Barretts Easophagous Guidelines

7/23/2019 Barretts Easophagous Guidelines

1/8

AGA

American Gastroenterological Association Medical Position Statement on

the Management of Barretts Esophagus

The AGA Institute Medical Position Panel consisted of the authors of the technical review (Stuart J. Spechler, MD, AGAF,

Prateek Sharma, MD, Rhonda F. Souza, MD, AGAF, John M. Inadomi, MD, AGAF, Nicholas J. Shaheen, MD, MPH,

AGAF), the chair of the Medical Position Panel (John I. Allen, MD, MBA, AGAF), the chair of the AGA Institute Practice

Management and Economics Committee and the AGA Institute CPT Advisor (Joel V. Brill, MD, AGAF), a community-based

gastroenterologist (Ronald E. Pruitt, MD, FACP, AGAF, FACG), an author of the AGA Institute Technical Review on the

Management of Gastroesophageal Reflux Disease (Peter J. Kahrilas, MD, AGAF), a general surgeon (Jeffrey H. Peters, MD),

a primary care physician (Kenneth Nix, MD), a pathologist (Elizabeth A. Montgomery, MD), a patient advocate (B. Donald

Mitchell), and an insurance provider representative (John Yao, MD, MBA, MPH, MPA, Senior Medical Director, Blue Shield

of California).

Podcast interview:www.gastro.org/gastropodcast.

This medical position statement considers a series ofbroad questions on the diagnosis, key clinical fea-tures, and management of Barretts esophagus. It is pub-lished in conjunction with a technical review1 on thesame subject, and interested readers are encouraged torefer to this publication for in-depth considerations oftopics covered by these questions. Topics were developed

during discussions among the authors of the technicalreview, representatives from the American Gastroenter-ological Association (AGA) Institute Council, and theAGA Institute Clinical Practice and Quality ManagementCommittee (CPQMC). The medical position statementconsiders major clinical issues encountered by physicianswho treat patients with Barretts esophagus. General is-sues related to the management of gastroesophageal re-flux disease (GERD), which often accompanies Barrettsesophagus, have been discussed in an earlier AGA medi-cal position statement2 and are not considered here.

For each question, the authors of the technical review

conducted a comprehensive literature search, reviewedpertinent reports, and analyzed the rigor and quality ofthe evidence. Once the technical review was completed,the authors submitted it to a medical position panel(MPP), which then developed the medical position state-ment. The MPP was composed of the authors of thetechnical review, gastroenterologists from a variety ofcommunity practice settings, a pathologist, a gastrointes-tinal surgeon whose area of expertise includes Barrettsesophagus, an author of the AGA technical review on themanagement of GERD, a health plan representative, anda patient with Barretts esophagus.

Developing recommendations for the management ofBarretts esophagus proved especially challenging given

the limited data available. The MPP ranked the strengthof each recommendation based on the strength of evi-

dence for each of its original recommendations afterconsidering the quality of evidence for a specific questionand the net health benefit, that is, the difference be-

tween estimated benefits and the risks of the interven-tion. The Grading of Recommendations Assessment, De-velopment and Evaluation (GRADE) process was used toassess the strength of recommendations, as outlined in a

previously published AGA document.3

Once the MPP approved the document, it was for-warded to the CPQMC for further review, comment, andmodification. The CPQMC met on September 24, 2010,

and reviewed the recommendations of the MPP. The finalcommittee-approved medical position statement thenwas forwarded to the AGA Institute Governing Board,

where endorsement was decided by majority vote of theboard. The final document is published without furthermodification in GASTROENTEROLOGYand becomes the of-

ficial position of the AGA Institute.The medical position statement presents information

by addressing clinically related questions and summariz-

ing key points from the technical review. When specificrecommendations about medical interventions or man-agement strategies for patients with Barretts esophagusare stated, a strength of recommendation and quality

Abbreviations used in this paper:AGA, American Gastroenterologi-cal Association; CPQMC, Clinical Practice and Quality ManagementCommittee; EMR, endoscopic mucosal resection; GEJ, gastroesopha-geal junction; GERD, gastroesophageal reflux disease; GRADE, Gradingof Recommendations Assessment, Development and Evaluation;MPP, medical position panel; PDT, photodynamic therapy; PPI, protonpump inhibitor; RFA, radiofrequency ablation.

