HPI: D.S. is a 40 year old white male who presented to an outside emergency department on 3/29/96 with a complaint of the sudden onset of sharp epigastric pain radiating to his back that began in the morning after his breakfast. He also had nausea and vomiting but denied any fever, chills, hematemesis, hematochezia or melena. He stated that he had never had pain like this before and denied ever drinking alcohol.

PMH: 1. Hyperlipidemia 2. Hypertension 3. Gout 4. S/P cornea transplant

Meds: 1. Penbutolol 20 mg po qd 2. Amlodipine 5 mg po qd

All: PCN (rash)

Social: Self-employed, no tobacco, no alcohol, unmarried, lives alone

Family history: Unknown

ROS: Recent intentional 30 pound weight loss

P.E. General: Obese white male, A & O x 3, mild distress due to pain

Vital Signs: Temp 99, HR 90, regular, RR 20, unlabored, BP 160/85,

Chest: CTA bilaterally

CV: RRR w/o m/g/r

Abdomen: Soft, epigastric tenderness w/o guarding or rebound tenderness, no masses, bowel sounds present but diminished

Rectal: Guiac negative

Labs: Na 136, K 4.3, Cl 98, HCO₃ 17, BUN 13, Cr 1.0, Glc 386

Ca 8.8, Albumin, LDH and SGOT not measured, Amylase 440, Cholesterol 813, Triglycerides >5,000, Serum ketones negative

WBC 11.1, Hgb 15.4, Plt 203K, 80% N, 10% Bands

PT 13.8, INR 1.2, PTT 27.5

ABG: 7.302/ 27.6/ 77.4/ 13.3 on room air

U/A: pH 5.0, S.G. 1.020, 3+ glucose, 3+ ketones.

Radiology: CXR: No acute cardiopulmonary disease

Abdominal ultrasound: Negative for gallstones

Hospital Course: The patient was admitted to a floor bed at the outside hospital with a presumptive diagnosis of acute pancreatitis. He was made NPO and treated with I.V. fluids, insulin drip, electrolyte replacement, and analgesics. At 48 hours, his labs were as follows:

Na 140, K 3.6, Cl 109, HCO₃ 20, BUN 8.5, Cr 0.8, Glc 269

Ca 5.2, Albumin 2.0, phosphate .5, SGOT 34

WBC 7.1, Hgb 13.8, Plt 196K

ABG: 7.416/ 24.1/ 84.5/ 15.3 on room air

At the request of his family, the patient was transferred to NCBH on 4/2/96 (H.D.5). He was maintained on NS at 200 ml/hr and begun on Ranitidine 50mg IV q8, Ciprofloxacin 400 mg IV q12, sliding scale insulin and meperidine prn. The following day, the antibiotics were discontinued. A CT scan was performed which revealed diffuse pancreatic inflammation but no evidence of necrosis. A fasting lipid profile showed cholesterol 324, TG 507, HDL 16. On H.D. 7, the patient was clinically improving and was begun on a clear liquid diet. His insulin was discontinued and he was begun on an oral hypoglycemic and gemfibrizol 600 mg po bid. He had no further complications and was discharged to home on H.D. 11. His discharge medications were captopril 12.5 mg po tid, gemfibrizol 600 mg po bid and glyburide 5 mg po qd.

The patient returned to clinic 6 weeks from discharge requesting a repeat FLP which showed cholesterol 170, triglycerides 125, HDL 38 and LDL 117.

