Professional Documents
Culture Documents
review article
Dan L. Longo, M.D., Editor
Barrett’s Esophagus
Stuart J. Spechler, M.D., and Rhonda F. Souza, M.D.
I
From the Esophageal Diseases Center, t has been estimated that 5.6% of adults in the United States have
Department of Medicine, Veterans Affairs Barrett’s esophagus,1 the condition in which a metaplastic columnar mucosa
(VA) North Texas Health Care System,
and the University of Texas Southwestern that confers a predisposition to cancer replaces an esophageal squamous mu-
Medical Center, Dallas. Address reprint cosa damaged by gastroesophageal reflux disease (GERD).2 GERD and Barrett’s
requests to Dr. Spechler at the Division esophagus are major risk factors for esophageal adenocarcinoma, a deadly tumor
of Gastroenterology and Hepatology
(111B1), Dallas VA Medical Center, 4500 whose frequency in the United States has increased by a factor of more than 7 dur-
S. Lancaster Rd., Dallas, TX 75216. ing the past four decades.3,4 The metaplastic columnar mucosa of Barrett’s esopha-
N Engl J Med 2014;371:836-45. gus causes no symptoms, and the condition has clinical importance only because
DOI: 10.1056/NEJMra1314704 it confers a predisposition to cancer.
Copyright © 2014 Massachusetts Medical Society.
Metaplasia, the process wherein one adult cell type replaces another, is a conse-
quence of chronic tissue injury.5 In patients with chronic esophageal injury from
GERD, Barrett’s metaplasia develops when mucus-secreting columnar cells replace
reflux-damaged esophageal squamous cells. The cells that give rise to this metapla-
sia are not known. It has been proposed that GERD might induce alterations in the
expression of key developmental transcription factors, causing mature esophageal
squamous cells to change into columnar cells (transdifferentiation) or causing im-
mature esophageal progenitor cells to undergo columnar rather than squamous
differentiation (transcommitment).5,6 In a rat model of reflux esophagitis, metapla-
sia develops from bone marrow stem cells that enter the blood and settle in the
reflux-damaged esophagus.7 Studies in mouse models have suggested that meta-
plasia might result from upward migration of stem cells from the proximal stom-
ach (the gastric cardia)8 or from proximal expansion of embryonic-type cells at the
gastroesophageal junction.9 It is not clear which of these processes contribute to
the pathogenesis of Barrett’s esophagus in humans.
Di agnosis
Esophagus
Squamous-lined
Squamous mucosa esophagus
Intestinal metaplasia
Columnar-lined
esophagus
Goblet cells
Gastroesophageal
junction
Squamous mucosa
Cardiac mucosa
Stomach
esophageal junction and lines the distal esophagus, plus esophageal-biopsy results that confirm Draft 4
the presence of
7/9/14
columnar metaplasia. Endoscopically, the gastroesophageal junction is identified as the most Authorproximal
Spechlerextent of the
gastric folds (dashed white line). Salmon-colored columnar mucosa extends in tongue-shaped Fig # projections
1 above the
gastroesophageal junction, lining the distal esophagus. A biopsy specimen obtained at theTitle level Barrett’s
of the upperEsphaguswhite
dot reveals the junction between esophageal stratified squamous epithelium and intestinal metaplasia
ME
with distinctive,
intestinal-type goblet cells, which establishes the diagnosis of Barrett’s esophagus. Intestinal
DE metaplasia
Longo may not be
uniformly distributed throughout the entire columnar-lined esophagus, however. In this example,
Artist NaKoscal
biopsy specimen
taken from the columnar-lined esophagus closer to the gastroesophageal junction (at the levelFigure
of AUTHOR
the lowerPLEASE white
NOTE: dot)
has been redrawn and type has been reset
shows cardiac mucosa composed of mucus-secreting columnar cells without goblet cells. Some gastroenterology Please check carefully
societies (e.g., the British Society of Gastroenterology) accept evidence of cardiac mucosaIssue
alonedate as 8/28/2014
diagnostic of
Barrett’s esophagus, but U.S. gastroenterology societies require evidence of intestinal metaplasia for a definitive di-
agnosis. Photomicrographs (hematoxylin and eosin) provided by Drs. Kevin Turner and Robert Genta.
