Gastroesophageal Reflux Disease

Gastroesophageal reflux disease (GERD) affects 7% of the U.S. population each day. Another 29% of this population has symptoms of heartburn weekly to monthly. Americans spend $2 billion per year for over-the-counter H₂-blockers and antacids and another $6 billion per year on prescription H₂-blockers and proton-pump inhibitors. Yet only a small proportion of persons with GERD seek help from physicians. Endoscopic evidence of esophagitis is found in about 2% of patients with symptoms of GERD but in approximately 10% to 15% of the subset of patients with these symptoms who are referred to gastroenterologists.

Proton-Pump Inhibitors Sped Healing Rate

Chiba N, De Gara CJ, Wilkinson JM, et al. Speed of healing and symptom relief in grade II to IV gastroesophageal reflux disease: a meta-analysis. Gastroenterology. 1997; 112:1798-810.

In assessing GERD, it has been shown that about 25% of patients with endoscopic evidence of esophagitis heal with placebo. Therapy with H₂-blockers increases that proportion to about 50%, and proton-pump inhibitors increase it to about 80%. These percentages are often incorrectly called the "healing rate." For patients, the critical product of therapy is the speed of healing and symptom relief, which reflects a true rate: healing over a given period of time. How rapidly various agents relieve symptoms and heal lesions on endoscopy has not been known, especially in patients with higher grades of esophagitis, which are more resistant to therapy.

This meta-analysis attempted to compare different drug classes with respect to the speed of healing and symptom relief. The authors found 43 studies that included a total of 7635 patients with endoscopically proven erosive esophagitis. For each drug class, they performed linear regression analysis to estimate the average percentage of patients who were healed and free of heartburn per week. The results are summarized in Table 1.

These data provide further evidence that patients with endoscopically proven esophagitis are more likely to remain free of heartburn and to show esophageal healing if they keep receiving proton-pump inhibitors rather than H₂-receptor antagonists and promotility drugs, although the latter have some degree of efficacy.

Transition of Barrett Esophagus Occurred Slowly

Drewitz DJ, Sampliner RE, Garewal HS. The incidence of adenocarcinoma in Barrett's esophagus: a prospective study of 170 patients followed 4.8 years. Am J Gastroenterol. 1997; 92:212-5.

Barrett esophagus, a complication of GERD, is intestinal metaplasia of the esophagus. It is important clinically because it may be a precursor to adenocarcinoma of the esophagus, a disease with a 5-year survival rate of less than 10%. The overall risk for adenocarcinoma is about 2% to 5% in patients who have developed metaplasia. However, if dysplastic changes are present, this risk is 25% to 50%. There is no correlation between the severity of GERD symptoms and the likelihood of developing Barrett esophagus; duration of GERD is more important.

How often should surveillance endoscopy be performed in patients known to have metaplastic changes in the esophagus? In this 11-year cohort study, patients at a Veterans Affairs medical center who were undergoing endoscopy were surveyed for Barrett esophagus. Endoscopists identified a total of 177 patients with Barrett esophagus, 7 (4%) of whom were found to have adenocarcinoma initially or within 6 months. The remainder were followed for a mean of 4.8 years. Adenocarcinoma developed in 4 of these patients, for an annual incidence of about 0.5%.

An editorial accompanying this article [1] concluded that annual endoscopy for patients with metaplastic changes is probably too frequent and recommended endoscopy at 2- to 5-year intervals. However, if dysplasia is present on initial biopsy specimens, more frequent endoscopy is prudent.

These studies and others allow us to address two common clinical questions with considerable confidence. The first question is, Which drugs should be used for GERD symptoms? For mild, episodic GERD, H₂-blockers are probably superior because proton-pump inhibitors induce parietal cell proliferation; therefore, when proton-pump inhibitor therapy is stopped, rebound hyperacidity occurs. For persons with more severe disease, proton-pump inhibitors used continuously are superior. So far, the theoretical risk for an association between proton-pump inhibitors and carcinoid tumors and atrophic gastris has not been borne out in practice. We know that these inhibitors are safe for at least 5 years, but it is prudent to keep the dose to a minimum, and many patients do well with every-other-day, or even every-third-day, therapy.

