Chapter 38 Gastroenterology

Epidemiology and Social Impact of Gastrointestinal Disease 843

Approach to the Patient with Gastrointestinal Complaints 843

The Abdominal Examination 844

Common Pediatric Gastrointestinal Disorders 844

Infantile Regurgitation 844

Diarrhea and Dehydration 844

Appendicitis 846

Common Adult Gastrointestinal Disorders 847

Stomach and Duodenum 848

Gallbladder 855

Cholelithiasis and Cholecystitis 855

Liver 856

Hepatitis 856

Cirrhosis 859

Pancreas 861

Acute Pancreatitis 861

Chronic Pancreatitis 862

Pancreatic Cancer 863

Lower Gastrointestinal Tract 863

Inflammatory Bowel Disease 863

Irritable Bowel Syndrome 866

Lower Gastrointestinal Bleeding 868

Diverticular Disease 870

Celiac Disease 872

Colorectal Cancer 873

He who does not mind his belly will hardly mind anything else.

Samuel Johnson (1763)

Thought depends absolutely on the stomach, but in spite of that, those who have the best stomachs are not the best thinkers.

Voltaire (1770)

Epidemiology and Social Impact of Gastrointestinal Disease

Although diseases of other organ systems (e.g., cardiovascular disease) may appear to be more dramatic illnesses with higher rates of morbidity and mortality, the overall impact of gastrointestinal (GI) disorders is often underestimated from both a biopsychosocial and a resource standpoint. Typically, diseases of the GI tract are misdiagnosed, mistreated, misunderstood, or missed altogether, ultimately leading to substantial psychological morbidity and tremendous direct and indirect expense. Digestive diseases cost an estimated $91 billion annually in U.S. health care costs, lost days from work, and premature deaths. More than 70 million Americans are diagnosed each year with disorders of the digestive tract, including gastroesophageal reflux disease, peptic ulcer disease, inflammatory bowel disease, GI cancers, motility disorders, hepatitis, cirrhosis, and food-borne illness (Foundation, 2009).

This chapter serves as an overview of common GI diseases and disorders encountered in family medicine practices, encompassing both adult and pediatric populations. The most recent evidence-based diagnostic and therapeutic guidelines and reviews are highlighted, and radiographs, endoscopy photos, and video segments are integrated where applicable.

Approach to the Patient with Gastrointestinal Complaints

Psychosocial Factors of Gastrointestinal Disease

Locke’s review in Sleisenger and Fordtran’s Gastrointestinal and Liver Disease (8th edition) provides an excellent framework of the biopsychosocial approach to the patient with GI complaints. In caring for patients with both acute and chronic GI disorders, family physicians should obtain a patient-centered, nondirected history, using open-ended questions that enable patients to tell the history in their words. Medical and social histories should be integrated so that symptomatic complaints are described in the context of the psychosocial events surrounding the presenting illness, including the setting of symptom onset and exacerbation. Throughout the encounter, the provider’s questions should communicate a sincere willingness to address both biologic and psychologic aspects of the illness. Evaluation and treatment of GI symptoms depend on a strong physician-patient therapeutic relationship, allowing the family physician to elicit, evaluate, and communicate the role of potential psychosocial factors in the disease state. Patient reassurance, acknowledgment of patient adaptations to chronic illness, reinforcement of healthy behaviors, and the consideration of psychopharmacologic medication are paramount (Locke, 2006).

The failure to find a specific structural etiology for a patient’s GI symptoms is usually the rule rather than the exception in the ambulatory care setting. Because functional GI disorders may represent an “illness without an evident disease,” some providers do not regard these as legitimate, especially within the biopsychosocial model of disease. This phenomenon often leads physicians (and patients to coerce physicians) to pursue unnecessary, costly, and invasive diagnostic tests to find the etiology of a patient’s symptoms, rather than focusing directly on symptom management and potential psychological comorbidities. Family physicians must establish clear boundaries with their patients to prevent unnecessary workups and, when indicated, should consider a referral to a mental health professional skilled in the care of patients with functional GI conditions, to assist in symptom management (Locke, 2006).

