2011 03 revision tecnica de barrett

8/4/2019 2011 03 Revision Tecnica de Barrett

1/35

American Gastroenterological Association Technical Review on theManagement of Barretts Esophagus

Learning Objectives

This article has an accompanying continuing med-ical education activity on page e13. Learning Objective: Atthe end of this activity, the successful learner should:

1. Identify the risk factors associated with developmentof esophageal adenocarcinoma.

2. Determine who should undergo surveillance after be-ing diagnosed with Barretts esophagus.

3. Assess the role of endoscopic therapy for patients withBarretts esophagus.

American GastroenterologicalAssociation Institute Process forDevelopment of Technical Reviews

The aim of evidence-based medicine is to im-prove the quality of health care by integrat-ing the best research evidence with clinical expertise andpatient values. Evidence-based clinical guidelines are setsof recommendations intended to assist health care pro-viders and patients in selecting the best management forcommon clinical situations while accounting for patient-specific circumstances. In addition to providing optimal,patient-centered care and improved outcomes, guidelinescan reduce practice variability, identify gaps in evidence,enhance efficiency of resource use, and facilitate develop-ment of outcome and performance measures.

The American Gastroenterological Association Insti-tute (AGAI) Medical Position Statement Procedure Man-ual, released in 2007, endorses the 2003 version of the USPreventive Services Task Force system to grade strengthof recommendations. Although an excellent standard forproducing recommendations regarding preventive ser-vices, the US Preventive Services Task Force has limita-tions when used to assess interventions that are notbased on prevention. The Grading of Recommendations

Assessment, Development and Evaluation (GRADE) sys-tem (http://www.gradeworkinggroup.org/index.htm) hasbeen adopted by several national and international soci-eties and was constructed to address the shortcomings ofexisting grading systems, including the US PreventiveServices Task Force system. GRADE separates quality ofevidence from the strength of recommendation to ensurethat the magnitude of benefits and harms is assessed asrigorously as the efficacy of interventions. With regard tostrength of recommendations, GRADE has 2 categories:strong and weak (Table 1). Strong recommendations aremeant to signify interventions that should be received by

most individuals with a particular condition and can beadopted as policy in most circumstances. Weak recom-

mendations require individualized scrutiny of the evi-dence and policy making would require substantial de-bate and involvement from multiple stakeholders. Theclassification requires consideration of 4 factors: qualityof evidence, uncertainty about the balance between de-sirable and undesirable effects, variability in values andpreferences, and uncertainty about whether the interven-tion represents a wise use of resources (Table 2). Ofimportance to our current health care debate is thatinterventions receiving a strong recommendation may betargets for development of performance measures.

Quality of evidence is assessed on a 4-point scale: high,

moderate, low, and very low. Instead of being classifiedstrictly on the basis of study design (ie, randomized, con-trolled clinical trials automatically receiving high qualitymarks), these levels reflect the likelihood that further re-search would change our confidence in the estimate of thebeneficial effect of a particular intervention. Five factorsthat determine quality include study limitations, inconsis-tency of results between studies, indirectness (generalizabil-ity) of results, imprecision, and publication bias. For thisreason, randomized, controlled clinical trials that havemethodological flaws may be downgraded, whereas well-done observational studies that have large effect sizes (ie,

relative risk [RR] 25 or 0.50.2) may be upgraded.

AGAI Procedure for Construction ofTechnical Reviews

The AGAI Clinical Practice and Quality Manage-ment Committee (CPQMC) chooses a topic by consensusdiscussion, votes after reviewing a list of potential topicsderived from AGAI member recommendations, and develops the specific questions the guideline will answer.The CPQMC committee chair, with support of AGA staff,then contacts the AGAI clinical counsel chair and re-

Abbreviations used in this paper: AFI, autofluorescence imaging;AGAI, American Gastroenterological Association Institute; APC, argonplasma coagulation; CI, confidence interval; CPQMC, Clinical Practiceand Quality Management Committee; EMR, endoscopic mucosal re-section; EUS, endoscopic ultrasonography; GEJ, gastroesophageal

junction; GERD, gastroesophageal reflux disease; GRADE, Grading ofRecommendations Assessment, Development and Evaluation; LOH,loss of heterozygosity; MPEC, multi-polar electrocoagulation; NBI, nar-row band imaging; PDT, photodynamic therapy; PPI, proton pumpinhibitor; QOLRAD, Quality of Life in Reflux and Dyspepsia; RFA, radio-frequency ablation; RR, relative risk; SEER, Surveillance, Epidemiologyand End Results; SF-36, 36-Item Short Form Health Survey; TR, tech-nical review.

2011 by the AGA Institute

0016-5085/$36.00doi:10.1053/j.gastro.2011.01.031

AGA

GASTROENTEROLOGY 2011;140:e18e52
http://linkinghub.elsevier.com/retrieve/pii/S001650851100093Xhttp://linkinghub.elsevier.com/retrieve/pii/S001650851100093Xhttp://www.gradeworkinggroup.org/index.htmhttp://www.gradeworkinggroup.org/index.htmhttp://www.gradeworkinggroup.org/index.htmhttp://linkinghub.elsevier.com/retrieve/pii/S001650851100093X


2/35

quests the input of the counsel for authorship and ex-ternal reviewers.

Authors are selected and write a technical review (TR),which is an evidence-based document that provides thebasis for clinical practice recommendations. For each ofthe specific questions raised by the CPQMC, authors

conduct an independent systematic review of the litera-ture using published guidelines (PRISMA). Articles se-lected for inclusion in the TR are based on a prioriinclusion and exclusion criteria agreed on by all authors.Data extraction is shared among TR authors, and theindividual study and summary results are reviewed andapproved by all authors. The search terms for each topicincluded in the TR are included in the Appendix. It is notthe function of the TR to provide a summary estimate foreach variable included in the review. For this reason,results are summarized in the text of the TR and notsubjected to a formal meta-analysis. The draft TR iscompiled by the lead author and approved by all authorsbefore submission for publication.

A medical position panel composed of the TR authors,additional content experts, practicing gastroenterolo-gists, other specialists (eg, surgeon, pathologist), a pa-tient representative, a payer representative, and AmericanGastroenterological Association staff meet through a se-ries of face-to-face and telephone meetings to constructthe medical position statement, which is based on the TRbut also reflects these discussions by the medical positionpanel. The medical position panel approves the medicalposition statement, after which this document and the TRare reviewed by the CPQMC. Based on the vote of the

committee, a recommendation is submitted to the AGAIGoverning Board, which provides final approval. When ap-proval is granted, the medical position statement is pub-lished in GASTROENTEROLOGY and is also posted on theAmerican Gastroenterological Association web site.

The objectives of the AGAI TR on the management ofpatients with Barretts esophagus were to evaluate diag-nostic and management strategies for patients at risk foror diagnosed with Barretts esophagus. Specifically, 10broad questions were developed by interaction amongthe authors, the AGAI, the Clinical Practice and Qual-ity Management Committee, and representatives from

the AGAI Council. The questions were designed toencapsulate the major management issues leading to

consultations for Barretts esophagus and esophagealadenocarcinoma in clinical practice in 2010. For eachquestion, a comprehensive literature search was con-ducted, pertinent evidence reviewed, and a summary ofrelevant data produced. The conclusions of this review

were based on the best available evidence or, in theabsence of quality evidence, the expert opinion of theauthors of the TR.

What Landmark Identifies theGastroesophageal Junction? WhatEpithelial Type Is Required for theDiagnosis of Barretts Esophagus?What Is the Definition of BarrettsEsophagus? Should EndoscopistsMeasure the Extent of BarrettsMetaplasia?

Authorities generally have defined Barretts esoph-

agus conceptually as the condition in which metaplasticcolumnar epithelium replaces the stratified squamousepithelium that normally lines the distal esophagus.14

Unfortunately, this deceptively simple conceptual defini-tion does not translate readily into clinically useful diag-nostic criteria for 2 major reasons. (1) There are nouniversally accepted, precise, and validated landmarksthat delimit the distal extent of the esophagus (ie, thatidentify the gastroesophageal junction [GEJ]). If it cannotbe determined precisely where the esophagus ends andthe stomach begins, then it may not be possible to ascer-tain the type of epithelium that lines the most distal

esophagus. (2) There is no way to verify that gastric-typecolumnar epithelia found in the distal esophagus aremetaplastic. These 2 factors become major confounderswhen attempting to establish a diagnosis of Barrettsesophagus for patients with only short segments ofesophageal columnar epithelium.

What Landmark Identifies theGastroesophageal Junction?

The diagnosis of Barretts esophagus can be sus-pected when, during endoscopic examination, columnarepithelium (which has a characteristic endoscopic ap-pearance) is observed to extend above the GEJ into theesophagus. Of course, this diagnostic suspicion is based on

Table 1. GRADE: Strength of Recommendation

Strength of

recommendation

Clinical

implication Policy implication

Strong Do it or Dont

do it

Adherence to this

recommendation could be

used as a quality orperformance measure

Weak Probably do it

or Probably

dont do it

Recommendation not

suitable for quality or

performance measure

Table 2. GRADE: Quality of Evidence

Quality of evidence Estimate of certainty of effect

High Further research is very unlikely to change the

estimate of effect

Moderate Further research is likely to have an important

impact and may change the estimate of

effect

Low Further research is very likely to have an

important impact and is likely to change the

estimate of effect

Very low Any estimate of effect is uncertain

AGA

March 2011 AGA e19


3/35

the assumption that the endoscopist can identify the GEJ.Few studies have addressed specifically the issue of how tolocalize the GEJ, and even those that have the accuracy ofthe criteria used cannot be assessed meaningfully in theabsence of a validated landmark (ie, a gold standard).