In 1579, Jacques Aubert published one of the earliest reports of acute pancreatitis and its complications describing:

" a rich merchant, who in his virile age did enjoy all his faculties very briskly, as eating and drinking and the like; but when he endeavored to sleep, he had both a cold sweat seized his whole body, and fell into a swoon, the Physicians who saw him in these fits did conjecture his disease might arise from his stomach being ill affected, and therefore did prescribe him Hiera Galeni, but without any success, he being much worse after the taking the same than before; after this they prescribed Cordials for him, but these had little success in him; and all the remedies which they subscribed no ways lessening his pain, he in this miserable condition put an end to his trouble, by leaving this world: His Body being opened in his pancreas was seen a perfect abscess, accompanied with much putrefied matter, which infected his whole body; and this was the only occasion of his speedy departure."³

The presence of fatty necrosis in acute pancreatitis was first postulated by W. Balser (1882) and the autodigestive process was suspected by H. Chiari (1896). Various forms of the condition were described by Senn (1886) a Chicago surgeon and Fitz (1889) a Boston physician. Senn suggested that surgical treatment would be helpful in patients with pancreatic gangrene or abscess, whereas Fitz suggested that early surgical intervention was hazardous. The connection between cholelithiasis and acute pancreatitis was described by Opie and Halsted (1901).

Definition

A wide spectrum of clinical , pathologic and etiologic entities have been grouped under the term "acute pancreatitis." By definition, an attack of pancreatitis is acute if the patient becomes asymptomatic following recovery. Early attempts to classify acute pancreatitis resulted in pathologic definitions dividing acute pancreatitis into edematous or interstitial pancreatitis and necrotizing or hemorrhagic pancreatitis. However, this is an impractical definition for most patients as laparotomy is contraindicated in acute pancreatitis. In order to establish a clinically based classification system for acute pancreatitis which would be useful for both practitioners and medical researchers, an International Symposium on Acute Pancreatitis was held in Atlanta in 1992. The following definitions were proposed:⁶

Acute pancreatitis is an acute inflammatory process of the pancreas, with variable involvement of other regional tissues or remote organ systems.

Severe acute pancreatitis is associated with organ failure and/or local complications, such as necrosis, abscess, or pseudocyst.

Mild acute pancreatitis is associated with minimal organ dysfunction and an uneventful recovery and it lacks the described features of severe acute pancreatitis (see above).

Acute fluid collections occur early in the course of acute pancreatitis, are located in or near the pancreas, and always lack a wall of granulation or fibrous tissue.

Pancreatic necrosis is a diffuse or focal area of nonviable pancreatic parenchyma, which is typically associated with peripancreatic fat necrosis.

A pseudocyst is a collection of pancreatic juice enclosed by a wall of fibrous or granulation tissue, which arises as a consequence of acute pancreatitis, pancreatic trauma, or chronic pancreatitis.

A pancreatic abscess is a circumscribed intra-abdominal collection of pus, usually in proximity to the pancreas, containing little or no pancreatic necrosis, which arises as a consequence of acute pancreatitis or pancreatic trauma.

Acute pancreatitis is a process of autodigestion caused by the premature activation of potent proteolytic and lipolytic enzymes. As these enzymes are capable of autodigestion of pancreatic tissue, the pancreas usually performs the following protective functions:

          3)Intracellular protease inhibitors (e.g. pancreatic secretory trypsin inhibitor or PTSI) made by the acinar cells are capable of inactivating proteolytic enzymes.

In pancreatitis, there appears to be a loss of any of these normal functions. The trigger that initiates this event is poorly understood.

Once pancreatitis is initiated, it is marked by interstitial edema and inflammatory infiltration of the pancreas. The accumulation of neutrophils in pancreatic and extrapancreatic tissue with the release of leukocyte enzymes as well as other vasoactive peptides leads to further progression of the disease both locally and systemically. Locally, the disease may progress to coagulation necrosis (a.k.a. necrotizing pancreatitis) of the glandular elements of the pancreas and surrounding tissues The systemic complications include hypoxemia, adult respiratory distress syndrome (ARDS), acute renal failure, disseminated intravascular coagulation (DIC), gastrointestinal bleeding and circulatory collapse.¹⁰

Etiology

Etiologic factors in acute pancreatitis vary widely in different parts of the world and in differing populations of patients.¹⁴