The Abdominal Examination

For descriptive purposes of location, the abdomen has been divided into four quadrants, constructed by an imaginary vertical line from the tip of the xyphoid process to the pubic symphysis and an imaginary horizontal line bridging the anterior superior iliac crests, referred to as the right upper (RUQ), right lower (RLQ), left upper (LUQ), and left lower (LLQ) quadrants (Bickley, 2008). The abdomen can be further divided into the epigastric, umbilical, hypogastric or suprapubic, and right and left flank regions. Knowledge of anatomic structures that lie in each of these quadrants and regions is imperative for the clinician to form an accurate differential diagnosis in relation to a patient’s presenting symptoms (Box 38-1).

Box 38-1 Differential Diagnosis for Abdominal Pain Based on Location

Modified from Differential diagnosis. In Ferri FF. Ferri’s Clinical Advisor 2010. Philadelphia, Saunders-Elsevier, 2010.

Other regions

Periumbilical

Intestinal: Small bowel obstruction or gangrene, early appendicitis

Vascular: Mesenteric thrombosis, dissecting, ruptured, or leaking aortic aneurysm

Clinicians should remember that a comprehensive physical examination, not just abdominal, should be performed in patients presenting with digestive complaints. The clinician should ensure patient comfort and modesty because the abdominal examination (and pelvic, if indicated) may cause significant pain, anxiety, and embarrassment to the patient. Infants, young children, and pregnant women may require additional care during the abdominal examination. A thorough explanation of examination techniques and findings is necessary to minimize patient anxiety.

The anorectal examination is an important component of the abdominal examination that may often be omitted. Although uncomfortable and embarrassing to the patient, it should not be neglected and should be approached with a calm, gentle attitude on the part of the clinician. The examiner should comment on both external and internal components of the anorectal examination, specifically anal sphincter tone; presence or absence of hemorrhoids, fissures, fistulas, or masses; prostatic abnormality in the male patient; and consistency of the stool. Anoscopy should be considered as an adjunct to the anorectal examination when indicated, allowing for direct visualization of the internal anorectal canal. In female patients, the pelvic examination may provide additional information in the diagnosis of abdominal symptoms, as it is often challenging to differentiate between GI and genitourinary complaints.

Common Pediatric Gastrointestinal Disorders

Infantile Regurgitation

In the infant with recurrent vomiting, a thorough history and physical examination are often sufficient to establish a diagnosis of uncomplicated gastroesophageal reflux disease (GERD), labeling the infant as “the happy spitter.” Diagnostic evaluation is indicated if there are signs of poor weight gain, GI obstruction, excessive crying and irritability, disturbed sleep, or feeding or respiratory problems suggesting suspected asthma or recurrent pneumonia. In the infant with uncomplicated GERD, parental education, reassurance, and anticipatory guidance are recommended; no specific intervention is necessary because the process is usually self-limited. Thickened formula and a trial of a hypoallergenic formula are the best treatment options. A trial of time-limited acid-suppression therapy, usually with histamine-2 receptor antagonists (H₂RAs), is useful in determining if GERD is causing vomiting and regurgitation. If symptoms worsen or do not improve by 18 to 24 months of age, reevaluation for complications of GERD is recommended, including an upper GI series (barium swallow study) and consultation with a pediatric gastroenterologist. In otherwise normal children who have recurrent vomiting or regurgitation after age 2 years, management options include an upper GI series, upper endoscopy with biopsy, and antisecretory therapy (Rudolph, 2001).

Diarrhea and Dehydration

Diarrhea is exceedingly common in pediatric and adult populations worldwide. Dehydration and electrolyte (e.g., sodium, potassium, bicarbonate) losses associated with severe diarrhea account for significant morbidity and may lead to mortality in cases of acute gastroenteritis, especially in countries with poor access to adequate healthcare. Rotavirus infection is the most common cause of diarrhea in U.S. infants and children, especially in winter months and temperate climates; Norwalk-like virus is the most common agent in adults (King et al., 2003).

A thorough history should be the first step in evaluating the patient who presents with a significant diarrheal illness, including the following (Guerrant et al., 2001):

When and how the illness began (abrupt or gradual onset and duration of symptoms)

Stool characteristics (watery, bloody, mucous, purulent, greasy)

Frequency of bowel movements and relative quantity of stool produced

Presence of dysenteric symptoms (fever, tenesmus, blood/pus in stool)

Symptoms of volume depletion (increased thirst, tachycardia, orthostasis, decreased urine output, lethargy, decreased skin turgor, decreased tear production)

Associated symptoms and their frequency and intensity (nausea, vomiting, abdominal pain, cramps, headache, myalgias, altered sensorium)

In addition, all patients should be asked about potential epidemiologic risk factors for diarrheal diseases, including the following:

Travel to an underdeveloped area

Daycare center attendance or employment

Consumption of unsafe foods (e.g., raw meats, eggs, or shellfish; unpasteurized milk or juices) or swimming in or drinking untreated fresh surface water from a lake or stream

Visiting a farm or petting zoo or having contact with reptiles or with pets with diarrhea

Knowledge of other ill persons (e.g., in a dormitory, office, or social function)

Recent or regular medications (e.g., antibiotics, antacids, antimotility agents)

Underlying medical conditions predisposing to infectious diarrhea (AIDS, immunosuppressive medications, prior gastrectomy, extremes of age)

Receptive anal intercourse or oral-anal sexual contact

Occupation as a food-handler or caregiver

In the majority of patients with acute gastroenteritis, the “gold standard” stool cultures and ova and parasite testing are seldom required, because the disease is most often viral in etiology and self-limited. In more severe cases with dehydration, metabolic derangement, longer duration, bloody stools (dysentery) and mucus in the stool, or known or suspected transmission of a pathogen, these tests are often required to identify the pathogen and to direct appropriate antimicrobial therapy (Table 38-1). Viral cultures are rarely performed and are unnecessary, except in rare cases and immunocompromised patients. Although fecal leukocytes and lactoferrin often suggest an inflammatory etiology of diarrhea, there is no consensus regarding the routine use of these tests in the initial testing of patients with either community-acquired or nosocomial diarrhea. Hospitalized patients with diarrhea, especially those with abdominal pain, should be tested for Clostridium difficile toxin (Guerrant et al., 2001). Oral metronidazole and oral vancomycin have been shown to be equally effective in the treatment of C. difficile–associated diarrhea (CDAD). Vancomycin is significantly more expensive and may select for colonization with vancomycin-resistant enterococci (VRE), and thus metronidazole should be recommended as first-line therapy.

Table 38-1 Common Pathogens and Recommended Therapy for Acute Diarrhea

Pathogen	Therapy (Adult Doses)
Campylobacter jejuni	Azithromycin, 500 mg qd for 3 days or Ciprofloxacin,∗ 500 mg bid for 7 days
Clostridium difficile	Metronidazole, 500 mg tid, or 250 mg qid, for 10 14 days or Vancomycin, 125 mg qid for 10-14 days
Entamoeba histolytica	Metronidazole, 500-750 mg tid for 10 days or Tinidazole, 2 g qd for 3 days followed by Paromomycin, 500 mg orally tid for 7 days or Iodoquinol, 650 mg tid for 20 days
Escherichia coli 0157:H7	No treatment with antimicrobials or antimotility drugs
E. coli (toxigenic)	Azithromycin, 1 g in 1 dose or Rifaximin, 200 mg tid for 3 days or Levofloxacin,^∗ 500 mg for 1 dose
Giardia lamblia	Tinidazole, 2 g in 1 dose or Nitazoxanide, 500 mg bid for 3 days or Metronidazole, 500-750 mg tid for 5 days
Salmonella spp. (non-typhi)^†	Ciprofloxacin, 500 mg bid for 5-7 days or Azithromycin, 1 g for 1 day, then 500 mg qd for 6 days
Shigella spp.	Ciprofloxacin, 500 mg bid for 5-7 days or Levofloxacin, 500 mg qd for 3 days or TMP-SMX-DS, 1 tablet bid for 3 days or Azithromycin, 500 mg for 1 dose, then 250 mg qd for 4 days
Staphylococcus aureus (food poisoning)	No treatment with antimicrobials or antimotility drugs
Vibrio cholerae	Ciprofloxacin, 1 g in 1 dose, plus aggressive fluid hydration
Vibrio parahaemolyticus	No treatment with antimicrobials or antimotility drugs
Yersinia enterocolitica	No treatment unless severe; if severe: Doxycycline, 100 mg IV bid, and Tobramycin or gentamicin, 5 mg/kg/day qd, or TMP-SMX and fluoroquinolones as alternatives

TMP-SMX, Trimethoprim-sulfamethoxazole; DS, double strength; IV, intravenously; qd, once daily; bid, twice daily; tid, three times daily; qid, four times daily.

∗ Avoid fluoroquinolones in pediatric patients and pregnant women.

† Antimicrobial therapy not indicated in asymptomatic patients or those with mild illness. Treatment advised in patients less than 1 year old or greater than 50 years old, if immunocompromised, or if patient has a vascular graft or prosthetic joints.

