Pharmacotherapy A Pathophysiologic Approach, 9th Ed.

20. Peptic Ulcer Disease

Bryan L. Love and Matthew N. Thoma


 Images Patients with peptic ulcer disease (PUD) should reduce psychological stress, cigarette smoking, and nonsteroidal antiinflammatory drug (NSAID) use and avoid foods and beverages that exacerbate ulcer symptoms.

 Images Eradication is recommended for all Helicobacter pylori–positive patients, especially those patients with an active ulcer, a documented history of a prior ulcer, or a history of ulcer-related complications.

 Images The selection of an H. pylori eradication regimen should be based on efficacy, safety, antibiotic resistance, cost, and the likelihood of medication adherence. Treatment should be initiated with a proton pump inhibitor (PPI)–based three-drug regimen. If a second course of H. pylori therapy is required, the regimen should contain different antibiotics.

 Images PPI cotherapy reduces the risk of NSAID-related gastric and duodenal ulcers and is at least as effective as recommended dosages of misoprostol and superior to the histamine-2 receptor antagonists (H2RAs).

 Images Standard PPI dosages and a nonselective NSAID are as effective as a selective cyclooxygenase-2 (COX-2) inhibitor in reducing the risk of NSAID-induced ulcers and upper GI complications.

 Images Patients with PUD, especially those receiving H. pylori eradication or misoprostol cotherapy, require patient education regarding their disease and drug treatment to successfully achieve a positive therapeutic outcome.

 Images The recommended treatment for severe peptic ulcer bleeding after appropriate endoscopic treatment is the IV administration of a PPI loading dose followed by a 72-hour continuous infusion with a goal of maintaining an intragastric pH of 6 or greater.

 Images Critically ill patients at the highest risk of developing stress-related mucosal bleeding (SRMB) who require prophylactic drug therapy include those with respiratory failure on mechanical ventilation or those with coagulopathy.

 Images There are limited data to support the selection of a PPI over an IV H2RA for SRMB prophylaxis. The decision should be based on appropriate individual patient characteristics (e.g., nothing by mouth, presence of nasogastric tube, renal failure).


Acid-related diseases (gastritis, erosions, and peptic ulcer) of the upper GI tract require gastric acid for their formation.13 Peptic ulcer disease (PUD) differs from gastritis and erosions in that ulcers typically extend deeper into the muscularis mucosa.1 There are three common forms of peptic ulcers: Helicobacter pylori positive, nonsteroidal antiinflammatory drug (NSAID) induced, and stress ulcers (Table 20-1). The term stress-related mucosal damage(SRMD) is preferred to stress ulcer or stress gastritis, because the mucosal lesions range from superficial gastritis and erosions to deep ulcers.

TABLE 20-1 Comparison of Common Forms of Peptic Ulcer


H. pylori–positive and NSAID-induced ulcers are chronic peptic ulcers that differ in etiology, clinical presentation, and tendency to recur (see Table 20-1). These ulcers develop most often in the stomach and duodenum of ambulatory patients (Fig. 20-1). Occasionally, ulcers develop in the esophagus, jejunum, ileum, or colon. The natural course of chronic PUD is characterized by frequent ulcer recurrence. The most important factors that influence ulcer recurrence are H. pylori infection and NSAID use. Other factors include cigarette smoking, alcohol use, ulcer-related complications, gastric acid hypersecretion, and patient noncompliance. The cause of ulcer recurrence is most likely multifactorial.


FIGURE 20-1 Anatomic structure of the stomach and duodenum and most common locations of gastric and duodenal ulcers.

Peptic ulcers are also associated with Zollinger-Ellison syndrome (ZES), radiation, chemotherapy, vascular insufficiency, and other chronic diseases (Table 20-2).1,3 Although a strong association exists between chronic pulmonary diseases, chronic renal failure, and cirrhosis, the pathophysiologic mechanisms of these associations remain unclear.1 In contrast, SRMD occurs primarily in the stomach in critically ill patients (see Table 20-1).1

TABLE 20-2 Potential Causes of Peptic Ulcer


This chapter focuses on chronic PUD associated with H. pylori and NSAIDs. A brief discussion of ZES and upper GI bleeding related to PUD and SRMD is included.


The epidemiology of PUD is complicated and difficult to estimate given the variability in the prevalence of H. pylori infection, NSAID use, and cigarette smoking as well as the various methods used to detect ulcers, for example, endoscopy, radiology, symptoms, or complications.1,4 The prevalence and incidence of PUD in the United States also reflects improvements in drug therapy, the dramatic shift to ambulatory management, and changes in the criteria and coding system for mortality and hospitalization data.1 Recent trends suggest a shift from predominance in men to a similar occurrence in men and women with increasing rates of disease in older individuals and a decrease in the younger population.1,4 Despite a modest decline in mortality, hospitalizations, and office visits, PUD remains one of the most common GI diseases, resulting in impaired quality of life, work loss, and high-cost medical care.


Most peptic ulcers occur in the presence of acid and pepsin when H. pylori, NSAIDs, or other factors (see Table 20-2) disrupt the normal mucosal defense and healing mechanisms.1 Hypersecretion of acid is the primary pathogenic mechanism in hypersecretory states such as ZES.3 Benign gastric ulcers can occur anywhere in the stomach, although most are located on the lesser curvature, just distal to the junction of the antral and acid-secreting mucosa (see Fig. 20-1). Most duodenal ulcers occur in the first part of the duodenum (duodenal bulb).

H. pylori

H. pylori infection causes chronic gastritis in infected individuals and is causally linked to PUD, mucosa-associated lymphoid tissue (MALT) lymphoma, and gastric cancer (Fig. 20-2).1,2,58 The majority of infected individuals remain asymptomatic, but 10% to 20% will develop PUD during their lifetime and about 1% will develop gastric cancer.1,2,8 Host-specific cofactors and H. pylori strain variability play an important role in the pathogenesis of PUD and gastric cancer.2,8 Although an association between H. pylori and PUD bleeding is less clear, there is evidence that eradication of H. pylori decreases recurrent bleeding.5,9 No specific link has been established between H. pyloriand dyspepsia, nonulcer dyspepsia (NUD), or gastroesophageal reflux disease (GERD).5,911 However, some patients with dyspepsia and NUD may have symptom improvement from H. pylori eradication.5,9 Although eradication of H. pylori may worsen GERD symptoms in some patients, eradication should not be withheld.5,10 An association between H. pylori infection and iron deficiency anemia has been established, but cause and effect have not been proven, and whether H. pylori eradication is beneficial is uncertain.5,11 There are insufficient data to support a link between H. pylori and extragastric manifestations including cardiovascular, hematologic, respiratory, hepatobiliary, and neurologic diseases.5,11


FIGURE 20-2 The natural history of Helicobacter pylori infection in the pathogenesis of gastric ulcer and duodenal ulcer, mucosaassociated lymphoid tissue (MALT) lymphoma, and gastric cancer.

The prevalence of H. pylori varies by geographic location, socioeconomic conditions, ethnicity, and age. In developing countries, H. pylori prevalence is more common than in industrialized countries and correlates with lower socioeconomic levels.2,5,6 The prevalence of H. pylori in the United States is 30% to 40% but is much higher in individuals over 60 years (50% to 60%) than in children under 12 years (10% to 15%) of age.2,5 Although most individuals in the United States acquire H. pylori in childhood, the rate of acquisition in children is declining and most likely will continue to fall as a consequence of improved socioeconomic conditions.2 Whites are infected with H. pylori less frequently than African Americans and Hispanic persons, but this is thought to be related to lower socioecomonic status and living conditions. Infection rates do not differ with gender or smoking status.

The most common route of H. pylori transmission is person to person by either gastro–oral (vomitus) or fecal–oral (diarrhea) contact that occurs primarily during childhood.2 Members of the same household are likely to become infected when someone in the same household is infected.2 H. pylori can also be transmitted by the use of inadequately sterilized endoscopes.

Nonsteroidal Antiinflammatory Drugs

NSAIDs (Table 20-3), including both prescription and nonprescription medications, are widely used in the United States, particularly in individuals over 60 years of age, to treat chronic pain and inflammation.1 Low-dose aspirin is used for cardiovascular and cerebrovascular risk reduction.1,1214 There is overwhelming evidence linking chronic NSAID (including aspirin) use to a variety of upper GI tract injuries.1,1216 NSAIDs cause superficial (topical) mucosal damage consisting of petechiae (intramucosal hemorrhages) within minutes of ingestion, and progress to erosions with continued use.1 These lesions typically heal within a few days and rarely cause ulcers or acute upper GI bleeding. Gastroduodenal ulcers develop in about 25% of chronic NSAID users with continued use.12 Gastric ulcers are most common, occur primarily in the antrum, and are of greater concern because of their potential to cause ulcer-related upper GI complications (see Table 20-1). As many as 2% to 4% of patients with an NSAID ulcer will bleed or perforate.12 Each year, NSAIDs account for at least 100,000 hospitalizations and between 7,000 and 10,000 deaths in the United States.12 NSAID-induced ulcers occur less frequently in the esophagus, small bowel, and colon.16,17 How NSAIDs damage the lower GI tract is unclear, but the enteropathy is associated with lower GI bleeding.

TABLE 20-3 Selected Nonsteroidal Antiinflammatory Drugs (NSAIDs) and Cyclooxygenase-2 (COX-2) Inhibitors


Table 20-4 lists the risk factors associated with NSAID-induced ulcers and upper GI complications. Combinations of factors confer an additive risk.1216,18 Advanced age is an independent risk factor, and the incidence of NSAID-induced ulcers increases linearly with the age of the patient.1 The high incidence of ulcer complications in older individuals may be explained by age-related changes in gastric mucosal defense. The relative risk of NSAID complications is increased for patients with a previous peptic ulcer and may be as high as 14-fold in those with a history of an ulcer-related complication.1,16 Although the risk of ulcer complications is greatest during the first few months after initiating continuous NSAID therapy, it does not vanish with long-term treatment.12

TABLE 20-4 Risk Factors Associated with Nonsteroidal Antiinflammatory Drug (NSAID)–Induced Ulcers and Upper GI Complicationsa


NSAID ulcers and related complications are dose dependent, but may occur with low doses of nonprescription NSAIDs and low cardioprotective dosages of aspirin (81 to 325 mg/day).1,1216 Factors such as NSAID potency, longer duration of effect, and a greater propensity to inhibit cyclooxygenase-1 (COX-1) versus cyclooxygenase-2 (COX-2) isoenzymes are associated with increased risk (see Table 20-3).1,16,18,19 NSAID-related dyspepsia, in itself, does not correlate directly with mucosal injury or clinical events. However, new-onset dyspepsia, changes in severity, or dyspepsia not relieved by antiulcer medications may suggest an ulcer or ulcer complication.1 Nonacetylated salicylates (e.g., salsalate) may be associated with decreased GI toxicity.1 Buffered or enteric-coated aspirin confers no added protection from upper GI events.13

NSAID ulcer and GI complication risk are increased with the use of multiple NSAIDs or the concomitant use of low-dose aspirin, oral bisphosphonates, corticosteroids, anticoagulants, antiplatelet drugs, and selective serotonin reuptake inhibitors.1,1216,20 The risk of an ulcer-related GI complication is greater when an NSAID or COX-2 inhibitor (see Table 20-3) is coadministered with low-dose aspirin than when either drug is taken alone.1,13,16 The NSAID may also reduce the antiplatelet effects of aspirin, although NSAIDs vary in their effects on platelet function.1315 Corticosteroids, when used alone, do not potentiate the risk of ulcer or complications, but the relative risk is increased twofold in corticosteroid users who are also taking concurrent NSAIDs.1,16 The relative risk of GI bleeding increases up to 20-fold when NSAIDs are taken concomitantly with anticoagulants (e.g., warfarin) and up to 6-fold with the concurrent use of serotonin reuptake inhibitors.16,17 When clopidogrel is taken in combination with aspirin, an NSAID, or an anticoagulant, the risk of GI bleeding is increased compared with when these agents are taken alone.13,15,16 Even when prescribed as monotherapy, clopidogrel increases the risk of rebleeding for patients with a history of a bleeding ulcer.13,15,16

H. pylori and NSAIDs act independently to increase ulcer risk and ulcer-related bleeding and appear to have additive effects.5,16 Thus, the incidence of peptic ulcer is higher in H. pylori–positive NSAID users. Whether H. pyloriinfection is actually a risk factor for NSAID ulcers remains controversial.1,5,16 However, eradication is reported to reduce the incidence of peptic ulcer if undertaken prior to starting the NSAID but does not reduce the risk for patients who were previously taking an NSAID.1,5,16

Cigarette Smoking

Epidemiologic evidence links cigarette smoking to PUD, but it is uncertain whether smoking causes peptic ulcers.1 Ulcer risk is proportional to the number of cigarettes smoked and is modest when fewer than 10 cigarettes are smoked per day. Cigarette smoking impairs ulcer healing, promotes ulcer recurrence, and increases ulcer risk.1 The exact mechanism by which cigarette smoking contributes to PUD remains unclear. Possible mechanisms include mucosal ischemia, inhibition of pancreatic bicarbonate secretion, and increases in gastric acid secretion, but these effects are inconsistent. Whether nicotine or other components of smoke are responsible for these physiologic alterations is unknown.

Psychological Stress

The importance of psychological factors in the pathogenesis of PUD remains controversial.1 Clinical observation suggests that ulcer patients are adversely affected by stressful life events. However, results from controlled trials are conflicting and have failed to document a cause-and-effect relationship.1 Emotional stress may induce behavioral risks such as smoking and the use of NSAIDs or alter the inflammatory response or resistance to H. pylori infection. The role of stress and how it affects PUD is complex and probably multifactorial.

Dietary Factors

The role of diet and nutrition in PUD is uncertain.1 Coffee, tea, carbonated beverages, beer, milk, and spices may cause dyspepsia but do not increase the risk for PUD. Beverage restrictions and bland diets do not alter the frequency of ulcer recurrence. Although caffeine is a gastric acid stimulant, constituents in decaffeinated coffee or tea, caffeine-free carbonated beverages, beer, and wine may also increase gastric acid secretion. In high concentrations, alcohol ingestion is associated with acute gastric mucosal damage and upper GI bleeding; however, there is insufficient evidence to confirm that alcohol causes ulcers.1


A physiologic imbalance between aggressive (gastric acid and pepsin) and protective (mucosal defense and repair) factors remains an important issue in the pathophysiology of gastric and duodenal ulcers.1,21Gastric acid is secreted by the parietal cells, which contain receptors for histamine, gastrin, and acetylcholine.21 Acid (as well as H. pylori infection and NSAID use) is an independent factor that contributes to the disruption of mucosal integrity.1 Increased acid secretion has been observed for patients with duodenal ulcers and may be a consequence of H. pylori infection.2,22 Patients with ZES (described in Zollinger-Ellison Syndrome below) have profound gastric acid hypersecretion resulting from a gastrin-producing tumor.3 In contrast, patients with gastric ulcer usually have normal or reduced rates of acid secretion (hypochlorhydria).

Acid secretion is expressed as the amount of acid secreted under basal or fasting conditions, basal acid output (BAO); after maximal stimulation, maximal acid output (MAO); or in response to a meal.21 Basal, maximal, and meal-stimulated acid secretion varies according to time of day and the individual’s psychological state, age, gender, and health status. The BAO follows a circadian rhythm, with the highest acid secretion occurring at night and the lowest in the morning. An increase in the BAO:MAO ratio suggests a basal hypersecretory state such as ZES. A review of gastric acid secretion and its regulation can be found elsewhere.21

Pepsin is an important cofactor that plays a role in the proteolytic activity involved in ulcer formation.21 Pepsinogen, the inactive precursor of pepsin, is secreted by the chief cells located in the gastric fundus (see Fig. 20-1). Pepsin is activated by acid pH (optimal pH of 1.8 to 3.5), reversibly inactivated at pH 4, and irreversibly destroyed at pH 7.