The second question is, When should endoscopy be undertaken? When patients are first seen, endoscopy should be done in those with danger signals: anorexia, nausea, vomiting, weight loss, dysphagia, anemia, blood in the stool, family history of peptic ulcer disease, or abnormal results on physical examination. Patients older than 45 years of age who have had symptoms for many years should also undergo endoscopy. Most other patients can simply receive medical therapy.

Peptic Ulcer Disease

Helicobacter pylori has been firmly established as a major cause of peptic ulcer disease, but clinicians still face numerous questions. One is whether eradication of H. pylori is permanent or whether infection can recur early. Another is, Which of the numerous options is the best treatment regimen in the United States?

Helicobacter pylori Infection Rarely Recurred

Van der Hulst RW, Rauws EA, Koycu B, et al. Prevention of ulcer recurrence after eradication of Helicobacter pylori: a prospective long-term follow-up study. Gastroenterology. 1997; 113:1082-6.

We now have strong evidence on how to successfully treat patients whose ulcers are associated with H. pylori infection. The next clinical question is whether eradication of H. pylori is permanent. A few short-term studies of H. pylori-related duodenal ulcers have shown relapse rates of 3% to 22%.

The goal of this prospective study was to determine the long-term outcome of ulcer disease after successful eradication of H. pylori. Van der Hulst and colleagues identified 186 patients with endoscopically proven H. pylori-associated gastric or duodenal ulcers who did not use aspirin or other non-steroidal anti-inflammatory drugs (NSAIDs). All underwent endoscopy every 3 months during the first year. By a median of 2.6 years, 90 patients (half with gastric ulcers and half with duodenal ulcers) continued to have endoscopy every 6 months. Some patients were followed for almost 10 years. In the group with duodenal ulcers, the relapse rate was 0% after 367 patient-years. Similarly, for the gastric ulcer group, the relapse rate was 0% after 113 patient-years.

These findings, combined with those of earlier studies, show that if H. pylori is eradicated and if patients avoid NSAIDs, the rate of relapse of ulcer disease is less than 2% per year, perhaps markedly less. However, many patients take NSAIDs surreptitiously, and clinicians may have to use intensive detective techniques, such as measurement of platelet cyclooxygenase activities, to assess compliance with advice in some patients with recurring ulcers.

Dual Regimen Was Ineffective in U.S. Trial

Laine L, Frantz JE, Baker A, et al. A United States multicentre trial of dual and proton pump inhibitor-based triple therapies for Helicobacter pylori. Aliment Pharmacol Ther. 1997; 11:913-7.

Each year, new drug regimens to eradicate H. pylori are reported. In the United States, 1-week therapy regimens composed of two antibiotics and a proton-pump inhibitor are popular. European studies have reported eradication rates of more than 90% with such therapy, but data on the use of these regimens in the United States have been scarce.

As a result, Laine and colleagues set up a clinical trial to compare the effectiveness of 1-week triple therapy with that of 1- and 2-week two-drug regimens. They randomly assigned 302 healthy persons with proven H. pylori infection to receive one of five regimens and assessed their infection status at 6 weeks. Results are summarized in Table 2.

Several conclusions can be drawn from the results. First, dual therapy seems to be inadequate for the eradication of H. pylori. Second, regimens that include metronidazole seem to be less effective, presumably because metronidazole resistance is more prevalent in the United States. Third, even the best triple-therapy approach may not be adequate if given for only 7 days. Indeed, several guidelines suggest that treatment should be as long as 10 days and should include two antibiotics and an antisecretory drug.

