Alan BR Thomson, MD,
Professor of Medicine, Division of
Gastroenterology, University of Alberta,
Canada
Nothing to disclose.
Coauthor(s)
Jon Meddings, MD,
Head, Department of Medicine, Professor,
Department of Internal Medicine, University of Alberta,
Canada
Nothing to disclose.
Shane M Devlin, MD, FRCP(C),
Clinical Assistant Professor, Department of
Internal Medicine, Peter Lougheed Center, University of
Calgary, Canada
Nothing to disclose.
Yvette PY Leung, MD,
Fellow in Gastroenterology, Department of
Internal Medicine, University of Calgary Medical
Clinic
Nothing to disclose.
Specialty Editor(s)
Alex J Mechaber, MD, FACP,
Associate Dean for Undergraduate Medical
Education, Associate Professor of Medicine, University of
Miami Miller School of Medicine
Nothing to disclose.
BS Anand, MD,
Professor, Department of Internal Medicine,
Division of Gastroenterology, Baylor College of
Medicine
Nothing to disclose.
Francisco Talavera, PharmD, PhD,
Senior Pharmacy Editor,
eMedicine
eMedicine Salary Employment
Waqar A Qureshi, MD,
Associate Professor of Medicine, Chief of
Endoscopy, Department of Internal Medicine, Division of
Gastroenterology, Baylor College of Medicine and Veterans
Affairs Medical Center
Nothing to disclose.
Chief Editor
Julian Katz, MD,
Clinical Professor of Medicine, Drexel
University College of Medicine; Consulting Staff, Department
of Medicine, Section of Gastroenterology and Hepatology,
Hospital of the Medical College of
Pennsylvania
Nothing to disclose.
Background
When we speak of duodenal ulcers, we often imply that these are part of what is known as peptic ulcer disease; duodenal ulceration may be only rarely due to other conditions. Most duodenal ulcers are associated with a Helicobacter pylori infection, or the use of gastric irritating medications such as aspirin, non-steroidal anti-inflammatory drugs (NSAIDs), or bisphosphonates.
Duodenal ulcers are a common condition characterized by the presence of a well-demarcated break in the mucosa that may extend into the muscularis propria of the duodenum (see images below). More than 95% of duodenal ulcers are found in the first part of the duodenum; most are less than 1 cm in diameter.[1]
View Image
Duodenal ulcer in an elderly patient who presented with melena and hypotension.
View Image
Duodenal ulcer in a 35-year-old woman who presented with tarry stools and a hemoglobin level of 75 g/L.
View Image
Duodenal ulcer in a 65-year-old man with osteoarthritis who presented with hematemesis and melena stools. The patient took naproxen on a daily basis.
Proper diagnosis of duodenal ulcers is important because prompt initiation of treatment can effectively prevent potentially serious complications.
For excellent patient education resources, visit eMedicine's Esophagus, Stomach, and Intestine Center. Also, see eMedicine's patient education articles Peptic Ulcers, Helicobacter Pylori (H. Pylori), and Gastritis.
The duodenal mucosa resists damage from the effect of aggressive factors, such as gastric acid and the proteolytic enzyme pepsin, with the help of several protective factors, such as a mucous layer, bicarbonate secretion, and protective prostaglandins.
The epithelial cells of the stomach and duodenum secrete mucus in response to irritation of the epithelial lining and as a result of cholinergic stimulation. A portion of the gastric and duodenal mucus exists in the form of a gel layer, which is impermeable to acid and pepsin. Other gastric and duodenal cells secrete bicarbonate, which aids in buffering acid that lies near the mucosa. Prostaglandins of the E type (PGE) have an important protective role, because PGE increases the production of both bicarbonate and the mucous layer.
In the event of acid and pepsin entering the epithelial cells, additional mechanisms are in place to reduce injury. Within the epithelial cells, ion pumps in the basolateral cell membrane help to regulate intracellular pH by removing excess hydrogen ions. Through the process of restitution, healthy cells migrate to the site of injury. Mucosal blood flow removes acid that diffuses through the injured mucosa and provides bicarbonate to the surface epithelial cells.
A duodenal ulcer occurs when an alteration occurs in the aggressive and/or protective factors such that the balance is in favor of gastric acid and pepsin. Any process that increases gastric acidity (eg, individuals with increased maximal and basal acid output), decreases prostaglandin production (eg, NSAIDs), or interferes with the mucous layer (eg, H pylori infection) can cause such an imbalance and lead to peptic ulcer disease.
Full understanding of the pathophysiology and pathogenesis of duodenal ulcers requires a brief discussion of the 2 major etiologies: NSAID use and H pylori infection. NSAIDs are pathogenic through their inhibition of the cyclooxygenase-1 (COX-1) pathway, which normally produces protective prostaglandins. These prostaglandins are protective because they augment both bicarbonate and mucous production, as mentioned above. However, perhaps more important, prostaglandins augment mucosal blood flow, and their inhibition leads to impairment of blood flow, leaving the mucosa vulnerable to damage.
Infection with H pylori is likely pathogenic by means of a variety of indirect mechanisms as the organism does not generally colonize the duodenum. The mechanisms are described as follows[2] :
H pylori infection that follows an antral predominant pattern leads to an inflammatory state in which high levels of tumor necrosis factor-alpha (TNF-alpha) and other cytokines are produced. These stimulate gastric acid production directly by increasing gastrin release from G cells and inhibit somatostatin production by antral D cells. This leads to a net increase in gastric acid secretion, which leads to an increased acid load in the duodenum, overwhelming the mucosal defense.[3, 4]
Duodenal acid exposure can lead to gastric metaplasia, whereby the duodenal mucosa can take on characteristics of gastric mucosa. H pylori can then colonize the duodenal mucosa and adhere to cells. This adherence leads to a variety of second-messenger signals, which invoke an immunologic response against those cells causing mucosal damage by host neutrophils and other inflammatory cells.