2011 by the AGA Institute

0016-5085/$36.00doi:10.1053/j.gastro.2011.01.030

AGA

GASTROENTEROLOGY 2011;140:10841091
http://www.gastro.org/gastropodcasthttp://www.gastro.org/gastropodcasthttp://www.gastro.org/gastropodcast


2/8

of evidence is stated. The strength of recommendation isscored either weak or strong and quality of evidenceis ranked as high, moderate, low, or very low in accor-dance with GRADE criteria. Recommendations are high-lighted by appearing within a text box.

A strong recommendation implies that benefits out-weigh risks for most patients. A weak recommendationimplies that benefits, risks, and the burden of interven-tion are balanced among several legitimate managementoptions or that appreciable uncertainty existed.

Clinicians are aware of the importance of patient val-ues and preferences in making clinical decisions. TheMPP and CPQMC recognized that fully informed pa-tients could make different choices about medical inter-ventions. When a clear recommendation was not possiblebased on scientific evidence but rational clinicians andpatients might opt for different management choices, a

weak recommendation was assigned. Applying this ap-proach, high-quality evidence does not always result instrong recommendations and, conversely, strong recom-mendations can emerge from lower-quality evidence.

Areas of Debate

There are several specific issues with regard toBarretts esophagus that generated significant debate andfor which there was not uniform consensus. We acknowl-edge that some conclusions concerning strengths of rec-ommendation do not represent full consensus of theMPP and/or the CPQMC and/or the authors of thetechnical review. Conclusions will impact managementdecisions for individual patients and may be used by avariety of entities to make decisions about cost-effective-ness and insurance coverage.

In the 5 decades since Norman Barrett first describedthe condition that bears his name, multiple definitions ofBarretts esophagus have been used to establish studypopulations for research and by clinicians to managepatients. As a result, the true incidence of Barretts esoph-agus in the general population and the risk of progres-sion to cancer both continue to be areas of uncertaintyand debate. Current recommendations are based in largepart on data derived from recent treatment trials, where

clear endoscopic definitions of landmarks are recordedand expert pathologists performed biopsy analysis withina centralized pathology laboratory.

Given the continuing variability in diagnosis of Bar-retts esophagus and dysplasia, the MPP based their rec-ommendations for patient management on the assump-tion that a patients diagnosis of Barretts esophagus andthe presence or absence of low-grade and high-gradedysplasia would be accurate to the highest degree possi-ble using best current standards of practice. Such stan-dards would include an endoscopic definition of Barrettsesophagus as defined in this medical position statement

and confirmation of a diagnosis of dysplasia by at leastone additional pathologist, preferably one who is an

expert in esophageal histopathology. We recognize that,at this time, additional qualifications that establish apathologist as one with specialized expertise in gastroin-testinal pathology and Barretts histology do not existand furthermore that this recommendation will result in

a financial impact on the health care system that webelieve is justified given the clinical implications associ-ated with making an accurate diagnosis.

A second area of debate concerned the clinical end pointneeded to recommend an intervention. On one hand, sev-eral experts advocated using reductions in cancer incidenceor cancer deaths as the sole criteria upon which to recom-mend certain management strategies (such as endoscopicsurveillance of Barretts esophagus or endoscopic ablationof Barretts mucosa). Others considered intermediate endpoints (for example, 5-year sustained elimination of Bar-retts mucosa in the case of endoscopic ablation) to be an

acceptable clinical outcome. There is no consensus in themedical literature about this issue, and it remains an area ofcontroversy influenced by various and legitimate but differ-ing points of view.

The MPP recognized current limits of scientific knowl-edge and the importance of the relationship between atreating physician and an informed patient in whichknown risks and potential benefits of a proposed medicalintervention are discussed. Ultimately, patients shouldshare in decisions about what care they receive. Withinareas of uncertainty, physicians and patients should haveoptions of therapy available if the net health benefit isacceptable.

