Shaheen N, Ransohoff DF. Gastroesophageal Reflux, Barrett Esophagus, and Esophageal CancerClinical Applications. JAMA. 2002;287(15):1982-1986. doi:10.1001/jama.287.15.1982
Author Affiliations: Division of Digestive Diseases and Nutrition, the Center for Esophageal Diseases and Swallowing, and the Center For Gastrointestinal Biology and Disease, University of North Carolina, Chapel Hill.
Scientific Review and Clinical Applications Section
Editor: Wendy Levinson, MD, Contributing Editor.
Gastroesophageal reflux disease (GERD), a condition commonly encountered
in the primary care setting, is a risk factor for adenocarcinoma of the esophagus.
Despite the ubiquity of the complaint, considerable uncertainty exists with
respect to several basic questions, including when to perform endoscopy in
patients with chronic reflux symptoms and how to address the cancer risk associated
with GERD. These clinical vignettes illustrate common clinical questions encountered
in caring for patients with GERD, especially as they relate to the issue of
cancer risk. Applying data reviewed in the companion article, we propose practical
answers to common clinical situations regarding care of patients with reflux.
We also present an algorithm for treatment of patients with chronic GERD symptoms.
Gastroesophageal reflux disease (GERD) is one of the most common problems
confronted by primary care physicians.1 The
most common symptom of GERD is heartburn, but GERD may present with a variety
of symptoms. Appropriate recognition and management of GERD relieves symptoms,
increases quality of life, and forestalls complications of reflux disease,
such as stricture formation.2- 7
A major concern of patients and physicians is the increased risk of esophageal
adenocarcinoma associated with reflux symptoms.8- 10
The following cases illustrate common questions faced in caring for
patients with GERD. The suggested treatment of the patients demonstrates one—but
often not the only—approach to the individual's care. We have attempted
to identify alternative diagnostic or therapeutic options that could be used
A 36-year-old Asian woman complains of 8 months of intermittent rising
substernal chest burning, occurring approximately 3 times a week, worse at
night and after large meals. Throughout the last 3 months, she has been taking
nonprescription ranitidine and initially had good symptom relief, but now
she is experiencing substantial periods of discomfort. She has been reading
about reflux on the Internet and is especially concerned regarding the cancer
risk with reflux.
This individual displays many of the classic features of uncomplicated
gastroesophageal reflux. The description of her discomfort, the associated
exacerbating factors, and the partial response of symptoms to H2
receptor antagonists (H2RAs) suggest that reflux disease is the
most likely diagnosis.
Given the prevalence of reflux in the population1,11,12
and the typical nature of this patient's symptoms, further testing to confirm
this diagnosis is not recommended.13 Instead,
initiation of appropriate therapy can be used as a diagnostic test and a therapeutic
measure.14,15 If this patient
responds to an empirical trial of proton pump inhibitors with resolution of
her symptoms, the diagnosis is confirmed and no further testing is necessary.
In typical reflux presentations such as this, further diagnostic testing is
reserved for patients demonstrating so-called alarm symptoms, such as dysphagia,
bleeding, anemia, and weight loss,16 as well
as those who do not respond as expected to empirical therapy.
Barrett esophagus is an endoscopically detectable metaplastic change
of the lining of the esophagus so that some portion is lined with specialized
columnar epithelium instead of the normal squamous epithelium (Figure 1). With respect to this patient's cancer risk, several demographic
and symptomatic features argue against further invasive measures to rule out
esophageal adenocarcinoma or Barrett esophagus. Although symptom severity
is not a reliable predictor of the presence of Barrett esophagus, the chronicity
of symptoms is.17 Individuals who are symptomatic
for more than 5 years are at increased risk compared with the general public
and those with symptoms of shorter duration. Additionally, epidemiologic studies
show that Barrett esophagus and adenocarcinoma of the esophagus are diseases
most prevalent in white men.18 With respect
to esophageal adenocarcinoma, the incidence in men is 4 times that in women
and is approximately 8 times as likely in whites as in other races. The incidence
of esophageal adenocarcinoma increases markedly with age, making a 36-year-old
unlikely to be affected. Finally, and most important, the absolute risk of
adenocarcinoma in patients with uncomplicated reflux symptoms is low, and
screening endoscopy has not been demonstrated to further decrease this risk.