Western endoscopists generally identify the GEJ as themost proximal extent of the gastric folds, a landmarkfirst proposed in 1987 in a report of a small and meth-odologically flawed study.5 The location of the proximalextent of the gastric folds is affected by respiration, gutmotor activity, and the degree of distention of the esoph-agus and stomach, all of which can change from momentto moment. Endoscopists in Asia often use the distalextent of the palisade vessels, which are fine longitudinalveins located in the lamina propria of the distal esopha-gus, as their landmark for the GEJ.6,7 The palisade vesselscan be obscured by esophagitis, their level of termination

can be irregular and difficult to localize with precision,and conceptually it is not clear why the distal end of thepalisade vessels should be considered the end of theesophagus. Thus, the scientific validity of the 2 mostwidely used landmarks for the GEJ is dubious.

The issue of the best landmark for the GEJ is likely toremain controversial indefinitely because, ultimately, thechoice of any such landmark will be arbitrary. The ma-jority of published studies on Barretts esophagus con-ducted over the past 20 years have used the proximalextent of the gastric folds as the landmark for the GEJand, in the absence of compelling data for the use ofalternative markers, we presently recommend the contin-

ued use of this landmark despite its shortcomings.

What Epithelial Type Is Required for aDiagnosis of Barretts Esophagus?

Barretts esophagus is judged to develop throughthe process of metaplasia in which one adult cell typereplaces another. For reasons that are not clear, Barrettsmetaplasia is predisposed to cancer development. Threetypes of columnar epithelia have been described in Bar-retts esophagus: (1) a gastric fundic-type epithelium thathas mucus-secreting cells, parietal cells, and chief cells;(2) a cardia-type (also known as junctional-type) epithe-

lium composed almost exclusively of mucus-secretingcells; and (3) an intestinal-type epithelium (sometimescalled specialized columnar epithelium or specialized in-testinal metaplasia) that contains prominent gobletcells.8 The fundic- and cardia-type epithelia in Barrettsesophagus can be morphologically indistinguishablefrom columnar epithelia found in the stomach.

If biopsy specimens of suspected Barretts esophagusreveal only fundic- and cardia-type epithelia, then it canbe difficult to establish that those epithelial types aremetaplastic because (1) biopsy sampling error can resultin inadvertent biopsy of the stomach instead of the

esophagus, especially when only short segments of co-lumnar epithelium appear to extend above the GEJ, and

(2) some authorities have argued that the normal distalesophagus can have short segments of a gastric-typecolumnar lining.9 With no precise landmark for the GEJ,it is difficult to support or refute that contention. Ifbiopsy specimens of suspected Barretts esophagus reveal

intestinal-type epithelium, in contrast, then there is littledoubt that the epithelium is abnormal and metaplastic.This finding does not obviate the issue of biopsy sam-pling error, however, because intestinal metaplasia iscommon in the stomach that is chronically infected with

Helicobacter pylori.10

Intestinal-type epithelium can be readily identified bythe pathologist and, unlike the gastric-type epithelia,intestinal-type epithelium is clearly abnormal when lo-cated in the esophagus. Furthermore, early reports sug-gested that the intestinal-type epithelium in Barrettsesophagus was the one predisposed to malignancy. For

those reasons, practical more than scientific or concep-tual, investigators and clinicians adopted the policy ofrequiring the demonstration of intestinal metaplasia inesophageal biopsy specimens as a sine qua non for thediagnosis of Barretts esophagus. However, recent find-ings have challenged the validity of that practice. Thereare data to suggest that cardia-type epithelium may notbe normal, but rather may be a metaplastic lining thatdevelops as a consequence of chronic gastroesophagealreflux disease (GERD).11

Histochemical and genetic studies of cardia-type epi-thelium have revealed molecular abnormalities, similar to

those found in specialized intestinal metaplasia, that maypredispose to cancer development.12,13 Recent clinicalstudies also suggest that cardia-type epithelium has ma-lignant potential. In one such study of 141 patients whounderwent endoscopic mucosal resection (EMR) forsmall esophageal adenocarcinomas, 71% had cardia-typeepithelium, not intestinal metaplasia, found adjacent tothe cancer, and 57% had no intestinal metaplasia what-soever found in the EMR specimen.14

The columnar-lined esophagus has clinical importanceonly because it predisposes to the development of esoph-ageal cancer. The debate about whether patients who

have only cardia-type epithelium lining the distal esoph-agus have Barretts esophagus is primarily a semanticissue. The key unanswered clinical question for those pa-tients is this: What is the risk of developing esophagealcancer? The great majority of studies on the risk of cancer inBarretts esophagus have included patients with specializedintestinal metaplasia either primarily or exclusively.15 Al-though recent data suggest that cardia-type epithelium maywell predispose to malignancy, the magnitude of that risk isnot yet clear. A reasonable estimate of cancer risk is neededto make rational management decisions for patients withBarretts esophagus, and no such estimate is available for

patients who have only cardia-type epithelium in their co-lumnar-lined esophagus.

AGA

e20 AGA GASTROENTEROLOGY Vol. 140, No. 3


4/35

Despite reasonable arguments supporting the conceptthat Barretts esophagus might be defined by the pres-ence of cardia-type as well as by intestinal-type epithe-lium in the esophagus, there are substantial practicalreasons for not adopting this definition into clinical

practice at this time. The inclusion of patients withcardia-type epithelium under the rubric of Barrettsesophagus would substantially increase the number ofpatients with that disorder, which would substantiallyincrease treatment costs. The benefits of surveillance andtreatment programs for Barretts esophagus are debated,even for patients with intestinal metaplasia, whose cancerrisk is far better defined. The likelihood of finding intes-tinal-type epithelium in Barretts esophagus varies di-rectly with the extent of the esophageal columnar lining,and the issue of whether to consider cardia-type epithe-lium a marker for Barretts esophagus usually concerns

only patients with short segments of esophageal colum-nar epithelium (generally segments considerably lessthan 3 cm in extent). The clinical benefit of biopsy sam-pling for patients with such short segments of esopha-geal columnar epithelium has not been established.

What Is the Definition of BarrettsEsophagus?

Any definition of Barretts esophagus necessarilywill have an arbitrary component. If Barretts esophagusis to be considered a medical condition rather thanmerely an anatomic curiosity, then it should have clinical

importance. The columnar-lined esophagus has clinicalimportance only if it predisposes to esophageal cancer.Therefore, we believe that Barretts esophagus can bedefined conceptually as the condition in which any extentof metaplastic columnar epithelium that predisposes tocancer development replaces the stratified squamous ep-ithelium that normally lines the distal esophagus. Pres-ently, intestinal metaplasia is the only one of the 3 typesof esophageal columnar epithelia described that clearlypredisposes to malignancy; therefore, we suggest that theterm Barretts esophagus presently should be used onlyfor patients who have intestinal metaplasia in the esoph-

agus. Circumstantial evidence suggests that cardia-typeepithelium also may be predisposed to cancer develop-ment, but presently that predisposition has not beenestablished and there are insufficient data to make mean-ingful recommendations regarding the management ofpatients who have a columnar-lined esophagus with car-dia-type epithelium alone. If future studies establish amalignant predisposition for cardia-type epithelium,then those patients also can be considered to have Bar-retts esophagus by our proposed definition. For now,however, only patients with intestinal-type columnarmetaplasia in the esophagus are known to have an in-

creased cancer risk, and only those patients meet ourcriteria for the diagnosis of Barretts esophagus.

Should Endoscopists Measure the Extent of Barretts Metaplasia?

Barretts esophagus has been categorized as longsegment (when the metaplastic epithelium extends atleast 3 cm above the GEJ) or short segment (when there

is 3 cm of metaplastic epithelium lining the esopha-gus).16 Another more recently proposed system for cate-gorizing Barretts esophagus, the Prague C and M crite-ria, identifies both the circumferential extent (C) and themaximum extent (M) of Barretts metaplasia.17 One studyhas shown excellent interobserver agreement among en-doscopists using the Prague C and M criteria when co-lumnar epithelium extends at least 1 cm above the GEJbut poor agreement for shorter segments of esophagealcolumnar lining.17

There may be clinical value in measuring the extent ofBarretts metaplasia visualized during endoscopic exam-

ination (ie, the distance between the GEJ and the squa-mocolumnar junction in the esophagus). Data suggestthat the likelihood of finding intestinal metaplasia in thecolumnar-lined esophagus and the magnitude of thecancer risk vary directly with the extent of the metaplasticlining (see the following text). For patients found to havedysplasia in Barretts esophagus, furthermore, the extentof metaplasia is a factor that may influence the choiceamong the therapeutic options (see the following text).Therefore, we advocate the use of a system, like thePrague C and M criteria, that provides information onthe extent of Barretts metaplasia. However, the clinicalbenefit of using any proposed endoscopic system forclassifying Barretts esophagus has not been establishedby formal investigation and, presently, patients with anyextent of intestinal metaplasia are managed similarly.

What Is the Risk of Esophageal Cancerfor the General Population of PatientsWith Barretts Esophagus?

A number of older reports linking Barretts esopha-gus to esophageal adenocarcinoma described an inordi-nately high incidence of cancer, in the range of 20 to 40 per1000 person-years (2%4% per year).1820 A variety of con-founding factors, including selection bias, the inclusion of

prevalent cancers, and publication bias, may have contrib-uted to the overestimation of cancer risk in those studies.15

Subsequent reports of larger series generally have describeda substantially lower cancer risk for patients with Barrettsesophagus, but even some modern reports describe veryhigh cancer incidence rates.21 National health statistics can-not be used to estimate cancer risk because the denomina-tor (ie, the total number of persons with Barretts esophagusin the general population) is not known. Nevertheless, areasonable estimate of the incidence of cancer in Barrettsesophagus is needed to formulate rational managementstrategies for patients with this condition.