(Fig. 1).11-13 This intestinal metaplasia is a well- metaplasia in some, if not all, cases.17 Neverthe-
established risk factor for adenocarcinoma.12 less, it is not clear that cardiac mucosa is an
However, some other societies, including the important risk factor for adenocarcinoma.18
British Society of Gastroenterology, also con- Thus, the major issue underlying disagreement
sider esophageal biopsies that show cardiac on histologic criteria for the diagnosis of Bar-
mucosa (comprising mucus-secreting columnar rett’s esophagus is whether the condition should
cells without goblet cells) to be diagnostic of be defined as a histologic curiosity (a mucosal
Barrett’s esophagus.14 Cardiac mucosa, although metaplasia, irrespective of its clinical importance)
traditionally considered the normal lining of the or as a medical condition (a mucosal metaplasia
gastric cardia, can have intestinal-type histo- that confers a predisposition to cancer). U.S.
chemical features and abnormalities in DNA con- gastroenterology societies have taken the latter
tent,15,16 and it appears to be a GERD-induced position.
Table 1. Proposed Risk Factors and Protective Factors for Barrett’s Esophagus and Esophageal Adenocarcinoma.*
* A dash indicates that studies have not addressed the question of whether the specified factor is associated with an in-
creased risk or has a protective effect. Citations for the information in this table are provided in the Supplementary
Appendix, available with the full text of this article at NEJM.org. GERD denotes gastroesophageal reflux disease.
50 years of age or older, either by intention (dur- approximately twice that among women,36 the
ing screening endoscopy for GERD symptoms) or risk is greater with a longer segment of Barrett’s
by chance (during endoscopy for conditions un- metaplasia,38 and the risk is especially high
related to GERD). Barrett’s esophagus is two to among persons with certain familial forms of
three times as common in men as in women, is Barrett’s esophagus.39 In addition, the risk ap-
uncommon in blacks and Asians, and is rare in pears to decrease with follow-up endoscopies
children.23,24 Other important risk factors include showing no progression to dysplasia.40
obesity (with a predominantly intraabdominal
fat distribution) and cigarette smoking, and there Scr eening a nd Surv eil l a nce
is a familial form of Barrett’s esophagus, which for B a r r e t t ’s E soph agus
accounts for 7 to 11% of all cases.25 Most con-
ditions associated with Barrett’s metaplasia are For decades, the primary strategy for preventing
also risk factors for esophageal adenocarcinoma.26 deaths from esophageal adenocarcinoma has
Conversely, factors that might provide protection been to screen patients with GERD symptoms for
against Barrett’s esophagus include the use of Barrett’s esophagus with the use of endoscopy
nonsteroidal antiinflammatory drugs, gastric in- and, for patients with Barrett’s esophagus on en-
fection with Helicobacter pylori, and consumption doscopic screening, to perform regular endo-
of a diet high in fruits and vegetables. scopic surveillance to detect curable neoplasia.2
No single risk factor yet identified can account Unfortunately, there is no proof that this strategy
for the profound increase in the incidence of is effective, and with an annual cancer incidence
esophageal adenocarcinoma in Western countries of only 0.1 to 0.3%, the logistics of performing a
during the past 40 years, a period when GERD randomized trial to prove that screening and sur-
and Barrett’s esophagus appear to have increased veillance prevent deaths from esophageal cancer
only modestly in frequency.27,28 There has been are daunting. Observational studies have shown
a steep rise in the frequency of central obesity, that patients with Barrett’s esophagus–associated
which might contribute to Barrett’s carcinogen- cancers diagnosed by means of surveillance en-
esis by promoting GERD and by increasing the doscopy have earlier-stage tumors and higher sur-
production of hormones that promote cell pro- vival rates than those whose tumors are discov-
liferation, such as leptin and insulin-like growth ered because of symptoms such as dysphagia and
factors.29,30 H. pylori infection, which may protect weight loss.41,42 However, such studies are highly
the esophagus from GERD by causing a gastritis susceptible to biases that might exaggerate the
that reduces gastric acid production, has declined benefits of surveillance. Some computer-model-
in frequency during the same period when esoph- ing studies have concluded that screening and sur-
ageal adenocarcinoma has risen in developed veillance can be cost-effective under certain cir-
countries.31 Another hypothesis links the rising cumstances, but such studies are not definitive.43,44