For these reasons, the yield of endoscopic screening in this patient is low
and may be outweighed by the small risk of complication from upper endoscopy.19- 22
Appropriate treatment for this patient consists of counseling about
the pathophysiology of reflux and a review of conservative measures sometimes
helpful in avoiding reflux symptoms. These conservative measures, listed in
BOX 1, may decrease distal esophageal acid exposures.23- 25
Additionally, adding a pharmacological intervention would be reasonable. Because
the patient has had an incomplete symptomatic response to H2RAs,
therapy with a proton pump inhibitor could be initiated at standard daily
dosing. Alternatively, in this patient, increasing the dose of H2RA
as an initial step is also acceptable. A discussion of the risk factors and
demographics of esophageal adenocarcinoma and the risks and benefits of endoscopic
screening should help the patient understand why this measure is not recommended
in her situation. Finally, close follow-up to ensure response of symptoms
to therapy and to allow the tapering of pharmacological therapy to the lowest
dose at which the patient is symptom free should be arranged. Figure 2 demonstrates a suggested algorithm for the use of endoscopy
and pharmacological therapy for patients with classic reflux symptoms. This
figure demonstrates a step-down approach to pharmacological therapy, starting
with a proton pump inhibitor and decreasing acid suppression to the lowest
dosage of either proton pump inhibitor or, preferably, H2RA that
keeps the subject symptom free. Step-up approaches, starting with H2RAs and intensifying therapy as necessary, are also acceptable in patients
who have not tried H2RA before evaluation.
Elevate the head of the bed on 15-cm blocks.
Avoid eating within 4 hours before sleep.
Avoid eating large meals and chocolate.
Avoid consuming caffeinated products.
Avoid fatty foods.
Lose weight if overweight.
A 65-year-old white man visits your office for follow-up for chronic
reflux symptoms. He had an endoscopy 3 years ago, at which time 5 cm of Barrett
mucosa without dysplasia was discovered. He otherwise is well, with no weight
loss or dysphagia. He has been receiving therapy with a proton pump inhibitor
daily for the last 6 years and is worried about the long-term effects of these
medications on his system.
When omeprazole, the first proton pump inhibitor, was introduced into
the US market, there was concern that the potency of the drug and the increase
in serum gastrin levels associated with its use might lead to the development
of gastrinomas. Now, despite the ubiquitous nature of these agents and more
than 15 years of experience with them, no increased cancer risk associated
with proton pump inhibitor use has been demonstrated. The drugs appear to
be safe, even when taken long-term and at doses higher than those initially
approved for the healing of erosive esophagitis.26
Routine monitoring of serum gastrin levels is not recommended and may, in
fact, cause the physician and patient some distress if they are checked, because
they are often elevated in those undergoing therapy.27- 29
Because of the dearth of data supporting surveillance endoscopy in Barrett
esophagus, any consideration of enrolling a patient in an endoscopic surveillance
program should be preceded by a frank discussion of the risks and benefits
of surveillance. At the end of this discussion, the patient should understand
the rationale behind endoscopic surveillance (to detect cancer in a preclinical
and potentially more curable stage), the absolute risk of esophageal adenocarcinoma
(approximately 1 in 200-300 patient-years),30- 32
the risk of serial upper endoscopy (1 major complication in approximately
1000 procedures),19 the treatment options if
dysplasia occurs within the Barrett esophagus (esophagectomy, closer surveillance
to detect early cancer,33 and enrollment in
an experimental ablative protocol),34 and the
lack of endoscopic-surveillance evidence showing a survival benefit in those
with Barrett esophagus. If after this discussion the patient opts for enrollment
in an endoscopic surveillance program, referral to a gastroenterologist or
surgeon for upper endoscopy with biopsies is warranted.34
A 36-year-old man presents for follow-up of 2 years of reflux disease.