A number of systematic reviews of studies on theincidence of cancer in patients with Barretts esophagus

AGA

March 2011 AGA e21


5/35

have been published.15,2224 A recent review describedoutcomes for 47 studies that met inclusion criteria froman initial search that yielded 7780 publications fromMEDLINE and EMBASE databases (years 19502006).24

The overall estimate of cancer incidence was based on

209 cancers discovered in 11,279 patients with Barrettsesophagus who were followed up for 47,496 person-years.The average incidence of cancer was 6.1 per 1000 person-years, but this estimate may have been spuriously highbecause it included patients in whom cancer was discov-ered within the first year after the diagnosis of Barrettsesophagus. Such patients likely had prevalent cancersthat were missed on their initial endoscopic examina-tions. After adjusting the results to exclude those pa-tients, the incidence of cancer was 5.3 per 1000 person-years (0.5% per year), an estimate well aligned with theresults of prior systematic reviews.

Among 4 studies published since the aforementionedsystematic review was a report of a cohort of 502 patientswith Barretts esophagus from Leeds in the United King-dom.25 The annual incidence of cancer among patientswho had Barretts esophagus with specialized intestinalmetaplasia was 16 per 1000 person-years (1.6% per year). Another single-center cohort from Birmingham in theUnited Kingdom calculated an incidence of 3.1 per 1000person-years (0.3% per year) based on 3 cancers diagnosedamong 188 patients with Barretts esophagus.26 A thirdstudy from the United Kingdom reported statistics froma multicenter national registry involving 738 patientswith a combined follow-up of 3697 years. The overallannual incidence of esophageal adenocarcinoma was0.5% (95% confidence interval [CI], 0.30.8).27 Finally, anendoscopic and pathology-based database in Spain wasretrospectively examined to calculate a cancer risk inpatients with Barretts esophagus of 5.2 per 1000 person-years (0.5% per year).28

The risk of cancer in Barretts esophagus appears tovary with the extent of esophageal metaplasia; therefore,patients with long-segment disease may have a higherincidence of adenocarcinoma than those with short-seg-ment Barretts esophagus.23 In the aforementioned Span-ish cohort, for example, the annual risk of esophageal

adenocarcinoma was 0.57% for patients with long-seg-ment Barretts esophagus compared with only 0.26% forpatients with short-segment disease.28 The risk of cancerdevelopment also is lower in women than in men withBarretts esophagus (4.5 vs 10.2 cancers per 1000 person-years).24

The precise risk of cancer in Barretts esophagus re-mains unclear. Reported estimates of cancer risk con-tinue to vary widely, and it is not clear how biases,statistical aberrations, and regional differences in inci-dence rates contribute to the disparities among the pub-lished reports. Methodologically, larger studies (ie, 500 or

more person-years of observation) report lower cancerincidence rates than smaller studies, as do studies that

represent population-based as opposed to referral-basedcohorts.24 Estimates of cancer risk also are likely to beaffected by the proportion of patients with short-seg-ment Barretts esophagus and women in the cohort. Withthese limitations in mind, most modern systematic re-

views and large series suggest that the annual incidenceof cancer for the general population of patients withBarretts esophagus is approximately 0.5% per year. Forpatients with dysplasia in Barretts epithelium, the risk ofcancer is substantially higher (see the following text).

Does Barretts Esophagus Affect LifeExpectancy? How Does a Diagnosis ofBarretts Esophagus Affect Quality ofLife?

Does Barretts Esophagus Affect LifeExpectancy?

A review of the literature reveals some contradic-tory results for studies assessing the impact of a diagno-sis of Barretts esophagus on life expectancy. In a popu-lation-based study in Northern Ireland, Anderson et alcompared mortality rates for subjects with Barrettsesophagus with those for age- and sex-matched subjectsin the general population. The investigators found nosignificant differences in overall mortality rates betweenthe 2 groups.29 Although deaths from esophageal cancerwere more common in the group with Barretts esopha-gus, the total frequency of such deaths was so low that ithad little effect on overall mortality. Another study thatcompared survival for subjects with Barretts esophaguswith survival for 2 control groups (the general popula-tion and patients with Schatzkis rings) also found nodifference in life expectancy among the groups.30 In con-trast, Moayyedi et al in the United Kingdom found in-creased mortality for subjects who had Barretts esopha-gus diagnosed at 4 hospitals in Leicestershire comparedwith age- and sex-matched subjects in the general popu-lation.31 Interestingly, however, the excess mortality inthe patients with Barretts esophagus was primarily dueto extraesophageal diseases such as bronchopneumoniaand ischemic heart disease. In another large population-based study in which survival for a cohort of 1677 pa-

tients with Barretts esophagus was compared with anage- and sex-matched cohort of 13,416 individuals in thegeneral population, Solaymani-Dodaran et al found thatthe patients with Barretts esophagus had a 37% increasein mortality.32Approximately 45% of the excess mortalityin the patients with Barretts esophagus was due toesophageal cancer, whereas 55% was due to extraesopha-geal disorders such as cardiovascular disease. The excessdeaths from cardiovascular problems may be related tothe association of Barretts esophagus with obesity,which also is a risk factor for cardiovascular disease.

Despite the somewhat disparate findings of these stud-

ies on survival for patients with Barretts esophagus,several conclusions are warranted. First, even though

AGA



6/35

deaths from esophageal adenocarcinoma are more com-

mon in patients with Barretts esophagus than in indi-

viduals without that condition, adenocarcinoma remains

an uncommon cause of mortality in patients with Bar-retts esophagus nevertheless.33,34 Furthermore, because

this cancer generally occurs later in life,35 the impact ofsuch a death on the mean survival of the cohort is

lessened, because deaths due to extraesophageal diseases

such as cardiovascular disorders are far more common

and therefore drive overall mortality rates. It also appears

that mortality due to cardiovascular disease may be in-

creased in patients with Barretts esophagus, perhapsbecause of the association of the disorder with obesity.

How Does a Diagnosis of Barretts EsophagusAffect Quality of Life?

Numerous studies have attempted to quantify the

quality of life in patients with Barretts esophagus, but

those studies are limited in several important ways. First,

they lump all subjects with prevalent disease into a singlecategory and compare that group with controls such as

subjects who have GERD without Barretts esophagus or

the general population. It is likely that the quality of life

for patients with Barretts esophagus varies with a num-

ber of important factors, such as disease duration and the

number of surveillance endoscopies performed, that arenot accounted for in such studies. Second, the studies

generally use a convenience sample of subjects with Bar-

retts esophagus attending outpatient clinics or endos-

copy units at tertiary care centers. The impact of the

disease on such subjects may be very different from that

on the general population of individuals with Barrettsesophagus. For instance, one might expect subjects at-

tending repeated endoscopic surveillance sessions to be

more concerned about their risk of developing adenocar-

cinoma than those who forego such measures. Finally,

there is no validated, disease-specific, widely accepted

quality-of-life measure for Barretts esophagus. Investiga-tors have used generic quality-of-life measures as well as

organ-specific measures and tools developed specifically

for GERD populations. Those tools may fail to captureimportant domains of quality of life for subjects with

Barretts esophagus.

With these limitations in mind, a systematic reviewfound 25 articles in the English-language literature that

provide quantitative assessments of quality of life or

financial and psychological burdens of disease for sub-

jects with Barretts esophagus.36 Five studies assessedpatients using the generic quality-of-life measure 36-Item

Short Form Health Survey (SF-36),37 3 studies used Qual-

ity of Life in Reflux and Dyspepsia (QOLRAD),38 and 2

studies used Gastrointestinal Quality of Life Index.39

Four studies evaluated utility measures in patients withBarretts esophagus. Utilities rate the desirability of living

with a given disease state on a scale of 0 to 1, where arating of 1 indicates essentially no decrease in the desir-ability of life and 0 indicates a quality of life so undesir-able it is equal to death. Eight studies did not use tradi-tional quality-of-life instruments but quantitatively

assessed impact (psychological, financial, social, and soon) or burden of disease by other measures. of these 25studies, 9 included patients with Barretts esophagus as asubset of patients with GERD symptoms or a prioricompared patients with Barretts esophagus with pa-tients with GERD. The remaining 16 studies includedonly patients with Barretts esophagus.

On balance, these studies show that a diagnosis ofBarretts esophagus has a substantial negative impact onquality of life. All reported cohorts showed lower SF-36scores in subjects with Barretts esophagus comparedwith the population norms. Studies using organ-specific

measures such as QOLRAD and Gastrointestinal Qualityof Life Index also showed diminished quality of life insubjects with Barretts esophagus compared with thepopulation norms. One study found similar QOLRADscores in patients with Barretts esophagus and patientswith GERD. Whereas both groups reported substantialGERD symptoms, it is possible that a major componentof the decrease in quality of life experienced by thepatients with Barretts esophagus was due to their GERDsymptoms. Attempts to quantify quality of life in Bar-retts esophagus using health state utilities have repeat-edly shown diminished utility for life with this condition.