Definition of Barretts Esophagus

For the purposes of this statement, the definitionof Barretts esophagus is the condition in which anyextent of metaplastic columnar epithelium that predis-poses to cancer development replaces the stratified squa-mous epithelium that normally lines the distal esopha-gus. Presently, intestinal metaplasia is required for thediagnosis of Barretts esophagus because intestinal meta-plasia is the only type of esophageal columnar epitheliumthat clearly predisposes to malignancy.

DiscussionThe diagnosis of Barretts esophagus can be sus-

pected when, during endoscopic examination, columnarepithelium is observed to extend above the gastroesoph-ageal junction (GEJ) into the tubular esophagus. Numer-ous authorities have defined Barretts esophagus variablyfor more than 5 decades, and any definition of the con-dition will continue to have an arbitrary component. IfBarretts esophagus is to be considered a medical condi-tion rather than merely an anatomic curiosity, then itshould have clinical importance. The columnar-linedesophagus has clinical importance primarily because it

predisposes to esophageal cancer. Therefore, Barrettsesophagus can be defined conceptually as the condition

AGA

March 2011 AGA 1085


3/8

in which any extent of metaplastic columnar epitheliumthat predisposes to cancer development replaces thestratified squamous epithelium that normally lines thedistal esophagus.

Intestinal-type epithelium in the esophagus is clearly

abnormal and metaplastic and predisposes to cancer de-velopment. Therefore, an esophageal biopsy specimenthat shows intestinal-type epithelium above the GEJ es-tablishes the diagnosis of Barretts esophagus. Recentdata suggest that cardia-type epithelium in the esopha-gus is also abnormal and may predispose to malignancy.The key unanswered clinical question for patients whohave cardia-type epithelium in the distal esophagus isthis: What is the risk of developing esophageal cancer? Agreat majority of studies on the risk of cancer in Barrettsesophagus have included patients with esophageal intes-tinal metaplasia either primarily or exclusively. Although

cardia-type epithelium might be a risk factor for malig-nancy, the magnitude of that risk remains unclear.Presently, there are insufficient data to make meaning-

ful recommendations regarding management of patientswho have solely cardia-type epithelium in the esophagus,and we do not recommend use of the term Barrettsesophagus for those patients. Based on this lack of data,it is justified not to perform endoscopic surveillance forpatients solely with cardia-type epithelium in the esoph-agus.

The issue of what endoscopic landmark best identifiesthe GEJ (ie, the level at which the esophagus ends and thestomach begins) is likely to remain controversial because

currently available data do not support a universallyaccepted definition. A majority of published studies onBarretts esophagus conducted over the past 20 yearshave used the proximal extent of the gastric folds as thelandmark for the GEJ. In the absence of compelling datafor the use of alternative markers, we advocate the con-tinued use of this landmark.

There may be clinical value in measuring and recordingthe extent of Barretts metaplasia visualized during en-doscopic examination (ie, the distance between the GEJand the squamocolumnar junction in the esophagus).The likelihood of finding intestinal metaplasia in the

columnar-lined esophagus, the severity of underlyingGERD, and the magnitude of the cancer risk appear tovary directly with the extent of the metaplastic lining.Therefore, we advocate systematic recording of the cir-cumferential extent and maximum extent of metaplasiain endoscopic reports using a methodology such as thePrague C (circumferential extent in centimeters) and M(maximum extent in centimeters) system.

The Impact of Barretts Esophagus onthe Individual Patient

Although the risk of esophageal cancer or death

from Barretts esophagus is low for an individual, theimpact of this diagnosis on patients is great because

mortality from esophageal cancer remains high and psy-chological and financial consequences have been docu-mented.

The annual incidence of esophageal cancer for thegeneral population of patients with Barretts esoph-agus is approximately 0.5% per year.

Although deaths from esophageal adenocarcinomaclearly occur with increased frequency in patientswith Barretts esophagus, on an actuarial basis theimpact on overall life expectancy for the individualpatient is low.

Mortality from cardiovascular disease may be in-creased in patients with Barretts esophagus, perhapsbecause both are associated with obesity.

Patients with Barretts esophagus report a poorerquality of life than individuals in the general popu-lation. It is unclear whether this is due to anxietyabout cancer, discomfort due to GERD symptoms,or other factors.