incidence of esophageal adenocarcinoma with Despite the lack of high-quality evidence to
increased dietary intake of nitrate, which has support the practice, medical societies currently
resulted from the widespread use of nitrate- recommend endoscopic screening for Barrett’s
based fertilizers.32 esophagus in patients with chronic GERD symp-
Estimates of the annual incidence of esopha- toms who have at least one additional risk factor
geal adenocarcinoma among patients with non- for esophageal adenocarcinoma, such as an age
dysplastic Barrett’s esophagus have ranged from of 50 years or older, male sex, white race, hiatal
0.1 to 2.9%, with the highest estimates in stud- hernia, elevated body-mass index, intraabdominal
ies with evidence of publication bias.33 Recent, body-fat distribution, or tobacco use.2,12,13,45 If
better-quality studies suggest that the risk of the screening examination does not reveal Bar-
esophageal adenocarcinoma in the general pop- rett’s esophagus, no further endoscopic screen-
ulation of patients with nondysplastic Barrett’s ing for the condition is recommended.13,45 For
esophagus is only 0.1 to 0.3% per year.34-37 How- patients found to have nondysplastic Barrett’s
ever, a number of factors influence the risk of metaplasia, whether by screening or by chance,
cancer for individual patients. For example, can- medical societies recommend regular endoscop-
cer risk among men with Barrett’s esophagus is ic surveillance at intervals of 3 to 5 years.2,11-13
Nevertheless, there are a number of reasons to ties detected by means of fluorescence in situ
question the value of screening and surveillance hybridization (FISH) and biomarker panels that
for Barrett’s esophagus. identify multiple abnormalities in DNA content,
The screening prerequisite of GERD symptoms gene expression, and DNA methylation have
limits the usefulness of the practice, because shown promise as predictors of cancer risk, as
patients with short-segment Barrett’s esophagus have some risk-stratification models that incor-
often have no GERD symptoms, and approxi- porate a variety of clinical, histologic, and molecu-
mately 40% of patients with esophageal adeno- lar features.52-56 However, none of these methods
carcinoma report no history of GERD.46 Studies have yet been validated sufficiently to justify
have shown that less than 10% of patients with routine application in clinical practice.
esophageal adenocarcinoma have a prior diagno- There are adverse consequences of endoscopic
sis of Barrett’s esophagus, suggesting that current screening and surveillance, in addition to the
screening practices are highly ineffective.47,48 Fur- high cost of endoscopy and the small risk of
thermore, a recent case–control study has chal- endoscopic complications. Identification of in-
lenged the efficacy of surveillance for cancer pre- nocuous neoplastic lesions by means of these
vention among patients with Barrett’s esophagus.49 procedures might lead to the use of invasive
This study compared the frequency of surveil- therapies with serious or even fatal complica-
lance endoscopy during a 3-year period among tions. Studies have shown that a diagnosis of
38 case patients (those known to have Barrett’s Barrett’s esophagus causes psychological stress,
esophagus who subsequently died of esophageal has a negative effect on quality of life, and re-
adenocarcinoma) with that among 101 living, sults in higher premiums for health and life in-
control patients with Barrett’s esophagus who surance.12 To date, medical societies have taken
were matched for age, sex, and follow-up dura- the position that, in the absence of definitive
tion. The case patients and controls had nearly data, it is better to err by performing unneces-
identical frequencies of endoscopic surveillance sary screening and surveillance than by forgoing
(55% among case patients and 60% among con- the opportunity to identify curable esophageal
trols), and surveillance was not associated with neoplasms. It is not clear whether the new data
a decreased risk of death from esophageal can- discussed above will influence future recommen-
cer (adjusted odds ratio, 0.99; 95% confidence dations. Despite the many limitations and dubi-
interval [CI], 0.36 to 2.75). However, this rela- ous benefits of screening and surveillance for
tively wide confidence interval does not exclude Barrett’s esophagus, these practices are still rec-
the possibility that surveillance was beneficial. ommended by medical societies. In general, rec-
A primary rationale for screening has been to ommendations for surveillance of established
identify patients with Barrett’s esophagus, who Barrett’s esophagus are stronger and more ex-
then will benefit from surveillance. If, as the plicit than recommendations for initial screening.
