Figure 69.1. Natural history of H. pylori infection. Source: Reprinted with permission from Suerbaum S, Michetti P. Medical progress: Helicobacter pylori infection. N Engl J Med. 2002;347(15):1175–1186. © 2002 Massachusetts Medical Society. All rights reserved.

NONSTEROIDAL ANTIINFLAMMATORY DRUGS

After H. pylori, NSAIDs are the second most important etiologic factor in PUD. Common estimates are that up to one-quarter of patients using NSAIDs chronically (including aspirin) will develop duodenal or gastric ulcers. NSAID use is felt to correlate with a higher risk of bleeding complications and death in PUD. The injurious effects of NSAIDs are in part due to direct cytotoxic effects on epithelial cells, as well as inhibition of cyclo-oxygenase-1 (COX-1), with resulting reductions in prostaglandin production and bicarbonate and mucus secretion, and alterations in mucosal blood flow.

    The risks of NSAID-induced ulcer formation and bleeding are greatest in patients with prior PUD and in patients using anticoagulant medications such as warfarin. In patients with NSAID-related PUD, cessation of NSAIDs is the optimal strategy to reduce the chance of ulcer recurrence. Among patients who require continuing NSAID therapy (including aspirin), misoprostol and PPIs have both been shown to decrease the risk of GI complications, although PPI therapy is preferred because of tolerability and ease of use. Recommendations vary regarding which NSAID users benefit most from concomitant PPI therapy. Generally, PPI cotherapy should be considered among NSAID users over 65 years old and should be strongly considered for patients on anticoagulation. For patients with prior PUD, H. pylori eradication and continuing PPI cotherapy are mandatory if NSAIDs must be continued. COX-2–selective NSAIDs carry a smaller risk of GI ulceration, but the associated cardiac risk profile of COX-2 agents limits their practical usefulness. For patients requiring antiplatelet agents for cardiac disease, aspirin with PPI therapy has been shown to be associated with a lower risk of GI events than clopidogrel alone. For patients requiring ongoing clopidogrel therapy, PPI therapy can also be considered when clinically indicated. There were initial concerns regarding reduced efficacy of clopidogrel with PPI cotherapy, but these concerns have diminished somewhat after more robust trial data has become available supporting the safety of coadministering these medications.

ZOLLINGER-ELLISON SYNDROME (GASTRINOMA)

The finding of severe PUD, particularly in the absence of obvious risk factors such as NSAID use or H. pylori infection, should trigger an evaluation for Zollinger-Ellison syndrome (gastrinoma). Zollinger-Ellison syndrome is rare, probably accounting for <1% of PUD. Multiple ulcers, severe gastroesophageal reflux disease, and malabsorption (leading to diarrhea and weight loss) are all classic findings of this syndrome, although many patients have PUD alone. Gastrinomas can occasionally be a feature of multiple endocrine neoplasia (MEN-1) syndrome and may therefore occur along with parathyroid and pituitary tumors.

    If Zollinger-Ellison is suspected, a fasting serum gastrin level should be measured, preferably after a patient has discontinued acid-suppressive therapy for 7 days. Although a fasting gastrin level >1000 pg/mL is virtually diagnostic of Zollinger-Ellison, a moderately elevated gastrin level (>100 pg/mL at our institution) should be followed by a secretin-stimulation test, the provocative test of choice because of its high sensitivity. In Zollinger-Ellison syndrome, secretin causes an abnormal increase (of >200 pg/mL) in the serum gastrin level within minutes. Confirmatory testing then includes an octreotide scan (or somatostatin receptor scintigraphy), and an abdominal computed tomography (CT) scan or endoscopic ultrasound to localize the tumor, which is typically within the pancreas or duodenum in an anatomic region termed the gastrinoma triangle. Although PPI therapy may control acid hypersecretion, the risk of malignancy (i.e., local metastatic spread) is such that gastrinomas must be resected when possible.

OTHER CAUSES OF GASTRODUODENAL ULCERS

Although NSAID ingestion and H. pylori infection are the two most common causes of PUD, numerous other conditions can contribute to ulcer formation (box 69.1). The term “peptic ulcer disease” specifically refers to ulceration occurring due to acid/pepsin exposure in the GI tract; however, we have also included other conditions that can cause gastroduodenal ulceration that may be mistaken for PUD (i.e., Crohn’s disease and malignancy).