A diagnosis of Barretts esophagus causes psycholog-ical stress in many patients and may increase finan-cial burdens due to increased life and health insur-ance premiums.

Discussion

Early studies of Barretts esophagus tended tooverestimate cancer risk due to the size of studies andselection bias. More recent studies and large meta-analy-ses have reported risks of 0.5% per year. Thus, 1 in 200patients with Barretts esophagus will develop esophagealcancer each year. It is not known if this risk varies (eitherincreases or decreases) as time passes from the originaldiagnosis. Some reports suggest a higher risk for patientswith long-segment Barretts esophagus and a greater riskin men compared with women.

The impact of esophageal adenocarcinoma on an in-dividual is devastating because current treatment optionsare limited and odds of survival remain low. Therefore,patients with Barretts esophagus, especially those withdysplasia, justifiably experience an increased level of anx-

iety and emotional burden. From a population riskstandpoint, however, a diagnosis of Barretts esophagushas little impact on overall life expectancy because thelifetime risk of developing esophageal cancer, even withinthe Barretts population, is low on an actuarial basis.Increased risk of death can come from cardiovascularcauses because of the association of both cardiac diseaseand Barretts esophagus with obesity.

Despite the fact that, on a population basis, the risk ofcancer is not greatly increased for patients with Barrettsesophagus, there is published evidence of adverse insur-ance selection and increased premium rates after a diag-

nosis of Barretts esophagus. It is imperative, therefore,that treating physicians realize the impact of this diag-

AGA

1086 AGA GASTROENTEROLOGY Vol. 140, No. 3


4/8

nosis and make every attempt to provide accurate evalu-ation and appropriate counseling concerning risks ofcancer and expectations from surveillance or treatmentstrategies.

Barretts Esophagus Risk andScreening

In patients with multiple risk factors associated withesophageal adenocarcinoma (age 50 years or older,male sex, white race, chronic GERD, hiatal hernia,elevated body mass index, and intra-abdominal distri-bution of body fat), we suggest screening for Barrettsesophagus (weak recommendation, moderate-qualityevidence).

We recommend against screening the general popula-tion with GERD for Barretts esophagus (strong rec-ommendation, low-quality evidence).

Well-established risk factors for Barretts esophagusinclude age older than 50 years, male sex, white race,chronic GERD, hiatal hernia, elevated body massindex, and intra-abdominal distribution of body fat.Quality of Evidence: High

Endoscopic screening for Barretts esophagus maybe warranted on an individual basis for patients who

have multiple risk factors, as noted previously. Thereis inadequate evidence of benefit to recommend en-doscopic screening for Barretts esophagus in thegeneral population of patients with GERD who donot have risk factors. Quality of Evidence: Low

Discussion

Despite the dearth of studies showing clinicalbenefit from endoscopic screening for Barretts esopha-gus, the practice remains widespread among clinicians inthe United States. Cost-effectiveness analyses suggestthat endoscopic screening may be warranted if certain

predefined clinical parameters are met, but several con-ceptual and logistical difficulties diminish the utility ofscreening endoscopy as it is currently practiced in theUnited States. First and foremost, approximately 40% ofsubjects who have adenocarcinoma of the esophagusreport no history of chronic GERD symptoms, and thenumber of patients with GERD compared with the num-ber with esophageal adenocarcinoma makes screening allpatients with GERD cost-ineffective. A reasonable middleground would be to select individuals with multiple riskfactors for endoscopic examination once a discussionbetween physician and patient occurs concerning expec-

tations, risks, benefits, and implications of the examina-tion.

Risk of Progression in BarrettsEsophagus

The diagnosis of dysplasia in Barretts esophagusshould be confirmed by at least one additional pathol-ogist, preferably one who is an expert in esophagealhistopathology (strong recommendation, moderate-quality evidence).