aforementioned report suggests, surveillance has
little benefit, then the practice of screening
M a nagemen t of B a r r e t t ’s
might be based on a fundamentally flawed prem- E soph agus
ise. Clearly, better methods are needed for risk
stratification to identify those patients with Bar- A brief algorithm for endoscopic surveillance and
rett’s esophagus who could benefit most from eradication therapy in patients with Barrett’s
surveillance or other interventions. Advanced esophagus is provided in Figure 2.
endoscopic imaging techniques have been stud-
ied for this purpose, including dye-based chromo- Treatment of GERD
endoscopy, optical and digital chromoendoscopy, In patients with Barrett’s metaplasia, refluxed
autofluorescence endoscopy, and confocal laser gastric acid can cause chronic inflammation,
endomicroscopy.50 In biopsy specimens from pa- double-strand DNA breaks, and increased cell
tients with Barrett’s metaplasia, abnormalities in proliferation, all of which may contribute to car-
p53 expression and in cellular DNA content on cinogenesis.57 This suggests that GERD should be
flow cytometry have been associated with neo- treated aggressively in patients with Barrett’s
plastic progression.51,52 Cytogenetic abnormali- esophagus, and there is indirect evidence to sug-
Figure 2. Algorithm for the Screening, Surveillance, and Management of Barrett’s Esophagus.
Endoscopy for patients with dysplasia or intramucosal carcinoma should include four-quadrant biopsy sampling at 1-cm intervals and
endoscopic resection of mucosal irregularities. If dysplasia or intramucosal carcinoma is discovered and these procedures have not been
performed, then repeat endoscopy is recommended before endoscopic eradication therapy is initiated. BMI denotes body-mass index,
and GERD gastroesophageal reflux disease.
gest that proton-pump inhibitors (PPIs) decrease Just as in patients who have GERD without
the risk of cancer development. For example, a Barrett’s metaplasia, PPIs are used in patients
recent cohort study involving 540 patients with with Barrett’s esophagus to control GERD symp-
Barrett’s esophagus who were followed for a me- toms and heal reflux esophagitis. For patients
dian of 5.2 years showed that PPI use was asso- who have no symptoms or endoscopic signs of
ciated with a 75% reduction in the risk of neo- GERD, as is common in short-segment Barrett’s
plastic progression.58 Bile acids can also cause esophagus, the issue of whether to use PPIs for
double-strand DNA breaks and might contribute chemoprevention remains unresolved and con-
to carcinogenesis in patients with Barrett’s meta- troversial. We believe that the indirect evidence
plasia, and PPIs do not prevent bile reflux.59 An- supporting a cancer-protective role for PPIs in
tireflux surgery can prevent reflux of all gastric Barrett’s esophagus is strong enough to warrant
contents (acid and bile), but the best available conventional-dose PPI treatment for asymptom-
data suggest that surgery is not more effective atic patients after they have been informed of
than PPI therapy in preventing cancer.57 Thus, the potential risks and benefits, although this
antireflux surgery is not advised solely for pro- approach is not specifically endorsed by medical
tection against cancer. societies.