The negative impact on utility varies with the degree ofdysplasia in Barretts epithelium and has been reportedto be as low as 0.77 for patients with high-grade dyspla-sia.40

Although subjects with Barretts esophagus are consis-tently found to have a poorer quality of life than thegeneral population, it is unclear to what extent this isattributable to anxiety regarding cancer risk, discomfortdue to GERD symptoms, or other factors. Patients withBarretts esophagus appear to greatly overestimate theircancer risk, and this overestimation is associated with anincrease in their utilization of health care.41 In addition

to this psychological distress, patients with Barrettsesophagus face higher individual costs for life insuranceand may be unable to obtain health insurance.42

In summary, by generic and organ-specific quality-of-life measures, subjects with Barretts esophagus repeat-edly have been shown to have substantially lower scoresthan population norms. Subjects with Barretts esopha-gus consistently report their utility of living with thedisease lower than without it, and the decrease in utilitycorrelates with the degree of dysplasia in Barretts epithe-lium. A diagnosis of Barretts esophagus appears to causepsychological stress and may be associated with substan-

tial, but incompletely understood, additional costs suchas increased life and health insurance premiums.

AGA

March 2011 AGA e23


7/35

Who Is at Risk for BarrettsEsophagus? Who Should Be Screenedfor Barretts Esophagus?

Who Is at Risk for Barretts Esophagus?

A systematic review of original literature on theepidemiology of Barretts esophagus shows that mostpublished studies describe hospital-based or endoscopyclinicbased cohorts; relatively few describe large popu-lation-based cohorts. It is important to consider thispotential bias when assessing those epidemiologic stud-ies, which have identified a number of risk factors forBarretts esophagus. Selected risk factors are reviewedbriefly in the following text.

It is not clear when Barretts esophagus develops, butmost recognized cases are diagnosed in the sixth decade oflife or later.43,44Although the utilization of endoscopy (andhence the chance of finding Barretts esophagus) is higher in

older subjects, cohort studies suggest that, as age increases,so does the likelihood that a subject with GERD symptomswill have Barretts esophagus.43 Whether the condition re-sults from a single catastrophic insult to the esophagealmucosa, which may be more likely to occur in older sub-jects, or whether Barretts esophagus is the cumulative re-sult of years of reflux-induced damage is not clear. Longi-tudinal studies, which have found that the extent ofBarretts metaplasia does not increase with time (at least insubjects on therapy), provide some support for the conceptthat the condition develops all at once.45

There is a male predominance for Barretts esophagus, as

there is for esophageal adenocarcinoma. In case-control andcohort studies, the risk of Barretts esophagus among menwith GERD symptoms is 1.5- to 3-fold higher than that ofwomen.44,4648 A recent meta-analysis of cohort studiescomprising consecutively enrolled subjects with Barrettsesophagus shows a case mix of men and women of approx-imately 2:1.49 Whether this male preponderance is the resultof differences between men and women in hormonal effectson the esophagus, body fat distribution, or other as-yetunidentified factors is not clear.

Most cohort studies show that the majority of subjectswith Barretts esophagus are white.48 Because most of

these studies are clinic or hospital based, some of theapparent ethnic predisposition may be biased by theunderlying demographics of the patients who attendthose facilities. Several studies have attempted to quan-tify how the proportion of subjects with heartburn whohave Barretts esophagus varies among different ethnicgroups. Abrams et al found that, among subjects whounderwent endoscopy at their institution, white subjectswere approximately 4 times as likely to have Barrettsesophagus as Hispanic or black subjects.43 Similarly, acohort study of subjects presenting for screening colono-scopy who were invited to undergo upper endoscopy

found that white subjects were more likely to have Bar-retts esophagus than black subjects.50 In a community-

based study, Corley et al found that the incidence ofBarretts esophagus in non-Hispanic white subjects wasmore than 6 times greater than that in black subjects.48

In contrast, Eloubeidi et al did not find race to be apredictor of Barretts esophagus in a Veterans Adminis-

tration population.51 If there are significant differences inthe prevalence of Barretts esophagus between white andblack subjects, they are not likely the result of differencesin the underlying prevalence of GERD, because the dis-tribution of heartburn symptoms appears to be similarbetween those 2 groups.52 However, limited studies sug-gest that both GERD and Barretts esophagus are far lesscommon in Asian patients than in white subjects.53

GERD is strongly associated with Barretts esophagus.Case-control studies suggest that subjects with heartburnare 6 to 10 times more likely to have Barretts esophagusthan those without heartburn. Furthermore, there appears

to be a dose-response relationship in that subjects withmore frequent or chronic GERD symptoms are more likelyto have Barretts esophagus.51,5457 Hiatal hernia also isassociated with the presence of Barretts esophagus.46,55

However, the relationship between GERD symptom severityand Barretts esophagus is not as strong.51

Cohort and case-control studies assessing esophagealacid exposure in subjects who have GERD with andwithout Barretts esophagus showed that those with Bar-retts esophagus have, on average, greater acid exposurethan those without and that the extent of Barretts meta-plasia correlates directly with the duration of esophagealacid exposure.55 Increased bile reflux (as estimated by asystem that measures esophageal exposure to bilirubin)also has been documented in subjects with Barrettsesophagus,46,58 but the role of components of refluxateother than acid in the development of the conditionremains controversial.

A high body mass index and an intra-abdominal dis-tribution of body fat have been shown to be strong riskfactors for Barretts esophagus. Patients with Barrettsesophagus have, on average, a higher body mass indexthan either patients with GERD without Barretts esoph-agus or general population controls.5961 Interestingly,the distribution of body fat may be the key to this

association. For any given body mass index, subjects withhigher amounts of intra-abdominal obesity, manifest ei-ther radiographically or by increased waist-to-hip ratiomeasures, appear to have an increased risk of Barrettsesophagus.59,60,62 In fact, in a recent analysis of bodyanthropometry in subjects with Barretts esophagus,body mass index was no longer an independent predictorof the disorder once waist-to-hip ratio was factored.62

Whether the increased risk associated with intra-abdom-inal obesity is due to mechanical or hormonal factors ora consequence of yet-undescribed factors is not known.

Alcohol and smoking are not nearly as strongly asso-

ciated with Barretts esophagus as they are with squa-mous cell cancer of the esophagus, and studies on the

AGA



8/35

association of these habits with Barretts esophagus haveyielded inconsistent results. In a population-based study,Ronkainen et al showed that subjects who used tobaccoor alcohol were approximately 3 times as likely to haveBarretts esophagus as subjects who did not.63 Other

studies have found an association between either smok-ing44 or alcohol64 and Barretts esophagus, whereas anumber of investigations have not.6567 In fact, recentdata suggest that consumption of wine actually mayprotect against Barretts esophagus.65,66 High vegetableand fruit intake appear to diminish the risk of Barrettsesophagus, although the mechanism is not known.68,69

In summary, well-established risk factors for Barrettsesophagus include advanced age, male sex, white race,GERD, hiatal hernia, elevated body mass index, and apredominantly intra-abdominal distribution of body fat.Moderate consumption of wine and a diet high in fruits

and vegetables may protect against the disorder.

Who Should Be Screened for BarrettsEsophagus?

Despite the considerable published data availableon risk factors for Barretts esophagus, few attempts havebeen made to apply this information systematically in thedesign of guidelines on who to screen for the condition.Furthermore, despite the dearth of studies showing clin-ical benefit resulting from endoscopic screening for Bar-retts esophagus, the practice remains widespread amongclinicians in the United States.70 Professional organiza-

tions are divided on whether to recommend endoscopicscreening for Barretts esophagus, however, with somesuggesting that the practice may be appropriate71 andothers not endorsing it routinely.72,73

Most recommendations on screening for Barrettsesophagus have focused on subjects with chronic GERDsymptoms, because GERD was one of the first and stron-gest risk factors for Barretts esophagus identified andbecause chronic esophageal inflammation due to GERDhas been thought to contribute to esophageal carcino-genesis. Several cohort and case-control studies suggestthat endoscopic screening and surveillance for Barretts

esophagus can have beneficial effects.74,75

Subjects whohave esophageal adenocarcinoma discovered as the resultof an endoscopic screening or surveillance program forBarretts esophagus present with earlier-stage tumors, areless likely to have lymph node involvement, and havebetter short-term life expectancies than patients whopresent with symptoms of esophageal cancer such asdysphagia and weight loss. Cost-effectiveness analysessuggest that endoscopic screening may be cost-effective ifcertain predefined clinical parameters are met.76,77

Although these data may seem compelling, several con-ceptual and logistical difficulties diminish the utility of

screening endoscopy as it is currently practiced in theUnited States. First and foremost, approximately 40% of

subjects who have adenocarcinoma of the esophagusreport no history of chronic GERD symptoms.78 UsingGERD symptoms as an entrance criterion to endoscopicscreening programs immediately excludes those subjects,decreasing by almost one-half our ability either to pre-

vent the cancer or to detect it at an early, presymptomaticstage. Second, even among subjects with GERD symptoms,the risk of adenocarcinoma is very low in absolute terms.Studies show that up to 40% of the adult US populationexperience GERD symptoms on a monthly basis and 20%on a weekly basis.79 Although the incidence of adenocarci-noma of the esophagus has increased 6-fold in the past 3decades,80 fewer than 10,000 Americans develop esophagealadenocarcinoma each year, representing a minute fractionof the total number of individuals with GERD symptoms.Even among patients with Barretts esophagus, cohort stud-ies show that more than 90% never develop esophageal

adenocarcinoma.81 Therefore, the vast majority of individ-uals who undergo endoscopic screening for Barretts esoph-agus based on the presence of GERD symptoms will notbenefit from the procedure.