Helicobacter pylori Eradication Decreased Complications from NSAIDs

Chan FK, Sung JJ, Chung SC, et al. Randomised trial of eradication of Helicobacter pylori before non-steroidal anti-inflammatory drug therapy to prevent peptic ulcers. Lancet. 1997; 350:975-9.

Patients who take NSAIDs have a fourfold increased risk for peptic ulcers. Ulcers most often develop in elderly persons and early in the course of NSAID therapy. The cost of this adverse effect is estimated to be almost $4 billion a year. Studies have shown that prophylaxis with misoprostol, famotidine, and omeprazole is effective in preventing ulcers, but it is expensive. About 50% of NSAID users who develop ulcer disease have H. pylori.

In this trial, Chan and colleagues investigated whether eradication of H. pylori before the initiation of NSAID therapy could reduce the occurrence of ulcers in patients without previous exposure to NSAIDs. They randomly assigned 100 otherwise healthy patients who were taking NSAIDs for musculoskeletal pain to receive naproxen alone (750 mg/d) or a 1-week course of triple therapy with bismuth (120 mg four times daily), tetracycline (500 mg four times daily), and metronidazole (400 mg four times daily) before beginning naproxen therapy. Endoscopy was done at study entry and at 8 weeks.

Helicobacter pylori was eradicated in 89% of patients receiving triple therapy and in none of those receiving naproxen alone. Endoscopic evidence of ulcers was seen in 7% of patients in the triple-therapy group, and symptoms were reported in 3%. In the group receiving naproxen alone, the figures were 26% and 14%, respectively (relative risk reduction for symptomatic ulcers, 79%; number needed to treat [NNT] to prevent symptoms in one patient, 9).

This study supports the hypothesis that prophylactic eradication of H. pylori can reduce the rate of NSAID-associated ulcers, but other studies have not demonstrated this outcome. To add to the confusion, recent reports [2, 3] suggest that H. pylori infection may actually protect NSAID users from developing ulcers. Studies are now in progress to determine whether the increased incidence of esophageal adenocarcinoma is associated with the treatment-induced reduction in the prevalence of certain strains of H. pylori in the population.

One clinical question remaining is whether the use of low-dose aspirin is associated with mucosal injury. Even dosages of 30 mg/d can cause erosions early in the course of use, but these erosions tend to dissipate over time. If a patient has no symptoms, low-dose aspirin is fairly safe and is excellent protection against vascular disease and, possibly, against colonic neoplasms.

Ulcers Bled Less with Omeprazole

Khuroo MS, Yattoo GN, Javid G, et al. A comparison of omeprazole and placebo for bleeding peptic ulcer. N Engl J Med. 1997; 336:1054-8.

Patients with bleeding ulcers are often treated with endoscopic techniques. A medical alternative is attractive, and evidence shows that a more neutral pH improves the function of platelets, which enhance clotting. However, trials of H₂-blockers in patients with bleeding ulcers have not shown benefit.

In this clinical trial, a group of researchers from India sought to determine whether the proton-pump inhibitor omeprazole could benefit patients with bleeding peptic ulcers. They randomly assigned 220 patients with gastric, duodenal, or stomal ulcers to receive either placebo or omeprazole, 40 mg orally every 12 hours for 5 days. After 5 days, they assessed the rate of further bleeding, surgery, or death. Results are shown in Table 3.

These results are striking. However, previous studies using H₂-blockers in doses adequate to neutralize gastric acid failed to influence transfusion requirements and death in patients with bleeding peptic ulcers. These discordant results mean that we need more and better data.

Appendicitis

Each year, about 250 000 patients have surgery for acute appendicitis. The incidence of this condition has decreased about 50% over the past 60 years. In at least 15% of operations, however, the appendix is normal. More worrisome, perforation occurs in more than 20% of procedures [4].

Appendiceal Computed Tomography Helped with Diagnosis

Rao PM, Rhea JT, Novelline RA, et al. Effect of computed tomography of the appendix on treatment of patients and use of hospital resources. N Engl J Med. 1998; 338:141-6.