Box 69.1 CAUSES OF GASTRIC AND DUODENAL ULCERS

    It is well known, although perhaps overstated, that gastric or duodenal ulcers can arise during critical illness. “Stress ulcers” are generally superficial and not a significant cause of major GI bleeding. The pathophysiology of stress ulcer formation is probably distinct from that of typical PUD and is thought to involve mucosal ischemia, hypoperfusion, and reperfusion. Historical terms have included Cushing ulcers, occurring in the setting of intracranial pathology, and Curling ulcers, occurring in burn patients. Known risk factors for stress ulcer formation include head trauma/neurosurgery, >30% burns, as well as mechanical ventilation and coagulopathy. Acid-suppression therapy with PPIs is indicated for stress ulcer prophylaxis in these specific settings. For the majority of patients hospitalized on general medical or surgical wards, however, stress ulcer prophylaxis is generally not necessary.

    Corticosteroids are also frequently implicated in peptic ulcer formation; however, it is not clear that corticosteroid use alone increases the risk of PUD. Concomitant use of NSAIDs and corticosteroids, however, does seem to confer an increased risk of ulcer formation when compared to NSAID use alone. Other medications, including sirolimus and bisphosphonates, appear to be associated with PUD in some cases.

    Chronic mesenteric ischemia can cause mucosal ulceration anywhere in the GI tract and should be a consideration in the evaluation of patients with nonhealing H. pylori–negative gastroduodenal ulcers. Advanced age, known atherosclerotic disease, and smoking are all risk factors for chronic mesenteric ischemia. Atherosclerotic disease of the GI tract may be clinically silent until two of the three major splanchnic vessels are involved because of the extensive collateral network within the GI vasculature.

    Rare causes of gastric or duodenal ulcer formation include cocaine use (possibly due to vasoconstriction and/or thrombosis), herpes simplex virus and cytomegalovirus infections (largely in immunosuppressed patients), and Crohn’s disease. Finally, gastric adenocarcinoma is always included in the differential diagnosis of gastric ulcers, and multiple endoscopic biopsies are mandatory to rule out malignancy in the setting of a nonhealing gastric ulcer.

CLINICAL PRESENTATION

Abdominal pain is a classic feature of PUD, but it is a highly nonspecific symptom. The abdominal pain of PUD is typically epigastric, nonradiating, and may occur in the postprandial period. Some patients will report improvement with antacid medications. Dyspepsia, a broader term that refers to a constellation of (usually epigastric) abdominal symptoms, including pain, bloating, and nausea, can be caused by a variety of disorders including PUD, nonulcer dyspepsia (symptoms in the absence of mucosal ulceration), pancreaticobiliary diseases, gastroesophageal reflux, and malignancy.

    The three primary complications of PUD are hemorrhage, perforation, and obstruction. All three complications are occurring less frequently in the era of antisecretory therapy and H. pylori eradication; however, PUD remains the most common cause of significant upper-GI tract hemorrhage. Bleeding typically occurs when an ulcer in the stomach or duodenum erodes into a small or medium-sized blood vessel within the submucosa. Massive bleeding may be particularly likely in the rare circumstances when a posterior duodenal bulb ulcer penetrates the gastroduodenal artery or a gastric ulcer penetrates the left gastric artery. Melena (black tarry stool) is usually indicative of bleeding from the upper gastrointestinal tract (above the ligament of Treitz), and hematemesis may also occur. Massive upper-GI bleeding (typically >500 mL of blood loss) may also cause hematochezia (red blood in the stool). A number of clinical scoring systems exist that may predict outcome, mortality, and the need for endoscopic intervention in PUD bleeding. Of these, the Rockall score and Blatchflord score are most commonly used. Most scoring systems identify advanced age, medical comorbidities, unstable hemodynamic status, and/or endoscopic findings of recent/active bleeding as the key features predictive of poor outcome. Appropriate triage to an intensive care unit, rapid resuscitation with fluid and blood products, PPIs, and early endoscopic therapy are mainstays in the treatment of PUD with bleeding, a scenario in which mortality can approach 10%.