Mucosal defense and repair mechanisms (mucus and bicarbonate secretion, intrinsic epithelial cell defense, and mucosal blood flow) protect the gastroduodenal mucosa from noxious endogenous and exogenous substances.1,21The viscous nature and near-neutral pH of the mucus–bicarbonate barrier protect the stomach from the acidic contents in the gastric lumen. Mucosal repair after injury is related to epithelial cell restitution, growth, and regeneration. The maintenance of mucosal integrity and repair is mediated by the production of endogenous prostaglandins (PGs). The term cytoprotection is often used to describe this process, but mucosal defense and mucosal protection are more accurate terms, as PGs prevent deep mucosal injury and not superficial damage to individual cells. Gastric hyperemia and increased PG synthesis characterize adaptive cytoprotection, the short-term adaptation of mucosal cells to mild topical irritants. This phenomenon enables the stomach to initially withstand the damaging effects of irritants. Alterations in mucosal defense that are induced by H. pylori or NSAIDs are the most important cofactors in the formation of peptic ulcers.

H. pylori

H. pylori is a spiral-shaped, pH-sensitive, gram-negative, microaerophilic bacterium that resides between the mucus layer and surface epithelial cells in the stomach, or any location where gastric-type epithelium is found.2,22 The combination of its spiral shape and flagellum permits it to move from the lumen of the stomach, where the pH is low, to the mucus layer, where the local pH is neutral. H. pyloriproduces large amounts of urease, which hydrolyzes urea in the gastric juice and converts it to ammonia and carbon dioxide.2 The local buffering effect of ammonia creates a neutral microenvironment within and surrounding the bacterium, protecting it from the lethal effect of gastric acid. H. pylori also produces acid-inhibitory proteins, which allow it to adapt to the low-pH environment of the stomach.2

H. pylori binds to specific regions within the stomach. It attaches to gastric-type epithelium by adherence pedestals, which prevent the organism from being shed during cell turnover and mucus secretion.2Colonization of the antrum and corpus (body) of the stomach is associated with gastric ulcer and cancer.1,2,22 Antral organisms colonize gastric metaplastic tissue (gastric tissue that develops in the duodenum secondary to changes in gastric acid or bicarbonate secretion) leading to duodenal ulcer (see Fig. 20-2).1,2 Although H. pylori causes chronic gastric mucosal inflammation in all infected individuals, only a minority actually develop an ulcer or gastric cancer.1,2,8 The difference in the diverse clinical outcomes is related to variations in bacterial pathogenicity and host susceptibility.2,22

Mucosal injury is produced by (a) elaborating bacterial enzymes (urease, lipases, and proteases), (b) adherence, and (c) H. pylori virulence factors.2,22 Lipases and proteases degrade gastric mucus, ammonia produced by urease may be toxic to gastric epithelial cells, and bacterial adherence enhances the uptake of toxins into gastric epithelial cells. H. pylori induces gastric inflammation by altering the host inflammatory response and damaging epithelial cells directly by cell-mediated immune mechanisms or indirectly by activated neutrophils or macrophages attempting to phagocytose bacteria or bacterial products.2,22 However, H. pylori strains are genetically diverse and account for differences in adaptation within the human host. Two of the most important are cytotoxin-associated gene protein (CagA) and vacuolating cytotoxin (VacA). About 60% of H. pylori strains in the United States possess CagA, but CagA-positive strains increase the risk for severe PUD, gastritis, and gastric cancer compared with CagA-negative strains.2,22 The VacA gene codes for the VacA cytotoxin, a vacuolating toxin. Although VacA is present in most H. pylori strains, strains vary in cytotoxicity and increased risk for peptic ulcer and gastric cancer.2,22 Host polymorphisms are important markers of disease susceptibility and may identify high-risk patients.2,22 Polymorphisms of interleukin (IL)-1β and its receptor antagonist, as well as tumor necrosis factor-α (TNF-α) and IL-10, may be associated with increased gastric acid secretion and duodenal ulcer or acid suppression and gastric cancer.2,22

Nonsteroidal Antiinflammatory Drugs

NSAIDs, including aspirin (see Table 20-3), cause gastric mucosal damage by two important mechanisms: (a) direct or topical irritation of the gastric epithelium and (b) systemic inhibition of endogenous mucosal PG synthesis.1,13Although the onset of injury is initiated topically by the acidic properties of many of the NSAIDs, systemic inhibition of the protective PGs limits the ability of the mucosa to defend itself against injury and thus plays the predominant role in the development of gastric ulcer.1,13

Topical irritant properties are predominantly associated with acidic NSAIDs (e.g., aspirin) and their ability to decrease the hydrophobicity of the mucous gel layer in the gastric mucosa. Most non-aspirin NSAIDs have topical irritant effects, but aspirin is the most damaging. Although NSAID prodrugs, enteric-coated aspirin tablets, salicylate derivatives, and parenteral or rectal preparations are associated with less acute topical gastric mucosal injury, they can cause ulcers and related GI complications as a result of their systemic inhibition of endogenous PGs.1

Cyclooxygenase (COX) is the rate-limiting enzyme in the conversion of arachidonic acid to PGs and is inhibited by NSAIDs (Fig. 20-3). Two similar COX isoforms have been identified: COX-1 is found in most body tissue, including the stomach, kidney, intestine, and platelets; COX-2 is undetectable in most tissues under normal physiologic conditions, but its expression can be induced during acute inflammation and arthritis (Fig. 20-4).1,13 COX-1 produces protective PGs that regulate physiologic processes such as GI mucosal integrity, platelet homeostasis, and renal function. COX-2 is induced (unregulated) by inflammatory stimuli such as cytokines and produces PGs involved with inflammation, fever, and pain. It is also constitutionally expressed in organs such as the brain, kidney, and reproductive tract. Adverse effects (e.g., GI or renal toxicity) of NSAIDs are primarily associated with the inhibition of COX-1, whereas antiinflammatory actions result primarily from NSAID inhibition of COX-2.1,13


FIGURE 20-3 Metabolism of arachidonic acid after its release from membrane phospholipids. Broken arrow indicates inhibitory effects. (ASA, aspirin; HPETE, hydroperoxyeicosatetraenoic acid; NSAIDs, nonsteroidal antiinflammatory drugs; PG, prostaglandin.)


FIGURE 20-4 Tissue distribution and actions of cyclooxygenase (COX) isoenzymes. Nonselective nonsteroidal antiinflammatory drugs (NSAIDs) including aspirin (ASA) inhibit COX-1 and COX-2 to varying degrees; COX-2 inhibitors inhibit only COX-2. Broken arrow indicates inhibitory effects.

The COX-1-to-COX-2 inhibitory ratio determines the relative GI toxicity of a specific NSAID. Nonselective NSAIDs, including aspirin (see Table 20-3), inhibit both COX-1 and COX-2 to varying degrees and are associated with an increased propensity to cause gastric ulcers.1,13 In contrast, the selective COX-2 inhibitors are associated with a reduction in ulcers and related GI complications, but the benefit of celecoxib is less than that of rofecoxib and valdecoxib (see Table 20-3). The addition of aspirin to a selective COX-2 inhibitor reduces its ulcer-sparing benefit and increases ulcer risk.1,13 Aspirin and nonaspirin NSAIDs irreversibly inhibit platelet COX-1, resulting in decreased platelet aggregation and prolonged bleeding times, thereby increasing the potential for upper and lower GI bleeding.1,13,15Coadministration of selected NSAIDs may reduce the antiplatelet effects of aspirin.1315Clopidogrel and other medications that impair angiogenesis do not cause ulcers, per se, but may impair healing of gastric erosions leading to ulceration.13,15


Upper GI bleeding, perforation, and obstruction occur with H. pylori–associated and NSAID-induced ulcers and constitute the most serious, life-threatening complication of chronic PUD.1,23 Bleeding is caused by the erosion of an ulcer into an artery. It may be occult (hidden) and insidious or may present as melena (black-colored stools) or hematemesis (vomiting of blood). The use of NSAIDs (especially in older adults) is the most important risk factor for upper GI bleeding. Deaths occur primarily in patients who continue to bleed or in those patients who rebleed after the initial bleeding has stopped (see Upper GI Bleeding below).

Ulcer-related perforation into the peritoneal cavity is generally considered a surgical emergency.1,23 About one third to one half of perforated ulcers are associated with the use of NSAIDs, with the highest mortality reported in the elderly.23 The pain of perforation is usually sudden, sharp, and severe, beginning first in the epigastrium, but quickly spreading over the entire abdomen. Most patients experience ulcer symptoms prior to perforation. However, older patients who experience perforation in association with NSAID use may be asymptomatic. Penetration occurs when an ulcer burrows into an adjacent structure (pancreas, biliary tract, or liver) rather than opening freely into a cavity.

Gastric outlet obstruction is related to mechanical obstruction caused by scarring, muscular spasm, or edema of the duodenal bulb usually resulting from chronic ulceration.1,23 Symptoms occur over several months and include early satiety, bloating, anorexia, nausea, vomiting, and weight loss. Perforation, penetration, and gastric outlet obstruction occur most often with long-standing PUD.

Treatment of PUD has improved so that even the most virulent ulcers can be managed with medication. Intractability to drug therapy is an infrequent manifestation of PUD and an infrequent indication for surgery.


The clinical presentation of PUD varies depending on the severity of epigastric pain and the presence of complications (Table 20-5).1 Ulcer-related pain in duodenal ulcer often occurs 1 to 3 hours after meals and is usually relieved by food, but this is variable. In gastric ulcer, food may precipitate or accentuate ulcer pain. Antacids usually provide immediate pain relief in most ulcer patients. Pain usually diminishes or disappears during treatment; however, recurrence of epigastric pain after healing often suggests an unhealed or recurrent ulcer.

TABLE 20-5 Clinical Presentation of Peptic Ulcer Disease


The presence or absence of epigastric pain does not define an ulcer.1 Ulcer healing does not necessarily render the patient asymptomatic. Why symptoms remain is unclear, but it may relate to sensitization of afferent nerves in response to mucosal injury.1 Conversely, the absence of pain does not preclude an ulcer diagnosis especially in the elderly who may present with a “silent” ulcer complication. The reasons for this are unclear, but may relate to differences in the way the elderly perceive pain or the analgesic effect of NSAIDs.

Dyspepsia in itself is of little clinical value when assessing subsets of patients who are most likely to have an ulcer. Patients taking NSAIDs often report dyspepsia, but dyspeptic symptoms do not directly correlate with an ulcer. Individuals with dyspeptic symptoms may have either uninvestigated (no upper endoscopy) or investigated (underwent upper endoscopy) dyspepsia. If an ulcer is not confirmed in a patient with ulcer-like symptoms at the time of endoscopy, the disorder is referred to as NUD.9 Ulcer-like symptoms may occur in the absence of peptic ulceration in association with H. pylori gastritis or duodenitis. There is no one sign or symptom that differentiates between H. pylori–positive and NSAID-induced ulcer.


Routine laboratory tests are not helpful in establishing the diagnosis of PUD (see Table 20-5).1

Tests for H. pylori

The diagnosis of H. pylori infection can be made using endoscopic or nonendoscopic tests (Table 20-6).2,5,24 The tests that require upper endoscopy are invasive, more expensive, and usually require a mucosal biopsy for histology, culture, or detection of urease activity. At least four tissue samples are taken from specific areas of the stomach, as patchy distribution of H. pylori infection can lead to false-negative results. Because certain medications may decrease the sensitivity of rapid urease test, antibiotics and bismuth salts should be withheld for 4 weeks and proton pump inhibitors (PPIs) for 1 to 2 weeks prior to endoscopic testing.2,5 If the patient has been taking these medications, then a gastric biopsy for histology should be performed.5

TABLE 20-6 Tests for Detection of Helicobacter pylori


Two types of nonendoscopic tests are available: tests that identify active infection and tests that detect antibodies (see Table 20-6). Antibody tests do not differentiate between active infection and previously eradicated H. pylori. The nonendoscopic tests include the urea breath test (UBT), serologic antibody detection tests, and the fecal antigen test. These tests are less invasive, more convenient, and less expensive than the endoscopic tests.2,5,24

The UBT is the most accurate noninvasive test and is based on H. pylori urease activity.5 The 13Carbon (nonradioactive isotope) and 14Carbon (radioactive isotope) tests require that the patient ingest radiolabeled urea, which is then hydrolyzed by H. pylori (if present in the stomach) to ammonia and radiolabeled bicarbonate. The radiolabeled bicarbonate is absorbed in the blood and excreted in the breath. A mass spectrometer is used to detect 13Carbon, whereas 14Carbon is measured using a scintillation counter. The fecal antigen test is less expensive and easier to perform than the UBT, and may be useful in children.

Serologic tests are a cost-effective alternative for the initial diagnosis of H. pylori infection in the untreated patient.2,5 Antibodies to H. pylori usually develop about 3 weeks after infection and remain present after successful eradication.5 Therefore, serology should not be used to confirm H. pylori eradication.2,5 Office-based tests are less expensive, widely available, and provide rapid results, but the results are less accurate and more variable than the laboratory-based tests. Salivary and urine antibody tests are under investigation.2

Testing for H. pylori is only recommended if eradication is planned. Serologic antibody testing is a reasonable choice if endoscopy is not planned. The diagnostic accuracy of H. pylori tests for patients with an active bleeding ulcer has been questioned because of the potential for false-negative results. However, endoscopic biopsy-based tests such as the rapid urease test have a high degree of specificity in these patients (see Peptic Ulcer–Related Bleeding below).5

Confirmation of eradication is indicated posttreatment of active ulcers, previous ulcers, MALT lymphoma, endoscopic resection of gastric cancer, and uninvestigated dyspepsia, but routine testing for all patients is neither cost-effective nor practical.5 The decision to test posttreatment should be patient-specific and take into consideration the patient’s diagnosis, age, and ulcer history. The UBT and fecal antigen are preferred nonendoscopic tests to confirm H. pylori eradication but must be delayed at least 4 weeks after the completion of treatment to avoid confusing bacterial suppression with eradication. The term eradication or cure is used when posttreatment tests conducted 4 weeks after the end of treatment do not detect the organism. Quantitative antibody tests are impractical for posttreatment as antibody titers remain elevated for long periods of time. A negative posttreatment antibody test, however, is considered reliable.

Imaging and Endoscopy

The diagnosis of PUD depends on visualizing the ulcer crater by either upper GI radiography or upper endoscopy (see Table 20-5).1 In the past, radiography was the initial diagnostic procedure of choice because of its lower cost, greater availability, and greater safety. Today, upper endoscopy has replaced radiography because it provides a more accurate diagnosis and permits direct visualization of the ulcer.