H pylori also affects the gastric and duodenal mucous layer, because this organism produces proteases that degrade the protective mucous layer. Moreover, H pylori infection decreases the production of epidermal growth factor, which normally promotes healing of gastric and duodenal mucosa.
H pylori organisms produce urease. Urease hydrolyzes urea to ammonia and carbon dioxide. Hydroxide ions produced by equilibration of ammonia with water may damage the gastric and duodenal mucosa. H pylori produces proteins that may serve as chemotactic factors for neutrophils and monocytes, which act as proinflammatory cells. H pylori also affects the gastric and duodenal mucous layer, because these organisms produce proteases that degrade the protective mucous layer. H pylori does not lead to the development of gastric and duodenal ulcers through alteration of the bacterial flora.
H pylori gene cagA and s1 or m1 forms of VacA (especially the intermediate region) are more common in disease-associated strains,[5] especially cagPA1 and vacA s1 genotypes.[6] CagA may be a risk factor for intestinal metaplasia.[7]
The cellular immune response in the H pylori– infected gastric mucosa is predominantly the T helper cell type 1 (Th1) type. Interleukin (IL)-18 levels are increased in this setting.[8] The acidic environment of the gastric lumen induces the gastric mucosal production of IL-8, which is involved in the proinflammatory pathways.[9]
The receptors for bacterial adhesins include the Lewis b structures of the secretory M4C 5AC mucin, and in H pylori –positive duodenal ulcer patients, the initial low level of MUC1 mucin in gastric juice increases with eradication of this organism.[10]
The secretion of soluble triggering receptor is increased in H pylori –positive persons with gastric but not duodenal ulcers,[11] suggesting that the host response may determine the site of ulceration. The age of the host may also influence the type of immune mechanism(s) involved in the pathogenesis of the H pylori –associated gastroduodenal disease.[12]
The 3 putative regions of H pylori include strain-specific genes, phase-variable genes, and genes with variable structures/genotype.[13] Some strain-specific genes are in the plasticity regions, away from the cag pathogenicity island.[14]
Enhanced expression and release of endothelins (ET-1), an increase in the expression of the immediate early gene EGR-1, and the upregulation of angiogenic growth factors necessarily initiated by hydrochloric acid and proteolytic enzymes are also mechanisms in duodenal ulceration.[15]
There are new controversies emerging in the H pylori story, including investigators that question whether H pylori does cause duodenal ulcers and whether H pylori infection persists until it is pharmacologically eradicated.[16]
The prevalence of duodenal ulcers is estimated to be 6-15% in the general population. Most individuals do not have clinically significant ulcer disease, peptic ulcer disease is decreasing,[17] and ulcers have become a rare cause for hospital admission.[18] The prevalence is linked to the presence of H pylori. Approximately only 10% of young persons have H pylori infection, and the proportion of people with the infection increases steadily with age.
Approximately 10% of the US population has evidence of a duodenal ulcer at some time. Of those infected with H pylori, the lifetime prevalence is approximately 20%. Overall, the incidence of duodenal ulcers has been decreasing over the past 3-4 decades.
International
As in the US, duodenal ulcer disease prevalence is linked to H pylori infection. The prevalence of H pylori infection varies widely among countries and even in regions within countries.
Mortality/Morbidity
Duodenal ulcers cause significant morbidity, which is mainly related to pain, and hospitalization for complications, such as ulcer hemorrhage, perforation, penetration, and obstruction.
Rates of duodenal ulcer complications and mortality are generally increased in elderly patients, perhaps because of the high incidence of comorbid diseases in this group and their increased use of NSAIDs.
With NSAID-related ulcers, the incidence of perforation is approximately 0.3% per patient year and of obstruction, approximately 0.1% per patient year. Combining both duodenal ulcers and gastric ulcers, the rate of any complication in all age groups combined is approximately 1-2% per ulcer per year.
Over the last 20 years, the mortality rate in the setting of ulcer hemorrhage has not changed appreciably despite the advent of histamine-2 receptor antagonists (H2RAs) and proton pump inhibitors (PPIs). However, evidence from meta-analyses and other studies has shown a decreased mortality rate from bleeding peptic ulcers when intravenous PPIs are used after successful endoscopic therapy.[19, 20, 21, 22] In general, if one considers all patients with duodenal ulcers, the mortality rate due to ulcer hemorrhage is approximately 5%.
The mortality rate of patients requiring surgical intervention for complications of duodenal ulcers, such as perforation and obstruction, is significantly higher than the general rate and related to the age of the patient. Most deaths in this setting result from postoperative complications.
Giant duodenal ulcers have high rates of morbidity and mortality.[23]
Race
No specific relationship between race and the occurrence of duodenal ulcers exists. In general, areas with a high prevalence of H pylori infection have a high prevalence of duodenal ulcers.
Sex
Over the last several years, a trend toward an increasing incidence of duodenal ulcers in females and decreasing incidence in males has been observed, especially in younger males, in whom the prevalence of H pylori infection is decreasing.
Historically, duodenal ulcers were believed to be more common in men than in women. Today, the prevalence is probably equal in men and women.
Age
The prevalence of duodenal ulcers increases with age. This is probably related to the increased prevalence of H pylori infection in older age groups, coupled with increased use of NSAIDs.
Patients with duodenal ulcers have a variety of clinical presentations, ranging from individuals who are completely asymptomatic to those who develop severe complications, such as gastrointestinal (GI) hemorrhage. Some generalizations can be made with respect to common clinical presentations of duodenal ulcers.
Some common symptoms in patients with duodenal ulcers follow:
Epigastric pain can be sharp, dull, burning, or penetrating.
Many patients experience a feeling of hunger.
The pain may radiate into the back.
About 20-40% of patients describe bloating, belching, or symptoms suggestive of gastroesophageal reflux.
Ulcer-related pain generally occurs 2-3 hours after meals and often awakens the patient at night. This pattern is believed to be the result of increased gastric acid secretion, which occurs after meals and during the late night and early morning hours when circadian stimulation of gastric acid secretion is the highest.