Dr. Spechler reports receiving consulting fees from Takeda otential conf lict of interest relevant to this article was re-
p
Pharmaceuticals USA, Ironwood Pharmaceuticals, and Torax ported.
Medical. Dr. Souza reports receiving consulting fees from Disclosure forms provided by the authors are available with
Ironwood Pharmaceuticals and Otsuka America. No other the full text of this article at NEJM.org.
References
1. Hayeck TJ, Kong CY, Spechler SJ, Ga- 14. Fitzgerald RC, di Pietro M, Ragunath esophageal reflux disease: a systematic
zelle GS, Hur C. The prevalence of Bar- K, et al. British Society of Gastroenterol- review. Clin Gastroenterol Hepatol 2007;
rett’s esophagus in the US: estimates ogy guidelines on the diagnosis and man- 5:17-26.
from a simulation model confirmed by agement of Barrett’s oesophagus. Gut 28. Coleman HG, Bhat S, Murray LJ, Mc-
SEER data. Dis Esophagus 2010;23:451-7. 2014;63:7-42. Manus D, Gavin AT, Johnston BT. Increas-
2. Spechler SJ. Barrett esophagus and 15. Hahn HP, Blount PL, Ayub K, et al. ing incidence of Barrett’s oesophagus:
risk of esophageal cancer: a clinical re- Intestinal differentiation in metaplastic, a population-based study. Eur J Epidemiol
view. JAMA 2013;310:627-36. nongoblet columnar epithelium in the 2011;26:739-45.
3. Pohl H, Sirovich B, Welch HG. Esoph- esophagus. Am J Surg Pathol 2009;33: 29. El-Serag H. The association between
ageal adenocarcinoma incidence: are we 1006-15. obesity and GERD: a review of the epide-
reaching the peak? Cancer Epidemiol Bio- 16. Liu W, Hahn H, Odze RD, Goyal RK. miological evidence. Dig Dis Sci 2008;53:
markers Prev 2010;19:1468-70. [Erratum, Metaplastic esophageal columnar epithe- 2307-12.
Cancer Epidemiol Biomarkers Prev 2010; lium without goblet cells shows DNA con- 30. Greer KB, Thompson CL, Brenner L,
19:2416.] tent abnormalities similar to goblet cell- et al. Association of insulin and insulin-
4. Thrift AP, Whiteman DC. The inci- containing epithelium. Am J Gastroenterol like growth factors with Barrett’s oesoph-
dence of esophageal adenocarcinoma con- 2009;104:816-24. agus. Gut 2012;61:665-72.
tinues to rise: analysis of period and birth 17. Chandrasoma PT. Histologic defini- 31. Parsonnet J. The incidence of Helico-
cohort effects on recent trends. Ann On- tion of gastro-esophageal reflux disease. bacter pylori infection. Aliment Pharmacol
col 2012;23:3155-62. Curr Opin Gastroenterol 2013;29:460-7. Ther 1995;9:Suppl 2:45-51.
5. Burke ZD, Tosh D. Barrett’s metapla- 18. Westerhoff M, Hovan L, Lee C, Hart J. 32. Iijima K, Henry E, Moriya A, Wirz A,
sia as a paradigm for understanding the Effects of dropping the requirement for Kelman AW, McColl KE. Dietary nitrate
development of cancer. Curr Opin Genet goblet cells from the diagnosis of Bar- generates potentially mutagenic concen-
Dev 2012;22:494-9. rett’s esophagus. Clin Gastroenterol Hep- trations of nitric oxide at the gastroesoph-
6. Wang DH, Clemons NJ, Miyashita T, atol 2012;10:1232-6. ageal junction. Gastroenterology 2002;122:
et al. Aberrant epithelial-mesenchymal 19. Fass R, Hell RW, Garewal HS, et al. 1248-57.