His condition, made manifest by substernal chest burning, is well controlled
by twice-daily proton pump inhibitors, for which he pays out of pocket. He
wonders how long he will have to take these medications and whether there
are other treatment options for him. He has heard that cancer can be associated
with reflux, and his neighbor recently underwent a surgical antireflux procedure.
He wonders if this procedure would protect him from cancer.
Longitudinal studies of reflux disease demonstrate that, for most patients,
the condition is chronic.26,35,36
Furthermore, in most of those who develop erosive esophagitis, maintenance
therapy will be necessary after healing to avoid recurrence of esophagitis.26,37 However, many patients requiring
proton pump inhibitors to heal mucosal disease can continue taking H2RAs. Therefore, in individuals with erosive disease, an 8-week course
of proton pump inhibitors to heal mucosal lesions may be followed by an attempt
to continue giving the patient H2RAs. In reflux patients who again
become symptomatic while receiving H2RAs, long-term maintenance
therapy with proton pump inhibitors may be initiated. In addition to good
relief of symptoms and high healing rates of erosive esophagitis, proton pump
inhibitor treatment effectively delays the development of peptic strictures
in patients with a history of strictures.7,38
However, medical therapy has not been demonstrated to avert the development
of Barrett esophagus or decrease the risk of cancer in individuals with long-term
Although a surgical antireflux procedure is a safe and effective treatment
for GERD in appropriately selected patients, it should not be pursued to decrease
the risk of cancer in patients with reflux symptoms. Although the risk of
mortality from a laparoscopic antireflux procedure is low (approximately 0.2%),39- 41 it is still likely
higher than the lifetime risk of death from adenocarcinoma in patients with
chronic reflux disease because of the rareness of this cancer in the GERD
population. Also, although it is intuitive to expect that decreasing exposure
of the esophagus to gastric acid might halt the development of neoplasia,
no data suggest that surgical antireflux procedures decrease the already low
risk of esophageal adenocarcinoma among patients with GERD. The limited data
that do exist suggest that surgical antireflux procedures do not decrease
the cancer risk in subjects with GERD.8,42,43
Whether surgical antireflux procedures decrease the incidence of cancer in
the subgroup with Barrett esophagus is a debated and unsettled issue. However,
for patients with typical reflux symptoms, cancer prevention should not be
the impetus for a surgical antireflux procedure.
Despite the commonness of GERD, it is often undiagnosed or undertreated
in practice. In patients with classic substernal chest burning, further diagnostic
testing is usually unnecessary, and empiric therapy with antisecretory medications
may be instituted. Testing is appropriate for individuals who display alarm
symptoms and those who do not respond as expected to therapy. The cancer risk
in patients with GERD is low, and screening endoscopy has not been demonstrated
to be effective in decreasing cancer incidence or increasing life expectancy.
Potent anti-acid medications, such as proton pump inhibitors, make total relief
of symptoms attainable in most patients. Long-term therapy with proton pump
inhibitors may be necessary and, given the available evidence, appears safe
and well tolerated. Surgical antireflux procedures are effective in appropriately
chosen patients but should not be pursued solely in an attempt to decrease
the already low risk of esophageal cancer in those with reflux.
Several other resources for physicians and patients to learn more about
the evaluation and management of gastroesophageal reflux disease are available
on the Internet (BOX 2).
American College of Gastroenterology's Web site, available at:http://www.acg.gi.org/acg-dev/patientinfo/frame_gerd.html
National Institutes of Health's information page, available at:http://www.niddk.nih.gov/health/digest/pubs/heartbrn/heartbrn.htm
Mayo Clinic's heartburn information site, available at:http://www.mayoclinic.com/findinformation/diseasesandconditions/invoke.cfm?id=DS00095
National Library of Medicine's online tutorial, available at:http://www.nlm.nih.gov/medlineplus/tutorials/gerd/id159101.html
American College of Gastroenterology's physician forum on GERD, available at:http://www.acg.gi.org/phyforum/gifocus/2evi.html
American College of Gastroenterology's guidelines for the diagnosis
and treatment of GERD, available at:http://www-east.elsevier.com/ajg/issues/9406/ajg1123fla.htm