About 50-80% of patients with duodenal ulcers experience nightly pain, as opposed to only 30-40% of patients with gastric ulcers and 20-40% of patients with nonulcer dyspepsia (NUD).
Pain is often relieved by food, a finding often cited as being specific for a duodenal ulcer. However, this symptom is present in only 20-60% of patients and is probably not specific for duodenal ulcers.
The pain of duodenal ulcers is generally episodic; however, the pain can evolve into a chronic, daily occurrence in some patients.
A change in the patient's usual pattern of ulcer pain should be considered serious, because it may herald an imminent complication. When food or antacids fail to relieve the pain or when the pain begins to radiate to new anatomic locations, a high index of suspicion of a complication is warranted.
Concern is especially warranted in the setting of new-onset nausea and vomiting, decreased appetite, and weight loss.
GI bleeding is a common complication of duodenal ulcers and can have serious consequences.
Patients may present with melena, coffee-ground emesis, or hematemesis.
The passage of frank blood in the stool or maroon-colored stool in the presence of a bleeding duodenal ulcer suggests precipitous GI bleeding.
Patients who develop gastric outlet obstruction as a result of a chronic, untreated duodenal ulcer usually report a history of fullness and bloating associated with nausea and emesis that occurs several hours after food intake. A common misconception is that adults with gastric outlet obstruction present with nausea and emesis immediately after a meal.
A few individuals with duodenal ulcers are completely asymptomatic.
According to one study, typical epigastric pain was rare in patients older than 65 years with peptic ulcer disease (ie, gastric ulcer and duodenal ulcer).
Elderly patients are more likely than younger patients to present in an asymptomatic fashion, which is especially common in the setting of NSAID use.
No characteristic physical findings are associated with duodenal ulcers. In general, most patients have tenderness over the epigastrium, but this finding has a low sensitivity and specificity. Less often, tenderness is present over the right upper quadrant (RUQ), left upper quadrant (LUQ), or supraumbilical region. Most patients with an uncomplicated duodenal ulcer do not have any other physical findings.
In the presence of a complication, such as gastric outlet obstruction, the physician may note upper abdominal distention and hear a succussion splash on auscultation.
Perforation of peptic ulcers is common in older persons using aspirin (ASA)/NSAIDs.[24]
Perforation usually results in classic findings of diffuse peritonitis with abdominal rigidity, guarding, and rebound tenderness. Bowel sounds may initially be hyperactive but, with time, become absent.
Risk factors for mortality from peptic ulceration include preoperative metabolic acidosis, renal insufficiency at the time of admission, reduced serum albumin concentrations on admissions or poor postoperative nutrition.[25]
The understanding of the etiology of duodenal ulcers changed dramatically in the latter part of the 20th century. Historically, duodenal ulcers were thought to be a disease related to diet and environmental stress alone. Subsequent studies revealed the importance of pepsin and acid secretion in the pathogenesis of duodenal ulcers. The most revolutionary change in the knowledge of duodenal ulcers was the discovery in 1982 that the bacterium H pylori was present in most affected patients.
NSAIDs
Long before the discovery of H pylori, NSAIDs were known to be associated with GI toxicity, including the formation of gastric ulcers and duodenal ulcers.[26]H pylori enhances the damaging effect of NSAIDs on the gastroduodenal mucosa,[27] and NSAID GI-complications also occur in children.[28]
As many as 4-10% of patients on daily therapeutic-dose NSAIDs develop a duodenal ulcer within 3 months of initiation of therapy, and up to 1% of these duodenal ulcers are clinically significant.
A clear dose-response relationship exists, with high NSAID doses associated with increased risk of duodenal mucosal damage. Clinically, NSAID-induced duodenal ulcers are most likely to bleed. In one study, NSAID use was associated with a relative risk for bleeding of 8.4, as opposed to 1.5 for H pylori –associated duodenal ulcers. Patients taking NSAIDs, especially elderly patients, are most likely to present with an asymptomatic bleeding duodenal ulcer.
Factors associated with an increased risk of duodenal ulcers in the setting of NSAID use are a history of previous peptic ulcer disease, advanced age, female sex, high doses or combinations of NSAIDs, long-term NSAID use, concomitant use of anticoagulants, and severe comorbid illnesses. Corticosteroids do not increase the risk of duodenal ulcers by themselves, but the risk of duodenal ulcers is increased when corticosteroids are used in combination with NSAIDs, as compared to NSAID use alone.
Aspirin is a significant risk factor for damage to the gastroduodenal mucosa.[29, 30] ; however, its damaging effect does not depend on an H pylori infection.[31]
Giving a PPI with the episodic use of the naproxen (500 mg bid) reduces gastroduodenal ulcers from 46.9% to 11.8%,[32] as well as reduces the risk of ulcers in persons with continuous low dose aspirin.[33]
H pylori eradication in persons on long-term therapy with an NSAID has no beneficial effect on the development of ulcers, erosions, or dyspepsia.[34]
Using the antiplatelet drug clopidrogel rather than aspirin plus a PPI causes fewer GI complications in persons with no pervious history of duodenal or gastric ulcers, but the difference is very small and the number-needed-to-treat (NNT) to prevent 1 episode of bleeding per year is approximately 200.[35]
Acetaminophen given with a PPI is associated with the same increased risk of hospitalization as with a traditional NSAID plus a PPI.[36]
Although initially controversial, most evidence now supports the assertion that H pylori and NSAIDs are synergistic with respect to the development of peptic ulcer disease. Eradication of H pylori in the setting of chronic NSAID use is associated with a decreased risk of ulcer bleeding.[37] A meta-analysis found that H pylori eradication in NSAID-naive users before the initiation of NSAIDs was associated with a decrease in peptic ulcers.[38]
The potential for decreased GI mucosal injury with newer COX-2–selective inhibitors, celecoxib and rofecoxib, has been emphasized. On September 30, 2004, Merck & Co, Inc, announced a voluntary withdrawal of rofecoxib (Vioxx) from the US and worldwide markets because of its association with an increased rate of cardiovascular events (including heart attacks and strokes) compared with placebo. The newer NSAIDs do not inhibit COX-1 and, therefore, do not have the disadvantage of reducing the synthesis of protective prostaglandins. Overall, selective COX-2 inhibitors are associated with, at best, a modest decrease in the risk of ulcer bleeding. One study showed that the combination of a traditional NSAID with a daily PPI had the same risk of bleeding as that of a COX-2 inhibitor alone.[39]
H pylori bacteria
H pylori bacteria are small, microaerophilic, spiral-shaped, gram-negative rods. The presence of H pylori in the stomach and duodenum is probably the most common bacterial infection in the world. Areas with a high prevalence of H pylori infection have a high incidence of duodenal ulcer.
