 protein blocks the active binding site of trypsin, rendering it inactive. However, approximately 1 in 100 persons in the United States are at least heterozygotes for SPINK1. It is an autosomal recessive gene pattern of inheritance. Therefore, although mutations of the SPINK1 gene are not usually enough to cause pancreatitis, they are likely to be a cofactor responsible for pancreatitis in some individuals.

For practical purposes, the infinite spectrum of pancreatitis severity is usually subdivided into mild and severe categories as follows:

• Mild pancreatitis - The gland exhibits interstitial edema and an inflammatory infiltrate without hemorrhage or necrosis, usually with minimal or no organ dysfunction
• Severe pancreatitis - Extensive inflammation and necrosis of the pancreatic parenchyma are present, often associated with severe gland dysfunction and multiorgan system failure

At surgery, peripancreatic fatty tissue is predominantly involved by necrosis, whereas the gland is usually less affected; hence, the extent of pancreatic necrosis is commonly overestimated. In very severe cases, arterial thrombosis may lead to panlobular infarction, in which the gland becomes a hemorrhagic, necrotic, gangrenous mass. The natural history of fat necrosis depends on its location and extent; small areas (< 1 cm) may resolve entirely, whereas large areas (>5 cm) may liquefy within a fibrotic capsule.

Various strategies have been used to predict the severity and outcome of acute pancreatitis, including the Ranson, Acute Physiology and Chronic Health Evaluation (APACHE) II, Glasgow, and Imrie scoring systems. Each has advantages and disadvantages, and none is currently recognized as a criterion standard.

For research purposes, the Atlanta classification of acute pancreatitis has been used to differentiate between severe and mild cases of acute pancreatitis. According to this classification, patients are diagnosed with severe acute pancreatitis if they show the following:

• Evidence of organ failure (eg, systolic blood pressure below 90 mm Hg, arterial partial pressure of oxygen [P_a O₂] 60 mm Hg or lower, serum creatinine level 2 mg/dL or lower, GI bleeding amounting to 500 mL or more in 24 hours)
• Local complications (eg, necrosis, abscess, pseudocyst)
• Ranson score of 3 or higher or APACHE score or 8 or higher

Ranson criteria

Ranson used a series of different criteria for the severity of acute pancreatitis to formulate a scoring system that is still widely used.^[14]

Criteria present on admission include the following:

• Patient older than 55 years
• WBC count higher than 16,000/µL
• Blood glucose level higher than 200 mg/dL
• Serum LDH level higher than 350 IU/L
• AST level higher than 250 IU/L

Criteria developing during the first 48 hours include the following:

• Hematocrit fall more than 10%
• BUN level increased by more than 8 mg/dL
• Serum calcium level lower than 8 mg/dL
• P_a O₂ less than 60 mm Hg
• Base deficit higher than 4 mEq/L
• Estimated fluid sequestration higher than 600 mL

Each of the above criteria counts for 1 point toward the score. A Ranson score of 0-2 has a minimal mortality, and the patient is admitted to the regular ward for medical therapy and support. A Ranson score of 3-5 has a 10-20% mortality rate, and the patient should be admitted to the intensive care unit (ICU). A Ranson score higher than 5 after 48 hours has a mortality of more than 50% and is associated with more systemic complications.

Although the Ranson criteria are perhaps the best known scoring system, they have several drawbacks. First, 11 criteria are used, some of which are evaluated on day 1 and others on day 2. The Ranson score is valid only at 48 hours after onset and not at any other time during the disease. Second, the threshold for an abnormal value depends on whether the pancreatitis is caused by alcohol or gallstones. Finally, the sensitivity is only 73% and the specificity is 77% for predicting mortality.

BUN

An elevated BUN level at admission has been associated with an increase in severe acute pancreatitis and/or death. This corresponds well with the APACHE 2 score. The rise in BUN is thought to be due to hemoconcentration, which is a surrogate marker for intravascular depletion. Intravascular depletion is thought to be a key mediator of the inflammatory reponse in acute pancreatitis.

APACHE II

The APACHE score has the advantage of being able to assess the patient at any point during the illness; however, it is very cumbersome for routine clinical use. Attempts have been made to make this evaluation user friendly (eg, with APACHE II, the Simplified Acute Physiology Score [SAPS], and the Imrie score), but it remains cumbersome. The sensitivity is 77%, and the specificity is 84%.

Biological markers

The most widely available biological marker for staging acute pancreatitis is the hematocrit value. Admission hemoconcentration to a hematocrit value greater than 47% had been reported as a sensitive predictor pancreatic necrosis at admission. However, subsequent study has revealed admission hematocrit to be useful only as a negative predictor for necrosis in patients without hemoconcentration.

CRP, a nonspecific acute-phase reactant produced by the liver in response to interleukin (IL)–6, is a useful marker at 36-48 hours. A CRP level greater than 6 at 24 hours or greater than 7 at 48 hours is consistent with severe acute pancreatitis. The sensitivity of this test is 73%, and the specificity is 71%.

IL-6 levels begin to rise in the first several hours of pancreatitis, stimulating the release of CRP. Early studies of IL-6 as a biological marker appear promising, indicating that this may be a reliable indicator of pancreatitis severity. However, this possibility is not yet validated, and IL-6 is not yet commercially available for clinical use in this setting.

Several other blood tests show promise in predicting the severity of acute pancreatitis. These include TAP, polymorphonuclear elastase, and phospholipase A₂. Like IL-6, they are generally not used in clinical practice and are more expensive than typically used tests. Some are only slightly better than using CRP.

Genetic markers

Polymorphisms in the chemokine monocyte chemotactic protein 1 (MCP-1) gene may play a role in predisposing patients to severe acute pancreatitis, although this marker is still under investigation and is not used clinically.