Published rates of progression from low-grade dys-plasia to either high-grade dysplasia or esophageal

adenocarcinoma range from 0.5% to 13.4% per pa-

tient per year, depending on the rigor of pathologic

confirmation of dysplasia. Quality of Evidence:

Low

A recent meta-analysis of multiple historical stud-ies reported an overall risk of progression from

high-grade dysplasia to cancer of 6% per patient

per year. Quality of Evidence: Moderate

Discussion

The risk of progression from low-grade dyspla-

sia in Barretts esophagus to either high-grade dyspla-

sia or adenocarcinoma remains controversial, based in

large part on the difficulty in distinguishing dysplasia

from nondysplastic Barretts esophagus and determin-

ing with reproducible accuracy the degree of dysplasia.Criteria used to distinguish low-grade dysplasia from

nondysplastic Barretts esophagus (especially in the

presence of esophageal inflammation) and low-grade

dysplasia from high-grade dysplasia are based primar-

ily on the degree of architectural and cytologic aber-

rations, which are, to some degree, subjective in nature.

Because dysplasia progresses to cancer in a manner

that lacks definitive markers of progression, there are

no well-defined cutoff points that separate low-grade

from high-grade dysplasia at this time. There is activeresearch focused on biomarkers of progression, and ifsuch markers are defined, management recommenda-

tions will likely change.

Low-grade dysplasia tends to be overcalled among

community pathologists, especially during an initial

endoscopic examination where esophageal inflamma-

tion may be present. This situation has been docu-

mented during controlled trials in which pathologists

with extensive experience in Barretts histology re-

viewed biopsy specimens obtained from general com-

munity practice and the diagnosis of dysplasia was

made only when consensus among the expert panelwas reached.

AGA

March 2011 AGA 1087


5/8

Endoscopic Surveillance in PatientsWith Barretts Esophagus

We suggest that endoscopic surveillance be performedin patients with Barretts esophagus (weak recommen-dation, moderate-quality evidence).

We suggest the following surveillance intervals (weakrecommendation, low-quality evidence):

No dysplasia: 35 years Low-grade dysplasia: 612 months High-grade dysplasia in the absence of eradication

therapy: 3 months.

Biopsy specimens obtained as part of an endo-scopic surveillance program can detect curableneoplasia in patients with Barretts esophagus.Whether endoscopic surveillance reduces cancerincidence or mortality is not known because nolong-term trial designed to answer this questionhas been performed.

Discussion

Endoscopic surveillance has become the stan-dard of practice for patients with Barretts esophagus

based on the unproven assumption that the practicewill reduce deaths from esophageal adenocarcinomaand thereby prolong survival. Societal guidelines gen-erally have recommended endoscopic surveillance forpatients with Barretts esophagus at intervals that varywith grade of dysplasia found in the metaplastic epi-thelium. Intervals of 3 to 5 years have been suggestedfor patients who have no dysplasia, 6 to 12 months forthose found to have low-grade dysplasia, and every 3months for patients with high-grade dysplasia whoreceive no ablation therapy. Studies suggest that ad-herence to recommended surveillance biopsy protocols

(for details, refer to the technical review) is associ-ated with higher rates of dysplasia and cancer detec-tion, but many practicing gastroenterologists do notadhere to those guidelines and adherence appears to bepoorest for the population at highest risk for cancerdevelopment (ie, patients with extensive Barrettsmetaplasia).

The preponderance of evidence, virtually all derivedfrom retrospective or poorly controlled studies, sug-gests that endoscopic surveillance can reduce mortalityfrom esophageal adenocarcinoma through the earlydetection of treatable cancers when attention is paid to

recommended biopsy techniques (see the followingtext).

Biomarkers in the Management ofBarretts Esophagus

We suggest against the use of molecular biomarkers to

confirm the histologic diagnosis of dysplasia or as amethod of risk stratification for patients with Barrettsesophagus at this time (weak recommendation, low-quality evidence).

Although biomarkers show promise, they cannotbe used to confirm the diagnosis of Barretts dysplasiaand have not been shown to predict which patients withBarretts are at risk for progression. To date, neitherindividual biomarkers nor panels of markers can be rec-ommended.Quality of Evidence: Low

Biopsy Protocol for EndoscopicSurveillance of Barretts Esophagus

For patients with Barretts esophagus who are under-going surveillance:

We recommend endoscopic evaluation be performedusing white light endoscopy (strong recommendation,moderate-quality evidence).

