Excess acid production is common in duodenal ulcer disease, with significant increases noted in basal and peak acid outputs, parietal and chief cell masses, and responses to food and hormonal stimulation. Zollinger-Ellison syndrome (with its associated hypergastrinemia and parietal cell overproduction) is the prototypical acid hypersecretory condition resulting in ulcer formation. Some patients with duodenal ulcer demonstrate rapid gastric emptying, which raises the acid exposure of the proximal duodenum. Pepsin secretion is also elevated in duodenal ulcer disease. Gastric acid production is relatively normal in patients with gastric ulcers.

Peptic ulcer disease results from compromise of the major determinants of gastric mucosal integrity, namely, the secretion of mucus, the production of bicarbonate, and cellular repair. The mucous barrier may become compromised by increased degradation of mucus, decreased secretion, or the production of defective mucus. Bile acids, pepsin, pancreatic enzymes, and mechanical forces contribute to the degradation of mucus. Gastric prostaglandin production appears to be important to sustaining the production of mucus, the secretion of bicarbonate, and mucosal repair. By helping to maintain a neutral pH and aqueous environment at the surface of the gastric epithelium, mucus and bicarbonate protect the mucosa from acid, pepsin, and other potentially injurious agents (see later discussion).

Aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) impair the principal mechanisms of mucosal protection and account for a major portion of peptic ulcer disease in modern societies. By inhibiting cyclooxygenases (COX) 1 and 2 and their conversion of arachidonic acid to prostaglandins, aspirin and the so-called nonselective NSAIDs block production of not only the COX-2-dependent prostaglandins important to inflammation, fever, and pain but also the COX-1-derived prostaglandins involved in mucosal protection, platelet aggregation, and renal function. The selective NSAIDs that block only COX-2 (the so-called COX-2 drugs) avoid much of the mucosal injury associated with the use of nonselective preparations.

With long-term use, all nonselective NSAIDs (including enteric-coated and nonaspirin salicylates and NSAID prodrugs) are capable of producing gastric ulcers. Established risk factors include advanced age, concurrent use of corticosteroids, prior history of ulcer, high NSAID doses, and possibly concurrent H. pylori infection, alcohol use, and smoking. Despite some relative in vitro differences in COX selectivity among the nonselective agents, the clinical risks of major gastrointestinal (GI)
injury and clinically important ulceration are similar, estimated by the U.S. Food and Drug Administration (FDA) to be 2% per patient-year of NSAID use. NSAID-associated risk generally increases with dose and duration of therapy. A higher risk with increasing dose is also seen with aspirin, but low doses (e.g., 81 mg/d) are not risk-free. About 15% of long-term NSAID users demonstrate gastric ulceration at endoscopy; up to one fourth of complications have been observed within the first month of therapy. The elderly are at greatest risk (reported relative risk, about 4.0).

In addition to inhibiting prostaglandin synthesis, many NSAID preparations produce acute diffuse mucosal injury by means of a direct erosive effect. Endoscopic study has shown that significant mucosal injury results from both plain and tablet forms of buffered aspirin, although not with entericcoated aspirin preparations, unless gastric emptying is delayed. Similar acute erosions occur with uncoated NSAIDs, but not with enteric-coated preparations or prodrug formulations. Such diffuse acute injury is rarely associated with symptoms or clinically significant ulceration, although minor occult bleeding may ensue. Concurrent H. pylori infection, smoking, alcohol use, alcohol-related disease, and preexisting peptic ulcer disease greatly increase the risk for ulcer and ulcer complications in NSAID-treated patients.

NSAID use is associated primarily with gastric ulceration and not with new duodenal ulceration, but preexisting duodenal ulceration may be exacerbated and complicated by NSAID use. It is estimated that more than half of ulcer complications associated with NSAID use occur in patients with preexisting duodenal ulcers. Much of the ulcer morbidity associated with NSAID use may be related to preexisting, subclinical disease.

Infection with H. pylori has emerged as a major, if not the major, precipitant of peptic ulcer disease. The organism’s flagellated anatomy and urea-splitting capability make it well suited to survive in the acidic mucoid environment of the gastric mucosa. It attaches to epithelial cells and, without invading, induces apoptosis and an inflammatory response. These reactions lead to cell death and ulcer formation. H. pylori infection is associated with 95% to 99% of cases of duodenal ulcer and ulcer recurrence and with more than 90% of gastric ulcers unrelated to NSAID use. Helicobacter infection also is strongly associated with antral gastritis and is a major risk factor for gastric carcinoma and mucosaassociated lymphoid tissue lymphoma of the stomach; its eradication results in remission of the latter. Cancer risk appears to increase when H. pylori is present in the setting of atrophic gastritis.

Cure of Helicobacter infection can speed ulcer healing and greatly reduce rates of recurrence. The organism does not appear to cause ulcers per se but rather seems to enhance mucosal susceptibility to the injurious actions of acid and pepsin. Acid hypersecretors may be particularly vulnerable to ulcer formation in the context of Helicobacter infection and subject to slow healing and a high rate of recurrence. The prevalence of H. pylori infection increases with age, approaching 90% in ulcer patients older than age 65 years. The mode of transmission is unclear, although person-to-person fecal-oral spread is suspected because clusters occur in families.