    A peptic ulcer that erodes fully through the gastric or duodenal wall and into the peritoneum is termed a perforating ulcer. This is in distinction to the term penetrating, which implies erosion into adjacent organs (the pancreas, liver, etc.). The anterior wall of the duodenum and the lesser curvature of the stomach are the most common sites for perforation. Spillage of luminal contents into the peritoneum typically leads to severe abdominal pain due to peritonitis, which is accompanied by exquisite tenderness and abdominal rigidity on physical examination. Perforation can be confirmed by identifying subdiaphragmatic air on upright chest radiography or by identifying extraluminal air on abdominal CT scan. Urgent surgical intervention is mandatory.

    Gastric outlet obstruction can occur either acutely or chronically in relation to PUD. Acute ulceration leading to edema within or near the pylorus or duodenal bulb can cause obstruction. Chronic ulceration with scarring can have similar effects. The typical symptoms of obstruction include nausea, vomiting, early satiety, and, in the chronic setting, weight loss. Physical examination may reveal distension and/or a succussion splash. Treatment of gastric outlet obstruction is initially supportive, including intravenous hydration, nasogastric tube decompression, and PPI therapy. Conservative therapy may be enough to relieve obstruction related to acute ulceration and edema in some cases, whereas obstruction due to chronic ulceration or scarring is more likely to require endoscopic dilation or surgery. The presence of gastric outlet obstruction mandates a thorough evaluation for malignancy with endoscopic biopsies.

DIAGNOSTIC EVALUATION OF PUD

In the absence of significant complications such as bleeding, perforation, or obstruction, the symptoms of PUD can be difficult to differentiate from other causes of abdominal pain or dyspepsia. If the history is highly suggestive of acid peptic disease (i.e., postprandial epigastric discomfort, alleviation with antacids, etc.), empirical acid-suppressive therapy with a PPI or H₂-receptor antagonist may be reasonable in patients who are under 55 years old in the absence of “alarm” features (box 69.2). H. pylori serology testing is also reasonable in this group of patients. Additional studies starting with upper endoscopy should be undertaken if symptoms do not subside with empiric therapy or if alarm symptoms are present.

Box 69.2 ALARM FEATURES DURING EVALUATION OF DYSPEPIA

TESTING FOR HELICOBACTER PYLORI

The first question to consider in discussing diagnostic tests for H. pylori is: who should be tested? Should all patients with dyspepsia be evaluated for H. pylori, or can specific subgroups of patients be identified for whom testing will yield the greatest benefit? The American College of Gastroenterology advocates a “test and treat” approach for H. pylori among certain patients who present with previously uninvestigated dyspepsia. This approach is acceptable for patients under the age of 55 years and without “alarm” features (box 69.2). Noninvasive H. pylori testing (i.e., serology) is reasonable in this population, particular in regions where the prevalence of H. pylori is high. An alternative to the H. pylori test and treat approach in this group of patients is to begin with empirical antisecretory therapy (with H₂-receptor antagonist or PPI) and cessation of any NSAID use for 4 weeks, followed by H. pylori testing if symptoms persist. Endoscopy is indicated for persistent symptoms, for patients with alarm features, and in patients older than 55 years.

    The most common noninvasive diagnostic tests for H. pylori are the H. pylori serology test and the urea breath test (table 69.1). The serologic test for antibodies against H. pylori is widely available and approximately 80% sensitive, but it cannot necessarily distinguish between active and prior infection. The urea breath test is more laborious and involves ingestion of radiolabeled urea, followed by measurement of the exhaled carbon isotope, which is released only in the presence of H. pylori urease activity. The urea breath test will only detect active H. pylori infection and may therefore be useful in order to document eradication. Fecal antigen testing for H. pylori is a third option, with sensitivity and specificity similar to those of the urease breath test. Furthermore, like the urease breath test, a positive fecal antigen test should generally reflect active infection; however, the test may remain positive for several weeks after H. pylori eradication therapy.

Stay updated, free articles. Join our Telegram channel

Join