H pylori infection is generally regarded as the most important etiologic factor in the development of duodenal ulcer.[40] Most authors regard H pylori as the cause of 85-95% of duodenal ulcers. Even in chronic scarred duodenal ulcers, H pylori may be found in 63% of subjects.[41]
The bacterium can induce duodenal mucosal damage by means of several mechanisms (see Pathophysiology).
All evidence supports the assertion that H pylori is the major cause of duodenal ulcers. However, the risk of developing a duodenal ulcer in an individual infected with H pylori is only about 1% per year, and only 10-15% of individuals with H pylori infection develop a duodenal ulcer at any point in life. Therefore, other pathogenic factors must function either independently or in concert with H pylori to produce duodenal ulcers.
Factors other than H pylori are likely involved in the pathogenesis of peptic ulcer disease in persons with hepatic cirrhosis[42] and chronic renal failure on dialysis.[43] The eradication of H pylori does not prevent ulcer formation or complications in persons with cirrhosis.[44] Endoscopic lesions of the upper GI tract are less common in those persons as compared with individuals without end-stage renal disease (ESRD).[45] Chronic obstructive pulmonary disease (COPD) increases the 30-day mortality in patients with bleeding or perforated peptic ulcers.[46]
Duodenal ulcers that are not associated with H pylori or NSAIDs may include Crohn disease,[47] lymphoma or adenocarcinoma, selective internal radiation therapy,[48] or localized ischemia from sickle cell disease.[49]
H pylori –negative duodenal ulcer
In one study, among 608 patients with duodenal ulcers, 42 (6.9%) were classified as idiopathic, but 3% had isolated duodenal colonization.[50] Some studies have shown that the proportion of H pylori –related duodenal ulcers is significantly less than the commonly reported 85-95%. One group examined nearly 2400 cases of endoscopically proven, non–NSAID-related duodenal ulcers and found that only 73% patients were positive for H pylori.[51] However, other studies have produced conflicting results, and another study showed that antibiotic use within 1 month of diagnosis may have resulted in false-negative results.
Epstein-Barr virus either alone or in combination with H pylori infection may be associated with peptic ulcer disease.[52]
In up to one third of patients with duodenal ulcers, basal acid output (BAO) and maximal acid output (MAO) are increased. In one study, increased BAO was associated with an odds ratios [OR] of up to 3.5, and increased MAO was associated with an OR of up to 7, for the development of duodenal ulcers. People at especially high risk are those with a BAO greater than 15 mEq/h. The increased BAO may reflect the fact that, in a significant proportion of patients with duodenal ulcers, the parietal cell mass is increased to nearly twice that of the reference range.[53]
In addition to the increased gastric and duodenal acidity observed in some patients with duodenal ulcers, accelerated gastric emptying is often present. This acceleration leads to a high acid load delivered to the first part of the duodenum, where 95% of all duodenal ulcers are located. Acidification of the duodenum leads to gastric metaplasia, which indicates replacement of duodenal villous cells with cells that share morphologic and secretory characteristics of gastric epithelium. Gastric metaplasia may create an environment that is well suited to colonization by H pylori.
Tramadol also increases a person’s risk of perforation and mortality from peptic ulcers.[54]
Duodenal ulcers occur in 3.6% of dyspeptic persons infected with the human immunodeficiency virus (HIV) and treated with highly active antiretroviral therapy (HAART); gastric ulcers occur in 2.7%.[55]H pylori becomes more prevalent as the CD4 cell count rises, and endoscopy becomes more useful for persons with CD4 cell counts less than or equal to 200, when opportunistic infections and malignancies are more common.
Lifestyle factors
Smoking: Evidence that tobacco use is a risk factor for duodenal ulcers is not conclusive. Evidence supporting a pathogenic role for smoking comes from the finding that smoking may accelerate gastric emptying and decrease pancreatic bicarbonate production. However, studies have produced contradictory findings. In one prospective study of more than 47,000 men with duodenal ulcers, smoking did not emerge as a risk factor.[56] However, smoking in the setting of H pylori infection may increase the risk of relapse of PUD.[57]
Smoking is harmful to the gastroduodenal mucosa, and H pylori infiltration is denser in the gastric antrum of smokers.[58]
Alcohol use: Ethanol is known to cause gastric mucosal irritation and nonspecific gastritis. Evidence that consumption of alcohol is a risk factor for duodenal ulcer is inconclusive. A prospective study of more than 47,000 men with duodenal ulcer did not find an association between alcohol intake and duodenal ulcer.[56]
Caffeine intake: Little evidence suggests that caffeine intake is associated with an increased risk of duodenal ulcers.
Diet: Historically, diet was considered one of the primary causes of peptic ulcer disease. However, current knowledge indicates that diet probably has little influence on the pathogenesis of duodenal ulcers. Deficiency of certain essential fatty acids necessary for prostaglandin production has been examined as a possible risk factor. Moreover, some physicians hypothesize that regional variability of duodenal ulcer prevalence not directly related to H pylori prevalence may be related to diet. In general, evidence linking diet and duodenal ulcer is weak.
Genetics
More than 20% of patents have a family history of duodenal ulcers, compared with only 5-10% of control groups. In addition, weak associations have been observed between duodenal ulcers and blood type O. Furthermore, patients who do not secrete ABO antigens in their saliva and gastric juices are known to be at higher risk. The reason for these apparent genetic associations is unclear.