Computed tomography (CT) has been shown to be about 93% to 98% accurate in confirming or excluding appendicitis.

The aim of this trial was to prospectively determine the effect of appendiceal CT on the management of patients in whom appendicitis was suspected. One hundred such patients had focused appendiceal CT. The outcomes were determined by surgery and pathologic examination in 59 patients and by clinical follow-up in 41. The costs of surgery, hospitalization, and testing were also identified.

Ninety-six patients had right lower-quadrant pain and tenderness, 66 had leukocytosis (>10 000 leukocytes/mm³), 63 had fever (body temperature > 37.5 °C), 62 had nausea, and only 26 had rebound tenderness. Fifty-three patients had appendicitis. All had follow-up until the problems were resolved. Computed tomography proved to be 98% sensitive and 98% specific. In this study, CT findings led to a change in treatment in 59 patients and to savings of $47 000 from appendectomies not performed in 13 patients and $20 000 from a decrease of 50 days of hospitalization. After the $22 000 cost of appendiceal CT was deducted, $447 was saved per patient evaluated.

The authors conclude that the routine use of appendiceal CT for patients in whom acute appendicitis is suspected improves patient care by avoiding unnecessary appendectomies and averting delays in needed therapy. However, the clinical suspicions of these investigators may have been lower than those of many physicians because only 53 patients had acute appendicitis. Moreover, the cost of their focused CT scan, about $220, is less than the cost at most institutions.

Diarrheal Diseases

Diarrhea is a symptom with scores of causes, including Crohn disease. In 1997, it was found that symptoms do not always adequately estimate the recurrence rate of this disease. In another study, Crohn disease was successfully treated with antibody against tumor necrosis factor. A fourth bacterium was added to the list of common causes of bacterial diarrhea.

Recurrence of Crohn Disease Was Greater Than Previously Thought

McLeod RS, Wolff BG, Steinhart AH, et al. Risk and significance of endoscopic/radiological evidence of recurrent Crohn's disease. Gastroenterology. 1997; 113:1823-7.

About 500 000 persons in the United States are known to have Crohn disease, and many more persons with diarrheal or abdominal pain disorders doubtless have this disease but have not yet received a diagnosis.

One of the worst aspects of Crohn disease is its propensity to recur. McLeod and colleagues designed a prospective study to determine the risk for recurrence of Crohn disease, as shown on endoscopy or radiography, in asymptomatic patients who had undergone resection surgery. They also measured outcomes in patients who developed symptomatic recurrent disease.

They recruited 76 patients: Fifty had a terminal ileo-ileocolic resection, 19 had a colonic procedure alone, and 7 had small-bowel resection. The overall endoscopic or radiologic recurrence rate was 27% at 1 year, 60% at 2 years, and 77.3% at 3 years. These figures are higher than earlier estimates of a 50% rate of recurrence at 10 years. However, the symptomatic recurrence rate was 18% at 1 year, 38% at 2 years, and 45% at 3 years. Not surprisingly, patients with more severe findings on endoscopy or radiology tended to develop symptoms.

In summary, the overall recurrence rate in Crohn disease is considerable-about 75% at 3 years. However, many patients with endoscopic evidence of recurrence are either asymptomatic or minimally symptomatic. The important, unanswered clinical question is whether maintenance therapy will have a greater effect on symptoms than it has on endoscopic evidence of recurrent disease. Several clinical trials looking at both outcomes are in progress. One meta-analysis has shown that mesalamine seems to be effective in reducing the frequency of recurrent disease [5].

Anti-Tumor Necrosis Factor Antibody Quieted Symptoms of Crohn Disease

Targan SR, Hanauer SB, van Deventer SJ, et al. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor for Crohn's disease. Crohn's Disease cA2 Study Group. N Engl J Med. 1997; 337:1029-35.