The natural history of PUD is characterized by periods of exacerbations and remissions.1 Ulcer pain is usually recognizable and episodic, but symptoms are variable, especially in older adults and for patients taking NSAIDs. Antiulcer medications, including the histamine-2 receptor antagonists (H2RAs), PPIs, and sucralfate, relieve symptoms, accelerate ulcer healing, and reduce the risk of ulcer recurrence, but they do not cure the disease. Both duodenal and gastric ulcers recur unless the underlying cause (H. pylori or NSAID) is removed. Successful H. pylori eradication markedly decreases ulcer recurrence and complications. Prophylactic cotherapy or a COX-2 inhibitor decreases the risk of upper GI events for patients who are taking NSAIDs. GI bleeding, perforation, and obstruction remain troublesome complications of chronic PUD. Mortality for patients with gastric ulcer is slightly higher than in duodenal ulcer and the general population. The development of gastric cancer in H. pylori–infected individuals is a slow process that occurs over 20 to 40 years and is associated with a lifetime risk of less than 1%.2,8


Desired Outcome

The treatment of chronic PUD varies depending on the etiology of the ulcer (H. pylori or NSAID), whether the ulcer is initial or recurrent, and whether complications have occurred (Fig. 20-5). Overall treatment is aimed at relieving ulcer pain, healing the ulcer, preventing ulcer recurrence, and reducing ulcer-related complications. The goal of therapy for H. pylori–positive patients with an active ulcer, a previously documented ulcer, or a history of an ulcer-related complication is to eradicate H. pylori, heal the ulcer, and cure the disease. Successful eradication heals ulcers and reduces the risk of recurrence for most patients. The goal of therapy for a patient with an NSAID-induced ulcer is to heal the ulcer as rapidly as possible. Patients who are at high risk of developing NSAID ulcers should receive prophylactic cotherapy or be switched to a selective COX-2 inhibitor NSAID (if available) to reduce ulcer risk and related complications. When possible, the most cost-effective drug regimen should be used.


FIGURE 20-5 Algorithm. Guidelines for the evaluation and management of a patient who presents with dyspeptic or ulcer-like symptoms. (COX-2, cyclooxygenase-2; GERD, gastroesophageal reflux disease; H2RA, H2-receptor antagonist; NSAID, nonsteroidal antiinflammatory drug; NUD, nonulcer dyspepsia; PPI, proton pump inhibitor.)

General Approach to Treatment

Antimicrobials such as clarithromycin, metronidazole, amoxicillin, bismuth salts, and antisecretory drugs (PPIs or H2RAs) relieve ulcer symptoms, heal the ulcer, and eradicate H. pylori infection. PPIs are preferred to H2RAs or sucralfate for healing H. pylori–negative NSAID-induced ulcers because they accelerate ulcer healing and provide more effective relief of symptoms. Treatment with a PPI should be extended from 4 to 8-12 weeks if the NSAID must be continued. A PPI-based H. pylori eradication regimen is recommended when the patient with an active ulcer is taking an NSAID and is H. pylori positive. Prophylactic cotherapy with either a PPI or misoprostol decreases ulcer risk and upper GI complications for patients taking nonselective NSAIDs. Selective COX-2 inhibitor NSAIDs (if available) may be used as an alternative to a nonselective NSAID, but their beneficial GI effect when taken with low-dose aspirin is negated and their association with adverse cardiovascular effects reduces their usefulness.

Dietary modifications are important for patients who are unable to tolerate certain foods and beverages. Lifestyle modifications such as reducing stress and stopping cigarette smoking are encouraged. Surgery may be necessary for patients with ulcer-related complications.

Nonpharmacologic Therapy

Images Patients with PUD should eliminate or reduce psychological stress, cigarette smoking, and the use of NSAIDs (including aspirin). Although there is no “antiulcer diet,” the patient should avoid foods and beverages (e.g., spicy foods, caffeine, and alcohol) that cause dyspepsia or that exacerbate ulcer symptoms. If possible, alternative agents such as acetaminophen or nonacetylated salicylate (e.g., salsalate) should be used for relief of pain. Elective surgery for PUD is rarely performed today because of highly effective medical management. A subset of patients, however, may require emergency surgery for bleeding, perforation, or obstruction. In the past, surgical procedures were performed for medical treatment failures and included vagotomy with pyloroplasty or vagotomy with antrectomy.23 Vagotomy (truncal, selective, or parietal cell) inhibits vagal stimulation of gastric acid. A truncal or selective vagotomy frequently results in postoperative gastric dysfunction and requires a pyloroplasty or antrectomy to facilitate gastric drainage. When an antrectomy is performed, the remaining stomach is anastomosed with the duodenum (Billroth I) or with the jejunum (Billroth II). A vagotomy is unnecessary when an antrectomy is performed for gastric ulcer. Postoperative consequences include postvagotomy diarrhea, dumping syndrome, anemia, and recurrent ulceration.

Pharmacologic Therapy


Images Table 20-7 presents guidelines for the eradication of infection in H. pylori–positive individuals. Table 20-8 lists regimens used to eradicate H. pylori infection.

TABLE 20-7 Guidelines for the Eradication of Helicobacter pylori Infection


TABLE 20-8 Drug Regimens Used to Eradicate Helicobacter pylori


Images First-line therapy is usually initiated with a PPI-based three-drug regimen for 10 to 14 days. If a second course of treatment is required, the PPI-based three-drug regimen should contain different antibiotics or a four-drug regimen with a bismuth salt, metronidazole, tetracycline, and a PPI should be used.

Patients with NSAID-induced ulcers should be tested to determine their H. pylori status. If H. pylori positive, treatment should be initiated with a PPI-based three-drug regimen. If H. pylori negative, the NSAID should be discontinued, and the patient treated with a PPI, H2RA, or sucralfate (see Table 20-9). If the NSAID is continued, treatment should be initiated with a PPI (if H. pylori negative) or with a PPI-based three-drug regimen (if H. pyloripositive). Cotherapy with a PPI or misoprostol or switching to a selective COX-2 inhibitor (if available) is recommended for patients at risk of developing an ulcer-related complication.

TABLE 20-9 Drug Dosing Table


Maintenance therapy with a PPI or H2RA should be limited to high-risk patients with ulcer complications, patients who fail eradication, and those with H. pylori–negative ulcers. Treatment failure is associated with poor medication adherence, antimicrobial resistance, NSAID use, cigarette smoking, acid hypersecretion, or tolerance to the antisecretory effects of an H2RA.

Treatment of H. pylori–Positive Ulcers

This chapter focuses on the eradication of H. pylori in adults.2529 A discussion of the treatment of H. pylori infection in children is found elsewhere.30,31

The treatment of H. pylori–positive PUD should be effective, well tolerated, easy to adhere to, and cost-effective. Historically, none of these factors have been addressed in a systematic way making it difficult to identify the best evidence-based treatment regimens.1 Successful eradication depends on the drug regimen, resistance to the antibiotics used, duration of therapy, medication adherence, and genetic polymorphism.2529 H. pylori regimens should have eradication (cure) rates of at least 80% based on intention-to-treat analysis or at least 90% based on per-protocol analysis, and they should minimize the potential for antimicrobial resistance.1,25,29 Not one antibiotic, bismuth salt, or antiulcer drug achieves this goal, but clarithromycin is the single most effective antibiotic. Two-drug regimens that combine a PPI and either amoxicillin or clarithromycin have yielded marginal and variable eradication rates in the United States and are not recommended.1,5 In addition, the use of only one antibiotic is associated with a higher rate of antimicrobial resistance.

Drug regimens (see Table 20-8) that combine an antisecretory drug with two antibiotics (triple therapy) or with two antibiotics and a bismuth salt (quadruple therapy) usually increase eradication rates to acceptable levels and reduce the risk of antimicrobial resistance.5,2529 When selecting an initial eradication regimen, an antibiotic combination should be used that permits second-line treatment (if necessary) with different antibiotics. The antibiotics that have been most extensively studied and found to be effective in various combinations include clarithromycin, amoxicillin, metronidazole, and tetracycline.1Because of insufficient data, ampicillin should not be substituted for amoxicillin, doxycycline should not be substituted for tetracycline, and azithromycin or erythromycin should not be substituted for clarithromycin. Antisecretory drugs enhance antibiotic activity and stability by increasing intragastric pH and by decreasing intragastric volume thereby enhancing the topical antibiotic concentration.27

Proton Pump Inhibitor–Based Three-Drug Regimens

PPI-based triple therapy (see Table 20-8) is the initial treatment of choice for eradicating H. pylori (see Table 20-7).5,2529 The regimens that combine either clarithromycin and amoxicillin or clarithromycin and metronidazole are more effective than the amoxicillin–metronidazole regimen. In most cases, increasing the antibiotic dosage does not improve eradication rates. The clarithromycin–amoxicillin regimen is preferred initially (see Table 20-7), but metronidazole should be substituted for amoxicillin for penicillin-allergic patients unless alcohol is consumed.5,2527 Unfortunately, eradication rates for PPI-based triple therapy have declined substantially in recent years in North America and Europe due primarily to an increase in clarithromycin-resistant H. pylori strains (see Factors that Predict H. pylori Eradication Outcomes below).5.2527 Other antibiotics and antibiotic combinations have been investigated, but these regimens should not be used as initial treatment in the United States until well-designed trials confirm their effectiveness.5,27

The recommended duration of therapy in the United States is 10 to 14 days, but the 14-day regimen is preferred in light of the decreasing eradication rate with the PPI-based triple-therapy regimens containing clarithromycin.5Although a 7-day course has been approved by the FDA and is used in Europe, the longer treatment periods favor higher eradication rates and are less likely to be associated with antimicrobial resistance.5,2527

Clinical Controversy…

Although clinical guidelines recommend a 10- or 14-day treatment course, some clinicians favor an initial 7-day H. pylori regimen. These clinicians believe that the shorter treatment period enhances the compliance of a complicated drug regimen.

The PPI is an integral part of the three-drug regimen and should be taken 30 to 60 minutes before a meal along with the two antibiotics (see Table 20-8).5,32 Prolonged PPI treatment beyond 2 weeks after eradication is usually not necessary for ulcer healing. A single daily dose of a PPI may be less effective than a twice-daily dose.33 Substitution of one PPI for another is acceptable and does not enhance or diminish H. pylori eradication.32,34 An H2RA should not be substituted for a PPI, as H2RA is associated with lower eradication rates.35,36 Pretreatment with a PPI does not influence H. pylori eradication.37

Bismuth-Based Four-Drug Regimens

Bismuth-based quadruple therapy (see Table 20-8) is recommended as an alternative first-line eradication therapy (see Table 20-7) for those allergic to penicillin.5,2529 Although this regimen may be used initially, it is often reserved as a second-line therapy after treatment failure with the PPI-based clarithromycin–amoxicillin regimen (see Eradication of H. pylori After Initial Treatment Failure below). Eradication rates for bismuth-based quadruple therapy (bismuth salicylate, metronidazole, tetracycline, and either a PPI or H2RA) are similar to those achieved with PPI-based triple therapy.5,27,38 Eradication rates are comparable when bismuth subcitrate potassium (biskalcitrate) is substituted for bismuth subsalicylate (see Table 20-8).39 Substitution of clarithromycin 250 to 500 mg four times a day for tetracycline yields similar results but increases adverse effects. Bismuth salts have a topical antimicrobial effect.1 The antisecretory drug hastens ulcer healing and relieves pain in patients with an active ulcer. All medications except the PPI should be taken with meals and at bedtime.

The original bismuth-based regimens contained an H2RA in place of a PPI, but a meta-analysis indicated that quadruple therapy with a PPI provides greater efficacy and permits a shorter treatment duration (7 days) when compared with the H2RA-based regimens (10 to 14 days).40 However, a 10- to 14-day duration is recommended in the United States as it generally provides higher eradication rates.5 When treating an active ulcer, the antisecretory drug is usually continued for 2 (PPI) to 4 (H2RA) weeks after stopping bismuth and antibiotics.

Bismuth-based quadruple therapy is the treatment of choice when medication costs are of overriding importance. However, major concerns include a four-times-a-day dosing regimen (see Table 20-8), poor medication adherence, and frequent adverse effects. Although minor adverse effects are more common, the frequency of moderate or severe adverse effects is similar to those reported for the PPI-based triple therapy.41

Sequential Therapy Sequential therapy is a new form of eradication therapy whereby the antibiotics are administered in a sequence rather than all together.5,26,29 The rationale for sequential therapy is to initially treat with antibiotics that rarely promote resistance (e.g., amoxicillin) to reduce the bacterial load and preexisting resistant organisms and then to follow with different antibiotics (e.g., clarithromycin and metronidazole) to kill the remaining organisms.1Treatment consists of a PPI and amoxicillin for 5 days followed by a PPI, clarithromycin, and tinidazole (or metronidazole) for an additional 5 days (see Table 20-8).5,26,42 Although this regimen has achieved eradication rates that are superior to the PPI-based three-drug regimens containing clarithromycin,42 the regimen requires a change in medication midtreatment, which may contribute to nonadherence.43 The advantages of sequential therapy need to be confirmed in the United States before it can be recommended as a first-line H. pylori eradication therapy (see Table 20-7).5,26,27

Eradication of H. pylori After Initial Treatment Failure

H. pylori eradication is often more difficult after initial treatment fails and successful eradication after retreatment is extremely variable.5,44 Treatment failures should be referred to a gastroenterologist for further diagnostic evaluation. Second-line (salvage) treatment should (a) use antibiotics that were not previously used during initial therapy; (b) use antibiotics that are not associated with resistance; (c) use a drug that has a topical effect such as bismuth; and (d) extend the duration of treatment to 14 days.5,44,45 The most commonly used second-line therapy, after unsuccessful initial treatment with a PPI–amoxicillin–clarithromycin regimen, is a 14-day course of the PPI-based quadruple therapy (see Table 20-8).5,26,27,45 A levofloxacin-containing regimen (see Table 20-8) may be an alternative second-line eradication regimen and may be better tolerated than PPI-based quadruple therapy (see Table 20-7).46 Additionally, the levofloxacin regimen may serve as an alternative to PPI–clarithromycin–metronidazole usually recommended in penicillin-allergic patients.47 However, concerns about using fluoroquinolones for H. pylorieradication are related to the development of resistance and adverse effects such as tendonitis and hepatotoxicity.26 Other second-line regimens that include rifabutin and furazolidone are discussed elsewhere.5,2527

Factors that Predict H. pylori Eradication Outcomes

Factors that predict H. pylori eradication outcomes include antibiotic resistance, poor medication adherence, short duration of therapy, CagA status, high bacterial load, low intragastric pH, and genetic polymorphism.5,45,4850Medication adherence decreases with multiple medications, increased frequency of administration, intolerable adverse effects, and costly drug regimens. One meta-analysis reported that CYP2C19 polymorphism may alter the effect of PPIs on gastric acid secretion thereby influencing eradication outcomes.49 Tolerability varies with different regimens.1,5 Common adverse effects include nausea, vomiting, abdominal pain, diarrhea, and taste disturbances (metronidazole and clarithromycin). Adverse effects with metronidazole are dose related (especially when >1 g/day) and include a disulfiram-like reaction with alcohol. Tetracycline may cause photosensitivity and should not be used in children because of possible tooth discoloration. Bismuth salts may cause darkening of the stool and tongue. Antibiotic-associated diarrhea and Clostridium difficile–associated disease can occur. Oral thrush and vaginal candidiasis may also occur.

An important predictor of H. pylori eradication is the presence or absence of resistant microorganisms.5,5052 U.S. data from 1993 to 1999 report resistance rates among H. pylori strains for metronidazole (37%), clarithromycin (10%), and amoxicillin (1.4%).51 Data from 1998 to 2002 reveal rates of 25% for metronidazole, 13% for clarithromycin, and 0.9% for amoxicillin.52 In one study, the proportion of clarithromycin resistance increased with increasing courses of macrolide antibiotics, from 7% resistance with no prior exposure to 80% resistance with ≥5 courses of macrolides. It is possible that the increased rate of clarithromycin resistance partially explains the decrease in efficacy of clarithromycin-containing regimens. The clinical importance of metronidazole resistance remains uncertain, as resistance can be overcome by using higher dosages and combining metronidazole with other antibiotics.5 Resistance to tetracycline and amoxicillin is uncommon.5 Resistance to bismuth has not been reported. The role of antibiotic sensitivity testing prior to initiating H. pylori treatment has not been established.