Stress has long been considered a key precipitant, a view supported by a higher incidence of chronic stress in ulcer patients than in controls, increased acid production in response to stress, and a more prolonged course and poorer prognosis in patients with chronic severe anxiety. Patients in whom ulcers develop view life stresses more negatively than do controls. Acid hypersecretion and ulcer formation have been observed in small-scale studies of patients undergoing severe emotional stress; with the subsidence of stress, acid secretion falls and ulcers heal. Confirmation of these small-scale observations will strengthen the association between stress and ulcer disease.

Smoking is an important risk factor identified by epidemiologic studies. For example, ulcers are twice as likely to develop in cigarette smokers as in nonsmokers. The risk for gastric ulcer correlates with the number of cigarettes smoked, and patients with ulcers have increased rates of smoking. The rates of recurrence are dramatically increased in patients who smoke, and healing is markedly slowed. Impaired prostaglandin production has been demonstrated in the gastric mucosa of smokers.

Alcohol and coffee have also been implicated. Coffee, including decaffeinated forms, stimulates acid secretion, as do other caffeine-containing beverages, but evidence proving ulcer causation is lacking. Ethanol can compromise the mucosal barrier and cause gastritis, and beer is almost as potent a stimulant of acid secretion as is gastrin. Nonetheless, the data on the link between alcohol use and ulcer disease are conflicting. However, patients with alcohol-related cirrhosis are at an increased risk for ulcer formation and complications.

Their contribution has been debated ever since these agents first became available, with randomized, controlled trials and meta-analyses producing conflicting results. Discrepancies are related in part to a failure to control for concurrent NSAID use. The risk appears to be nil if the patient is not taking NSAIDs concurrently and the dose is less than 30 mg/d. However, highdose steroid therapy (>30 mg/d) may confer an increment of risk, as might the underlying disease for which the prednisone is being taken. In the absence of concurrent NSAID use, the overall risk appears to be very modest.

Heredity plays some role. The incidence of ulcer in the parents, siblings, and children of ulcer patients is increased, and studies of twins show a greater concordance (e.g., both twins affected) among identical than among fraternal twins. Increased mealstimulated gastrin release and pepsin secretion have been found to be hereditary traits among ulcer patients and their families.

The clinical course prior to treatment of H. pylori was characterized by the majority of patients becoming pain-free within the first 4 weeks and their ulcers healing completely by 4 to 12 weeks (depending on the size of the lesion). Of interest, there was little correlation between the cessation of pain and objective healing. The 5-year recurrence rates were high (30% to 90%). No correlation was found between the recurrence rate and ulcer size, duration of symptoms, or location. Recurrent ulcers healed just as rapidly and completely as original lesions. The rate of development of a major complication, such as hemorrhage, perforation, or obstruction, was less than 1% annually. Bleeding was slightly more common from duodenal than from gastric ulcers and two to three times more common than perforation.

With the application of therapies to eradicate H. pylori infection and markedly suppress acid production, the clinical course of peptic ulcer disease has improved substantially. Complete healing is now achieved in up to 95% of cases within 4 to 6 weeks of initiation of treatment, and the 1-year risk for relapse is now about 5% with the eradication of Helicobacter (but >50% when acid suppression is the only form of initial therapy).

Serologic testing is used by many primary care physicians in lieu of endoscopic examination or breath testing because it is readily available, convenient, reasonably sensitive and specific, and inexpensive. Sensitivity and specificity for regionally adjusted testing (there is some geographic variation in strains) approach those for breath testing. This approach has been found to be cost-effective for the detection of infection in persons who have never been treated, but it cannot be used definitively to determine cure or recurrence of infection because the test result often remains positive (although antibody titers do decline) for years after the eradication of the organism. Positivity means only prior exposure to the organism. Testing for eradication can be achieved noninvasively by performing a “¹³C-urea breath test or a stool antigen test.

Based on the use of ¹³C- or ¹⁴C-labeled carbon dioxide, breath testing takes advantage of urea splitting by live organisms and, unlike serology, provides direct evidence of active infection. Consequently, breath testing is very useful as a noninvasive method for determining eradication of infection, which becomes important in the settings of relapse, recurrence, and disease complicated by bleeding. Test sensitivity and specificity approach those of endoscopic measures and can be used in lieu of invasive testing. The test is costly and not widely available, limiting it is use in primary care practice, both for diagnosis and for the determination of eradication, but its use can obviate the need for invasive testing. Testing for cure or relapse requires waiting 4 weeks after treatment.

The detection of H. pylori antigen in the stool provides a convenient alternative to other modes of testing, especially for determining the eradication of infection. Under study conditions, the sensitivity and specificity of the monoclonal antibody stool antigen test were 97% and 94%, respectively. Negative and positive predictive values were 97% and 95%, respectively. The diagnostic accuracy was 96%; the likelihood ratio for a positive test was 17 and for a negative test was 0. Although the ¹³C-urea breath test is the most accurate among the noninvasive tests for the determination of cure, this test offers a reasonable alternative.

Any program to speed healing, limit recurrences, and prevent complications must address both the short-term and prolonged use of agents injurious to the mucosal barrier.