Modified from Gilbert DN, Moellering RC, Eliopoulos GM, et al. The Sanford Guide to Antimicrobial Therapy, 39th ed. Vermont, Antimicrobial Therapy, 2009.

Nonpathogenic causes of diarrhea should be considered when a viral etiology is unlikely and a diagnostic evaluation has not identified a pathogen. Differential diagnosis includes irritable bowel syndrome, inflammatory or ischemic bowel disease, laxative abuse, partial bowel obstruction, rectosigmoid abscess, Whipple’s disease, pernicious anemia, diabetes mellitus, malabsorption syndromes (e.g. celiac disease), small bowel diverticulosis, and scleroderma in primarily adult patients, and an appropriate workup should be considered.

KEY TREATMENT

Oral metronidazole and vancomycin are equally effective in the treatment of Clostridium difficile–associated diarrhea (Guerrant et al., 2001) (SOR: A).

In cases of traveler’s diarrhea, (e.g. enterotoxigenic Escherichia coli, Shigella, Salmonella, or Campylobacter), prompt treatment with a fluoroquinolone or, in children, trimethoprim-sulfamethoxazole (TMP-SMX) has been shown to reduce the duration of the illness from 3 to 5 days to less than 1 to 2 days (Guerrant et al., 2001) (SOR: A).

Appendicitis

A diagnosis of appendicitis should be considered in any pediatric patient presenting with acute abdominal pain. Acute appendicitis often presents with a constellation of signs and symptoms, including fever, anorexia, nausea, vomiting, tenesmus, migratory RLQ abdominal pain, abdominal tenderness and guarding, and signs of peritoneal irritation. Classically, hours after onset, the pain migrates to McBurney’s point, defined as the point two-thirds the distance from the umbilicus along a straight line toward the anterosuperior iliac spine of the pelvis. Rovsing’s sign (referred tenderness from LLQ to RLQ during palpation), psoas sign (pain elicited by extending hip posteriorly with patient lying prone), and obturator sign (pain elicited by abducting right hip with patient lying supine) are often conducted but are of little diagnostic value. No sign or combination of signs has accurately predicted acute appendicitis in children (Cincinnati Children’s Hospital [CCH], 2002).

While none has proved adequately predictive of acute appendicitis in the pediatric population, laboratory studies are typically performed in the emergency department because other diagnoses may need to be excluded. A series of studies discovered an elevated white blood cell (WBC) count in 87% to 92% of patients with acute appendicitis, although 8% to 13% of patients with appendicitis had a normal WBC count (CCH, 2002). The abdominal examination is particularly unreliable in women of reproductive age, so pelvic exam, urinalysis, and urine pregnancy test represent a reasonable clinical strategy to exclude genitourinary pathology.

Diagnostic imaging is not routinely recommended in patients with a high or a low probability of appendicitis, because it can alter management strategies and has not proved cost-effective; imaging is most helpful when the clinical assessment is equivocal. Controversy exists over the superiority of ultrasound versus computed tomography (CT) in accurately diagnosing appendicitis, and both tests have a positive predictive value approaching 100%. Although ultrasound may be advantageous in thin patients, CT is preferred in the evaluation of a more obese child (Halter et al., 2004). Typical radiographic findings in the evaluation of acute appendicitis include appendicoliths (Fig. 38-1), dilation of the appendix with adjacent hazy fat (Fig. 38-2), and periappendiceal abscesses (Fig. 38-3). If the abdominal CT does not show evidence of acute appendicitis, the patient may either be admitted for observation or discharged at the discretion of the examiner and parents, with instructions for follow-up if symptoms worsen. Expert opinion states that if there is high suspicion of appendicitis on the basis of history, physical examination, and laboratory studies, the patient should go directly to the operating room for an exploratory laparotomy to evaluate the appendix, without an imaging study.

Figure 38-1 Appendicolith.

(Courtesy Dr. Perry Pernicano, Clinical Assistant Professor, Department of Radiology, University of Michigan Medical School, Ann Arbor.)

Figure 38-2 Appendicitis. Dilated thickened appendix (A), with adjacent hazy fat (B).

(Courtesy Dr. Perry Pernicano.)

Figure 38-3 Periappendiceal abscess (arrow).

(Courtesy Dr. Perry Pernicano.)