Probiotics (e.g., strains of Lactobacillus and Bifidobacterium) and foodstuffs (e.g., cranberry juice and some milk proteins) with bioactive components have been used proactively to control H. pyloricolonization in at-risk individuals and, when taken as a supplement to eradication therapy, may have a role in improving H. pylori eradication and reducing the adverse effects of PPI-based triple therapy.5355However, the administration of probiotics alone does not eradicate H. pylori infection. In the future, the regular intake of probiotics may constitute a low-cost alternative for individuals who are at risk for H. pylori infection and, in combination with antibiotics, augment eradication rates. These preliminary data are encouraging and warrant more research in this area.

Treatment of NSAID-Induced Ulcers

Nonselective NSAIDs should be discontinued (when possible) on confirmation of an active ulcer. If the NSAID is stopped, most uncomplicated ulcers heal with standard regimens of an H2RA, PPI, or sucralfate (see Table 20-9).1,13,29 However, PPIs are usually preferred because they provide more rapid symptom relief and ulcer healing. If the NSAID is continued despite ulceration, consideration should be given to reducing the NSAID dose, switching to acetaminophen or a nonacetylated salicylate, or using a more selective COX-2 inhibitor (see Table 20-3). PPIs are the drugs of choice when the NSAID is continued, as potent acid suppression is required to accelerate ulcer healing.1,13,29 If the ulcer is H. pylori positive, eradication should be initiated with a regimen that contains a PPI.1,13,29

Strategies to Reduce the Risk of NSAID Ulcer and GI Complications There are three therapeutic approaches to reducing the risk of NSAID ulcers and related upper GI complications (see Table 20-10). Medical cotherapy with either a PPI or misoprostol decreases ulcer risk and GI complications in high-risk patients.1214,16,19,29 The use of a selective COX-2 inhibitor instead of a nonselective NSAID also decreases risk of ulcers and upper GI events.1214,16,18,19,29 Unfortunately, these strategies do not completely eliminate ulcers and complications for patients at the “highest risk.” When selecting a gastroprotective strategy, the GI benefits must be balanced against the cardiovascular risks associated with selective COX-2 inhibitor NSAIDs, nonselective NSAIDs, and concomitant antiplatelet therapy.1216 Strategies aimed at reducing the topical irritant effects of nonselective NSAIDs, for example, prodrugs, slow-release formulations, and enteric-coated products, do not prevent ulcers or GI complications.

TABLE 20-10 Drug Monitoring Table


Misoprostol Cotherapy Misoprostol, 200 mcg orally four times per day, reduces the risk of NSAID-induced gastric and duodenal ulcer, and related upper GI complications, but diarrhea and abdominal cramping limit its use.12,16,56Because a dosage of 200 mcg three times per day is comparable in efficacy to 800 mcg/day, the lower dosage should be considered for patients unable to tolerate the higher dose.12,16Reducing the misoprostol dosage to 400 mcg/day or less to minimize diarrhea compromises its gastroprotective effects. A fixed combination of misoprostol 200 mcg and diclofenac (50 or 75 mg) may enhance compliance, but the flexibility to individualize drug dosage is lost. A large clinical trial in rheumatoid arthritis patients provided the most compelling evidence that misoprostol reduces the risk of upper GI complications for high-risk patients.57

Proton Pump Inhibitor Cotherapy Images PPI cotherapy reduces NSAID-related gastric and duodenal ulcer risk and is better tolerated than misoprostol.1214,16,56 All PPIs are effective when used in standard dosages (see Table 20-9). Although head-to-head comparative trials are few, there are limited data to indicate that PPIs are superior to standard H2RA dosages.12,13,16 When lansoprazole (15 or 30 mg/day) was compared with misoprostol 800 mcg/day or placebo, both dosages of lansoprazole and misoprostol effectively reduced ulcer recurrence, although the PPI was better tolerated.58 A greater proportion of those in the misoprostol group reported treatment-related adverse events and withdrew early from the study. Results from observational studies and meta-analyses indicate the PPIs reduce the risk of NSAID-related ulcer bleeding.12,16,19,59

H2-Receptor Antagonist Cotherapy Standard H2RA dosages (e.g., famotidine 40 mg/day) are effective in reducing NSAID-related duodenal ulcer but not gastric ulcer (the most frequent type of ulcer associated with NSAIDs).12,13,16 Higher dosages (e.g., famotidine 40 mg twice daily, ranitidine 300 mg twice daily) may reduce the risk of gastric and duodenal ulcer, but studies comparing double dosages with PPIs or misoprostol are not available.12,13 One study suggests that famotidine 20 mg twice daily may be an alternative to PPIs for patients taking low cardioprotective dosages of aspirin, but additional studies are required to confirm these findings.60 The H2RAs are not recommended as prophylactic cotherapy because it is likely that they are not as effective as the PPIs or misoprostol in preventing NSAID-induced gastric ulcer and related GI complications.16 An H2RA, however, may be used to relieve NSAID-related dyspepsia.

Cyclooxygenase-2 Inhibitors Two large outcome trials have compared celecoxib61 and rofecoxib62 with nonselective NSAIDs. Patients in the Celecoxib Long-Term Arthritis Safety Study (CLASS) trial who were taking celecoxib and required cardioprotection (antiplatelet effects of aspirin) were permitted to take low-dose aspirin.61 Although a 6-month analysis found a nonsignficant reduction in ulcer complications with celecoxib when compared with ibuprofen and diclofenac, results after 1 year found no difference between the groups. Today, celecoxib is not considered a selective COX-2 inhibitor (Table 20-3) by the FDA as it contains the same GI warnings as the nonselective and partially selective NSAIDs.63 A post hoc analysis confirmed that any gastroprotective benefits of celecoxib were negated in aspirin users. Similar effects have been observed with rofecoxib. Additionally, an increased number of nonfatal myocardial infarctions and thrombotic stroke were observed in studies of rofecoxib leading to its withdrawal from the market.64 Subsequently, valdecoxib was withdrawn from the market amid concerns about cardiovascular risk.

Cardiovascular safety was also evaluated in the CLASS trial, but serious cardiovascular thromboembolic events were no different between celecoxib and the comparative nonselective NSAIDs. In contrast, the results of a meta-analysis of randomized trials of COX-2 inhibitor NSAIDs reported a dose-dependent increase in cardiovascular events with all COX-2 inhibitor NSAIDs, including celecoxib.65 Increased cardiovascular risk appears to be dependent on a number of factors including increased COX-2 selectivity, higher dosages, and a longer duration of treatment.12,18,19 Thus, the lowest effective celecoxib dose should be used for the shortest duration of time. Dyspepsia and abdominal pain, fluid retention, hypertension, and renal toxicity are associated with the COX-2 inhibitors and nonselective NSAIDs.1

COX-2 Inhibitor Versus NSAID Plus PPI ImagesThere are limited data that suggest that for high-risk, H. pylori–negative patients, a COX-2 inhibitor NSAID may be as beneficial as a nonselective NSAID plus a PPI in reducing NSAID-related ulcer complications.12,18,19 However, neither the COX-2 inhibitor NSAID nor the NSAID plus a PPI will eliminate upper GI events for these patients. Combining a COX-2 inhibitor NSAID with a PPI may be considered for very high-risk patients, but this regimen is likely to be of modest benefit.12,18

GI and Cardiovascular Safety Issues

There is no difference in cardiovascular risk between the selective COX-2 inhibitor NSAIDs and the nonselective or partially selective NSAIDs, with the exception of naproxen.12,18,65 Thus, individual patient risk factors for NSAID-related GI bleeding and cardiovascular events must be weighed when determining treatment (see Table 20-11). Naproxen is preferred compared with other nonselective NSAIDs and COX-2 inhibitors because of its comparative cardiovascular safety and not because of its GI safety profile. There is insufficient evidence regarding the preferred NSAID for patients also taking low-dose aspirin.14 Clopidogrel should not be substituted for low-dose aspirin in order to reduce recurrent GI bleeding as it is inferior to a PPI plus low-dose aspirin.13 Despite limited evidence to suggest an interaction via the hepatic cytochrome P450 (CYP450) pathway, combining a PPI and clopidogrel with or without low-dose aspirin results in less GI bleeding.13 Ongoing studies for patients with cardiovascular disease should provide the necessary information to help resolve these issues. The lowest possible daily dose of a COX-2 inhibitor should be used as the cardiovascular risk may be dose dependent. However, no studies, to date, have evaluated the safety of low-dose COX-2 inhibitor NSAIDs for patients with or at risk for cardiovascular disease. In the future, there will be new formulations and classes of NSAIDs and COX-2 inhibitors with an improved GI and cardiovascular safety profile.66 Until then patients who take NSAIDs or COX-2 inhibitors should be counseled about the signs and symptoms of upper GI bleeding and major cardiovascular events and what they should do if they occur.

TABLE 20-11 Guidelines for Reducing GI Risk for Patients Receiving Chronic NSAID Therapy


Treatment of Non–H. pylori, Non-NSAID Ulcers

Very few individuals have non–H. pylori, non-NSAID (idiopathic) ulcers.29,67,68 Patients should be double-checked to verify that they are H. pylori negative and that they are not taking ulcerogenic medications. Possible explanations for non–H. pylori, non-NSAID ulcers include gastric hypersecretion, gastric outlet obstruction, genetic predisposition, concomitant diseases (see Table 20-2), and heavy tobacco use. Treatment should be initiated with conventional ulcer healing therapy (see Table 20-9). Although standard H2RA or sucralfate dosage regimens heal the majority of gastric and duodenal ulcers in 6 to 8 weeks, PPIs provide comparable ulcer healing rates in 4 weeks.32 A higher daily dose or a longer treatment duration is sometimes needed to heal larger gastric ulcers. Antacids are not used as single agents to heal ulcers because of the high volume and frequent doses required. When conventional antiulcer therapy is discontinued after ulcer healing, most patients develop a recurrent ulcer within 1 year.32Maintenance therapy may be required to prevent ulcer recurrence.

Long-Term Maintenance of Ulcer Healing

Continuous antiulcer therapy is aimed at the long-term maintenance of ulcer healing and the prevention of ulcer-related complications. Because H. pylori eradication dramatically decreases ulcer recurrence, continuous maintenance therapy is primarily used to treat high-risk patients who failed H. pylori eradication, have a history of ulcer-related complications, have frequent recurrences of H. pylori–negative ulcers, and are heavy smokers or NSAID users. For most patients, standard maintenance dosages (see Table 20-9) are effective.32

Treatment of Refractory Ulcers

Ulcers are considered refractory to therapy when symptoms, ulcers, or both persist beyond 8 to 12 weeks despite conventional treatment or when several courses of H. pylori eradication fail.1,45 Poor patient compliance, antimicrobial resistance, cigarette smoking, NSAID use, gastric acid hypersecretion, or tolerance to the antisecretory effects of an H2RA (see Antiulcer Agents below) may contribute to refractory PUD. Patients with refractory ulcers should undergo upper endoscopy to confirm a nonhealing ulcer, exclude malignancy, and assess H. pylori status. H. pylori–positive patients should receive eradication therapy (see Treatment of H. pylori–Positive Ulcers above). In H. pylori–negative patients, higher PPI dosages (e.g., omeprazole 40 mg/day) heal the majority of ulcers. Continuous treatment with a PPI is often necessary to maintain healing, as refractory ulcers recur when therapy is discontinued or the dose is reduced. Switching from one PPI to another is not beneficial. Patients with refractory gastric ulcer may require surgery because of the possibility of malignancy.

Antiulcer Agents

Proton Pump Inhibitors The PPIs (omeprazole, esomeprazole, lansoprazole, dexlansoprazole, rabeprazole, and pantoprazole) dose-dependently inhibit basal and stimulated gastric acid secretion.32 When PPI therapy is initiated, the degree of acid suppression increases over the first 3 to 4 days of therapy, as more proton pumps are inhibited.32 Because PPIs inhibit only those proton pumps that are actively secreting acid, they are most effective when taken 30 to 60 minutes before meals.32 The duration of acid suppression is a function of binding to the H+/K+-adenosine triphosphatase (ATPase) enzyme and is longer than their elimination half-lives.32 Symptomatic acid rebound on withdrawal of a PPI has been reported in healthy volunteers after 8 weeks of treatment.1,69

Various PPI dosage forms and formulations exist (see Table 20-12) and include the delayed-release enteric-coated dosage forms that have pH-sensitive granules contained in gelatin capsules (omeprazole, esomeprazole, prescription and nonprescription lansoprazole, and dexlansoprazole), rapidly disintegrating tablets (lansoprazole), and delayed-release enteric-coated tablets (rabeprazole, pantoprazole, and nonprescription omeprazole).32 The pH-sensitive enteric coating prevents degradation and premature protonation of the drug in stomach acid but dissolves at a higher pH in the duodenum where the drug is absorbed. Dexlansoprazole is formulated with a dual-release mechanism that results in inhibition of proton pumps that become activated after initial release of the medication.70,71 Omeprazole is also available as an immediate-release formulation (oral suspension, oral capsule) containing sodium bicarbonate, which raises intragastric pH and protects omeprazole from acid degradation in the stomach thus permitting rapid absorption from the duodenum.72 IV products available in the United States include pantoprazole and esomeprazole.

TABLE 20-12 Proton Pump Inhibitor Formulations and Options for Administration


Five of the PPIs provide similar rates of ulcer healing (dexlansoprazole has not been labeled at this time for these indications), maintenance of ulcer healing, and symptom relief when used in recommended dosages (see Table 20-9). When higher dosages are indicated, the daily dose should be divided in order to obtain better 24-hour control of intragastric pH. A dosage reduction is unnecessary for patients with renal impairment or in older adults but should be considered in severe hepatic disease.32 The short-term adverse effects of the PPIs are similar to those observed with the H2RAs and include headache, nausea, and abdominal pain.32 Because the immediate-release formulations contain sodium bicarbonate, they are contraindicated for patients with metabolic alkalosis and hypokalemia.72 The sodium should also be taken into consideration for patients who are on sodium-restricted diets, for example, congestive heart failure.

Drug Interactions All PPIs increase intragastric pH and may alter the bioavailability of orally administered drugs, such as ketoconazole (weak bases) and digoxin, or pH-dependent dosage forms.32,73 This interaction is especially important with atazanavir, a protease inhibitor. Concomitant use with a PPI can significantly reduce the oral bioavailability of atazanavir and potentially lead to therapeutic failure and viral resistance in patients infected with HIV.74Omeprazole and esomeprazole selectively inhibit the hepatic CYP2C19 pathway and may decrease the elimination of phenytoin, warfarin, diazepam, and carbamazepine.32 The PPIs may increase the metabolic clearance and decrease the GI absorption of levothyroxine resulting in increased thyroid-stimulating hormone levels and a corresponding increase in the levothyroxine dose.75 Clinically significant drug interactions with PPIs are rare and usually do not constitute a major clinical risk.76,94

Clinical Controversy…

Some clinicians believe that the antiplatelet effect of clopidogrel is attenuated by omeprazole resulting in an increased risk of adverse cardiac events when these medications are taken concomitantly. Strategies to avoid this interaction for patients at risk of NSAID-related GI events include switching omeprazole to another PPI or substituting an H2RA for the PPI. Others believe that the use of a PPI (if indicated) and clopidogrel is not a safety concern and that it is not necessary to switch omeprazole to another PPI.