Hedgehog signaling characterizes Bar- Correlation of oesophageal acid exposure 33. Shaheen NJ, Crosby MA, Bozymski
rett’s metaplasia. Gastroenterology 2010; with Barrett’s oesophagus length. Gut EM, Sandler RS. Is there publication bias
138:1810-22. 2001;48:310-3. in the reporting of cancer risk in Barrett’s
7. Sarosi G, Brown G, Jaiswal K, et al. 20. Taylor JB, Rubenstein JH. Meta-analy- esophagus? Gastroenterology 2000;119:
Bone marrow progenitor cells contribute ses of the effect of symptoms of gastro- 333-8.
to esophageal regeneration and metapla- esophageal reflux on the risk of Barrett’s 34. Desai TK, Krishnan K, Samala N, et al.
sia in a rat model of Barrett’s esophagus. esophagus. Am J Gastroenterol 2010;105: The incidence of oesophageal adenocarci-
Dis Esophagus 2008;21:43-50. 1729-37. noma in non-dysplastic Barrett’s oesoph-
8. Quante M, Bhagat G, Abrams JA, et al. 21. Fletcher J, Wirz A, Henry E, McColl agus: a meta-analysis. Gut 2012;61:970-6.
Bile acid and inflammation activate gas- KE. Studies of acid exposure immediately 35. Wani S, Falk G, Hall M, et al. Patients
tric cardia stem cells in a mouse model of above the gastro-oesophageal squamoco- with nondysplastic Barrett’s esophagus
Barrett-like metaplasia. Cancer Cell 2012; lumnar junction: evidence of short seg- have low risks for developing dysplasia or
21:36-51. ment reflux. Gut 2004;53:168-73. esophageal adenocarcinoma. Clin Gastro-
9. Wang X, Ouyang H, Yamamoto Y, et al. 22. Spechler SJ, Zeroogian JM, Antonioli enterol Hepatol 2011;9:220-7.
Residual embryonic cells as precursors of DA, Wang HH, Goyal RK. Prevalence of 36. Bhat S, Coleman HG, Yousef F, et al.
a Barrett’s-like metaplasia. Cell 2011;145: metaplasia at the gastro-oesophageal junc- Risk of malignant progression in Bar-
1023-35. tion. Lancet 1994;344:1533-6. rett’s esophagus patients: results from a
10. Sharma P, Morales TG, Sampliner RE. 23. Wang A, Mattek NC, Holub JL, Lieber- large population-based study. J Natl Can-
Short segment Barrett’s esophagus — the man DA, Eisen GM. Prevalence of compli- cer Inst 2011;103:1049-57. [Erratum, J Natl
need for standardization of the definition cated gastroesophageal reflux disease and Cancer Inst 2013;105:581.]
and of endoscopic criteria. Am J Gastro- Barrett’s esophagus among racial groups 37. Hvid-Jensen F, Pedersen L, Drewes
enterol 1998;93:1033-6. in a multi-center consortium. Dig Dis Sci AM, Sørensen HT, Funch-Jensen P. Inci-
11. Wang KK, Sampliner RE. Updated 2009;54:964-71. dence of adenocarcinoma among patients
guidelines 2008 for the diagnosis, sur- 24. El-Serag HB, Gilger MA, Shub MD, with Barrett’s esophagus. N Engl J Med
veillance and therapy of Barrett’s esopha- Richardson P, Bancroft J. The prevalence 2011;365:1375-83.
gus. Am J Gastroenterol 2008;103:788-97. of suspected Barrett’s esophagus in chil- 38. Anaparthy R, Gaddam S, Kanakadandi
12. Spechler SJ, Sharma P, Souza RF, In- dren and adolescents: a multicenter endo- V, et al. Association between length of
adomi JM, Shaheen NJ. American Gastro- scopic study. Gastrointest Endosc 2006; Barrett’s esophagus and risk of high-
enterological Association technical re- 64:671-5. grade dysplasia or adenocarcinoma in pa-
view on the management of Barrett’s 25. Orloff M, Peterson C, He X, et al. tients without dysplasia. Clin Gastroen-
esophagus. Gastroenterology 2011;140(3): Germline mutations in MSR1, ASCC1, and terol Hepatol 2013;11:1430-6.