Common Adult Gastrointestinal Disorders

Esophagus

Barrett’s Esophagus and Esophageal Adenocarcinoma

Barrett’s esophagus is a premalignant condition related to chronic GERD. The hallmark is a change in the mucosal lining of the distal esophagus from the normal squamous epithelium to columnar-appearing mucosa resembling that of the stomach and small intestines, referred to as intestinal metaplasia (Fig. 38-4). The estimated risk of progression to adenocarcinoma of the esophagus with Barrett’s esophagus is approximately 0.5% per year, whereas without Barrett’s esophagus the risk is 0.07%, prompting development of clinical practice guidelines for surveillance endoscopy. Adenocarcinoma of the esophagus has had the fastest-rising incidence of any cancer in the United States and Western Europe over the last two decades. Family and other primary care physicians who see the vast majority of patients with GERD in its nonerosive and more complicated forms are charged with the task of suspecting and appropriately referring patients with Barrett’s esophagus for EGD. Although risk factors for Barrett’s esophagus and adenocarcinoma are not evidence-based, there is suggestive evidence that male gender, white race, older age, dysplasia, smoking, and obesity place patients at a higher risk (Sampliner, 2002).

Figure 38-4 Barrett’s esophagus.

(Courtesy Dr. Erik-Jan Wamsteker.)

Currently, no evidence-based guidelines exist for the assessment and surveillance of patients with Barrett’s esophagus; routine screening will not be cost-effective unless criteria can be identified to select patients at high risk. Recommendations from the American Society for Gastrointestinal Endoscopy (ASGE) state that screening esophagogastroduodenoscopy (EGD) for Barrett’s esophagus should be considered in select patients with chronic, long-standing GERD. After a negative screening examination, further screening endoscopy is not indicated. For patients with established Barrett’s esophagus of any length and with no dysplasia, after two consecutive examinations within 1 year, an acceptable interval for additional surveillance is every 3 years. Surveillance in patients with low-grade dysplasia is recommended, although the optimal interval and biopsy protocol have not been established. A follow-up EGD at 6 months should be performed, and if low-grade dysplasia is confirmed, surveillance at 12 months and yearly thereafter as long as dysplasia persists is advised (Hirota et al., 2006).

Stomach and Duodenum

Dyspepsia

Dyspepsia (“bad digestion”) accounts for approximately 5% of all visits to family practitioners and is the most common reason for referral to a gastroenterologist in the United States, accounting for 20% to 40% of consultations (Jones and Lacy, 2004). The term dyspepsia refers to episodic or recurrent pain or discomfort arising from the proximal GI tract related to meals and is associated with heartburn, reflux, regurgitation, indigestion, bloating, early satiety, and weight loss. The lack of a standardized definition affects accurate prevalence data, given the challenge of clearly defining dyspepsia as either functional or nonulcer dyspepsia (~60% of cases), or that caused by structural or biochemical disease (40%) (Dickerson and King, 2004). Regardless of cause, dyspepsia has a profoundly negative impact on patients’ health-related quality of life (HRQOL) and results in significant economic burden.

Nonulcer dyspepsia (NUD) is defined in patients who have undergone either formal radiographic or endoscopic evaluation and who do not have an organic lesion (e.g., ulcer, tumor) to explain their symptoms. Potential etiologies for NUD include gastric acid hypersecretion, gastroduodenal dysmotility, visceral hypersensitivity, emotional stress, and psychological factors. As with other functional GI disorders, the potential for underlying psychosocial and lifestyle factors must be addressed. There is no current recommendation on the role of prokinetics, cytoprotectives, or antidepressants in the management of functional dyspepsia; similarly, no clear evidence supports specific diet and lifestyle modifications or psychosocial interventions.

Peptic Ulcer Disease

Key Points

• A noninvasive H. pylori “test and treat” strategy is as effective as endoscopy in the initial management of patients under age 45 with uncomplicated dyspepsia.

• Eradication of H. pylori infection in patients with a duodenal or gastric ulcer reduces symptom recurrence.

• Ulcer prophylaxis with an H₂RA or PPI should be considered in patients at high risk for NSAID-associated PUD, including those with a history of PUD, elderly patients, and patients taking corticosteroids or anticoagulants.

Peptic ulcer disease (PUD) is the most common cause of upper gastrointestinal bleeding (UGIB) and a leading cause of dyspepsia (Box 38-2), with a cumulative lifetime prevalence of 8% to 14% (Fig. 38-5). Although up to 70% of patients with gastric and duodenal ulcers are 25 to 64 years old, the peak prevalence of complicated ulcer disease requiring hospitalization is age 65 to 74 (Saad and Scheiman, 2004).