One of the most perplexing potential PPI drug interactions is with the antiplatelet drug clopidogrel. This interaction is especially important given recent consensus guidelines that recommend the use of a PPI for high-risk patients on antiplatelet therapies to prevent ulcers and related GI bleeding.13 Clopidogrel, a prodrug, is converted to its active form through CYP2C19. PPIs may attenuate the antiplatelet effect of clopidogrel by inhibiting or competing for this metabolic pathway. FDA safety guidelines recommend that the coadministration of omeprazole, omeprazole/sodium bicarbonate, or esomeprazole with clopidogrel be avoided because they reduce the effectiveness of clopidogrel.7779 Details of new studies performed by the manufacturer and submitted to the FDA, as well as warnings regarding omeprazole, esomeprazole, and other interacting drugs (e.g., cimetidine), are contained in the clopidogrel package insert.80 A reduced antiplatelet effect of clopidogrel may also result from genetic polymorphisms of the CYP2C19 pathway leading to decreased biotransformation of the drug to its active form.81,82 Whether the use of other PPIs such as pantoprazole, lansoprazole, dexlansoprazole, and rabeprazole interacts with clopidogrel remains uncertain as the capacity to inhibit CYP2C19 varies among these PPIs.83,84 While some reports suggest a “class effect” among the different PPIs, other pharmacodynamic studies suggest an interaction with omeprazole and esomeprazole but not with pantoprazole.83,84 Because of the uncertainty of the effect of the PPI–clopidogrel interaction on clinical outcomes and the extent to which other PPIs may interact with clopidogrel, caution is warranted. The only randomized double-blind trial comparing clopidogrel with or without omeprazole yielded no apparent increase in cardiovascular events due to clopidogrel and omeprazole cotherapy; however, there was a significant reduction in the rate of upper GI bleeding. This trial has been criticized by many due to the low number of cardiovascular events, formulation of omeprazole used, and premature termination of the study due to loss of funding by the sponsor. Patients should have an acceptable indication for a PPI recognizing that risk versus benefit must be weighed on an individual basis. If a PPI is absolutely necessary with clopidogrel, many clinicians continue to avoid the use of omeprazole and esomeprazole.

Potential Risks and Long-Term Safety Issues Numerous potential risks and safety issues (see Table 20-13) have been associated with the long-term use of PPIs as a consequence of prolonged hypergastrinemia and chronic hypochlorhydria.32,8588 In most cases, causality is difficult to ascertain because of the study design, confounding variables, and subject selection. All of the PPIs dose-dependently increase serum gastrin concentrations twofold to fourfold as a function of their potent acid-inhibitory effect.32,85 Fasting gastrin elevations are usually within the normal range and return to baseline within 1 month of discontinuing the drug. In humans, PPIs may lead to enterochromaffin-like (ECL) hyperplasia as a result of the hypergastrinemia, but there is no evidence that these changes result in dysplasia, carcinoid tumors, or gastric adenocarcinoma.85,86 Although long-term PPI therapy in H. pylori–positive individuals is associated with progressive atrophic gastritis, there are insufficient data to link chronic PPI use with gastric cancer in H. pylori–positive patients.85,86 Despite theoretical and in vitro data, there is no evidence to support an association between PPIs and colonic polyps or colorectal cancer.85,86 Bacterial overgrowth occurs in the stomach as a consequence of hypochlorhydria and may lead to carcinogenic N-nitroso compounds in animals but is unlikely to result in significant gastric nitrosation in humans.85,88

TABLE 20-13 Potential Risks and Safety Issues Associated with the Proton Pump Inhibitors


Chronic PPI therapy may be associated with an increased risk of infection and nutritional deficiences.8587 Gastric acid (low stomach pH) plays an important role in the defense against bacterial colonization of the stomach and in nutrient absorption. Acid suppression has been implicated as a risk factor for community-acquired pneumonia (CAP) and enteric infections (C. difficileSalmonellaCampylobacter).86Three case-controlled studies demonstrate a higher adjusted relative risk of CAP for patients currently using PPIs compared with controls.8991 The results of these retrospectively designed studies, however, need to be interpreted cautiously because of the variability in the length of therapy for current PPI users and the inclusion of older (>60 years of age) patients with concomitant comorbidities. A systematic review of the literature has linked PPIs with various enteric infections, but the most convincing data were with C. difficile.92 It is likely that sustained elevations in intragastric pH facilitate the survival of C. difficile spores. However, the magnitude of risk varies and causality is difficult to establish. The risk of various infections associated with PPI therapy cannot be firmly established until the results of large prospective studies are made available.

The absorption of vitamin B12, dietary iron, and calcium requires an acidic environment and may be adversely affected by PPI-induced prolonged hypochlorhydria (see Table 20-10).86 Although a few studies have investigated the long-term use of PPIs on vitamin B12 and iron absorption, the clinical importance of their effect on absorption has not been established, and monitoring of B12 and iron levels cannot be recommended.86 However, adequate supplementation and monitoring should be considered in high-risk populations (e.g., older patients, vegetarians, alcoholism) who may be already depleted.85,86 High PPI dosage and long-term therapy have been associated with an increased risk of hip, wrist, and spine fractures related to reduction in calcium absorption.26,93 Although the results of the studies vary, the FDA has revised the warnings and precautions of prescription and nonprescription PPIs to reflect this potential risk.93 Bone density tests for osteoporosis screening, calcium supplementation, or other precautions cannot be recommended solely based on chronic PPI therapy.94 However, it is appropriate to screen and treat older patients for osteoporosis regardless of whether they are receiving long-term PPI therapy.

On the basis of numerous case reports of hypomagnesemia, the FDA has revised the warnings and precautions of prescription and nonprescription PPIs. Hypomagnesemia, both symptomatic and asymptomatic, has been reported with serious adverse events including tetany, arrhythmias, and seizures (see Table 20-10). In most cases it occurs in patients taking PPIs more than 1 year, but can occur with as little as 3 months of therapy.

H2-Receptor Antagonists Ulcer healing is comparable among H2RAs (cimetidine, famotidine, nizatidine, and ranitidine) with equipotent multiple daily doses or a single full dose given after dinner or at bedtime (see Table 20-9), but tolerance to their antisecretory effect may occur.95 Twice-daily administration suppresses daytime acid and benefits patients with daytime ulcer pain. Cigarette smokers may require higher doses or a longer duration of treatment. H2RAs are eliminated renally and therefore a dosage reduction is recommended for patients with moderate-to-severe renal failure.96 The short- and long-term safety of all four H2RAs is similar.1 Thrombocytopenia, the most common hematologic adverse effect, is reversible and occurs with all four H2RAs (see Table 20-10). However, the propensity for H2RAs to cause thrombocytopenia is likely overestimated.97 Cimetidine inhibits several CYP450 isoenzymes, resulting in numerous drug interactions (e.g., theophylline, lidocaine, phenytoin, warfarin, and clopidogrel).77,96Ranitidine has less potential for hepatic CYP450 drug interactions, while famotidine and nizatidine do not interact with drugs metabolized by the hepatic CYP450 pathway.96 The H2RAs decrease acid secretion and may alter the bioavailability of orally administered drugs, similar to that seen with the PPIs.

Sucralfate Sucralfate heals peptic ulcers, but is not widely used today for this indication.1 Deterrents to its use include the requirement for multiple doses per day, large tablet size, and the need to separate the drug from meals and potentially interacting medications. Drug interactions can be minimized by giving the interacting drug at least 2 hours before sucralfate. Alternative therapy is warranted for patients taking oral fluoroquinolones. Constipation may be troublesome especially in older individuals. Seizures may occur in dialysis patients taking aluminum-containing antacids. Hypophosphatemia may develop with long-term treatment. Gastric bezoar formation has also been reported (see Table 20-10).

Prostaglandins Misoprostol, a synthetic PGE1 analogue, moderately inhibits acid secretion and enhances mucosal defense.1,98,99 Antisecretory effects are dose dependent over the range of 50 to 200 mcg; cytoprotective effects occur in humans at doses of greater than 200 mcg. Because protective effects occur at higher doses, it is difficult to establish the protective effect independent of the antisecretory action. A dose of 200 mcg four times daily or 400 mcg twice daily (although not recommended in the United States) heals duodenal ulcers and gastric ulcers comparable to standard H2RA or sucralfate regimens. Diarrhea, the most troublesome adverse effect, is dose dependent and develops in 10% to 30% of patients.1,98,99 Abdominal cramping, nausea, flatulence, and headache typically accompany the diarrhea. Taking the drug with or after meals and at bedtime may minimize the diarrhea (see Table 20-10). Misoprostol is contraindicated in pregnant women because it is uterotropic and produces uterine contractions that may endanger pregnancy.98,99 If misoprostol is prescribed to women in their childbearing years, contraceptive measures must be confirmed and a negative serum pregnancy test should be documented within 2 weeks of initiating treatment (see Table 20-10). Patients should be counseled about the GI effects and the need to avoid magnesium antacids, as they may increase the propensity for GI adverse effects.

Bismuth Preparations Bismuth subsalicylate and bismuth subcitrate potassium (biskalcitrate) are the only available bismuth salts in the United States.1,96 Possible ulcer healing mechanisms include an antibacterial effect, a local gastroprotective effect, and stimulation of endogenous PGs. Bismuth salts do not inhibit or neutralize acid. Bismuth subsalicylate is regarded as safe and has few adverse effects when taken in recommended dosages. Because renal insufficiency may decrease bismuth elimination, bismuth salts should be used with caution in older patients and in renal failure. Bismuth subsalicylate may cause salicylate sensitivity or bleeding disorders and should be used with caution for patients receiving concurrent salicylate therapy. Bismuth salts impart a black color to stool and possibly the tongue (liquid preparations). Long-term use of bismuth salts is not recommended due to the potential for bismuth toxicity.

Antacids Antacids neutralize gastric acid, inactivate pepsin, and bind bile salts.96,98,99 Aluminum-containing antacids also suppress H. pylori and enhance mucosal defense. The GI adverse effects are most common and are dose dependent. Magnesium salts cause an osmotic diarrhea, whereas aluminum salts cause constipation. Diarrhea usually predominates with magnesium/aluminum preparations. Aluminum-containing antacids (except aluminum phosphate) form insoluble salts with dietary phosphorus and interfere with phosphorus absorption. Hypophosphatemia occurs most often for patients with low dietary phosphate intake (e.g., malnutrition or alcoholism). Combined treatment with sucralfate may amplify the hypophosphatemia and aluminum toxicity.

Magnesium-containing antacids should not be used for patients with a creatinine clearance of less than 30 mL/min (0.5 mL/s) because magnesium excretion is impaired, which may lead to toxicity. Hypercalcemia may occur for patients with normal renal function taking more than 20 g/day of calcium carbonate and for patients with renal failure who are taking more than 4 g/day. The milk-alkali syndrome (hypercalcemia, alkalosis, renal stones, increased blood urea nitrogen, and increased serum creatinine concentration) occurs with high calcium intake for patients with systemic alkalosis produced by either ingestion of absorbable antacids (sodium bicarbonate) or prolonged vomiting. Antacids may alter the absorption and excretion of drugs when administered concomitantly.96,98,99 Drug interactions may occur when antacids are administered with iron, warfarin, tetracycline, digoxin, quinidine, isoniazid, ketoconazole, or the fluoroquinolones. Most interactions can be avoided by separating the antacid from the oral drug by at least 2 hours.


The metabolism of PPIs occurs primarily through CYP2C19 and polymorphisms of CPY2C19 result in significant differences in enzymatic activity (poor, intermediate, or rapid metabolizers). For example, approximately 85% of white and nearly 100% of Asian populations have polymorphisms resulting in poor metabolism of substrates for CYP2C19. Eradication response rates of H. pylori are influenced by pharmacogenomics, with poor metabolizers achieving 100% eradication, intermediate metabolizers achieving 60%, and rapid metabolizers achieving 30% eradication.100 Prior knowledge of CYP2C19 genotype may help to optimize the PPI dose and interval to minimize therapeutic failure. Rapid metabolizers may need more frequent PPI dosing, up to four times daily, to ensure an optimal gastric pH. Further studies are required to determine if increased AUC achieved in poor metabolizers translates to an additional risk of adverse effects.101 Pharmacologic properties such as bioavailability and plasma concentrations of individual PPIs differ between individuals, but it remains unclear whether these differences impact the efficacy of H. pylori eradication.

Patients with high BMI have reduced antibiotic concentration at the gastric mucosal level and may result in higher risk of treatment failure. Likewise, prior allergy information and history of antimicrobial use is important in tailoring a regimen for H. pylori eradication. Tailoring eradication therapy based on H. pylori clarithromycin sensitivities is gaining popularity as molecular diagnostic testing improves. In one study, eradication rates improved from 70% in the control group to 94.3% in the treatment arm by tailoring eradication therapy from detection of clarithromycin-resistant H. pylori in feces.102

Smoking is a risk factor for treatment failure or ulcer recurrence; therefore, patients should be encouraged to quit smoking.


Images Table 20-14 lists the recommendations for treating and monitoring patients with PUD. Relief of epigastric pain should be monitored throughout the course of treatment for patients with either H. pylori– or NSAID-related ulcers. Ulcer pain typically resolves in a few days when NSAIDs are discontinued and within 7 days on initiation of antiulcer therapy. Most patients with uncomplicated PUD will be symptom free after treatment with any one of the recommended antiulcer regimens. The persistence, or recurrence, of symptoms within 14 days after the end of treatment suggests failure of ulcer healing or H. pylori eradication or an alternative diagnosis such as GERD. Most patients with uncomplicated H. pylori–positive ulcers do not require confirmation of ulcer healing or H. pylori eradication. However, eradication should be confirmed after treatment in individuals who are at risk for complications, for example, individuals who had a prior bleeding ulcer. The UBT is the preferred test to confirm H. pylorieradication when endoscopy is not indicated. Medication adherence should be assessed for patients who fail therapy. Because a large number of at-risk patients treated with NSAIDs do not receive adequate prophylaxis for GI complications, therapeutic outcomes can be improved by advocating preventive strategies. Patients on NSAIDs should be closely monitored for signs or symptoms of bleeding, obstruction, penetration, or perforation. A followup endoscopy is justified for patients with frequent symptomatic recurrence, refractory disease, complications, or suspected hypersecretory states.

TABLE 20-14 Recommendations for Treating and Monitoring Patients with Helicobacter pylori –Associated and Nonsteroidal Antiinflammatory Drug (NSAID)–Induced Ulcers



ZES is characterized by gastric acid hypersecretion and recurrent peptic ulcers that result from a gastrin-producing tumor (gastrinoma).21,45 In the United States, ZES accounts for 0.1% to 1% of patients with duodenal ulcer; however, this may be an underestimation of the true incidence because of the heterogeneity of clinical manifestations.21 Gastrinomas are classified as those associated with multiple endocrine neoplasia type 1 (MEN 1) or sporadic tumors, which have a greater tendency to behave as malignant tumors. In more than 80% of cases, gastrinomas are localized in an area referred to as the triangle of gastrinomas, which includes the convergence of the cystic duct and the common bile duct, the junction of the second and third portions of the duodenum, and the junction of the head and body of the pancreas.21 More than 50% of the gastrinomas are malignant, often with metastases to regional lymph nodes, liver, and bone.21

ZES is suspected for patients with multiple ulcers and recurrent or refractory PUD, often accompanied by esophagitis or ulcer complications.21,45 Ulcers occur most often in the duodenum but may involve the stomach or jejunum. Diarrhea occurs in about 50% of patients, and results from high concentrations of acid that overwhelm the duodenum’s buffering capacity and cause damage to the mucosa.21 Intraluminal acid also causes steatorrhea by inactivating pancreatic lipase and precipitating bile acids. Vitamin B12 malabsorption may result from reduced intrinsic factor activity. The diagnosis is established when the serum gastrin is higher than 1,000 pg/mL (ng/L; 481 pmol/L) and the BAO ≥15 mEq/h (≥15 mmol/h) for patients with an intact stomach (BAO ≥5 mEq/h [≥5 mmol/h] for patients with previous gastric surgery) or when hypergastrinemia is associated with a gastric pH value of ≤2.21 In situations in which the serum gastrin is between 100 and 1,000 pg/mL (ng/L; 48 and 481 pmol/L) and gastric pH is ≤2, a secretin or calcium proactive test is used to aid the diagnosis.21 Identification of the location of the tumor with imaging techniques is essential, as early surgical resection prior to liver metastases is often curative.21 The widespread use of PPIs, although effective in reducing symptoms, may mask the clinical presentation and PPI-related hypergastrinemia may further complicate the diagnosis.21

Treatment is based on the presence or absence of peptic ulcers, esophagitis, diarrhea, and a gastrinoma, which may be malignant. The PPIs are the oral drugs of choice for managing gastric acid hypersecretion. Treatment should be instituted with omeprazole 60 mg/day (or an equivalent dose of the available PPIs) and should be adjusted based on individual patient response.21 Dividing the daily dose and giving the PPI every 8 to 12 hours is most effective in controlling acid output and relieving symptoms. The goal of therapy for uncomplicated patients is to maintain BAO between 1 and 10 mEq/h (1 and 10 mmol/h), in the hour preceding the next dose of the PPI. In complicated cases, such as patients with MEN 1, GERD, or those who are undergoing partial gastrectomy, the BAO should be maintained below 5 mEq/h (5 mmol/h). A gradual reduction in PPI dose is recommended after the adequate control of gastric acid hypersecretion is achieved. IV PPIs are used to suppress gastric acid secretion in patients who are unable to take oral PPIs. The somatostatin analogues, octreotide and lanreotide, directly inhibit the release of gastrin and gastric acid secretion and have been used with varying degrees of success.21 However, these drugs are not considered first-line therapy as they are only available in the injectable form. Preliminary studies have found the long-acting depot formation of octreotide to be efficacious in the treatment of GI neuroendocrine tumors.21 Patients with metastatic gastrinoma require tumor resection or treatment with chemotherapeutic agents.