A rare genetic association exists between familial hyperpepsinogenemia type I (a genetic phenotype leading to enhanced secretion of pepsin) and duodenal ulcers. However, H pylori can increase pepsin secretion, and a retrospective analysis of the sera of one family studied before the discovery of H pylori revealed that their high pepsin levels were more likely related to H pylori infection.
Other causes
Acid hypersecretory syndromes
Gastrinoma (Zollinger-Ellison syndrome): First described in 1955, Zollinger-Ellison syndrome is caused by a tumor of pancreatic islet cells that produces gastrin. It is associated with gastric acid hypersecretion and development of peptic ulcer disease. From 0.1% to 1% of duodenal ulcers are thought to be secondary to an underlying gastrin-secreting tumor.[59] With the shortage of secretion for diagnostic testing in persons suspected of Zollinger-Ellison syndrome, the glucagon provocative test may prove to be a suitable alternative.[60]
Systemic mastocytosis: Systemic mastocytosis is a disease associated with diffuse infiltration of the skin, GI tract, bone marrow, spleen, and liver with mastocytes. Gastric acid hypersecretion occurs in response to histamine production by mastocytes.
Basophilia: In the setting of a myeloproliferative disorder, basophilia can be associated with duodenal ulcer secondary to histamine production, as is systemic mastocytosis. This tends to occur more frequently after chemotherapy-induced cell lysis that causes increased release of histamine from cells.
Other factors
Infection: Some evidence suggests that herpes simplex virus-1 (HSV-1) and cytomegalovirus (CMV) may be associated with duodenal ulcers and gastric ulcers in a minority of patients.
Chemotherapy: Chemotherapeutic agents, such as 5-fluorouracil (5-FU), methotrexate (MTX), and cyclophosphamide, have been associated with development of duodenal ulcers.
Radiation: Local radiation can result in mucosal damage, which may lead to the development of duodenal ulcers.
Crack cocaine: Use of crack cocaine causes localized vasoconstriction, and the reduced blood flow may lead to mucosal damage.
The risk of upper GI tract bleeding may be increased in users of the diuretic spirolactone,[61] or moderate and high affinity serotonin reuptake inhibitors.[62]
In general, laboratory studies are of little help in evaluating a patient with suspected duodenal ulcer, but such tests as complete blood cell (CBC) count, amylase and lipase determinations, and liver function tests (LFTs) may be useful in excluding other causes of upper abdominal pain.
Several noninvasive laboratory tests are available to aid in the diagnosis of H pylori infection.
Urea breath test: The patient ingests radiolabeled urea, which, in the presence of urease produced by H pylori, is metabolized to carbon dioxide and ammonia. A laboratory assay is then used to detect the radiolabeled carbon dioxide. This test has a sensitivity of 90-95%. The urea breath test can be used to diagnose infection, but it is more often used to evaluate the success of treatment of H pylori infection.
Serology: Enzyme-linked immunoassay (ELISA) can detect both immunoglobulin G (IgG) and A (IgA) antibodies directed against H pylori. The sensitivity of most serologic tests is approximately 95%, but the specificity is low (41-71%), so that the clinical utility of a negative H pylori serology test is good, but that of a positive test is poor.[63]
Fecal antigen test: The detection of H pylori in feces is emerging as a noninvasive method of detection. This test has mainly been used in pediatric settings.
Several investigative techniques can be used to diagnose duodenal ulcers.
Single-contrast barium radiography can detect 70-80% of duodenal ulcers (see image below).
View Image
Deformity of duodenal cap caused by recurrent ulceration. Single-contrast view.
The sensitivity increases to greater than 90% when double contrast radiography is performed and evaluated by an experienced radiologist (see the images below, which are of the same patient).
View Image
Double-contrast upper gastrointestinal series. Posterior wall duodenal ulcer (same patient in Images 5 and 6).
View Image
Lateral view of a posterior wall ulcer (same patient in Images 5 and 6).
The disadvantage of radiographic studies is that biopsy specimens of the lesion, either to test for H pylori infection or to evaluate for the presence of malignancy, cannot be obtained.
Other tests used to help diagnose H pylori infection include the rapid urease test, tissue culture, and histology.
The rapid urease test requires an endoscopically acquired biopsy specimen that is inserted into a receptacle containing a urea substrate and a pH indicator. In the presence of H pylori, the urea is metabolized and the color-sensitive pH indicator demonstrates a positive result.
Culture of tissue for the presence of H pylori is not a rapid or practical test and is available only in research settings.
As with culture, the presence of the organism can be detected histologically, but the expertise of a trained pathologist is required (see images below). Other tests are faster and more cost effective.
View Image
An antral gland of the stomach with a large Giemsa-stained colony of Helicobacter pylori in the lumen (arrow) at 250X power. Courtesy of Pantaleo Bufo....
View Image
A transverse section of the gastric lamina propria is shown in a patient with Helicobacter pylori infection. In the lower part, an antral gland of the....
Although not as sensitive as autopsy or visual inspection during surgery, esophagogastroduodenoscopy (EGD) is the most sensitive test available to detect duodenal ulcers. It has a sensitivity of greater than 95%.
EGD has a low specificity for the cause of dyspepsia (gastric or duodenal ulcer, or gastroesophageal reflux disease [GERD] vs normal endoscopy reflux disease or non-ulcer dyspepsia).
With EGD, the ulcer can be visualized, a biopsy specimen can be obtained, and, if required, bleeding ulcers can be treated directly. EGD is a more invasive test and requires conscious sedation in many patients.
Treatment of duodenal ulcers varies depending on the etiology and clinical presentation. The initial management of a stable patient with dyspepsia differs from the management of an unstable patient with upper GI hemorrhage. In the latter scenario, failure of medical management not uncommonly leads to surgical intervention.