Figure 38-5 Gastric ulcer.

(Courtesy Dr. Erik-Jan Wamsteker.)

Numerous options exist for the management of uninvestigated dyspepsia in primary care (Fig. 38-6). Given the high cost and often limited access to endoscopy testing, not all patients with uninvestigated dyspepsia should undergo an invasive investigation of their symptoms. The presence of “alarm symptoms” for PUD raises concern for a gastric malignancy. The American Gastroenterological Association (AGA, 2005) endorses prompt endoscopic evaluation in any patient over age 45 with new-onset dyspepsia.

Figure 38-6 Evaluation of uninvestigated dyspepsia.

(Modified from Saad R, Scheiman JM. Diagnosis and management of peptic ulcer disease. Clin Fam Pract 2004;6:569-587.)

The most common complications of PUD include UGIB, perforation, penetration, and gastric outlet obstruction. An upper GI hemorrhage can occur in up to 15% of patients with PUD and is most common in patients older than 60, with mortality as high as 10%. Perforation occurs in approximately 7% of patients with PUD, again classically in elderly patients receiving long-term nonsteroidal anti-inflammatory drugs (NSAIDs). Perforation can be confirmed on plain abdominal radiography; barium contrast studies and upper endoscopy are contraindicated with suspected perforation, and urgent surgical consultation is mandatory. Mortality may be as high as 30% to 50% in patients with perforation, particularly in elderly and debilitated patients. Penetration occurs when the ulcer crater erodes through and into adjacent organs, including the small bowel, pancreas, liver, and biliary tree. Often subtle, it typically presents as acute pancreatitis. Gastric outlet obstruction occurs in 1% to 3% of PUD patients, resulting from acute inflammation or mechanical obstruction caused by scarring at the gastroduodenal junction (Saad and Scheiman, 2004).

Approaches to confirm a diagnosis of PUD include double-contrast barium esophagography (upper GI series) and upper endoscopy. Although the upper GI series has accuracy of 80% to 90%, upper endoscopy is the preferred diagnostic modality, with multiple studies showing consistent diagnostic superiority in identifying gastric and duodenal ulcers. Despite a higher procedural cost and a slightly increased risk in procedure-related complications (bleeding, perforation, oversedation), upper endoscopy should be the initial diagnostic study performed in suspected PUD. Most importantly, upper endoscopy provides the distinct advantage of permitting biopsies and brushings to identify underlying pathology.

Fecal-oral infection with the bacterium Helicobacter pylori is a major risk factor for the development of PUD. Its prevalence and association with PUD is higher in populations who have a lower standard of living than in the United States, especially in Africa and Central America, although this may be declining. Approximately 90% of patients worldwide with duodenal ulcers are infected with the H. pylori pathogen, whereas 30% to 40% of U.S. ulcer patients are infected (Chey and Wong, 2007). The strongest evidence to support the role of H. pylori as an etiology of PUD is the elimination of ulcer recurrence when the infection has been successfully eradicated.

The evidence-based standard of care dictates that patients with dyspepsia and no alarm symptoms should be tested for H. pylori infection and then given eradication therapy if positive, (“test and treat”) (Chey and Wong, 2007). H. pylori–positive or H. pylori–negative patients who do not undergo endoscopy initially should do so if their symptoms persist. The effectiveness of this approach depends on the prevalence of H. pylori infection in patients with ulcers in the community; as in some geographic areas, prevalence may be too low to make this approach effective.

Nonendoscopic testing for H. pylori includes a quantitative assay for serum immunoglobulin G (IgG) antibodies, the radiolabeled urea breath test, and the stool antigen test. The median sensitivity and specificity for serologic IgG tests are 92% and 83%, respectively (SIGN, 2003). Some patients may have persistently positive IgG antibodies for months to years after eradication therapy, yielding a false-positive result during that period if retested. In comparative studies, urea breath tests are more accurate than serologic tests. The stool antigen test is recommended as the preferred initial noninvasive diagnostic test. The urea breath test is the recommended standard to determine if H. pylori has been successfully eradicated (Chey and Wong, 2007).