There are about 160 cases of upper GI bleeding per 100,000 adults annually in the United States.103 Despite a decreased incidence of PUD and improvements in the management of upper GI bleeding, the mortality rate associated with acute hemorrhage remains between 5% and 15%.103105 Upper GI bleeding is categorized as variceal or nonvariceal bleeding. Two common types of nonvariceal bleeding are bleeding from chronic peptic ulcers and bleeding from SRMD (stress gastritis, stress ulcer, or stress erosions), both of which are acid peptic complications.103107 Upper GI bleeding associated with chronic PUD usually precedes hospital admission, whereas bleeding associated with SRMD develops in severely ill patients during hospitalization.103,107

The underlying pathophysiology of bleeding from a peptic ulcer or from SRMD is similar in that impaired mucosal defense in the presence of gastric acid and pepsin leads to mucosal damage. In chronic PUD, H. pylori infection and NSAID use are the most important etiologic factors, whereas the primary pathogenic factor of SRMD in critically ill patients is thought to be mucosal ischemia, which is a result of reduced gastric blood flow (decreased oxygen and nutrient delivery) resulting from splanchnic hypoperfusion.1,103,107109 Defense mechanisms are normally in place to protect the gastric mucosa against the damaging effects of gastric acid, pepsin, and bile (see Pathophysiology above). However, these defense mechanisms become diminished in the face of the overwhelming physiologic stress from critical illness and coupled with mucosal ischemia and subsequent reperfusion injury along with gastric acid result in the rapid development of mucosal lesions.107,108 In contrast to chronic PUD, stress-related mucosal lesions are characteristically asymptomatic, numerous, located in the proximal stomach, and unlikely to perforate.108Bleeding from SRMD occurs from superficial mucosal capillaries, whereas bleeding associated with chronic PUD usually results from a single vessel.108

The mortality rate associated with clinically important stress-related mucosal bleeding (SRMB) is approximately 50% and is related to the underlying severity of disease and comorbidities in this patient population.106,108,110 In contrast, the mortality associated with chronic PUD-related bleeding is approximately 10% but can increase dramatically in select patient populations.104,111,112 Although the initial management of acute upper GI bleeding focuses on aggressive resuscitative measures and ensuring hemodynamic stability, the medical management of PUD-related bleeding and SRMB is distinctly different.103,104,107,108

Peptic Ulcer–Related Bleeding

The most common presenting signs and symptoms of PUD-related bleeding are hematemesis (vomiting up blood) or melena (dark, tarry stools) or possibly both. When evaluating patients with PUD-related bleeding, the degree of risk for adverse outcomes must be rapidly assessed in order to determine if the patient’s condition constitutes a medical emergency.103,104 Two risk stratification tools exist for early assessment and triage.113,114 The Blatchford score is a newer risk stratification scale and is used to evaluate the need for urgent endoscopic intervention for patients presenting with PUD-related bleeding.113The scale values range from 0 to 23, with higher scores indicating higher risk. The most well-known scale, however, is the Rockall Score.114 This validated risk assessment instrument is composed of two assessments: the clinical score, which is performed prior to endoscopy, and the endoscopic score. The use of these risk stratification tools can reduce the requirement of endoscopic procedures and lead to early discharge for low-risk patients while ensuring rapid intervention for patients at higher risk.103 These data will also allow the pharmacist to make appropriate pharmacotherapy decisions based on the patient’s assessed level of risk. When considering the risk of death due to PUD bleeding, the following patients generally have poorer prognoses and usually require more aggressive intervention including admission to an intensive care unit (ICU)103,104,107,108:

   1. Older than 60 years of age

   2. Comorbid conditions (e.g., ischemic heart disease, congestive heart failure, renal failure, hepatic failure, metastatic cancer)

   3. High transfusion requirements

   4. Ongoing blood loss

   5. Presence of hypovolemic shock (i.e., tachycardia with a pulse of ≥100 beats/min, hypotension with a systolic blood pressure of <100 mm Hg with concomitant orthostatic changes, such as an increase in the heart rate of ≥20 beats/min or decrease in systolic blood pressure of ≥20 mm Hg on standing from a sitting position)

   6. Prolonged prothrombin time (or increased international normalized ratio [INR])

   7. Erratic mental status

Initial therapy for patients with defined hemostatic instability should focus on correcting fluid volume loss though appropriate volume resuscitative measures. This is usually accomplished with a continuous 0.9% sodium chloride infusion (or blood products if clinically indicated) through two large-bore peripheral IV catheters (i.e., 16 to 18 gauge).103,105 The use of nasogastric (NG) tubes remains controversial but may aid in early assessment and gastric lavage.103

Diagnostic endoscopy is usually performed as early as possible (preferably within 24 hours of presentation) to identify the source of the bleeding, assess the potential risk for rebleeding, and, if appropriate, employ therapeutic interventions to promote hemostasis.103,105 Several endoscopic treatment approaches (e.g., thermocoagulation, argon plasma coagulation therapy, injection sclerotherapy, hemoclipping, and ligation) can be used; however, to maximize the likelihood of positive outcomes, patients should be treated with a combination of at least two endoscopic modalities, such as thermocoagulation and injection of lesions with epinephrine.103105 The appearance of the ulcer at the time of endoscopy is a prognostic indicator for the risk of rebleeding.103,105,111 Clean-based and flat spot (pigmented) ulcers are most commonly seen and are associated with a low risk of rebleeding (5% and 10%, respectively).105,111 In most cases, patients with clean-based ulcers can be immediately discharged after endoscopy on antiulcer therapy (usually twice-daily PPI), while patients with flat spot ulcers may be admitted to the general hospital ward for a brief observation period. Patients with an adherent clot overlying the ulcer base are at intermediate risk of rebleeding (22%), and controversy exists as to the appropriate management of these patients.105,111Adherent clots can be removed, and then the lesion be reclassified based on what is observed following clot removal.105,111 Patients with a visible vessel or active bleeding are at the highest risk of rebleeding (43% and 55%, respectively) and should be managed within an ICU for at least 24 hours and then monitored on a general medical/surgical service for an additional 48 hours (total of 72 hours), as rebleeding significantly increases mortality.103,104,111

Images Antisecretory therapy is often used as adjuvant therapy to endoscopic procedures to prevent PUD rebleeding in high-risk patients because acid impairs clot stability.103 However, endoscopic hemostatic techniques remain the treatment of choice for patients with life-threatening bleeding, as this has been associated with better outcomes when compared with either placebo or pharmacotherapy alone.103105,110H2-receptor antagonists are ineffective in preventing PUD rebleeding because they do not achieve an intragastric pH of 6 (which is needed to promote hemostatis and clot stability) and tolerance to their antisecretory effect develops rapidly (especially with high-dose or IV therapy).103105,109,115,116 In contrast, PPIs reduce the incidence of rebleeding and need for surgery but have no significant impact on overall mortality.103,104,115,116 Interestingly, subgroup analysis from two differing meta-analyses has suggested a mortality benefit with high-dose IV PPIs in a subpopulation of patients with the highest risk of rebleeding (i.e., nonbleeding visible vessel or active bleeding).115,117

The precise route (oral or IV) and the dose of PPI should be based on the clinical situation, risk of rebleeding, endoscopic identification of the lesion, and patient risk.115,116 Because of the theoretical goal of maintaining intragastric pH values >6, and data from randomized controlled trials, practice guidelines recommend that high-dose continuous-infusion PPI (equivalent to omeprazole 80 mg given IV as a loading dose, followed by 8 mg/h continuous infusion for 72 hours) be used to reduce the risk of rebleeding in high-risk patients who have undergone endoscopy hemostasis.103,104,115,116 Because IV omeprazole is not available in the United States and three small randomized controlled trials have not demonstrated any evidence of improved outcomes between the available IV PPIs and omeprazole, most clinicians consider IV pantoprazole and esomeprazole to be equivalent to IV omeprazole on a milligram-per-milligram basis. Thus, IV pantoprazole and esomeprazole can be interchanged as there is currently no evidence to suggest that one PPI is superior to another.116,118 The administration of high-dose continuous-infusion PPIs given prior to endoscopy hastens resolution of bleeding stigmata.119 However, PPI therapy is not a replacement for interventional endoscopy for patients who are at high risk of rebleeding, as data demonstrate that the combination of a high-dose PPI continuous IV infusion with therapeutic endoscopy is superior to either strategy alone.116,120 High-dose oral PPI therapy (omeprazole 80 mg/day for 5 days) is also effective; however, concerns exist as to whether critically ill patients will absorb the medication.110,115 The risk of rebleeding is greatest within the first 72 hours (especially the first 24 hours), and it is during this time that antisecretory therapy to prevent rebleeding in high-risk patients should be employed.103,104 Patients should be transitioned to an oral PPI on completion of IV therapy. Somatostatin and octreotide have not demonstrated any significant benefit in treating nonvariceal bleeding and are not recommended at this time.103,104

Patients with upper GI bleeding should be tested for H. pylori at the time of endoscopy (see Tests for H. pylori above). However, the tests are associated with an increased rate of false-negatives when obtained during acute bleeding episodes.121 If the initial results of the rapid urease test and/or histology are negative, a confirmatory test (14C-UBT or serology) should be performed following the acute bleeding episode. There is no rationale for using IV therapy to eradicate H. pylori. Ulcer treatment, including H. pylori eradication, if appropriate, should be initiated after the acute bleeding episode has resolved (see Treatment of H. pylori–Positive Ulcers and Treatment of NSAID-Induced Ulcers above).

Stress-Related Mucosal Bleeding

Critically ill patients may develop SRMD leading to SRMB because of the homeostatic compromise that is associated with severe illness.107,108 More than 75% (some studies suggesting up to 100%) of critically ill patients develop SRMD within the first 1 to 3 days of admission to an ICU, but the incidence of clinically important SRMB (defined as overt bleeding with concomitant hemodynamic instability and likely requirement for blood products) is 1% to 4%.107109 Clinically important bleeding increases the length of ICU stay by up to 11 days, results in excessive healthcare costs, and is associated with increased mortality.107,108,122 Thus, attempts to prevent SRMB are warranted in high-risk patients. Prophylactic therapy to prevent bleeding is most effective if initiated early in the patient’s course.

Images Patients who are at risk for SRMB include those with respiratory failure (need for mechanical ventilation for longer than 48 hours), coagulopathy (INR >1.5, platelet count <50,000 mm3 [<50 × 109/L]), hypotension, sepsis, hepatic failure, acute renal failure, high-dose corticosteroid therapy (>250 mg/day hydrocortisone or equivalent), multiple trauma, severe burns (>35% of body surface area), head injury, traumatic spinal cord injury, major surgery, prolonged ICU admission (>7 days), or history of GI bleeding.106108,122 The relative importance of the various risk factors remains controversial, but most clinicians concur that patients with respiratory failure or coagulopathy should receive prophylaxis, as these two factors are independent risk factors for SRMB.106,107 In the absence of these risk factors, some clinicians only administer prophylaxis to patients who have two of the aforementioned risk factors.106 Since not all patients in a hospital or ICU are at increased risk of SRMB, a cost-effective approach should be developed to target prophylactic therapy at high-risk patients.

Clinical Controversy…

Some clinicians feel that SRMB prophylaxis is indicated for all critically ill patients given the associated mortality and increased length of stay for patients who develop this complication. Others adhere to clinical guidelines that suggest only high-risk patients such as those with respiratory failure requiring ventilation or patients with coagulopathy and perhaps those patients with two or more other risk factors will benefit from prophylaxis.

Prevention of SRMB includes resuscitative measures that restore mucosal blood and pharmacotherapy that either maintains an intragastric pH of >4 or provides gastric mucosal protection.106108 Although the benefits of enteral nutrition to patient outcome (e.g., improved nutritional status enhances mucosal integrity) are of overall clinical importance, its precise role as a sole modality to prevent SRMB remains controversial.107,108 Therapeutic options for the prevention of SRMB include antacids (which are of historical interest, as they are no longer used because of cumbersome dosage schedules and side effects), antisecretory drugs (H2RAs and PPIs), and sucralfate (mucosal protectant).106108,110,122

Antisecretory therapy is generally preferred for SRMB prophylaxis for several reasons. First, a large landmark study demonstrated that IV ranitidine was superior to oral sucralfate in preventing SRMB.123Second, ranitidine did not increase the risk for nosocomial pneumonia, as the incidence of pneumonia was not different between the two treatment groups.123 Third, although sucralfate is an evidence-based option, it is cumbersome, requiring multiple daily dosage administration (up to four times daily), which may occlude NG tubes, cause constipation, interact with several medications, and/or increase the potential for aluminum toxicity in patients with renal dysfunction. Finally, sucralfate does not have any appreciable effect on reducing intragastric pH. Although data published in 2004 indicated that H2RAs were the most commonly prescribed antisecretory agents used to prevent SRMB,124 PPIs have become the mainstay of therapy.107 Regardless, numerous studies and years of experience support the use of H2RAs, and they remain a recommended option for the prevention of SRMB.106108,124

Parenteral H2RAs may be administered as either continuous infusions or intermittent bolus doses (see Table 20-15). Cimetidine, given as a continuous IV infusion, is the only FDA-labeled H2RA for the prevention of SRMB. Despite evidence suggesting that continuous infusions of the H2RAs are superior to intermittent dosing at maintaining an intragastric pH >4, there is no evidence to suggest better outcomes with respect to prevention of SRMB.107Thus, intermittent bolus doses of H2RAs are used more commonly than continuous infusions.108,110 Drug interactions are more common with cimetidine, and for this reason the other H2RAs (famotidine, ranitidine) are used more frequently.107 Adverse events associated with the use of H2RAs for the critically ill patient include thrombocytopenia, mental status changes (more common in older patients or individuals with renal or hepatic compromise), and tachyphylaxis (especially with parenteral or high-dose therapy).107 Given that the H2RAs are renally eliminated, dosage reductions are recommended for patients with renal dysfunction.