Although the therapeutic principles of these distinct clinical scenarios share some similarities, they differ sufficiently to warrant separate discussions. Age is an independent risk factor for the incidence and mortality from bleeding peptic ulcer, with the risk increasing in persons older than 65 years, and increasing further in those older than age 75 years.[64] In one study, at least 2 risk factors (previous duodenal ulcer, H pylori infection, use of ASA/NSAID, and smoking) were present in two thirds of persons with acute gastroduodenal bleeding.[65]
The principles of management of bleeding peptic ulcers outlined below are equally applicable to both gastric and duodenal ulcers.
Medical management of bleeding duodenal ulcers
Urgent EGD is the treatment of choice in the setting of a bleeding duodenal ulcer for diagnostic and therapeutic reasons. Endoscopy provides an opportunity to visualize the ulcer, to determine the degree of active bleeding, and to attempt hemostasis by direct measures. Primary endoscopic hemostatic therapy (EHT) is successful in about 90% of patients; when this fails, transcatheter embolization may be useful.[66] (See Surgical Care for a full discussion of endoscopic therapy.) Medical management usually serves as an adjunct to direct endoscopic therapy.
Risk factors to predict rebleeding following EHT for nonvariceal upper GI bleeding include failure to use a PPI after the procedure; endoscopically demonstrated bleeding, especially peptic ulcer bleeding; EHT monotherapy; post-EHT use of heparin; and bleeding in the patient with moderate or severe cirrhosis.[67] Other predictors of rebleeding after EHT include pre-endoscopic hemodynamic instability, comorbid illness, active bleeding at endoscopy, large ulcer size, or posterior wall duodenal ulcer.[68] These high risk persons might be considered for initial care in the ICU, and follow-up ("second look") endoscopy, especially because many of these factors (advanced age, comorbidities, in-hospital bleeding, rebleeding, hypovolemic shock, need for surgery) are associated with hospital mortality.[69]
Acid suppression is the general pharmacologic principle of medical management of acute bleeding from a peptic ulcer. Reducing gastric acidity is believed to improve hemostasis primarily through the decreased activity of pepsin in the presence of a more alkaline environment. Pepsin is believed to antagonize the hemostatic process by degrading fibrin clots. By suppressing acid production and maintaining a pH above 6, pepsin becomes markedly less active.
Two classes of acid-suppressing medications currently in use are H2RAs and PPIs.[70] Both classes are available in intravenous or oral preparations. Examples of H2RAs include ranitidine, cimetidine, famotidine, and nizatidine. Examples of PPIs include omeprazole, pantoprazole, lansoprazole, and rabeprazole.
There is a very good safety profile for PPIs, although attention must continue to be focused on adverse effects, especially with long-term and/or high-dose therapy, such as Clostridium difficile infection, community-acquired pneumonia, hip fracture, and vitamin B12 deficiency.[71]
Long-term use of PPIs are also associated with decreased absorption of some medications. PPIs impair gastric secretion of acid; thus, absorption of any medication that depends on gastric acidity, such as ketoconazole and iron salt, is impaired on long-term PPI therapy. In fact, achlorhydria (absence of intragastric acidity) may be associated with iron deficiency anemia, because the ferric form of iron must be converted to the ferrous form by gastric acid. Most iron absorbed is in the ferrous form.
H2RAs are an older class of medications, and their use in the setting of an actively bleeding duodenal ulcer has been largely superseded by the use of PPIs. Many gastroenterologists assert that intravenous PPI therapy maintains hemostasis more effectively than intravenous H2RA. Thus, intravenous H2RA no longer has a role in the management of bleeding peptic ulcers.[72]
Andriulli et al evaluated the use of high- versus low-dose PPIs after endoscopic hemostasis in patients with peptic ulcer bleeding.[73] The primary end point was the in-hospital rebleeding rate (determined on repeat endoscopy). Patients with actively bleeding ulcers and those with nonbleeding visible vessel or adherent clot were treated with epinephrine injection and/or thermal coagulation, then randomized to receive an intensive regimen of 80-mg PPI bolus, followed by 8 mg/h as continuous infusion for 72 hours, or a standard regimen of a 40-mg PPI bolus daily, followed by saline infusion for 72 hours. After the infusion, all patients were given 20 mg PPI twice daily orally.[73]
Of 238 patients in the intensive PPI regimen group, bleeding recurred in 28 (11.8%); of 236 patients in the standard regimen group, bleeding recurred in 19 (8.1%) (P = 0.18).[73] Most of the rebleeding episodes occurred during the initial 72-hour infusion. The duration of hospital stay was < 5 days for 88 (37.0%) in the intensive regimen group and 111 patients (47.0%) in the standard group (P = 0.03), and there were fewer surgical interventions in the standard group (1) relative to the intensive regimen group (3). Five patients in each treatment group died.[73] The investigators concluded that standard-dose PPIs infusion was as effective as a high-dose regimen in reducing the risk of recurrent bleeding following endoscopic hemostasis of bleeding ulcers.
Parenteral PPI is indicated after successful endoscopic therapy for ulcers with high-risk signs, such as active bleeding, visible vessels, and adherent clots. Parenteral PPI use before endoscopy is common practice. A Canadian database (RUGBE) indicated some benefit in decreasing rebleed rates.[72] No randomized control trial provided evidence to support the use of parenteral PPI in this setting, but giving oral PPI both before and after EHT for those persons with peptic ulcers with signs of recent hemorrhage can be justified on the grounds of cost-effectiveness.[22]
Based on intragastric pH data, nonvomiting patients with bleeding ulcers may be treated with oral lansoprazole (120-mg bolus, followed by 30 mg every 3 h).[2]
When indicated, intravenous pantoprazole or omeprazole is administered as an 80-mg bolus followed by a continuous 8-mg/h infusion for 72 hours. This treatment is changed to oral PPI therapy after 72 hours if no rebleeding occurs.