There also are substantial problems with the executionof various facets of endoscopic screening and surveillanceprograms that diminish their effectiveness. These includedifficulties in the endoscopic recognition of importantlesions,82 the random nature of esophageal biopsy sam-pling that is subject to considerable sampling error,83 anddisagreement among pathologists in the histologic inter-pretation of the esophageal biopsy specimens.84 Finally,

endoscopic examinations are expensive, especially afterfactoring in costs not only for the endoscopy but also forthe tissue acquisition and interpretation.

No direct evidence substantiates the utility of endo-scopic screening for Barretts esophagus, and substantialconfounding factors such as lead time and length biascompromise the validity of the observational studies,suggesting that the practice is beneficial.85 Therefore,inadequate evidence exists to endorse endoscopic screen-ing for Barretts esophagus based solely on the presenceof GERD symptoms, and decisions on when to recom-mend endoscopic screening should continue to be indi-

vidualized. It is incumbent on the practitioner to ensurethat patients who elect to undergo endoscopic screeningfor Barretts esophagus understand not only the putativeadvantages of the procedure but also the substantialshortcomings and possible negative effects, which in-clude the expense and risks of the endoscopy and of theinvasive procedures that might be recommended to treatlesions found by endoscopy as well as the potential ad-verse impact on quality of life (see the previous text).Whether the development of new endoscopic technolo-gies that are more sensitive, less expensive, and easier toperform or the availability of improved endoscopic treat-

ments for Barretts esophagus eventually will tip thescales in favor of screening is not yet known.

AGA

March 2011 AGA e25


9/35

What Is the Natural History ofDysplasia in Barretts Esophagus?

During the process of carcinogenesis in Barrettsesophagus, some of the genetic alterations that endowcells with growth advantages (eg, activation of oncogenes,

inactivation of tumor suppressor genes) also cause mor-phologic changes in the tissue that the pathologist rec-ognizes as dysplasia (also called intraepithelial neoplasia).Thus, dysplasia can be viewed as the histologic expressionof genetic alterations that favor unregulated cellgrowth.86 Dysplasia can be categorized as low grade orhigh grade depending on the degree of histologic abnor-malities. Pathologists can have difficulty distinguishinglow-grade dysplasia in Barretts esophagus from reactivechanges caused by reflux esophagitis; consequently, in-terobserver agreement for the diagnosis of low-grade dys-plasia may be poor.

Dysplasia in Barretts esophagus sometimes causes noendoscopically apparent abnormalities, and dysplasia canbe patchy both in its extent and severity. These factorscontribute to the substantial problem of biopsy samplingerror in identifying dysplasia. Endoscopists traditionallyhave used a 4-quadrant biopsy sampling system (which isessentially a random sampling technique) to find dyspla-sia in Barretts esophagus, and it is clear that this systemcan miss areas of dysplasia and even cancer. In series ofpatients who underwent esophagectomies because endo-scopic examination revealed high-grade dysplasia in Bar-retts esophagus, for example, a number of studies havefound that invasive cancer is present in 30% to 40% of theresected esophagi.87 However, a recent critical review ofthose studies suggests that 13% is a more accurate esti-mate of the frequency of invasive cancer in this situation,and when a careful endoscopic examination excludes allvisible lesions, the frequency of finding invasive cancer atesophagectomy is only 3%.88

A meta-analysis published in 2008 focused on theincidence of esophageal adenocarcinoma in patients withhigh-grade dysplasia in Barretts esophagus.89 The litera-ture search of MEDLINE and associated sources yielded3843 citations, of which 196 were deemed potentiallyrelevant; on complete review, however, only 4 articles met

the inclusion criteria for the study (ie, study patients hadhistologically confirmed Barretts esophagus with high-grade dysplasia, no prevalent cancer, and no ablative orsurgical therapy). The crude incidence of esophageal ad-enocarcinoma among patients with high-grade dysplasiawas 55.7 per 1000 person-years (5.6% per year). Usingdifferent weighting algorithms, the incidence increasedto 65.8 per 1000 person-years (6.6% per year) with 95%CIs of 49.9 to 84.6 (Poisson) or 49.7 to 81.8 (binomial)per 1000 person-years of observation.

The incidence of cancer in patients who have low-gradedysplasia in Barretts esophagus is especially poorly de-

fined. Some studies have found a risk of cancer no greaterthan that for the entire population of patients with

Barretts esophagus,9092 whereas others have observedconsiderably higher rates.84,93 One reason for these dis-parities may be the poor interobserver correlation amongpathologists in the diagnosis of low-grade dysplasia.94

This possibility is supported by the observation that,

among patients whose diagnosis of low-grade dysplasia isconfirmed by 2 or more pathologists, the incidence ofcancer is substantially higher than that for patientswhose pathologists disagree on the diagnosis.84,93 Theextent of low-grade dysplasia defined as the proportionof crypts that exhibit dysplastic changes has been sug-gested to improve the ability to discern which patientsare at greater risk for development of cancer.95

Another unresolved issue is whether dysplasia can re-gress or whether the inability to demonstrate dysplasiaon follow-up endoscopic examinations is merely the re-sult of biopsy sampling error. In one of the largest mul-

ticenter longitudinal studies, 42% of patients initiallydiagnosed with low-grade dysplasia had no dysplasiafound on subsequent endoscopic examinations and anadditional 32% had low-grade dysplasia found only in-termittently during their course of surveillance.92 In thisstudy, the calculated incidence of esophageal adenocar-cinoma among patients with low-grade dysplasia was0.6% per year (95% CI, 0%2%), a rate of cancer develop-ment similar to that reported for the general populationof patients with Barretts esophagus. On the other hand,3 patients in whom cancer developed did not have dys-plasia diagnosed during prior endoscopic surveillanceexaminations. It is not clear whether those cancers devel-

oped de novo, without first manifesting dysplasia, orwhether the preceding dysplasia merely was missed dueto biopsy sampling error. Other reports also have de-scribed the apparent regression of dysplasia in the ab-sence of ablation.22

Does Endoscopic Surveillance ImproveSurvival for Patients With BarrettsEsophagus?

Endoscopic surveillance has been proposed forpatients with Barretts esophagus with the unproved as-sumption that the practice will reduce deaths from

esophageal adenocarcinoma and thereby prolong survival. Societal guidelines generally have recommendedendoscopic surveillance for patients with Barretts esoph-agus at intervals that vary with grade of dysplasia foundin the metaplastic epithelium. Intervals of 3 to 5 yearshave been suggested for patients who have no dysplasia,6 to 12 months for those found to have low-grade dys-plasia, and every 3 months for patients with high-gradedysplasia who receive no invasive therapy.72

During endoscopic surveillance, the endoscopist at-tempts to identify esophageal neoplasia in an early, cur-able stage, usually in the form of dysplasia. To find

dysplasia, endoscopists generally have relied on a system-atic, 4-quadrant biopsy sampling technique designed

AGA



10/35

with the intent of maximizing the chance for identifyingan inconspicuous lesion that may be randomly distrib-uted throughout the Barretts epithelium. The Seattleprotocol calls for obtaining such 4-quadrant biopsyspecimens at intervals of every 1 to 2 cm throughout the

columnar-lined esophagus. In addition, areas of mucosalirregularity (eg, nodules, masses, ulceration), which areespecially likely to be associated with dysplasia, are sam-pled separately.

A prospective study showed a significant increase inthe number of cases of high-grade dysplasia and invasivecancer detected after institution of a rigorous endoscopicsurveillance protocol like the one described previously.96

In the community, however, surveillance often is notperformed in this rigorous manner. A retrospective studyof endoscopic and pathology reports from 15 hospitals inThe Netherlands revealed that adherence to the Seattle

protocol was good (79%) for cases in which Barrettsmetaplasia involved only up to 5 cm of the distal esoph-agus but diminished with increasing extent of metaplasiato the point that there was only 30% adherence amongcases with metaplasia involving 10 to 15 cm of the esoph-agus.97 An American study using a large pathology data-base maintained by Caris Diagnostics identified 2245patients who had esophageal biopsy specimens taken forevaluation of Barretts esophagus and who had endos-copy reports that documented the extent of esophagealcolumnar lining.98 Overall, adherence to the Seattle pro-tocol was found in only 51% of cases. As in the previousstudy, adherence to the protocol varied inversely with theextent of Barretts metaplasia. In addition, failure toadhere to the protocol was associated with a significantlydecreased rate of detecting dysplasia (summary odds ra-tio, 0.53; 95% CI, 0.350.82). These studies suggest that,although adherence to recommended surveillance proce-dures is associated with higher rates of detection ofdysplasia, many practicing gastroenterologists do not ad-here to those guidelines and adherence appears to bepoorest for the population at highest risk for develop-ment of cancer (ie, patients with extensive Barretts meta-plasia).

Streitz et al studied 77 patients who they treated for

esophageal adenocarcinoma to explore whether prior en-doscopic surveillance was associated with better survival(Table 3).99 In 19 patients, the cancers were found duringsurveillance endoscopies performed because Barrettsesophagus had been discovered 8 to 120 months earlier(median, 24 months). The remaining 58 patients pre-sented to the hospital with symptoms of esophagealcancer, and Barretts esophagus was first diagnosed whentheir tumors were resected. Compared with the latterpatients, the patients whose tumors were discovered dur-ing endoscopic surveillance had cancers in significantlylower stages and had a significantly better 5-year actuarial

survival rate (62% vs 20%). It should be noted that 9patients in the surveillance group had esophagectomy

performed for carcinoma in situ or high-grade dysplasia,although 2 of these had invasive carcinoma found in theresected specimen.