TABLE 20-15 Pharmacotherapy Options for Prophylaxis of Stress-Related Mucosal Bleeding


The PPIs are more potent than H2RAs in inhibiting acid secretion and, unlike H2RAs, tolerance does not develop.107,108,110,122 Although there are limited data assessing the efficacy of PPIs for the prevention of SRMB, recent reports suggest that PPIs may be used as alternatives to H2RAs or sucralfate.122,125,126 Initial open-label studies of PPIs for SRMB prophylaxis were performed with omeprazole compounded suspensions given as 40 mg for two doses on the first day, followed by 20 mg/day thereafter in critically ill or trauma patients requiring mechanical ventilation and the presence of at least one additional risk factor.127,128 The results of these studies demonstrated significant pH control above 4 and no GI bleeding in either trial. Because of the small number of subjects, no firm clinical outcomes can be derived from these studies. Since then, numerous extemporaneously compounded sodium bicarbonate suspensions of PPIs (see Tables 20-12 and 20-15) have been used as cost-effective regimens for stress ulcer prophylaxis in patients with NG tubes.107,110,122,129 In one study, immediate-release omeprazole–sodium bicarbonate suspension (40 mg × 2 doses, and then 40 mg/day) was more effective than continuous-infusion cimetidine in maintaining intragastric pH >4 in critically ill patients, but there was no difference in the incidence of clinically important SRMD between the two groups.125 Based on meeting the prespecified criteria for noninferiority, immediate-release omeprazole gained FDA labeling for the prevention of SRMB.125 In another trial of ICU patients requiring mechanical ventilation, patients were randomized to receive either lansoprazole 30 mg given as a rapidly disintegrating tablet mixed in 10 mL of water and administered via an NG tube or lansoprazole 30 mg given as an IV infusion (no longer available in the United States) once daily for 72 hours.130 Enterally administered lansoprazole, despite having a lower bioavailability, resulted in superior intragastric pH control when compared with the IV-administered lansoprazole.

One of the most compelling dose-finding pilot studies to evaluate the use of IV PPIs for SRMB prophylaxis was performed in over 200 ICU patients examining five different dosing strategies of pantoprazole (40 mg given every 8 hours, every 12 hours, and every 24 hours or 80 mg given every 12 hours or every 24 hours) versus cimetidine given as a 300 mg IV bolus followed by a 50 mg/h continuous infusion.126Each regimen was given for at least 48 hours and for up to 7 days. The time the intragastric pH was ≥4 increased from day 1 to 2 in all the pantoprazole groups, whereas it actually decreased in the cimetidine group suggesting tachyphylaxis. No bleeding was identified in any of the treatment groups. The results suggest appropriate pH control can be obtained with doses of 80 mg given on day 1 and 40 mg given every 12 hours thereafter. The use of IV esomeprazole has also demonstrated efficacy in a small number of clinical studies.107 Given the relatively limited data, the overall efficacy, optimal dosage, frequency, and route of administration for PPIs in the prevention of SRMB remain to be fully elucidated.107,108 Based on the available evidence, several PPI dosing regimens for SRMB prophylaxis exist (see Table 20-15). Adverse events that have been described when the PPIs are used for this indication include an increased risk of enteric infections, including C. difficile–associated diarrhea and nosocomial pneumonia.107

Images When deciding on the most appropriate pharmacotherapy plan for the prevention of SRMB for a specific patient, the clinical presentation of the patient and medication costs should be used as a guide. Patients who can take oral medication or have a working NG tube in place may be placed on an oral or compounded PPI suspension as a cost-effective measure. For most patients who are not able to utilize one of these routes, an IV H2RA is appropriate. However, if the patient has renal dysfunction, develops thrombocytopenia, or mental status changes while on an H2RA, then an IV PPI may be the most appropriate prophylaxis option. Consideration should be given to both the patient’s clinical condition and the most cost-effective option when developing protocols for preventing SRMB.

Improvement in the patient’s overall medical condition (resolution of risk factors, discharge from the ICU, extubation, and oral intake) suggests that prophylactic therapy can be discontinued.107 Too often the patient is continued on SRMB prophylaxis on transition to the general medical/surgical unit and is often discharged on oral PPI therapy without an appropriate indication.87 This results in unnecessary costs for the patient and the healthcare system.87Pharmacists should identify patients in whom SRMB prophylaxis is no longer indicated.107 If a patient develops clinically important bleeding, endoscopic evaluation of the GI tract is indicated along with aggressive antisecretory therapy (see Peptic Ulcer–Related Bleeding above).




    1. Soll AH, Graham DY. Peptic ulcer disease. In: Yamada T, Alpers DH, Kalloo KN, et al., eds. Textbook of Gastroenterology, 5th ed. Hoboken, NJ: Wiley-Blackwell, 2009:936–981.

    2. Washington MK, Peek RM. Gastritis and gastropathy. In: Yamada T, Alpers DH, Kalloo KN, et al., eds. Textbook of Gastroenterology, 5th ed. Hoboken, NJ: Wiley-Blackwell, 2009:1005–1025.

    3. Del Valle J. Zollinger-Ellison syndrome. In: Yamada T, Alpers DH, Kalloo KN, et al., eds. Textbook of Gastroenterology, 5th ed. Hoboken, NJ: Wiley-Blackwell, 2009:982–1004.

    4. Everhart JE, Ruhl CE. Burden of digestive diseases in the United States part I: Overall and upper gastrointestinal diseases. Gastroenterology 2009;136:376–386.

    5. Chey WD, Wong BCY, Practice Parameters Committee of American College of Gastroenterology. Guideline on the management of Helicobacter pylori infection. Am J Gastroenterol 2007;102:1–18.

    6. Azevedo NF, Huntington J, Goodman KJ. The epidemiology of Helicobacter pylori and public health implications. Helicobacter 2009;14(Suppl 1):1–7.

    7. Talley NJ, Fock KM, Moayyedi P. Gastric cancer consensus conference recommends Helicobacter pylori screening and treatment in asymptomatic persons from high-risk populations to prevent gastric cancer. Am J Gastroenterol 2008;103:510–514.

    8. Leung WK, Ng EKW, Sung JJY. Tumors of the stomach. In: Yamada T, Alpers DH, Kalloo KN, et al., eds. Textbook of Gastroenterology, 5th ed. Hoboken, NJ: Wiley-Blackwell, 2009:1026–1053.

    9. Talley NJ, Holtmann G. Approach to the patient with dyspepsia and related functional gastrointestinal complaints. In: Yamada T, Alpers DH, Kalloo KN, et al., eds. Principles of Clinical Gastroenterology, 1st ed. Hoboken, NJ: Wiley-Blackwell, 2008:38–61.

   10. Furuta T, Delchier JC. Helicobacter pylori and non-malignant diseases. Helicobacter 2009;14(Suppl 1):29–35.

   11. Pellicano R, Franceschi F, Saracco G, et al. Helicobacters and extragastric diseases. Helicobacter 2009;14(Suppl 1):58–68.

   12. Lanza FL, Chan FKL, Quigley EMM, Practice Parameters Committee of the American College of Gastroenterology. Guidelines for prevention of NSAID-related ulcer complications. Am J Gastroenterol 2009;104:728–738.

   13. Bhatt DL, Scheiman J, Abraham NS, et al. ACCF/ACG/AHA 2008 expert consensus document on reducing the gastrointestinal risks of antiplatelet therapy and NSAID use: A report of the American College of Cardiology Foundation Task Force on clinical expert consensus documents. Am J Gastroenterol 2008;103:2890–2907.

   14. Chan FKL Abraham NS, Scheiman JM, Laine L. Management of patients on nonsteroidal anti-inflammatory drugs: A clinical practice recommendation from the First International Working Party on Gastrointestinal and Cardiovascular Effects of Nonsteroidal Anti-Inflammatory Drugs and Anti-Platelet Agents. Am J Gastroenterol 2008;103:2908–2918.

   15. Cryer B. Management of patients with high gastrointestinal risk on antiplatelet therapy. Gastroenterol Clin North Am 2009;38:289–303.

   16. Rostom A, Moayyedi P, Hunt R, Canadian Association of Gastroenterology Consensus Group. Canadian consensus guidelines on long-term nonsteroidal anti-inflammatory drug therapy and the need for gastroprotection: Benefits versus risks. Aliment Pharmacol Ther 2009;29:481–496.

   17. Lanas A, Sopena F. Nonsteroidal anti-inflammatory drugs and lower gastrointestinal complication. Gastroenterol Clin North Am 2009;38:333–352.

   18. Chan FKL. The David Y. Graham Lecture: Use of nonsteroidal antiinflammatory drugs in a COX-2 restricted environment. Am J Gastroenterol 2008;103:221–227.

   19. Scheiman JM. Balancing risks and benefits of cyclooxygenase-2 selective nonsteroidal anti-inflammatory drugs. Gastroenterol Clin North Am 2009;38:305–314.

   20. Loke YK, Trivedi AN, Singh S. Meta-analysis: Gastrointestinal bleeding due to interaction between selective serotonin uptake inhibitors and non-steroidal anti-inflammatory drugs. Aliment Pharmacol Ther 2008;27:31–40.

   21. Del Valle J, Tedisco A. Gastric secretion. In: Yamada T, Alpers DH, Kalloo KN, et al., eds. Textbook of Gastroenterology, 5th ed. Hoboken, NJ: Wiley-Blackwell, 2009:284–329.

   22. Costa AC, Figueiredo C, Touati E. Pathogenesis of Helicobacter pylori infection. Helicobacter 2009;14(Suppl 1): 15–20.

   23. Glasglow RE, Mulvihill SJ. Surgery for peptic ulcer disease and postgastrectomy syndromes. In: Yamada T, Alpers DH, Kalloo KN, et al., eds. Textbook of Gastroenterology, 5th ed. Hoboken, NJ: Wiley-Blackwell, 2009:1054–1070.

   24. Monteiro L, Oleastro M, Lehours P, Megraud F. Diagnosis of Helicobacter pylori infection. Helicobacter 2009;14(Suppl 1): 8–14.

   25. Malfertheiner P, Megraud F, O’Morain C, et al. Current concepts in the management of Helicobacter pylori infection—The Maastricht III Consensus Report. Gut 2007;56:772–781.

   26. O’Connor A, Gisbert J, O’Morain C. Treatment of Helicobacter pylori infection. Helicobacter 2009;14(Suppl 1):46–51.

   27. Vakil N. H. pylori treatment: New wine in old bottles? Am J Gastroenterol 2009;104:26–30.

   28. Rokkas P, Sechopoulos P, Robotis I, et al. Cumulative H. pylori eradication rates in clinical practice by adopting first and second-line regimens proposed by the Maastricht III Consensus and a third-line empirical regimen. Am J Gastroenterol 2009;104:21–25.

   29. Malfertheiner P, Chan FKL, McColl KEL. Peptic ulcer disease. Lancet 2009;374:1449–1461.

   30. Kindermann A, Lopes AI. Helicobacter pylori infection in pediatrics. Helicobacter 2009;14(Suppl 1):52–57.

   31. Francavilla R, Lionetti E, Cavallo L. Sequential treatment for Helicobacter pylori eradication in children. Gut 2008;57:1178.

   32. Boparai V, Rajagopalan J, Triadafilopoulos G. Guide to the use of proton pump inhibitors in adult patients. Drugs 2008;68:925–947.

   33. Villoria A, Garcia P, Calvet X, et al. Meta-analysis: High-dose proton pump inhibitors vs. standard dose in triple therapy for Helicobacter pylori eradication. Aliment Pharmacol Ther 2008;28:868–877.

   34. Vergara M, Vallve M, Gisbert JP, et al. Meta-analysis: Comparative efficacy of different proton-pump inhibitors in triple therapy for Helicobacter pylori eradication. Aliment Pharmacol Ther 2003;18:647–654.

   35. Graham DY, Hammoud F, el-Zimaity HM, et al. Meta-analysis: Proton pump inhibitor or H2-receptor antagonist for Helicobacter pylori eradication. Aliment Pharmacol Ther 2003;17:1229–1236.

   36. Gisbert JP, Khorrami S, Calvet X, et al. Meta-analysis: Proton pump inhibitors vs. H2-receptor antagonists—Their efficacy with antibiotics in Helicobacter pylori eradication. Aliment Pharmacol Ther 2003;18:757–766.

   37. Janssen MJR, Laheij RJF, Boer WA, et al. Meta-analysis: The influence of pre-treatment with a proton pump inhibitor on Helicobacter pylori eradication. Aliment Pharmacol Ther 2005;21:341–345.

   38. Luther J, Higgins PDR, Schoenfeld PS, et al. Empiric quadruple vs. triple therapy for primary treatment of Helicobacter pylori infection: Systematic review and meta-analysis of efficacy and tolerability. Am J Gastroenterol 2010;105:65–73.

   39. Laine L, Hunt R, el-Zimaity H, et al. Bismuth-based quadruple therapy using a single capsule of bismuth biskalcitrate, metronidazole, and tetracycline given with omeprazole versus omeprazole, amoxicillin, and clarithromycin for eradication of Helicobacter pylori in duodenal ulcer patients: A prospective, randomized, multicenter, North American trial. Am J Gastroenterol 2003;98:562–567.

   40. Fischbach LA, van Zanten S, Dickason J. Meta-analysis: The efficacy, adverse events, and adherence related to firstline anti-Helicobacter pylori quadruple therapies. Aliment Pharmacol Ther 2004;20:1071–1082.

   41. Laine L. Is it time for quadruple therapy to be first line? Can J Gastroenterol 2003;17(Suppl B):33B–35B.

   42. Jafri NS, Hornung CA, Howden CW. Meta-analysis: Sequential therapy appears superior to standard therapy for Helicobacter pylori infection in patients naive to treatment. Ann Intern Med 2008;148:1–9.

   43. O’Morain CA, O’Connor JP. Is sequential therapy superior to standard triple therapy for the treatment of Helicobacter pylori infection? Nat Clin Pract Gastroenterol Hepatol 2009;6:8–9.

   44. Megraud F, Lamouliatte H. Review article: The treatment of refractory Helicobacter pylori infection. Aliment Pharmacol Ther 2003;17:1333–1343.

   45. Napolitano L. Refractory peptic ulcer disease. Gastroenterol Clin North Am 2009;38:267–288.

   46. Saad RJ, Schoenfeld P, Kim HM, et al. Levofloxacin-based triple therapy versus bismuth-based quadruple therapy for persistent Helicobacter pylori infection: A meta-analysis. Am J Gastroenterol 2006;101:488–496.

   47. Gisbert JP, Perez-Aisa A, Castro-Fernandez M, et al. Helicobacter pylori first-line treatment and rescue option containing levofloxacin in patients allergic to penicillin. Dig Liver Dis 2010;42:287–290.

   48. Suzuki T, Matsuo K, Sawaki A, et al. Systematic review and meta-analysis: Importance of CagA status for successful eradication of Helicobacter pylori infection. Aliment Pharmacol Ther 2006;24:273–280.

   49. Padol S, Yuan Y, Thabane M, et al. The effect of CYP2C19 polymorphisms on H. pylori eradication rate in dual and triple first-line PPI therapies: A meta-analysis. Am J Gastroenterol 2006;101:1467–1475.

   50. Fischbach L, Evans EL. Meta-analysis: The effect of antibiotic resistance status on the efficacy of triple and quadruple first-line therapies for Helicobacter pylori. Aliment Pharmacol Ther 2007;26:343–357.

   51. Meyer JM, Silliman NP, Wang W, et al. Risk factors for Helicobacter pylori resistance in the United States: The surveillance of H. pylori antimicrobial resistance partnership (SHARP) study, 1993–1999. Ann Intern Med 2002;136:13–24.

   52. Duck WM, Sobel J, Pruckler JM, et al. Antimicrobial resistance incidence and risk actors among Helicobacter pylori-infected persons in the United States. Emerg Infect Dis 2004;10:1088–1094.