Whether acid suppression improves therapeutic outcomes of peptic ulcers compared with placebo may be more important than the issues raised above. Many researchers have compared parenteral PPI therapy with placebo. Overall, the results demonstrate a shorter period bleeding and a decreased incidence of rebleeding with PPI therapy. Some studies have demonstrated a decreased need for emergency surgery and transfusion; however, evidence that parenteral PPI reduces mortality from ulcer bleeding is relatively recent.[19]
Concomitant H pylori infection in the setting of bleeding peptic ulcers should be eradicated, as this lowers the rate of rebleeding.[74, 75]
Medical management of stable duodenal ulcers
Treatment of duodenal ulcers depends on the cause. Because the 2 most common causes are H pylori infection and NSAID use, the focus of this section is specific strategies aimed at treating duodenal ulcers in these settings.
H pylori infection: In general, patients with a documented duodenal ulcer who have H pylori infection should receive eradication therapy.[76, 77, 78, 79, 80, 81] Several studies have evaluated different regimens for H pylori eradication. In 1998, The American College of Gastroenterology (ACG) published practice guidelines for the management of H pylori infection (the 2007 guidelines can be found here[82] ).[83] The ACG guidelines recommended the following treatments[83] :
Lansoprazole 30 mg PO bid or omeprazole 20 mg PO bid, plus amoxicillin 1000 mg PO bid and clarithromycin 500 mg PO bid for 14 days (Other PPIs may also be substituted.)
Lansoprazole 30 mg PO bid or omeprazole 20 mg PO bid, plus metronidazole 500 mg PO bid and clarithromycin 500 mg PO bid for 14 days
Ranitidine bismuth citrate 400 mg PO bid, plus clarithromycin 500 mg PO bid and amoxicillin 1000 mg PO bid or metronidazole 500 mg PO bid or tetracycline 500 mg PO bid for 14 days
Bismuth subsalicylate 525 mg PO qid, plus metronidazole 500 mg PO tid and tetracycline 500 mg PO qid and a PPI (eg, lansoprazole 30 mg PO [optimal dose] or omeprazole 20 mg PO [optimal dose]) for 14 days
Bismuth subsalicylate 525 mg PO qid, plus metronidazole 250 mg PO qid and tetracycline 500 mg PO qid and any H2RA for 14 days
These first-line treatments may fail in over 20% of H pylori –positive individuals, especially those with non-ulcer dyspepsia as compared with those with duodenal ulcers, and second- or third-line therapies may be necessary.[84]
Dyspepsia is more common in H pylori –positive than H pylori –negative persons, and treatment offers a small incremental symptom benefit.[85]
Size of the ulcer, the presence of associated atrophic gastritis, and successful eradication of associated H pylori are independent predictive factors for the healing of peptic ulcers.[86] A 7-day treatment is adequate in those patients whose condition has not failed previous attempts at eradication.[80, 87]
Spouses and H pylori –positive family members of H pylori –positive persons should be considered for testing and treatment of H pylori infection,[88] since mother-to-child transmission may be a major route of H pylori infection.[89]
It is controversial whether new GERD and erosive esophagitis occur after eradication of H pylori in duodenal ulcer patients, but if they do occur, the inflammation is usually mild and short-lived, and maintenance therapy is usually not required.[90]
In a study by Liou et al, H pylori eradication rates were higher for a 7-day antibiotic regimen containing clarithromycin, amoxicillin, and lansoprazole (CAL) when used as first-line therapy compared with levofloxacin, amoxicillin, and lansoprazole (LAL).[91] Additionally, CAL did not achieve a higher rate of eradication than LAL as second-line therapy; thus, consideration of the sequence of administering antibiotic regimens for H pylori is important.
Medical management of NSAID ulcers
Discontinuation of NSAIDs is paramount, if it is clinically feasible.
Treat H pylori infection if it is present. For patients who must continue with their NSAIDs, PPI maintenance is recommended to prevent recurrences even after eradication of H pylori.[37, 92]
If NSAIDs must be continued, changing to a COX-2 selective inhibitor is an option. However, use of a traditional NSAID and once-daily PPI is comparable to a selective COX-2 inhibitor with respect to ulcer bleeding in patients with a history of peptic ulcer disease.[39]
In general, 6-8 weeks of therapy with a PPI is required for complete healing of a duodenal ulcer.
Endoscopic intervention is the primary mode of treating bleeding ulcers. Surgical management of duodenal ulcers is generally reserved for refractory ulcers and bleeding ulcers that fail to respond to medical management.
Endoscopic therapy
Endoscopic therapeutic intervention is indicated for bleeding duodenal ulcers with high-risk signs (eg, active bleeding, visible vessels, adherent clots).
Several tools are available to the endoscopist to achieve hemostasis; these tools include bipolar cautery, use of a heater probe or hemoclips, argon plasma coagulation, and local injection of epinephrine and other agents.
Bipolar cautery and use of a heater probe both apply heat to the ulcer and cauterize the bleeding vessel. Hemoclip placement is a promising therapy but requires a skilled endoscopist and a lesion amenable to clip placement.
Injection with epinephrine achieves hemostasis through the vasoconstrictive effect of epinephrine. However, some physicians argue that it is effective mainly through the tamponade effect of local fluid injection. This is supported by the fact that injection of saline achieves comparable hemostasis.
Argon plasma coagulation uses heat to achieve hemostasis. One group compared argon plasma coagulation with heater probe and found no difference in the incidence of rebleeding or the need for surgical intervention.[93]
In ulcers with high-risk signs, endoscopic combination therapy with epinephrine injection and hemoclip placement is superior to injection alone.[94]
Urgent surgical management
The indications for urgent surgery include the following: (1) failure to achieve hemostasis endoscopically, (2) recurrent bleeding despite endoscopic attempts at achieving hemostasis (many advocate surgery after 2 failed endoscopic attempts), and (3) perforation.
In general, 5% of bleeding ulcers eventually require operative management. Most emergent surgical procedures involve simple oversewing of the ulcer to achieve hemostasis.
Elective surgical management
The indications for elective surgical management include the following: (1) duodenal ulcer refractoriness to medical treatment, (2) patient intolerance to medications, and (3) patient noncompliance with medications.
With the advent of improved antisecretory therapy and with the discovery of H pylori, elective surgical management of duodenal ulcers has become much less common in areas where such treatment is readily available.