We recommend 4-quadrant biopsy specimens be takenevery 2 cm (strong recommendation, moderate-qualityevidence).

We recommend specific biopsy specimens of any mu-cosal irregularities be submitted separately to the pa-thologist (strong recommendation, moderate-qualityevidence).

We recommend 4-quadrant biopsy specimens be ob-tained every 1 cm in patients with known or suspecteddysplasia (strong recommendation, moderate-qualityevidence).

We suggest against requiring chromoendoscopy oradvanced imaging techniques for the routine surveil-lance of patients with Barretts esophagus at this time(weak recommendation, low-quality evidence).

During endoscopic surveillance of Barretts esopha-gus, careful inspection of the entire area of metapla-sia using white light endoscopy remains the stan-dard of care. Specific biopsy sampling of anymucosal irregularities should be performed. In areaswithout mucosal irregularities, 4-quadrant biopsysampling of the Barretts epithelium at intervals of

every other centimeter is recommended for patientswho are not known to have dysplasia. For patients

AGA



6/8

who are known to have dysplasia, 4-quadrant biopsysampling of the Barretts epithelium at intervals ofevery centimeter is recommended. Quality of Evi-dence: Moderate

For the routine endoscopic evaluation of Barrettsesophagus, the use of chromoendoscopy or elec-tronic chromoendoscopy or advanced imaging tech-niques such as confocal laser endomicroscopy is notnecessary. These technologies may be helpful inguiding the performance of biopsies in patients whoare known to have dysplasia and in patients whohave mucosal irregularities detected by white lightendoscopy. Quality of Evidence: Low

Prevention of Cancer in BarrettsEsophagus

We recommend against attempts to eliminate esoph-ageal acid exposure (proton pump inhibitors [PPIs] indoses greater than once daily, esophageal pH moni-toring to titrate PPI dosing, or antireflux surgery) forthe prevention of esophageal adenocarcinoma (strongrecommendation, moderate-quality evidence).

We recommend screening patients to identify car-diovascular risk factors for which aspirin therapy isindicated (strong recommendation, high-quality

evidence).

We suggest against the use of aspirin solely to preventesophageal adenocarcinoma in the absence of otherindications (weak recommendation, moderate-qualityevidence).

For patients with Barretts esophagus, GERD ther-apy with medication effective to treat GERD symp-toms and to heal reflux esophagitis is clearly indi-cated, as it is for patients without Barrettsesophagus. Quality of Evidence: High

Evidence to support use of acid-reducing agents,specifically PPIs, in patients with Barretts esoph-agus solely to reduce risk of progression to dys-plasia or cancer is indirect and has not beenproven in a long-term controlled trial. The risksand potential benefit of long-term PPI therapyshould be discussed carefully with patients withBarretts esophagus in the context of their overallhealth status and medication use. Quality of Ev-idence: Low

Evidence does not support greater than standard

doses of PPI therapy with an expectation of cancerrisk reduction. Quality of Evidence: Low

Antireflux surgery is not more effective than medicalGERD therapy for the prevention of cancer in Bar-retts esophagus. Quality of Evidence: Moderate

Although available evidence suggests that nonsteroi-dal anti-inflammatory drugs (NSAIDs) may decreasethe incidence of esophageal cancer, it is not clearthat this potential benefit outweighs the risks of themedications. However, given that cardiovasculardeaths are more common than deaths from esoph-ageal adenocarcinoma among patients with Barrettsesophagus, screening for cardiovascular risk factorsand interventions is warranted. Quality of Evi-dence: Moderate

The Role of Endoscopic Therapy inPatients With Barretts Esophagus

We recommend endoscopic eradication therapywith radiofrequency ablation (RFA), photodynamictherapy (PDT), or endoscopic mucosal resection(EMR) rather than surveillance for treatment ofpatients with confirmed high-grade dysplasia withinBarretts esophagus (strong recommendation, mod-erate-quality evidence).

We recommend EMR for patients who have dyspla-sia in Barretts esophagus associated with a visiblemucosal irregularity to determine the T stage of the

neoplasia (strong recommendation, moderate-qual-ity evidence).