Peters et al reported the results of a similar studycomparing outcomes for 17 patients who had esophagec-

tomy performed for high-grade dysplasia or adenocarci-noma discovered during endoscopic surveillance withthose for 35 patients who had esophagectomy for ade-nocarcinomas discovered outside a surveillance pro-gram.100 Although the endoscopic surveillance protocolwas not standardized, all 17 patients had a diagnosis ofBarretts esophagus without cancer established at least 6months before esophagectomy. Five of the 9 patients whounderwent resection for high-grade dysplasia were foundto have invasive cancer in the resected specimen. As in theprevious study, the patients whose tumors were discov-ered during surveillance had cancers in significantly

lower stages and longer survivals.In another single-center experience, Van Sandick et alalso compared outcomes for patients with adenocarci-noma of the esophagus or esophagogastric junction dis-covered in and outside endoscopic surveillance pro-grams.75 The 16 patients in the surveillance group wereknown to have had Barretts esophagus for a medianduration of 42 months, and they had been under endo-scopic surveillance for intervals ranging from 2 monthsto 2.5 years (median, 10 months). Five of those 16 pa-tients had esophagectomy for high-grade dysplasia; 1 wasfound to have invasive cancer, 2 were found to haveintramucosal carcinoma, and 2 had only high-grade dys-plasia found in the resected esophagus. Compared withthe 54 patients whose tumors were discovered outside asurveillance program, the pathologic stage of cancer wassignificantly lower for patients in the surveillance group,and their 2-year survival was significantly better (85.9% vs43.3%; P .0029). Surveillance was associated with sig-nificantly longer survival even when patients with a pre-operative diagnosis of high-grade dysplasia (without can-cer) were excluded from the analysis.

Fountoulakis et al studied a cohort of consecutivepatients who underwent esophagectomy for high-gradedysplasia or adenocarcinoma in Barretts esophagus,

comparing the survival of those whose neoplasms werediscovered in and outside a standardized endoscopic sur-veillance program.101 The surveillance group included 17patients who had at least one endoscopy performed atleast 6 months after their first diagnosis of Barrettsesophagus, whereas the no-surveillance group comprised74 patients who first presented to the hospital withadenocarcinoma. Overall survival was significantly longerfor patients in the surveillance group, with 1- and 3- yearsurvival rates for the surveillance and no-surveillancegroups of 88% versus 67% and 80% versus 31%, respec-tively.

Using the Northern California Kaiser Permanente can-cer registry, Corley et al studied outcomes for 589 pa-

AGA

March 2011 AGA e27


11/35

tients with adenocarcinoma of the esophagus or gastriccardia diagnosed from 1990 to 1998.74 Among only 23patients who were known to have had Barretts esopha-gus for at least 6 months before the cancer was diag-nosed, 15 had their tumors discovered during surveil-

lance endoscopy, whereas 8 had their cancers detectedduring endoscopy performed because of cancer symp-toms. As in the single-center observational studies dis-cussed previously, the patients whose tumors were dis-covered during surveillance had cancers at significantlylower stages and had significantly better 2-year survivalcompared with the patients whose tumors were discov-ered because they were symptomatic (73.3% vs 12.5%; P.02).

Cooper et al studied a cohort of 1633 patients in theSurveillance, Epidemiology and End Results (SEER) da-tabase who had adenocarcinoma of the esophagus or

gastric cardia.102

Through linkage with Medicare claimsdata, the investigators were able to identify those whohad endoscopy performed more than 1 year before thecancer was diagnosed. A record of prior endoscopy was

found in only 9.7% of the patients, and only 3.7% hadBarretts esophagus identified at least 1 year before thediagnosis of cancer. However, a prior diagnosis of Bar-retts esophagus was associated with a lower cancer stageand a higher probability of treatment by esophagectomy.

The median survival for patients with esophageal adeno-carcinoma who had a prior endoscopy was 7 months,compared with only 5 months for those who had noprior endoscopy (P .01). The association between priorendoscopy and prolonged survival remained significanteven after adjusting for age at diagnosis, sex, race, andnumber of comorbidities using Cox proportional hazardsmodeling (relative hazards, 0.73; 95% CI, 0.570.93).

A more recent analysis of the SEER-Medicare dataconducted by Cooper et al included 2754 patients with anew diagnosis of esophageal adenocarcinoma.103 Havingan endoscopy performed 3 years to 6 months before the

diagnosis of cancer was associated with an improvementin median survival from 7 months (for patients with noprior endoscopy) to 11 months, yielding a significantreduction in the hazard ratio for death. Additional inde-

Table 3. Studies Comparing Survival Based on Prior Endoscopy

No. of

subjects

Prior endoscopy No prior endoscopy

ReferenceSubjects Survival Subjects Survival Significance

77 11 stage 0I (5 HGD,

4 CIS, 2 stage I), 4

stage II, 3 stage III,

0 stage IV

62% 5-year survival 9 stage I, 15 stage II, 25

stage III, 4 stage IV

20% 5-year survival P .007 99

52 17: 4 HGD only, 9

intramucosal, 2

submucosal, 1

muscularis propria,

1 transmural

Approximate 85%

2-year survival

35: 10 stage 0I, 14

stage II, 11 stage IIIIV

Approximate 40%

2-year survival

log rank, 5.8;

P .05

100

70 16: 4 HGD only, 3

intramucosal, 5

submucosal, 1

muscularis propria,

3 transmural

85.9% 2-year

survival

54: 10 stage 0I, 14


43.3% 2-year

survival

P .0029 75

23 15: 1 stage 0, 8

stage I, 4 stage

2a, 2 stage 2b

73.3% 2-year

survival

8: 2 stage 1, 1 stage 2a,

2 stage 2b, 1 stage 3,

2 stage 4

12.5% 2-year

survival

P .02 74

1256 119: 9 in situ, 59

local, 29 regional,22 distant

7-month median

survival

1137: 14 in situ, 337

local, 373 regional,413 distant

5-month median

survival

P .01 102

2754 157 in situ or local,

160 regional/

distant/unstaged

11-month median

survival

690 in situ or local,

1747 regional/distant/

unstaged

7-month median

survival

P .001 103

91 17: 1 HGD, 13 stage

0I, 3 stage II, 1

stage IIIIV

80% 3-year survival 74: 11 stage 01, 26


31% 3-year survival P .008 101

155 25: 7 stage I, 11


4 stage IV

10% 5-year survival 130: 16 stage I, 46


38 stage IV

10% 5-year survival hazard ratio,

0.82 (0.521.29)

105

245 Not applicablea Odds ratio,

0.66 (0.450.96)

Not applicable Not applicable P .03 104

CIS, carcinoma in situ; HGD, high-grade dysplasia.a

Case-control study of 245 incident cases of death from esophageal/cardia adenocarcinoma compared with 980 controls with reflux, matchedfor age, sex, and race. Exposure of interest was esophagogastroduodenoscopy.

AGA



12/35

pendent factors associated with improved survival in-cluded a prior diagnosis of Barretts esophagus and re-ceipt of therapy, including surgery, radiation, orchemotherapy. of note, however, only 11.5% of the pa-tients with cancer were found to have had a prior endos-

copy, and Barretts esophagus was identified in only 8.1%of patients before their diagnosis of cancer.

In a case-control study using the Veterans Affairs da-tabase, Kearney et al found that a group of 245 patientswho had GERD and adenocarcinoma of the esophagus orgastric cardia were significantly less likely to have under-gone endoscopy in the 1 to 8 years before the index datethan 980 control subjects (matched by age, sex, and race)who had GERD without cancer (adjusted odds ratio,0.66; 95% CI, 0.450.96).104 Furthermore, a dose effectwas noted whereby endoscopy performed 2 to 4 yearsbefore the index date was more protective against cancer

death than endoscopy performed more than 4 years be-fore that date.

Countering these positive results is a retrospective, con-trolled, cohort study by Rubenstein et al that used theVeterans Affairs national administrative databases to iden-tify 155 subjects who had GERD associated with adenocar-cinoma of the esophagus or esophagogastric junction.105

Although the 25 patients who had endoscopy performed 1to 5 years before their diagnosis of cancer had tumors in alower stage than the 130 patients who had no endoscopyduring that same period, there was no significant differencein survival between the 2 groups (adjusted hazard ratio,

0.93; 95% CI, 0.581.50). Analysis of a subset of subjects forwhom complete endoscopic and histopathologic data wereavailable revealed that patients who had endoscopic surveil-lance performed according to guidelines proposed by the American College of Gastroenterology had better survivalthan those who did not. After adjustment for potentialconfounders, however, the improvement in survival did notachieve statistical significance. This study highlights thepotential for lead- and length-time biases that can exagger-ate the benefits of surveillance programs in observationalstudies.

In summary, the evidence to support endoscopic sur-

veillance as a means to improve survival for patients whodevelop neoplasia in Barretts esophagus relies on admin-istrative data that have been examined retrospectively.The preponderance of this evidence suggests that endo-scopic surveillance can reduce mortality from esophagealadenocarcinoma through the early detection of treatablecancers. However, such observational studies are prone toa number of biases, such as lead- and length-time biases,that could well exaggerate the benefits of surveillanceprograms. It remains unclear whether endoscopic surveil-lance is beneficial at all; consequently, it is not possible tomake meaningful recommendations regarding the opti-

mal intervals between endoscopic procedures or the op-timal surveillance biopsy procedures.

Can Biomarkers Be Used to Confirmthe Histologic Diagnosis of Dysplasia?Can Biomarkers Be Used Instead ofDysplasia for Risk Stratification inBarretts Esophagus?

Can Biomarkers Be Used to Confirm the Histologic Diagnosis of Dysplasia?