1) Alcoholic pancreatitis (40 - 50%) develops in patients after heavy ethanol for years. Studies show that oral administration of ethanol induces a transient stimulation of pancreatic secretion and the sphincter of Oddi is simultaneously contracted. Chronic alcoholism leads to an increase in protein secretion with the acinar cells producing proteinaceous plugs in the small pancreatic ducts, causing atrophy of the acini drained by the obstructed duct.¹¹ Studies show that these chronic, irreversible pathologic changes predate the first attack of acute pancreatitis.²⁰ The factors triggering an acute attack superimposed on this chronic process are not understood. Ten per cent of alcoholics develop pancreatic insufficiency without a recognized acute attack.

2) Gallstone pancreatitis (30%) is nearly always associated with fecal excretion of gallstones, whereas a much lower frequency of fecal stones is found in patients with gallstones who do not have pancreatitis.²¹ Apparently, passage of a gallstone through the ampulla of Vater somehow creates conditions favorable to the development of pancreatitis. This condition is not obstruction since ligation of the ampulla does not cause pancreatitis. Instead, it has been proposed that biliary reflux is necessary to produce pancreatitis. It should be noted that gallstone (or "large duct" ) pancreatitis differs from alcoholic (or "small duct") pancreatitis in that gallstone pancreatitis does not lead to chronic pancreatitis and is cured by cholecystectomy.

3) Postoperative pancreatitis (6-32%) may follow any type of abdominal surgery. In particular, biliary tract procedures (e.g. ERCP) and retroperitoneal node dissections are associated with this complication. Prompt recognition is difficult owing to the underlying surgery resulting in abdominal tenderness and elevated amylase. As a result, this type often carries a high mortality.

4) Hypertriglyceridemia (<10%) may occur with types I, IV, and V hyperlipidema. Clinically significant acute pancreatitis does not occur unless the serum triglyceride level reaches 1000-2000 mg/dl.²² Havel (1968) proposed that free fatty acids produced by pancreatic lipase are increased to toxic concentrations. Following therapy for hypertriglyceridemia, there is a decrease in the frequency of attacks of pancreatitis.

5) Drugs are associated with the development of pancreatitis.²¹ It is often difficult to determine whether the pancreatitis is a result of the drug itself or secondary to the condition for which it was prescribed. However, some drugs (esp. anti-metabolites and sulfonamides) have a strong association independent of clinical status, e.g. L-asparaginase is associated with a 10% frequency of pancreatitis.

6) Vascular insufficiency of the pancreas is generally underreported as many patients are asymptomatic secondary to their underlying catastrophic disease. In a post mortem series of patients who died with hemorrhagic shock, 10% of patients were found to have pancreatitis.

7) Pancreas divisum is a common (5%) anatomic variant resulting from the failure of the dorsal and ventral pancreatic ducts to fuse during development. The portion of the pancreas (tail, body, and part of the head) derived from the embryologic dorsal pancreas is drained through the duct of Santorini and the minor papilla rather than through the duct of Wirsung and the ampulla of Vater. This variant occurs in 20 to 25% of patients with otherwise unexplained pancreatitis. This variant drainage is presumed to be inadequate; however, sphincterotomy of the minor papilla provides no benefit in the prevention of pancreatitis.²¹

8) Pancreatitis may be inherited as an autosomal dominant trait. This is rare.

9) About 15% of pancreatitis is idiopathic.

Diagnosis

History and Physical

The initial clinical features of acute pancreatitis are varied and closely mimic a number of other conditions including mesenteric infarction, acute biliary disease and myocardial infarction. Abdominal pain is nearly universally present. The pain typically begins after a heavy meal or during a drinking binge. The time from onset to peak intensity of the pain ranges from a few minutes to an hour. The pain tends to be constant rather than colicky and varies in intensity from minor to severe. At the time of onset of the attack, the pain is often localized the epigastric area and left upper quadrant but later becomes diffuse and radiates to the back. The back pain results from retroperitoneal irritation and is partially relieved by flexing the trunk. Nausea and vomiting occur in 70 to 90% of patients, but this does not relieve the discomfort. Finally, a previous history of biliary disease or alcohol use is often present, and some patients complain of vague dyspepsia or flatulence during the weeks preceding the episode.