Elective surgical approaches
Vagotomy
Vagotomy involves resection of the vagus nerve, which eliminates the autonomic stimulation of the parietal cells. Historically, a truncal vagotomy was performed; however, this led to gastric atony and subsequent stasis in as many as 20% of patients. Currently, selective vagotomies are the procedures of choice.
Selective vagotomy preserves the celiac and hepatic branches of the vagus nerve, thus decreasing the incidence of gastric atony. However, a gastric drainage procedure (eg, pyloroplasty) remains an essential component of this surgical approach. Highly selective vagotomy results in denervation of the parietal cells but preserves nerves supplying the pyloroantral region.
The Billroth I and Billroth II procedures are the 2 types of truncal vagotomy and antrectomy. These surgical approaches carry a mortality rate of approximately 1% and are currently performed much less frequently.
Surgical consultation is recommended for all patients with bleeding ulcers, especially those patients who are at high risk of significant bleeding. Such ulcers include those that are causing hemodynamic instability, those that are actively bleeding, and those that are showing a visible vessel on endoscopy.
A special diet is not indicated in patients with duodenal ulcers. It is common sense to avoid any food or beverages which on a person-by-person basis aggravate symptoms. Although the link between duodenal ulcers and alcohol use and smoking is inconclusive, moderation of alcohol intake and cessation of smoking may be recommended for other health reasons.
Clinical Context:
Suppresses gastric acid secretion by specifically inhibiting H+/K+ -ATPase system at the secretory surface of gastric parietal cells. Administer 30 min before sucralfate to prevent reduction in bioavailability.
Clinical Context:
Suppresses gastric acid secretion by specifically inhibiting H+/K+ -ATPase system at the secretory surface of gastric parietal cells. Indicated for < 8-wk treatment of erosive esophagitis associated with GERD. May consider additional 8-wk course if ulcer does not heal. Safety and efficacy for maintenance therapy (eg, >16 wk) not established.
Clinical Context:
Suppresses gastric acid secretion by specifically inhibiting H+/K+ -ATPase system at the secretory surface of gastric parietal cells. Used for < 4 wk to treat and relieve the symptoms of an active duodenal ulcer.
Clinical Context:
Combination of ranitidine and bismuth citrate, compound with bactericidal effects against H pylori. Used in with clarithromycin, because drugs act synergistically against H pylori.
Clinical Context:
Semisynthetic penicillin antibiotic; interferes with synthesis of cell wall mucopeptides during active multiplication, resulting in bactericidal activity against susceptible bacteria.
Clinical Context:
Semisynthetic antibacterial agent derived from Streptomyces cultures. Effective against gram-positive and gram-negative organisms, as well as mycoplasmal, chlamydial, and rickettsial infections. Inhibits bacterial protein synthesis by binding with 30S and, possibly, 50S ribosomal subunit(s).
Clinical Context:
Macrolide antibiotic that inhibits bacterial growth, possibly by blocking the dissociation of peptidyl tRNA from ribosomes, arresting RNA-dependent protein synthesis.
In treatment of bleeding duodenal ulcers with intravenous PPI therapy, the treatment is generally continued for 72 hours unless persistent bleeding requires longer duration of therapy. If this occurs, a second endoscopy is indicated, and continued bleeding would mandate surgical intervention.
If a patient is infected with H pylori, initiate eradication therapy early.
Patients with a healing duodenal ulcer generally require maintenance PPI therapy for as long as 6-8 weeks.
Patients with refractory ulcers may continue receiving once-daily PPI therapy indefinitely. In this setting, if H pylori is absent, consider a secondary cause of duodenal ulcer, such as Zollinger-Ellison syndrome.
The most important preventive measure regarding ulcers is complete avoidance of NSAIDs. If NSAIDs are required, use of a COX-2 selective inhibitor may help prevent gastric and duodenal mucosal ulceration. An alternate approach is the use of a traditional NSAID with a once-daily PPI.
In certain high-risk patients, such as those taking NSAIDs and steroids or receiving chemotherapy, prophylactic use of PPIs may help prevent ulcer formation.
Bardou M, Toubouti Y, Benhaberou-Brun D, Rahme E, Barkun AN. High dose proton pump inhibition decrease both re-bleeding and mortality in high-risk patients with acute peptic ulcer bleeding. A series of meta-analyses [abstract]. Gastroenterology. 2003;123(suppl 1):A625.
Bardou M, Youssef M, Toubouti Y, et al. Newer endoscopic therapies decrease both re-bleeding and mortality in high risk patients with acute peptic ulcer bleeding: a series of meta-analyses [abstract]. Gastroenterology. 2003;123:A239.
Friedman L, Peterson W. Peptic ulcer and related disorders. In: Fauci AS, Braunwald E, Isselbacher KJ, et al, eds. Harrison's Principles of Internal Medicine. 14th ed. New York, NY: McGraw-Hill; 1998:1596-616.
Duodenal ulcer in an elderly patient who presented with melena and hypotension.
Duodenal ulcer in a 35-year-old woman who presented with tarry stools and a hemoglobin level of 75 g/L.
Duodenal ulcer in a 65-year-old man with osteoarthritis who presented with hematemesis and melena stools. The patient took naproxen on a daily basis.
Deformity of duodenal cap caused by recurrent ulceration. Single-contrast view.
Double-contrast upper gastrointestinal series. Posterior wall duodenal ulcer (same patient in Images 5 and 6).
Lateral view of a posterior wall ulcer (same patient in Images 5 and 6).
An antral gland of the stomach with a large Giemsa-stained colony of Helicobacter pylori in the lumen (arrow) at 250X power. Courtesy of Pantaleo Bufo, University of Foggia, Italy.
A transverse section of the gastric lamina propria is shown in a patient with Helicobacter pylori infection. In the lower part, an antral gland of the stomach is present with some H pylori in the lumen (red-blue arrow). In the upper part, a mast cell (yellow arrow) is present (Giemsa staining, 250X). Courtesy of Pantaleo Bufo, University of Foggia, Italy.