   53. Gotteland M, Brunser O, Cruchet S. Systematic review: Are probiotics useful in controlling gastric colonization by Helicobacter pylori. Aliment Pharmacol Ther 2006;23:1077–1086.

   54. Kim MN, Kim N, Lee SH, et al. The effects of probiotics on PPI-triple therapy for Helicobacter pylori eradication. Helicobacter 2008;13:261–268.

   55. Scaccianoce G, Zullo A, Hassan C, et al. Triple therapies plus different probiotics for Helicobacter pylori eradication. Eur Rev Med Pharmacol Sci 2008;12:251–256.

   56. Targownik LE, Metge CJ, Leung S, et al. The relative efficacies of gastroprotective strategies in chronic users of nonsteroidal anti-inflammatory drugs. Gastroenterology 2008;134:937–944.

   57. Silverstein FE, Graham DY, Senior JR, et al. Misoprostol reduces serious gastrointestinal complications in patients with rheumatoid arthritis receiving nonsteroidal anti-inflammatory drugs: A randomized, double-blind, placebo-controlled trial. Ann Intern Med 1995;123:241–249.

   58. Graham DY, Agrawal NM, Campbell DR, et al. Ulcer prevention in long-term users of nonsteroidal anti-inflammatory drugs: Results of a double-blind, randomized, multicenter, active- and placebo-controlled study of misoprostol vs lansoprazole. Arch Intern Med 2002;152:169–175.

   59. Arora G, Singh G, Triadafilopoulos G. Proton pump inhibitors for gastroduodenal damage related to nonsteroidal antiinflammatory drugs or aspirin: Twelve important questions for clinical practice. Clin Gastroenterol Hepatol 2009;7:725–736.

   60. Taha AT, McCloskey C, Prasad R, Bezlyak V. Famotidine for the prevention of peptic ulcers and oesophagitis in patients taking low-dose aspirin (FAMOUS): A phase III, randomised, double-blind, placebo-controlled trial. Lancet 2009;374:119–125.

   61. Silverstein F, Faich G, Goldstein JL, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal antiinflammatory drugs for osteoarthritis and rheumatoid arthritis. The CLASS study: A randomized controlled trial. JAMA 2000;284:1247–1255.

   62. Bombardier C, Laine L, Reicin A, et al. Comparison of upper intestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. VIGOR Study Group. N Engl J Med 2000;343:1520–1528.

   63. US Food and Drug Administration. Drug Information. COX-2 Selective and Non-Selective Non-Steroidal Anti-Inflammatory Drugs (NSAIDs).

   64. Bresalier RS, Sandler RS, Quan H, et al. Cardiovascular events associated with rofecoxib in colorectal adenoma chemoprevention trial. N Engl J Med 2005;352:1092–1102.

   65. Kearney PM, Baigent C, Godwin J, et al. Do selective cyclooxygenase-2 inhibitors and traditional non-steroidal antiinflammatory drugs increase the risk of atherothrombosis? Meta-analysis of randomized trials. BMJ 2006;322:1302–1308.

   66. Fiorucci S. Prevention of nonsteroidal anti-inflammatory drug-induced ulcer: Looking forward to the future. Gastroenterol Clin North Am 2009;38:315–322.

   67. McColl KEL. How I manage H. pylori-negative, NSAID/aspirin-negative peptic ulcers. Am J Gastroenterol 2009;104:190–193.

   68. Gisbert JP, Calvert X. Helicobacter pylori-negative duodenal ulcer disease. Aliment Pharmacol Ther 2009;30:791–815.

   69. Reimer C, Sondergaard G, Hilsted L, et al. Proton-pump inhibitor therapy induces acid-related symptoms in healthy volunteers after withdrawal of therapy. Gastroenterology 2009;137:80–87.

   70. Dexilant (dexlansoprazole) [package insert]. Deerfield, IL: Takeda Pharmaceuticals America Inc.

   71. Metz DC, Valiky M, Dixit T, et al. Review article: Dual delayed release formulation of dexlansoprazole MR, a novel approach to overcome the limitations of conventional single release proton pump inhibitor therapy. Aliment Pharmacol Ther 2009;29:928–937.

   72. Zegerid (omeprazole/sodium bicarbonate) [package insert]. San Diego, CA: Santarus Inc.

   73. Lahner E, Annibale B, Fave GD. Systematic review: Impaired drug absorption related to the co-administration of antisecretory therapy. Aliment Pharmacol Ther 2009;29:219–229.

   74. Béïque L, Giguère P, la Porte C, Angel J. Interactions between protease inhibitors and acid-reducing agents: A systematic review. HIV Med 2007;8:335–345.

   75. Sachmechi I, Reich DM, Aninyei M, et al. Effect of proton pump inhibitors on serum thyroid-stimulating hormone level in euthyroid patients treated with levothyroxine for hypothyroidism. Endocr Pract 2007;13:345–349.

   76. Parikh N, Howden CW. The safety of drugs used in acid-related disorders and functional gastrointestinal disorders. Gastroenterol Clin North Am 2010;39:529–542.

   77. U.S. Food and Drug Administration. Information for Healthcare Professionals: Update to the Labeling of Clopidogrel Bisulfate (Marketed as Plavix) to Alert Healthcare Professionals About a Drug Interaction with Omeprazole (Marketed as Prilosec and Prilosec OTC).

   78. U.S. Food and Drug Administration. Public Health Advisory: Updated Information About a Drug Interaction Between Clopidogrel Bisulfate (Marketed as Plavix) and Omeprazole (Marketed as Prilosec and Prilosec OTC). 2009,

   79. U.S. Food and Drug Administration. Information for Healthcare Professionals: Update to the Labelling of Clopidogrel Bisulfate (Marketed as Plavix) to Alert Healthcare Practitioners About a Drug Interaction with Omeprazole (Marketed as Prilosec and Prilosec OTC).

   80. Plavix (clopidogrel bisulfate tablets). Bridgewater, NJ: Bristol-Myers Squibb/Sanofi Pharmaceuticals Partnership. Revised 2009,

   81. Mega JL, Close SL, Wiviott SD, et al. Cytochrome P-450 polymorphisms and response to clopidogrel. N Engl J Med 2009;360:354–362.

   82. Freedman JE, Hylek EM. Clopidogrel, genetics, and drug responsiveness. N Engl J Med 2009;360:411–413.

   83. Last EJ, Sheehan AH. Review of recent evidence: Potential interaction between clopidogrel and proton pump inhibitors. Am J Health Syst Pharm 2009;66:2117–2122.

   84. Norgard NB, Mathews KD, Wall GC. Drug–drug interaction between clopidogrel and proton pump inhibitors. Ann Pharmacother 2009;43:1266–1274.

   85. Sheen E, Triadafilopoulos G. Adverse effects of long-term proton pump inhibitor therapy. Dig Dis Sci 2011;56:931–950.

   86. Ali T, Roberts DN, Tierney WM. Long-term safety concerns with proton pump inhibitors. Am J Med 2009;122:896–903.

   87. Heidelbaugh JJ, Goldberg KL, Inadomi JM. Overutilization of proton pump inhibitors: A review of cost-effectiveness and risk. Am J Gastroenterol 2009;104(Suppl):S27–S32.

   88. Williams C, McColl KE. Review article: Proton pump inhibitors and bacterial overgrowth. Aliment Pharmacol Ther 2006;23:3–10.

   89. Laheij RJF, Sturkenboom MCJM, Hassing R-J, et al. Risk of community-acquired pneumonia and use of gastric acid-suppressive drugs. JAMA 2004;292:1955–1960.

   90. Gulmez SE, Holm A, Frederiksen H, et al. Use of proton pump inhibitors and the risk of community-acquired pneumonia: A population-based case–control study. Arch Intern Med 2007;167:950–955.

   91. Sarkar M, Hennessy S, Yang YX. Proton-pump inhibitor use and the risk of community-acquired pneumonia. Ann Intern Med 2008;149:391–398.

   92. Leonard J, Marshall JK, Moayyedi P. Systematic review of the risk of enteric infection in patients taking acid suppression. Am J Gastroenterol 2007;102:2047–2056.

   93. US Food and Drug Administration. FDA Safety Communication: Possible Increase Risk of Fractures of the Hip, Wrist and Spine with the Use of Proton Pump Inhibitors.

   94. Kahrilas PJ, Shaheen NJ, Vaezi MF. American Gastroenterological Association Medical Position Statement on management of GERD. Gastroenterology 2008;135:1383–1391.

   95. Furuta K, Adachi K, Komazawa Y, et al. Tolerance to H2-receptor antagonist correlates well with the decline in efficacy against gastroesophageal reflux in patients with gastroesophageal reflux disease. J Gastroenterol Hepatol 2006;21:1581–1585.

   96. Zweber A, Berardi RR. Heartburn and Dyspepsia. In: Berardi RR, Ferreri SP, Hume AL, et al., eds. Handbook of Nonprescription Drugs, 16th ed. Washington, DC: American Pharmaceutical Association, 2009:231–246.

   97. Wade EE, Rebuck JA, Healey MA, Rogers FB. H2 antagonist-induced thrombocytopenia: Is this a real phenomenon? Intensive Care Med 2002;28:459–465.

   98. Chong YS, Su LL, Arulkumaran S. Misoprostol: A quarter century of use, abuse and creative misuse. Obstet Gynecol Surg 2004;59:128–140.

   99. Maton PN, Burton ME. Antacids revisited: A review of their clinical pharmacology and recommended therapeutic use. Drugs 1999;57:855–870.

  100. Kang JM, Kim N, Lee DH, et al. Effect of the CYP2C19 polymorphism on the eradication rate of Helicobacter pylori infection by 7-day triple therapy with regular proton-pump inhibitor dosage. J Gastroenterol Hepatol 2008;23:1287–1291.

  101. Ma Q, Lu AYH. Pharmacogenetics, pharmacogenomics, and individualized medicine. Pharmacol Rev 2011;63:437–459.

  102. Kawai T, Yamagishi T, Yagi K, et al. Tailored eradication therapy based on fecal Helicobacter pylori clarithromycin sensitivities. J Gastroenterol Hepatol 2008;23:S171–S174.

  103. Gralnek IM, Barkun AN, Bardou M. Management of acute bleeding from a peptic ulcer. N Engl J Med 2008;359:928–937.

  104. Barkun AN, Bardou M, Kuipers EJ, et al. International consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding. Ann Intern Med 2010;152:101–113.

  105. Bjorkman DJ. Endoscopic diagnosis and treatment of nonvariceal upper gastrointestinal hemorrhage. In: Yamada T, Alpers DH, Kalloo KN, et al., eds. Textbook of Gastroenterology, 5th ed. Hoboken, NJ: Wiley-Blackwell, 2009:3018–3031.

  106. American Society of Health-System Pharmacists therapeutic guidelines on stress ulcer prophylaxis. Am J Health Syst Pharm 1999;56:347–379.

  107. Ali T, Harty RF. Stress-induced ulcer bleeding in critically ill patients. Gastroenterol Clin North Am 2009;38:245–265.

  108. Stollman N, Metz DC. Pathophysiology and prophylaxis of stress ulcer in intensive care unit patients. J Crit Care 2005;20:35–45.

  109. Laine L, Takeuchi K, Tarnawski A. Gastric mucosal defense and cytoprotection: Bench to bedside. Gastroenterology 2008;135:41–60.

  110. Devlin JW, Welage LS, Olsen KM. Proton pump inhibitor formulary considerations in the acutely ill—Part 2: Clinical efficacy, safety, and economics. Ann Pharmacother 2005;39:1844–1851.

  111. Elmunzer BJ, Inadomi JM, Elta GH. Risk stratification in upper gastrointestinal bleeding. J Clin Gastroenterol 2007;41:559–563.

  112. Chiu PWY, Ng EKW. Predicting poor outcome from acute upper gastrointestinal hemorrhage. Gastroenterol Clin North Am 2009;38:215–230.

  113. Blatchford O, Murray WR, Blatchford M. A risk score to predict need for treatment for upper-gastrointestinal haemorrhage. Lancet 2000;356:1318–1321.

  114. Rockall TA, Logan RF, Devlin HB, et al. Risk assessment after acute upper gastrointestinal haemorrhage. Gut 1996;38:316–321.

  115. Leontiadis GI, Sharma VK, Howden CW. Proton pump inhibitor therapy for peptic ulcer bleeding: Cochrane Collaboration meta-analysis of randomized controlled trials. Mayo Clin Proc 2007;82:286–296.

  116. Leontiadis GI, Howden CW. The role of proton pump inhibitors in the management of upper gastrointestinal bleeding. Gastroenterol Clin North Am 2009;38:199–213.

  117. Bardou M, Toubouti Y, Benhaberou-Brun D, et al. Meta-analysis: Proton-pump inhibitors in high-risk patients with acute peptic ulcer bleeding. Aliment Pharmacol Ther 2005;21:677–686.

  118. Sung JYJ, Barkun A, Kuipers EJ, et al. Intravenous esomeprazole for prevention of recurrent peptic ulcer bleeding: A randomized trial. Ann Intern Med 2009;150:455–464.

  119. Lau JY, Leung WK, Wu JCY, et al. Omeprazole before endoscopy in patients with gastrointestinal bleeding. N Engl J Med 2007;356:1631–1640.

  120. Sung JJ, Chan FK, Lau JY. The effect of endoscopic therapy in patients receiving omeprazole for bleeding ulcers with nonbleeding visible vessels or adherent clots. Ann Intern Med 2003;139:237–243.

  121. Gisbert JP, Abraiira V. Accuracy of Helicobacter pylori diagnostic tests in patients with bleeding peptic ulcer: A systematic review and meta-analysis. Am J Gastroenterol 2006;101:848–863.

  122. Jung R, MacLaren R. Proton-pump inhibitors for stress ulcer prophylaxis in critically ill patients. Ann Pharmacother 2002;36:1929–1937.

  123. Cook D, Guyatt G, Marshall J, et al. A comparison of sucralfate and ranitidine for the prevention of upper gastrointestinal bleeding in patients requiring mechanical ventilation. N Engl J Med 1998;338:791–797.

  124. Daley RJ, Rebuck JA, Welage LS, Rogers FB. Prevention of stress ulceration: Current trends in critical care. Crit Care Med 2004;32:2008–2013.

  125. Conrad SA, Gabrielli A, Margolis B, et al. Randomized, double-blind comparison of immediate-release omeprazole oral suspension versus intravenous cimetidine for the prevention of upper gastrointestinal bleeding in critically ill patients. Crit Care Med 2005;33:760–765.

  126. Somberg L, Morris J, Fantus R, et al. Intermittent intravenous pantoprazole and continuous cimetidine infusion: Effect on gastric pH control in critically ill patients at risk of developing stress-related mucosal disease. J Trauma 2008;64:1202–1210.

  127. Phillips JO, Metzler MH, Palmirei MT, et al. A prospective study of simplified omeprazole suspension for the prophylaxis of stress-related mucosal damage. Crit Care Med 1996;24:1793–1800.

  128. Lasky MR, Metzler MH, Phillips JO. A prospective study of omeprazole suspension to prevent clinically significant gastrointestinal bleeding from stress ulcers in mechanically ventilated trauma patients. J Trauma 1998;44:527–533.

  129. Devlin JW, Welage LS, Olsen KM. Proton pump inhibitor formulary considerations in the acutely ill—Part 1: Pharmacology, pharmacodynamics and available formulations. Ann Pharmacother 2005;39:1667–1677.

  130. Olsen KM, Devlin JW. Comparison of the enteral and intravenous lansoprazole pharmacodynamic responses in critically ill patients. Aliment Pharmacol Ther 2008;28:326–333.