Discussion

The goal of endoscopic eradication therapy forpatients with dysplasia in Barretts esophagus is to per-manently eliminate all intestinal metaplasia and achievea complete reversion to squamous epithelium. Althoughendoscopic eradication therapy is not suggested for thegeneral population of patients with Barretts esophagusin the absence of dysplasia, we suggest that RFA, with orwithout EMR, should be a therapeutic option for select

individuals with nondysplastic Barretts esophagus whoare judged to be at increased risk for progression tohigh-grade dysplasia or cancer. Specific criteria that iden-tify this population have not been fully defined at thistime. When such criteria are identified from controlledtrials, then management recommendations should beupdated.

Endoscopic eradication therapy with RFA should alsobe a therapeutic option for treatment of patients withconfirmed low-grade dysplasia in Barretts esophagus. Werecognize the controversies surrounding both definitionand management of dysplasia in Barretts esophagus and

that the risk of progression to cancer in this populationof patients can vary greatly among individuals. The AGA

AGA

March 2011 AGA 1089


7/8

Institute strongly supports the concept of shared deci-sion making where the treating physician and patienttogether consider whether endoscopic surveillance oreradication therapy is the preferred management optionfor each individual.

The current literature is inadequate to recommendendoscopic eradication therapy with cryotherapy for pa-tients with confirmed low-grade or high-grade dysplasiawithin Barretts esophagus or patients judged to be athigh risk for progression to high-grade dysplasia oresophageal carcinoma. Further studies are needed to as-sess whether reversion to squamous epithelium can per-sist long-term after cryotherapy.

As stated in the introduction, recommendations for clin-ical management are based on the assumption that a pa-tients diagnosis of Barretts esophagus and the presence orabsence of low-grade and high-grade dysplasia is accurate to

the highest degree possible using best current standards ofpractice. The statement regarding confirmed low-grade orhigh-grade dysplasia refers to patients in whom the diagno-sis is confirmed by at least 2 pathologists, preferably one ofwhom is an expert in esophageal histopathology. We rec-ommend that the diagnosis of dysplasia be confirmed inthis manner before initiating endoscopic eradication ther-apy for any stage of dysplasia.

Current endoscopic techniques used to eradicateBarretts esophagus include RFA, PDT, cryotherapy,thermal energy application, and EMR. Details ofeach technology are found in the technical review.

The most commonly used technologies currently areRFA and EMR used alone or in combination. Evi-dence for their efficacy has emerged rapidly over thepast decade. Quality of Evidence: Moderate

The goal of endoscopic eradication therapy is theelimination of all Barretts epithelium to preventneoplastic progression. Complete eradication ap-pears to be more effective than therapy that removesonly a localized area of dysplasia in Barretts epithe-lium.Quality of Evidence: Low

Although RFA and PDT have not been compared

head-to-head in controlled trials, RFA appears tohave at least comparable efficacy and fewer seriousadverse effects compared with PDT. Quality of Ev-idence: Moderate

The second goal of eradication therapy is to achievereversion to normal-appearing squamous epitheliumwithin the entire length of the esophagus withoutislands of buried intestinal metaplasia. RFA can leadto reversion of the metaplastic mucosa to normal-appearing squamous epithelium in a high propor-tion of subjects at any stage of Barretts esophagus.The data to date show that reversion to squamous

epithelium can persist for up to 5 years. Quality ofEvidence: High

There are no data from controlled trials showingthat endoscopic eradication therapy, including RFAand cryotherapy, is more effective at reducing cancerrisk or more cost-effective than long-term endo-scopic surveillance in patients with Barretts esoph-

agus in the absence of dysplasia (nondysplastic Bar-retts metaplasia). Quality of Evidence: Very Low

RFA therapy for patients with low-grade dysplasialeads to reversion to normal-appearing squamous epi-thelium in90% of cases.Quality of Evidence: High

RFA therapy for patients with high-grade dysplasiareduces progression to esophageal cancer, as shownin a randomized sham-controlled trial. Several addi-tional uncontrolled trials have shown a similar re-duction in cancer development and sustained rever-sion to squamous mucosa in a large percentage of

patients.Quality of Evidence: High The current literature is inadequate to assess the

ability of cryotherapy to achieve sustained reversionof the metaplastic mucosa to normal-appearingsquamous epithelium in subjects at any stage ofBarretts esophagus. Further longitudinal studies areneeded.Quality of Evidence: Very Low