Several medication regimens have been developed for the treatment of confirmed H. pylori infection, based on data from randomized controlled trials (RCTs), meta-analyses, and systematic reviews (Table 38-2). The best evidence-based recommendation for H. pylori eradication is for 14-day triple therapy with the use of a proton pump inhibitor (PPI), clarithromycin, and either amoxicillin or metronidazole, yielding eradication rates from 70% to 85%. The most common salvage regimen in patients with persistent H. pylori is bismuth quadruple therapy. Recent data suggest that 10-day therapy with a PPI, levofloxacin, and amoxicillin is more effective and better tolerated than bismuth quadruple therapy for persistent H. pylori infection (Chey and Wong, 2007).

Table 38-2 First-Line Regimens for Helicobacter pylori Eradication

Another major etiology of PUD is the increasing and widespread use of both NSAIDs and aspirin. The use and overuse of these medications is the most common cause of PUD in H. pylori–negative patients, and up to 60% of unexplained cases of PUD are attributed to unrecognized NSAID use. In a meta-analysis of observational studies of GI bleeding risk from various NSAIDs, a fourfold increased risk associated with NSAID-use persisted throughout therapy and fell to baseline within 2 months of discontinuation of the drug (Hernandez-Diaz and Rodriguez, 2000). Although both H. pylori and aspirin/NSAIDs are involved in the overwhelming majority of PUD cases, other factors contribute to the remaining 1% to 5% (Box 38-3).

Independent risk factors that augment the impact of H. pylori and NSAID-related PUD risk and may promote ulcer complications include advancing age; history of PUD or complicated ulcer disease with perforation, penetration, or gastric outlet obstruction; use of multiple NSAIDs (including concomitant use of low-dose aspirin and an NSAID); and concurrent warfarin or corticosteroid use. Evidence suggests that smoking may increase the risk of PUD and ulcer complications by impairing gastric mucosal healing. Alcohol use may increase the risk of ulcer complications in NSAID users, but its overall effect in those patients without concomitant liver disease has not been clearly defined. Currently, no solid evidence implicates dietary factors in the development of PUD.

Use of NSAIDs and aspirin is frequently associated with symptoms of dyspepsia, even in the absence of PUD. Empiric antisecretory therapy with PPIs is an attractive strategy that involves subjecting only those patients to upper endoscopy who fail to respond to a 4-week course of pharmacotherapy. The rationale for this approach is that most patients with dyspepsia do not have H. pylori infection or PUD (most typically have GERD), and their response to antisecretory therapy eliminates the need for further, expensive and invasive diagnostic testing (Saad and Scheiman, 2004).

KEY TREATMENT

The most efficacious therapy for H. pylori eradication consists of 14-day triple therapy with a PPI, clarithromycin, and amoxicillin or metronidazole, yielding eradication rates of 70% to 85% (Chey and Wong, 2007) (SOR: A).

Gastroesophageal Reflux Disease

Key Points

• The PPIs provide the most rapid symptomatic relief and healing of esophagitis in the highest percentage of patients.

• Although less effective than PPIs, H₂RAs given in divided doses may be effective in some patients with less severe symptoms of GERD; continuous therapy to control symptoms and prevent complications is appropriate.

• Chronic acid suppression has been associated with malabsorption of iron, vitamin B₁₂, and calcium; small bowel bacterial overgrowth; increased risk of hip fracture; and community-acquired pneumonia.

A complex, chronic, and relapsing condition, GERD carries a risk of significant morbidity and resultant complications. Population-based studies revealed that 40% of U.S. adults experience heartburn at least once a month; age- and gender-adjusted prevalence of weekly heartburn or acid regurgitation approaches 20% (Heartburn, 1998; Locke 1997). Most patients with GERD self-treat with over-the-counter (OTC) medications and do not seek medical attention for their symptoms.

Most patients with GERD evaluated in primary care practices have nonerosive reflux disease (NERD), although some will progress to erosive esophagitis (Fig. 38-7), and even fewer to more severe disease resulting in esophageal strictures, Barrett’s esophagus, and adenocarcinoma of the esophagus (DeVault and Castell, 2005). Patients with NERD are prone to develop atypical or extraesophageal manifestations of disease (Box 38-4). Because of a small risk of disease progression, however, they generally do not require long-term surveillance despite persistent reflux symptoms. Symptom relapse rates in patients with NERD are similar to those in patients with erosive esophagitis, and although many will require continuous pharmacotherapy to control their symptoms, almost all continue to exhibit no definable erosive esophagitis on upper endoscopy (Fass, 2002).