Nonspecific physical findings include fever, tachycardia and hypotension. Although rarely palpable on initial examination, an abdominal mass will develop in 10 to 20% of patients. In hemorrhagic pancreatitis, retroperitoneal blood may dissect into the flanks or around the umbilicus , producing Grey Turner's or Cullen's sign, respectively.

Laboratory

An elevated serum amylase is the standard laboratory marker for acute pancreatitis. Amylase is produced in large quantities by the pancreas, salivary glands and certain malignant tumors, and in lesser quantities by the lungs and fallopian tubes. Most of the amylase secreted by the pancreas and salivary glands enters and is confined to the gut. However, a small fraction of amylase normally enters the plasma; two-thirds of which is salivary isoamylase and one third of which is pancreatic isoamylase. Amylase is cleared from the plasma with a half life of about two hours, with about 20 % excreted into the urine. The rate of excretion of amylase in the urine is increased out of proportion to the serum amylase level in pancreatitis.²

Conditions other than pancreatitis may cause an elevated serum amylase level. Salivary gland inflammation or trauma causes excessive release of salivary amylase into the serum. Chronic idiopathic elevations of serum amylase activity may occur and are almost always due to increases in salivary amylase, although there may be no evidence of salivary gland disease. In mesenteric infarction and perforated peptic ulcer, amylase from the intestinal lumen may leak into the circulation; these conditions should routinely initially be ruled out.

While amylase is derived from a number of organs, serum lipase, trypsin or pancreatic isoamylase are all derived almost entirely from the pancreas. These enzymes are technically more difficult to measure than amylase, thus limiting their widespread usage.

Imaging

Computed Tomography demonstrates a diffusely enlarged pancreas in greater than 90% of patients during an acute attack of pancreatitis, whereas ultrasound is less sensitive due to overlying bowel gas.⁴ CT is reserved only in cases where diagnosis is questionable or to evaluate for possible complications.

Risk Stratification

Pancreatitis occurs in 0.5% of the population and accounts for about 1 death annually per 100,000. The overall mortality of acute pancreatitis is about 10%, although the development of complications increases mortality to greater than 50%.¹⁰ About 25% of patients develop complications. These complications may be local (e.g. pancreatic necrosis, infection, phlegmon, fluid collection or pseudocyst) or systemic. Death within the first week is usually due to a systemic complication, such as hypovolemic shock, renal failure or ARDS, whereas death after the first week is caused by local complications, i.e. pancreatic necrosis complicated by infection.

Because clinical management of acute pancreatitis depends on its severity, a number of investigators have attempted to determine which patients are at greatest risk for developing severe acute pancreatitis. Systems to estimate severity include: 

1. Applications of different clinical scoring systems (e.g. Ranson's criteria)

2. Estimates of comorbid illness

3. Systems using pancreatic appearance on CT

4. Measures of systemic inflammatory markers or systemic release of pancreatic enzymes (e.g. C -reactive protein )

One of the early systems for determining severity of acute pancreatitis was developed by Ranson(Table 1)¹⁸. This system measures five features at admission and six additional criteria at forty-eight hours. These initial criteria were based primarily in patients with alcoholic pancreatitis so the criteria were revised to include two separate systems, one for alcoholic pancreatitis and one for gallstone pancreatitis (Table 2).¹⁸ When there are 0-2 positive criteria, the process is benign whereas if there are greater than six criteria, the prognosis is severe. However, when there are 3-5 of Ranson's criteria, there is still a 15% mortality, but the system does not predict which of these will have a fatal outcome. The modified Glasgow criteria (Table 3) is another system which was based primarily in patients with gallstone pancreatitis.⁵ This system analyzes eight criteria within 48 hours of admission. Finally, the APACHE-II (Acute Physiology and Chronic Health Evaluation) system is more complicated but has the advantage of being applicable at any point in the patient's course.¹³

All of these grading systems suffer from not allowing one to select which among the group is likely to suffer a complication or die. In addition, each system has a significant false positive rate with 25 to 50% of patients predicted to have severe acute pancreatitis with resultant complications and death, actually doing well.