Presently, cancer risk stratification for patientswith Barretts esophagus is based primarily on the histo-logic finding of dysplasia. However, dysplasia is a veryimperfect predictor of cancer risk for a number of rea-sons, including poor interobserver agreement among pa-thologists in distinguishing dysplasia from reactive epithe-lial changes (caused by reflux esophagitis) and in gradingthe severity of dysplastic change.84,106,107 Immunostainingfor p53 has been proposed as an adjunct to the diagnosisof dysplasia, but the utility of this technique is limited by

confounding factors such as high rates of false-positivestaining and staining characteristics that vary with thetype of p53 antibody used.108 Immunostaining for-methylacyl-CoA racemase (AMACR) and for a panel ofbiomarkers that includes -catenin, cyclin D1, and p53also has shown promise in preliminary studies for distin-guishing dysplasia from reactive changes and for distin-guishing among grades of dysplasia.109111 In contrast tothose promising reports, a study that attempted to cor-relate grades of dysplasia with messenger RNA expressionlevels for a panel of 10 genes believed to contribute tocarcinogenesis in Barretts esophagus found that the util-ity of the panel was severely limited by significant inter-patient and intrapatient variations in gene expressionlevels.112 At this time, data supporting the use of bio-markers to confirm the histologic diagnosis of dysplasiamust be considered preliminary, and biomarkers cannotyet be recommended for this purpose for routine clinicalpractice.

Can Biomarkers Be Used Instead of Dysplasiafor Risk Stratification in Barretts Esophagus?

A number of biomarkers other than the histologicfinding of dysplasia have been proposed to predict therisk of neoplastic progression and, hence, the need for

endoscopic surveillance or more invasive treatments inpatients with Barretts esophagus. In general, these puta-tive biomarkers reflect DNA abnormalities, acquired dur-ing the process of carcinogenesis, that either endow thecells with growth advantages directly or favor the devel-opment of growth-promoting mutations. Most of theproposed biomarkers for Barretts esophagus have beenevaluated in cross-sectional studies only, and no bio-marker yet has been validated in prospective, controlledclinical trials. However, some biomarkers have been eval-uated in studies in which biopsy specimens of Barrettsmetaplasia were collected in a prospective systematic

fashion, but the biomarker analyses were performed ret-rospectively. Promising biomarkers that have been so

AGA

March 2011 AGA e29


13/35

evaluated include aneuploidy/tetraploidy, 17p loss ofheterozygosity (LOH), and several multiple biomarkerpanels.

Aneuploidy/tetraploidy. Genomic instability, whichpredisposes to the development of cancer-causing muta-

tions, can be manifested by gains or losses in parts ofchromosomes, a condition called aneuploidy. Aneuploidycan be detected in fresh frozen tissues by flow cytom-etry and in esophageal brushings or paraffin-embeddedtissues by fluorescence in situ hybridization. Severalreports suggest that flow cytometric evidence of aneu-ploidy and/or increased tetraploidy (specimens inwhich the fraction of cells with 4 sets of chromosomesexceeds 6%) can predict neoplastic progression in Bar-retts esophagus more accurately than the histologicgrade of dysplasia.113115

One study found that the 5-year cumulative incidence

of cancer was 4% for patients with Barretts esophaguswho had biopsy specimens showing no dysplasia, indef-inite dysplasia, or low-grade dysplasia (95% CI, 1.69.0).113 If the flow cytometry in those same cases wasnormal, then the 5-year incidence of cancer was 0% (95%CI, 04.7); if the flow cytometry showed aneuploidy orincreased tetraploidy, then the 5-year incidence of cancerwas 28% (95% CI, 12.055.0). Thus, the results of flowcytometry provided more useful predictive informationon cancer than the histologic finding of no to low-gradedysplasia. For patients with high-grade dysplasia, how-ever, flow cytometry added little additional predictiveinformation. These findings were confirmed in a subse-quent study by the same investigators.114

The aforementioned studies suggest that aneuploidyor increased tetraploidy might be used to predict the riskof cancer in patients who have Barretts esophagus withno dysplasia or only low-grade dysplasia. However, thosestudies detected aneuploidy/tetraploidy by flow cytom-etry performed on frozen tissue specimens. Esophagealbiopsy specimens rarely are frozen in clinical practice,and high-quality flow cytometry may not be widely avail-able in clinical centers. These factors may have hinderedthe adoption of aneuploidy/tetraploidy as a clinical bio-marker for neoplastic progression in Barretts esophagus.

In small studies, aneuploidy/tetraploidy detected by au-tomated image cytometry and fluorescence in situ hy-bridization, techniques that might be more feasible forroutine clinical practice, have been found to predict neo-plastic progression in Barretts esophagus.116,117 Thesepreliminary findings require validation in high-qualityprospective studies before the tests can be recommendedfor routine clinical application.

17p LOH. LOH for chromosome 17p, which har-bors the p53 gene, also has shown promise as a biomarkerfor neoplastic progression in Barretts esophagus. In onestudy of patients with Barretts esophagus whose biopsy

specimens showed changes ranging from no dysplasia tohigh-grade dysplasia, the 3-year cumulative incidence of

cancer was 38% (95% CI, 26.054.0) for those with 17pLOH compared with only 3.3% (95% CI, 1.4 8.0) forthose with 2 intact 17p alleles.118 In the subset of patientswho had no dysplasia, indefinite dysplasia, or low-gradedysplasia, furthermore, 17p LOH was a significant pre-

dictor of progression to high-grade dysplasia (RR, 3.6;95% CI, 1.310). Other studies by the same group ofinvestigators confirm these findings.119

In the aforementioned studies, 17p LOH was detectedusing a labor-intensive technique in which cells that werepurified by flow cytometry were then subjected to whole-genome amplification followed by genotypic analyses forLOH. More recently, cross-sectional studies have shownpromising results using simpler fluorescence in situ hy-bridization analyses for 17p LOH on biopsy and brushcytology specimens of Barretts esophagus.120123 How-ever, one study that compared the techniques head to

head found that fluorescence in situ hybridization had alower sensitivity for detecting 17p LOH than the flowcytometric approach.121

Biomarker panels. Numerous genetic abnormal-ities are acquired in variable sequences during the processof carcinogenesis in Barretts esophagus. Therefore, it isnot surprising that combinations of biomarkers in panelsmay be better at predicting the risk of neoplastic progres-sion than individual biomarkers. For example, one studyfound that patients with Barretts esophagus who had 3biomarker abnormalities (aneuploidy/tetraploidy, 17pLOH, and 9p LOH) in their esophageal biopsy specimenshad an 80% incidence of cancer within 6 years, whereasthose who had none of those abnormalities had an inci-dence of cancer of only 12% at 10 years.119 Comparedwith the latter group, the RR of cancer progression forpatients with an abnormal biomarker panel was 38.7(95% CI, 10.8138.5).

A gene methylation-based biomarker panel also hasshown promise for predicting development of cancer inBarretts esophagus. Promoter methylation is a processthat can silence the expression of a number of genes,including cancer-preventing tumor suppressor genes. Astudy that evaluated methylation of a number of tumorsuppressor genes in biopsy specimens of Barretts esoph-

agus found that methylation of p16, RUNX3, and HPP1was associated with a significantly increased risk of pro-gression to high-grade dysplasia or cancer.124 The inves-tigators generated a mathematical model for predictingneoplastic progression in Barretts esophagus that usedthe results of their methylation-based biomarker panel(that included the aforementioned tumor suppressorgenes), the patients age, and the extent of Barretts meta-plasia. When applied in a retrospective longitudinal fash-ion to a cohort of patients with Barretts esophagus, thisprediction model was able to identify patients destined toprogress to high-grade dysplasia or cancer as early as 2

years before neoplasia was recognized. Another morerecent study has confirmed the ability of a methylation-

AGA



14/35

based biomarker panel to predict neoplastic progressionin Barretts esophagus.125

In summary, a number of individual biomarkers andpanels of biomarkers have been proposed to predict therisk of neoplastic progression for patients with Barretts

esophagus. To date, however, none of these has been validated in prospective, controlled clinical trials. Avail-able data on aneuploidy/tetraploidy and 17p LOH sug-gest that these biomarkers are no better than the histo-logic finding of high-grade dysplasia for predictingprogression of cancer in Barretts esophagus. However,aneuploidy/tetraploidy, 17p LOH, and methylation-based biomarker panels may be superior to histologyalone for risk stratifying those patients with Barrettsesophagus whose initial biopsy specimens show no dys-plasia, indefinite dysplasia, or low-grade dysplasia. Incertain circumstances, therefore, those biomarkers could

be used in combination with histology for risk stratifica-tion. Current data suggest that the identification of an-euploidy by flow cytometry or the identification of 17pLOH by the combination of flow cytometry, whole-ge-nome amplification, and genotypic analysis are the bestavailable biomarker techniques. Thus, the routine clinicaluse of biomarkers instead of dysplasia for risk stratifi-cation in Barretts esophagus cannot be recommendedat this time. Nevertheless, it seems likely that theresults of biomarker validation studies will be availablein the near future and that biomarkers eventually willbe used to determine which patients with Barrettsesophagus will benefit from endoscopic surveillance or

ablative techniques.

Should Chromoendoscopy orElectronic Chromoendoscopy BeUsed to Enhance the Detection ofMetaplasia and Dysplasia in BarrettsEsophagus?