The pathogenesis of Crohn disease is unknown, but a study published in 1997 may be a major advance toward an improved understanding of cause and treatment. In vitro studies have shown that the production of tumor necrosis factor is increased in the intestinal mucosa of patients with Crohn disease.

The goal of this clinical trial was to test the effect of monoclonal antibody against tumor necrosis factor as a possible treatment for patients with Crohn disease resistant to other treatment. In a 12-week, multicenter study, Targan and colleagues recruited 108 patients with moderate-to-severe disease that was resistant to multiple drug and nutritional regimens. Patients were randomly assigned to receive placebo or a 2-hour single infusion of monoclonal antibody against tumor necrosis factor. The outcome of interest was change in a Crohn disease activity index, which is based on a combination of symptoms, physical examination results, and laboratory variables.

Patients in the treatment groups had a 36% decrease in the index score, which was clinically significant. Of the patients who received antibody, 33% had remission, compared with 4% of the patients who received placebo (NNT to produce one remission, 3).

A single infusion of chimeric monoclonal antibody to tumor necrosis factor was an effective short-term treatment for patients with moderate-to-severe Crohn disease resistant to other therapy. This important study not only has implications for effective treatment but also provides insight into the pathogenesis of Crohn disease. It implies that the treatment of Crohn disease will evolve to resemble the treatment of cancer: Multiple drug regimens that attack different parts of the inflammatory cascade will be used.

Diarrhea Was Common after Diagnosis of Sprue

Fine KD, Meyer RL, Lee EL. The prevalence and causes of chronic diarrhea in patients with celiac sprue treated with a gluten-free diet. Gastroenterology. 1997; 112:1830-8.

Most physicians have learned that celiac sprue is a rare disease that is usually seen in persons of Irish descent. However, it may be much more common than previously assumed. For example, 3% to 5% of diabetic patients have evidence of sprue. Diarrhea is a common presenting symptom but is not universal in patients with celiac sprue. Even after patients have received a diagnosis and have been placed on a gluten-free diet, many still have diarrhea.

Fine and colleagues attempted to determine the prevalence and causes of chronic diarrhea in 78 patients with sprue who had been prescribed gluten-free diets for at least 1 year. Chronic diarrhea (passage of loose stools more than three times weekly for 6 months) was present in 79% of patients before diagnosis and in 17% after therapy began. After thorough evaluations, causes for diarrhea included microscopic colitis (3 patients), pancreatic insufficiency (2 patients), the irritable bowel syndrome (2 patients), fecal incontinence (2 patients), lactose malabsorption (1 patient), and excessive fructose ingestion (1 patient).

The conventional approach to patients with sprue who have continued diarrhea is to focus on misdiagnosis or noncompliance. Noncompliance has received particular attention because patients tend to gain weight after gluten is removed from the diet. This effect is disturbing to many, leading to deviations from the prescribed diet. This study suggests that a more thoughtful gastrointestinal evaluation may be warranted.

Escherichia coli O157:H7 Was an Important Bacterial Pathogen

Slutsker L, Ries AA, Greene KD, et al. Escherichia coli O157:H7 diarrhea in the United States: clinical and epidemiologic features. Ann Intern Med. 1997; 126:505-13.

Escherichia coli O157:H7 was isolated in 1982 and is now recognized as a cause of both sporadic and outbreak-associated bloody diarrhea in the United States. It received much attention after an outbreak in Washington State that was associated with undercooked hamburger, and it has been found in nonpasteurized apple juice and other poorly prepared products. Still, the true epidemiologic features of E. coli O157:H7 have not been appreciated.