The system using estimates of comorbid illness was developed by Rabeneck et al. (Table 4).¹⁶ Initially, this system appeared promising because it was easy to use, but like the prior clinical staging systems, it, too has a significant false positive rate. 

Table 1. Ranson’s Criteria

At admission or diagnosis

Age > 55 years

White blood cell count > 16,000/mm3

Blood glucose > 200mg/dl

Serum LDH > 350 IU/ml

SGOT > 250 Sigma-Frankel units/dl

During initial 48 hours

Hematocrit fall > 10%

BUN rise > 5 mg/dl

Serum calcium level < 8.0

Arterial oxygen pressure < 60 mm Hg

Base deficit > 4 mEq/L

Estimated fluid sequestration > 6,000 ml

 

Table 2. Modified Ranson’s Criteria

Alcoholic pancreatitis	Gallstone pancreatitis
On admission to hospital	On admission to hospital
Age > 55 years	Age > 70 years
WBC > 16,000	WBC > 18,000
Glucose > 200	Glucose > 220
LDH >350	LDH > 400
SGOT > 250	SGOT > 250
Within 48 hours of admission	Within 48 hours of admission
Decrease in Hct > 10	Decrease in Hct > 10
Increase in BUN > 5	Increase in BUN > 2
Serum calcium < 8	Serum calcium < 8
Base deficit > 4	Base deficit > 5

Estimated fluid deficit > 6L Estimated fluid deficit > 4L

PaO₂ < 60 PaO₂ < 60

 

Table 3. Modified Glasgow Criteria.

 

Within 48 hours of admission

Age > 55 years

White blood cell count > 15,000

Glucose > 180

BUN > 45

LDH > 600

Albumin < 3.3

Serum calcium < 8.0

PaO₂ < 60

Table 4. Rabeneck Clinical Prognostic Staging System

	Clinical Severity of Comorbidity
Acute Pancreatitis	Nonsevere	Severe
No ileus or Peritonitis	I	II
Ileus or peritonitis	II	III

The appearance of the pancreas on CT scanning has been used as a marker of severe acute pancreatitis.⁴ With high resolution CT scanning, pancreatic necrosis is identified based on the scan's ability to visualize pancreatic vascular enhancement. Lack of enhancement is evidence of pancreatic necrosis. Both pancreatic infection and death are significantly more common in patients with lack of pancreatic enhancement on CT scanning.

A number of laboratory markers (Table5) have been evaluated as predictors of prognosis.^2,10 In particular, C-reactive protein, leukocyte elastase and trypsinogen activator peptide have shown the greatest promise.  

            Table 5. Laboratory Tests as Predictors of Prognosis

Phospholipase A2	C-reactive protein
Lactate dehydrogenase	Pancreatitis-associated protein
Tumor necrosis factor	Serum endotoxin level
Alpha 2-macroglobulin	Alpha 1-antitrypsin
PMN elastase	Trypsinogen activation peptide

Management

The management of acute pancreatitis is directed at three therapeutic objectives:¹⁸

1. Limiting pancreatic inflammation

2. Interrupting pathogenesis

3. Supportive therapy and treatment of complications as they arise

The following objectives are discussed in detail below:

Limiting inflammation

1. Inhibition of pancreatic secretion

Nasogastric (NG) suction is usually initiated at admission to the hospital to reduce nausea, vomiting and abdominal distention. In addition, it was reported that aspiration of gastric contents would reduce pancreatic secretion by reducing secretin release. A meta analysis of nine prospective controlled trials (408 patients) of nasogastric suctioning compared to withholding of feeds showed no therapeutic advantage for NG suction ¹⁸, i.e. no reduction in morbidity or mortality. However, NG suction is recommended for patient comfort.