The Seattle biopsy protocol for endoscopic surveillance in Barretts esophagus, which involves 4-quad-rant biopsy sampling of every 1 to 2 cm of the columnar-lined esophagus, is time consuming, labor intensive,costly, and subject to considerable sampling error. A

number of alternative endoscopic techniques have beenproposed to enhance the detection of intestinal metapla-sia and dysplasia in the esophagus. The ultimate goals forthese advanced imaging techniques are to improve theendoscopic detection of curable neoplasia in Barrettsesophagus while reducing procedure time, expense, andsampling error.

Modern videoendoscopes use a charge-coupled device,which has a surface composed of photosensitive elements(pixels). In high-resolution endoscopes, the charge-cou-pled device has a large number of pixels (600,000 to1,000,000) that provide detailed images of the mucosal

surface. High-resolution endoscopy can be combinedwith magnification devices that enlarge the video image

up to 150.126128 High-definition television systems,which can generate up to 1080 scanning lines on a screen,enable the projection of a high-quality image onto a largescreen for ease of viewing.129 Compared with standardendoscopy, high-resolution endoscopy appears to have

higher sensitivity for detecting early neoplastic lesions inBarretts esophagus.130,131 Indeed, it is not clear that theimage enhancement techniques discussed in the follow-ing text add important information beyond that availableby careful white light inspection of the esophagus usinghigh-resolution endoscopy.

Chromoendoscopy involves the application of chemi-cal agents (eg, Lugols solution, methylene blue, indigocarmine, and acetic acid) that highlight various featuresof the esophageal mucosa in an attempt to improve thedetection of abnormalities.132140 Lugols solution, whichis taken up by esophageal squamous cells that contain

glycogen, has been used to highlight the squamocolum-nar junction and as an aid for identifying residual islandsof Barretts metaplasia (which are not stained by Lugolssolution) after endoscopic eradication therapy.141 Re-ports on the use of methylene blue, which is absorbed byintestinal-type epithelium that is not dysplastic, havedescribed variable results. In a prospective, randomized,crossover trial that compared methylene bluedirectedbiopsy with standard 4-quadrant biopsy in 48 patientswith Barretts esophagus, the techniques were found tobe similar for the detection of intestinal metaplasia anddysplasia, although the mean number of biopsies re-quired to detect those conditions was significantly lowerwith methylene blue staining.135 In contrast, anotherrandomized crossover study found that the 4-quadrantbiopsy technique detected dysplasia significantly moreoften than the methylene bluedirected biopsy tech-nique.136 A recent meta-analysis of 9 studies that in-cluded 450 total patients found that methylene bluestaining and 4-quadrant biopsy techniques have similarrates for detecting intestinal metaplasia and dysplasia.140

Further decreasing enthusiasm for methylene blue stain-ing is a study documenting that the technique causesDNA damage in Barretts epithelium.142

Different mucosal pit patterns in columnar epithelia

can be recognized by combining magnification endos-copy with the mucosal application of indigo carmine dyeor acetic acid. Using acetic acid, Guelrud et al described 4pit patterns in Barretts epithelium (round, reticular, vil-lous, and ridged) and found that the ridged and villouspatterns were associated with intestinal metaplasia.141

Sharma et al studied 80 patients with Barretts esophagususing indigo carmine and found that the ridged/villousmucosal pattern had high sensitivity (97%) and reason-able specificity (76%) for intestinal metaplasia.138 In ad-dition, all of 6 patients with high-grade dysplasia werefound to have a distorted or irregular glandular pattern.

However, a prospective, randomized, crossover study of28 patients found that indigo carmine chromoendoscopy

AGA

March 2011 AGA e31


15/35

did not increase the sensitivity for detecting early neo-plasia in Barretts esophagus beyond that of high-resolu-tion white light endoscopy.131

Electronic chromoendoscopy can be achieved bytechniques such as narrow band imaging (NBI), which

uses spectral narrow-band optical filters to highlight vas-cular patterns on the mucosal surface, or by optimalband imaging and I-scan (Pentax Medical Company,Montvale, NJ), which use a postprocessing technology tohighlight contrast between squamous and columnar ep-ithelia.143 In a preliminary study of 24 patients withhigh-grade dysplasia or early cancer in Barretts esopha-gus, optimal band imaging was found to detect neoplasiawith a sensitivity of 87% but with a positive predictive value of only 37%.144 Some single-center studies haveattempted to correlate the magnified NBI appearance ofthe mucosal glandular and vascular patterns with the

presence of metaplasia and dysplasia.145,146

In one suchprospective study of magnification NBI in 51 patientswith Barretts esophagus, a ridge/villous pattern pre-dicted the presence of intestinal metaplasia with a sensi-tivity of 93.5% and a specificity of 86.7%, whereas anirregular/distorted pattern predicted high-grade dyspla-sia with a sensitivity and specificity of 100% and 98.7%,respectively.145 The magnified NBI images could not dis-tinguish low-grade dysplasia from nondysplastic tissue,however. In another study, Kara et al showed that areas ofhigh-grade dysplasia in Barretts esophagus had at leastone of 3 abnormal patterns by NBI: (1) an irregular/disrupted mucosal pattern, (2) an irregular vascular pat-tern, and (3) abnormal blood vessels.146 A randomizedcrossover trial that compared chromoendoscopy and NBIin 28 patients found no significant difference betweenthe 2 techniques for the detection of high-grade dys-plasia and early cancer (93% vs 86% sensitivity), andneither technique was superior to high-resolution whitelight endoscopy in that regard.131 In another study, en-doscopists were asked to identify dysplasia in still imagesof Barretts epithelium taken during magnification en-doscopy. The yield for identifying dysplasia in imagestaken with high-resolution white light endoscopy was86%, and prediction rates did not increase significantly

with the addition of either chromoendoscopy or NBI.147Two recent reports describe prospective studies that

have compared the diagnostic yield of NBI (nonmagni-fied) with that of white light endoscopy.148,149 In onestudy of 65 patients who were known to have dysplasia inBarretts esophagus, standard-resolution white light en-doscopy was performed first, followed by NBI performedby another endoscopist who used NBI to detect andobtain biopsy specimens from areas suspicious for dys-plasia.148 The lesions initially detected by standard en-doscopy were then disclosed and biopsy was performed;finally, random 4-quadrant biopsy specimens were taken

throughout the columnar-lined esophagus. NBI wasfound to identify more patients with dysplasia than stan-

dard-resolution white light endoscopy with random biopsysampling (57% vs 43%; P .001). NBI also found highergrades of dysplasia significantly more often than standardendoscopy (18% higher grade with NBI than with standardvs 0% higher grade with standard than with NBI; P .001).

In a multicenter, randomized, crossover trial comparingNBI-targeted biopsies with high-resolution white lightendoscopy and 4-quadrant biopsies in 123 patients withBarretts esophagus, there were no significant differencesbetween the 2 techniques in the frequency of detectingintestinal metaplasia (85% for each technique) and dys-plasia (71% for NBI vs 55% for white light endoscopy; P.15).149 However, significantly fewer biopsies were re-quired to establish a diagnosis with NBI (3.6 vs 7.6 perprocedure; P .0001).

Chromoendoscopy for Barretts esophagus is time con-suming, fraught with technical problems (eg, achieving

uniform dye application), potentially hazardous (in thecase of methylene blue), poorly standardized (regardingthe interpretation of mucosal patterns), and subject toconsiderable interobserver variability. Studies comparingchromoendoscopy with standard-resolution endoscopyhave had contradictory findings, and studies have notestablished any diagnostic advantage for chromoendos-copy beyond that which can be achieved by high-resolu-tion white light endoscopy. Consequently, we do notadvocate the routine use of chromoendoscopy in Bar-retts esophagus. Electronic chromoendoscopy tech-niques such as NBI are less time consuming and techni-cally easier to perform than chromoendoscopy but are

still subject to problems of poor standardization andinterobserver variability. The studies discussed previouslysuggest that NBI may be superior to standard-resolutionwhite light endoscopy for detecting esophageal metapla-sia and dysplasia, but studies so far have not establisheda convincing advantage for NBI over high-resolutionwhite light endoscopy.

We conclude that endoscopic surveillance is best per-formed with careful inspection of the columnar-linedesophagus using high-resolution white light endoscopy,with biopsy sampling of any lesions or suspicious areasso identified followed by 4-quadrant biopsy sampling of

the Barretts metaplasia. The use of NBI or similar elec-tronic chromoendoscopy techniques cannot be advo-cated or discouraged at this time.

Should Advanced Endoscopic ImagingTechniques Such as AutofluorescenceImaging, Confocal LaserEndomicroscopy, Diffuse Reflectanceand Light Scattering Spectroscopy, andOptical Coherence Tomography Be Usedto Enhance the Detection of Metaplasiaand Dysplasia in Barretts Esophagus?

Cells contain endogenous fluorophores (eg, re-duced nicotinamide adenine dinucleotide, porphyrins)

AGA



16/35

that can absorb endoscopically delivered laser light andre-emit it as fluorescent light with distinctive spectro-scopic characteristics. Autofluorescence imaging (AFI) ex-ploits this phenomenon to highlight abnormal areas inBarretts esophagus that can be targeted for biopsy sam-

pling.150152 AFI attempts to distinguish normal fromneoplastic epithelia based on differences in their fluores-cence spectra. An initial feasibility study found that AFIwas a sensitive test that improved the rate of detectinghigh-grade dysplasia but with poor specificity that re-sulted in a positive predictive value of only 50%.150

In a multicenter study of 84 patients, Curvers et alexplored the diagnostic potential of tri-modal imagingin which the esophagus is first inspected by high-resolu-tion white light endoscopy, followed by AFI to rapidlyhighlight abnormal areas not detected by white light,followed by NBI to confirm the abnormality of areas

highlighted

2011 03 revision tecnica de barrett

Documents