The aim of this large prevalence study was to determine the frequency of isolation of E. coli O157:H7 and to identify clinical or epidemiologic features that might help clinicians distinguish infection with this organism from infection with Campylobacter, Salmonella, or Shigella species. Slutsker and colleagues reviewed 30 000 fecal specimens taken from inpatients and ambulatory patients at 10 hospitals over a 2-year period. They isolated E. coli O157:H7 in 0.4% of specimens, Campylobacter species in 2.3%, Salmonella species in 1.8%, and Shigella species in 1.1%. Although infection with E. coli O157:H7 is relatively infrequent compared with infection with the other three common pathogens, accounting for only 7% of infections, several clinical clues were identified that make it a more likely diagnosis. It is more common in patients who have visible blood in stools, are at the extremes of life, have abdominal tenderness or leukocytosis, or live in northern states.

Stool specimens from all patients with a history of bloody diarrhea should be cultured for E. coli O157:H7 because this organism can cause serious disease and person-to-person transmission is common. In addition, E. coli O157:H7 has two presentations that might mislead clinicians. First, it can simulate inflammatory bowel disease by presenting as a hemorrhagic colitis. Second, it can present as thrombocytopenia and fever, simulating thrombotic thrombocytopenia purpura, because it produces a toxin that damages endothelial cells and colonic mucosa.

Colon and Anal Cancer

Colon cancer will be diagnosed in about 131 600 persons and will kill approximately 56 500 persons in 1998. The lifetime risk for colon cancer is about 6%. Most cases of disease occur in those older than 60 years of age. Trends in survival are shown in Table 4, and the reason why African Americans die in disproportionate numbers remains unclear.

Diminutive Adenomas Were Markers of More Neoplasia

Read TE, Read JD, Butterly LF. Importance of adenomas 5 mm or less in diameter that are detected by sigmoidoscopy. N Engl J Med. 1997; 336:8-12.

Adenomatous polyps larger than 1 cm are known to increase the risk for colon cancer. However, the danger of smaller adenomas has been unknown, and treatment and follow-up have been controversial. Read and colleagues sought to determine the prevalence of proximal colonic neoplasms in asymptomatic patients with adenomatous polyps 5 mm or less on flexible sigmoidoscopy.

These investigators recruited 3496 consecutive patients undergoing sigmoidoscopy. All patients had average risk for colon cancer, and 137 of these patients had diminutive polyps ( 5 mm in diameter). These patients then underwent colonoscopy. Neoplasms were found in the proximal colon in 29% of patients. Two patients had proximal carcinoma in situ, and 2 had proximal stage 1 carcinomas.

The prevalence of proximal colonic neoplasms in asymptomatic patients with rectosigmoid adenomas less than 5 mm in diameter is substantial, and colonoscopy is warranted in such patients. This finding, along with those reported in last year's Update series, revises the risks for colon cancer (Table 5). In general, most patients should begin having fecal occult blood tests annually at about age 50 years. These same patients should also have flexible sigmoidoscopy every 5 to 10 years. Persons at higher risk should begin having studies earlier, and those at very high risk should have colonoscopy rather than the usual screening regimens. Any blood from the colon warrants colonoscopy to detect not only neoplasia but also inflammatory bowel disease. Patients who have colonic polyps removed at colonoscopy should have repeated colonoscopy every 3 years.

Human Papillomavirus Was Associated with Anal Cancer

Frisch M, Glimelius B, van den Brule AJ, et al. Sexually transmitted infection as a cause of anal cancer. N Engl J Med. 1997; 337:1350-8.

Epidermoid anal cancer is rare, but its incidence has increased markedly in recent years, especially among women and unmarried men. This has raised the question of whether this neoplasm could be the result of sexually transmitted infections.

These Danish researchers conducted a case–control study of 417 case-patients with anal cancer, 534 controls with rectal adenocarcinoma, and 554 controls from the general population. Surgical specimens were tested for human papillomavirus (HPV) DNA by polymerase chain reaction.

Major associations were found between the number of sexual partners and anal cancer. In women, receptive anal intercourse and venereal infections in partners were also associated with higher risk (odds ratios, 3.4 and 2.4, respectively). Fifteen percent of men with anal cancer and none of the controls reported having had homosexual contact. High-risk types of HPV (notably, HPV-16) were detected in 84% of the anal cancer specimens, whereas all of the rectal adenocarcinoma specimens tested were negative for HPV.