EMR is both a valuable diagnostic/staging proce-dure and a potentially therapeutic procedure thatshould be performed in patients who have dysplasiaassociated with visible mucosal irregularities in Bar-retts esophagus. Quality of Evidence: High

The Role of Esophagectomy

Most patients with high-grade dysplasia (70%80%)can be successfully treated with endoscopic eradica-tion therapy. Esophagectomy in patients with high-grade dysplasia is an alternative; however, currentevidence suggests that there is less morbidity withablative therapy. Quality of Evidence: Moderate

Before proceeding with esophagectomy, patientswith high-grade dysplasia or intramucosal carci-noma with Barretts esophagus should be referred

for evaluation by surgical centers that specialize inthe treatment of foregut cancers and high-gradedysplasia.Quality of Evidence: High

Supplementary Material

Note: To access the supplementary materialaccompanying this article, visit the online version ofGastroenterology at www.gastrojournal.org, and at doi:10.1053/j.gastro.2011.01.030.

References

1. Spechler SJ, Sharma P, Souza RF, et al. American Gastroenter-

ological Association technical review on the management of Bar-retts esophagus. Gastroenterology 2011;140:e18 e52.

AGA

http://www.gastrojournal.org/http://www.gastrojournal.org/http://dx.doi.org/10.1053/j.gastro.2011.01.030http://dx.doi.org/10.1053/j.gastro.2011.01.030http://dx.doi.org/10.1053/j.gastro.2011.01.030http://dx.doi.org/10.1053/j.gastro.2011.01.030http://dx.doi.org/10.1053/j.gastro.2011.01.030http://www.gastrojournal.org/


8/8

2. American Gastroenterological Association medical position state-

ment on the management of gastroesophageal reflux disease.

Gastroenterology 2008;135:13831391, 1391.e1-5.

3. Dassopoulos T, Inadomi JM, Lewis JD, et al. The development of

clinical guidelines by the American Gastroenterological Associa-

tion. Gastroenterology 2010;138:417418.

Reprint requests

Address requests for reprints to: Chair, Clinical Practice andQuality Management Committee, AGA National Office, 4930 Del RayAvenue, Bethesda, Maryland 20814. e-mail:[email protected];telephone: (301) 272-1189.

Conflicts of interest

The authors disclose the following: Dr Souza is a consultant forTakeda Pharmaceuticals and is a consultant for and receives grantsupport from AstraZeneca. Dr Inadomi receives grant/researchsupport from NIH and BARRX; and is a consultant for TakedaPharmaceuticals, AstraZeneca, and Ethicon Endo-Surgery, Inc. Dr

Peters receives grant/research support from Torax Medical Inc,Medigus Ltd, and Takeda Pharmaceuticals. Dr Shaheen is aconsultant to NeoGenomics Labs; AstraZeneca; CSA Medical;Oncoscope, Inc; and Takeda. He has also received research fundingfrom AstraZeneca; BRRX Medical, Inc; Oncoscope, Inc; CSAMedical; and Takeda. Dr Spechler receives grant/research support

from AstraZeneca, BARRX Medical, Inc, and Takeda and is aconsultant for Proctor & Gamble. Dr Sharma receives grant/research support from BARRX Medical, Inc, Mauna Kea, Olympus,and Takeda Pharmaceuticals. He is also a consultant forAstraZeneca, Santarus, and Takeda Pharmaceuticals. The remainingauthors disclose no conflicts.

Funding

Supported by the Office of Medical Research, Department ofVeterans Affairs (R.F.S., S.J.S.), the National Institutes of Health(R01-DK63621 to R.F.S., R01-CA134571 to R.F.S. and S.J.S., andR01 DC00646 to P.J.K.), and the National Institutes of Health,National Institute of Diabetes and Digestive and Kidney Diseases(K24 DK080941 to J.M.I.).

AGA

March 2011 AGA 1091
mailto:[email protected]:[email protected]:[email protected]

Documents

Barretts Easophagous Guidelines