H2 blockers are routinely used in most institutions. Most clinical and experimental trials use cimetidine. No clinical benefit has been demonstrated and in fact, a few studies report increased morbidity.

Anticholingeric drugs (e.g. atropine) are sometimes used in an effort to reduce vagally mediated pancreatic secretion. These drugs generally can only be used in patients with mild pancreatitis as those that have severe pancreatitis frequently are tachycardic which contraindicates their use. A controlled clinical study of patients with mild alcoholic pancreatitis demonstrated no benefit in the atropine-treated group^.7

2. Inhibition of pancreatic enzymes

A number of agents, especially protease inhibitors (e.g. gabexate mesilate) have been investigated. These studies remain equivocal.¹⁵

3. Anti-inflammatory agents

Steroids and non-steroidal anti-inflammatory agents have been administered to patients. Although animal studies indicate that steroids may be beneficial, these studies have not been duplicated in humans. Indomethacin suppositories have been shown to significantly reduce pain.

4. Operative biliary procedures

Operative biliary procedures have been investigated in patients with gallstone pancreatitis. Acosta et al. (1978) studied 132 patients; 86 patients were treated non-operatively and 46 patients were treated with early biliary surgery. Of the patients treated with the non-surgical approach, 14 (16%) died.¹ Of the patients treated with early biliary surgery, a gallstone was found impacted in the ampulla of Vater in 33(72%) patients and the overall mortality was 2%. This study has never been duplicated, probably because jaundice was one of the criteria in this study for gallstone pancreatitis and it has since become noted that many (as many as 79% in some series) with abdominal pain, gallstones and elevated serum amylase levels do not have significant pancreatic inflammation at surgery.

Kelly and Wagner (1988) studied 165 patients with gallstone pancreatitis (Table 6).¹² Patients were assigned to either early biliary surgery (<48 hours) or delayed surgery (>48 hours). There was a significant reduction in morbidity and mortality if surgery was postponed until evidence of pancreatitis had abated. The general consensus is now that all patients with gallstone pancreatitis should be managed non-operatively at first, unless there is evidence of gangrenous cholecystitis or obstructive cholangitis.

 

Table 6. Results of a Controlled Clinical Evaluation of Early Biliary Surgery in Patients with Gallstone Pancreatitis.

	Early Operation ( < 48 hr)	Delayed Operation (>48 hr)
All patients	n=83	n=82
Impacted ampullary stone (%)	26	5
Pancreatic hemorrhage or necrosis(%)	14	14
Morbidity (%)	30.1	5.1
Mortality (%)	15.1	2.4
< 3 Signs	n=60	n=65
Morbidity (%)	4	2
Mortality (%)	2	0
> 3 Signs	n=23	n=17
Morbidity (%)	82.6	17.6

Mortality (%) 47.8 11.8

5. Endoscopic biliary procedures

A large randomized clinical trial by Fan et al. (1988) found no benefit of ERCP or, when a stone was identified, ERCP with sphincterotomy and stone extraction in cases of mild pancreatitis.⁸ However, in patients with severe pancreatitis, a reduction in morbidity was reported for early ERCP with or without sphincterotomy.

Interruption of pathogenesis

1. Antibiotics

With necrotizing pancreatitis, the most frequently recovered organisms are E. coli (51%), Enterococcus (19%), Staphylococcus (18%), Proteus (10%), Klebsiella (10%), Pseudomonas (10%), Streptococcus faecalis (7%) and Bacteroides (6%). ²¹ A number of controlled clinical trials have evaluated ampicillin, primarily in alcoholic pancreatitis.¹⁸ There is no therapeutic benefit from this medication in this population. Because of the high frequency of positive biliary cultures in patients with gallstone pancreatitis, it would be logical to treat these patients with antibiotics, although there are no studies to prove this point. One study evaluated imipenem in 74 patients with necrotizing pancreatitis documented by CT.¹⁶ Pancreatic sepsis was significantly reduced ( 12.2% compared with 30.3%) although the overall mortality was not affected.