This degree of association between HPV infection and anal cancer suggests that anal cancer may be prevented through the avoidance of unsafe sexual practices.

Acute Pancreatitis

Acute pancreatitis remains difficult to diagnose because of poor sensitivity and specificity of both the physical examination and serum amylase and lipase measurements. Early diagnosis is important because about 20% of patients will have severe necrotizing disease, which is associated with a mortality rate of up to 40%.

Trypsinogen-2 Helped Rule Out Pancreatitis

Kemppainen EA, Hedstrom J, Puolakkainen PA, et al. Rapid measurement of urinary trypsinogen-2 as a screening test for acute pancreatitis. N Engl J Med. 1997; 336:1788-93.

Patients with acute pancreatitis have been noted to have elevated serum and urinary levels of trypsinogen-2, a precursor of trypsin. Therefore, Kemppainen and colleagues sought to determine the characteristics of urinary trypsinogen-2, measured by a dipstick test, as a diagnostic marker for acute pancreatitis. They recruited 500 consecutive patients who presented to an emergency department with abdominal pain. Patients had abdominal CT, which was considered the diagnostic standard, and assays for serum and urinary amylase levels. Fifty-three patients were found to have pancreatitis, and the results of the tests are shown in Table 6.

The dipstick test, which will be available in the United States late in 1998, will help rule out the diagnosis of acute pancreatitis because of its high sensitivity, but a positive result will probably require confirmation with more specific tests.

Treatments of Pancreatitis Complications Were Clarified

Baron TH, Morgan DE. The diagnosis and management of fluid collections associated with pancreatitis. Am J Med. 1997; 102:555-63.

The nomenclature surrounding the complications of pancreatitis has been complex, which makes it difficult to plan treatment. In 1992, the International Symposium on Acute Pancreatitis attempted to simplify the nomenclature, and this review paper summarizes that report, as well as the treatments for common complications.

Acute fluid collections appear within 48 hours of pain and affect up to 50% of patients. They consist of enzyme-rich fluid and frequently resolve without intervention.

Acute pseudo cysts are collections of pancreatic juice surrounded by nonepithelialized granulation tissue and take at least 4 weeks to develop. Patients who develop them often have improvement after an acute attack of pancreatitis and then return with increased abdominal pain. About 40% of cases resolve within 6 weeks of discovery; therefore, watchful waiting is prudent. Drainage should be done in patients who have pseudo cysts greater than 5 cm in diameter after 6 weeks or those who have severe abdominal pain; gastric outlet obstruction; or clinical evidence of infection, such as fever or leukocytosis.

Pancreatic necrosis, formerly called phlegm, occurs in about 20% of patients with acute pancreatitis. It is defined as at least one focal area of non-enhancing pancreatic parenchyma that makes up more than 30% of the pancreas, as detected by contrast CT. Because it has a high complication rate, necrosis must be carefully differentiated from the more benign pseudocyst, which sometimes requires CT-directed fine-needle aspiration. Treatment of this condition is controversial unless complications develop.

The most common complication of pancreatic necrosis is infection, which occurs in about 35% to 70% of patients with necrosis, often in the second to third week after abdominal pain begins. The infections are typically polymicrobial, and treatment is surgical drainage, which is complex and carries high morbidity and mortality rates.

In contrast, pancreatic abscesses are discrete collections of pus near the pancreas. Like pseudocysts, they take at least 4 weeks to form. Little debris is found in the pus. Patients tend to present with worsening abdominal pain, fever, and leukocytosis. The pathogens are usually polymicrobial, often involving enteric bacteria. Radiologically directed drainage with large catheters can be done in many patients, and antibiotics are directed at identified pathogens.

Dr. Roberts (Series Editor): Madrona Medical Group, 3199 Steller Court, Bellingham, WA 98226-7805.