2. Peritoneal lavage

Most studies of peritoneal lavage involve introducing fluid (usually 2 liters of isotonic saline with dextrose, potassium, heparin and ampicillin) through a percutaneous catheter in the abdominal wall. The fluid is introduced over fifteen minutes, left in the abdominal cavity for thirty minutes and drained out by gravity for the remaining fifteen minutes. This is repeated each hour over a 48 hour interval. In early studies, peritoneal lavage was equivocal. It did not appear to affect overall mortality, but it did influence the timing and cause of death. In patients treated with two days of lavage early in their course, almost all deaths were due to late pancreatic sepsis whereas in patients not treated with lavage, about half of the deaths were due to cardiovascular or respiratory failure. This anecdotal observation led Ranson's group to study long (7 day) lavage. Early studies of small numbers of patients show a decrease in mortality (Table 7).¹⁹ This matter needs further investigation.

 

Table 7. Incidence of Pancreatic Infection and Death after Lavage.

Number of Prognostic Criteria

	>3	3 - 4	>5
Long Lavage (7day)
Patients (n)	14	4	10
Deaths	2(14%)	0	2(20%)
Abscess	3(21%)	0	3(30%)
Abscess death	0	0	0
Short Lavage (2day)
Patients (n)	15	8	7
Deaths	3(20%)	0	3(43%)
Abscess	6(40%)	2(25%)	4(57%)
Abscess death	3(20%)	0	3(43%)

Supportive therapy and treatment of complications

1. Fluid resuscitation

As large volumes of fluid are often administered due to third-spacing, hemodynamic monitoring with a pulmonary artery catheter is often required. In cases of severe acute pancreatitis, a high cardiac output and low systemic vascular resistance is frequently observed. It is important to maintain tissue perfusion (i.e. avoid vasoconstriction) as ischemia may convert mild pancreatitis to more severe pancreatitis.

2. Respiratory support

Respiratory failure usually occurs early in the course of the disease. It is recommended that all patients with acute pancreatitis (regardless of clinical setting) have arterial blood gases measured at admission and 12 hour intervals for the first 48 hours.

3. Electrolyte replacement

Electrolyte abnormalities are usually encountered. Although hypocalcemia is usually present, the ionized calcium is often normal. Replacement of calcium should be judicious as hypercalcemia has been implicated in the pathogenesis of pancreatitis.

4. Treatment of pancreatic necrosis

Pancreatic infection, when it occurs, usually is recognized after two weeks of treatment An elevated white blood cell count and fever are often present. CT scan may show the extent of necrosis. Because the infected material is semisolid, percutaneous catheter drainage is unsuccessful and a surgeon must be consulted for laparotomy.⁹

5. Treatment of pancreatic pseudocysts

Pseudocysts occur in 10% to 15% of patients. Early studies indicated that up to 40% of patients with pseudocysts would develop complications and only 20% would resolve, usually within the first 6 weeks of observation. It now appears that pseudocysts less that 6 cm in diameter can be followed with observation and resolution may occur over months to years. Pseudocysts larger than 6 cm or those that develop complications can be drained via a catheter (unlike pancreatic necrosis) or managed surgically.⁹

Summary

In conclusion, the management of acute pancreatitis first involves establishing the correct diagnosis, searching for an etiology which may be eliminated thus preventing recurrent episodes, and stratifying the patient’s risk for significant complications. Medical managment is primarily supportive. Efforts at limiting inflammation have not been uniformly successful. Early studies of the empiric use of imipenem and long peritoneal lavage appear promising, but larger studies are needed to establish their therapeutic role.