e18-e52. CTHRC1 in patients with Barrett esopha- 39. Munitiz V, Parrilla P, Ortiz A, Martinez-
13. ASGE Standards of Practice Commit- gus and esophageal adenocarcinoma. de-Haro LF, Yelamos J, Molina J. High risk
tee, Evans JA, Early DS, et al. The role of JAMA 2011;306:410-9. of malignancy in familial Barrett’s esoph-
endoscopy in Barrett’s esophagus and 26. Lagergren J, Lagergren P. Recent de- agus: presentation of one family. J Clin
other premalignant conditions of the velopments in esophageal adenocarcinoma. Gastroenterol 2008;42:806-9.
esophagus. Gastrointest Endosc 2012;76: CA Cancer J Clin 2013;63:232-48. 40. Gaddam S, Singh M, Balasubramanian
1087-94. 27. El-Serag HB. Time trends of gastro- G, et al. Persistence of nondysplastic Bar-
rett’s esophagus identifies patients at low- tion of dysplasia and adenocarcinoma in esophagus. Gastrointest Endosc 2005;62:
er risk for esophageal adenocarcinoma: patients with Barrett’s esophagus. Hum 16-23.
results from a large multicenter cohort. Pathol 2008;39:1128-35. 66. Pech O, Behrens A, May A, et al. Long-
Gastroenterology 2013;145:548-53. 53. Alvi MA, Liu X, O’Donovan M, et al. term results and risk factor analysis for
41. Fountoulakis A, Zafirellis KD, Dolan DNA methylation as an adjunct to histo- recurrence after curative endoscopic ther-
K, Dexter SP, Martin IG, Sue-Ling HM. Ef- pathology to detect prevalent, inconspicu- apy in 349 patients with high-grade in-
fect of surveillance of Barrett’s oesopha- ous dysplasia and early-stage neoplasia in traepithelial neoplasia and mucosal ade-
gus on the clinical outcome of oesopha- Barrett’s esophagus. Clin Cancer Res nocarcinoma in Barrett’s oesophagus.
geal cancer. Br J Surg 2004;91:997-1003. 2013;19:878-88. Gut 2008;57:1200-6.
42. Corley DA, Levin TR, Habel LA, Weiss 54. Jin Z, Cheng Y, Gu W, et al. A multi- 67. Dunbar KB, Spechler SJ. The risk of
NS, Buffler PA. Surveillance and survival center, double-blinded validation study of lymph-node metastases in patients with
in Barrett’s adenocarcinomas: a popula- methylation biomarkers for progression high-grade dysplasia or intramucosal car-
tion-based study. Gastroenterology 2002; prediction in Barrett’s esophagus. Cancer cinoma in Barrett’s esophagus: a system-
122:633-40. Res 2009;69:4112-5. atic review. Am J Gastroenterol 2012;107:
43. Sonnenberg A, Soni A, Sampliner RE. 55. Rubenstein JH, Morgenstern H, Appel- 850-62.
Medical decision analysis of endoscopic man H, et al. Prediction of Barrett’s 68. Orman ES, Li N, Shaheen NJ. Efficacy
surveillance of Barrett’s oesophagus to esophagus among men. Am J Gastroen- and durability of radiofrequency ablation
prevent oesophageal adenocarcinoma. Ali- terol 2013;108:353-62. for Barrett’s esophagus: systematic review
ment Pharmacol Ther 2002;16:41-50. 56. Bird-Lieberman EL, Dunn JM, Cole- and meta-analysis. Clin Gastroenterol
44. Inadomi JM, Sampliner R, Lagergren J, man HG, et al. Population-based study Hepatol 2013;11:1245-55.
Lieberman D, Fendrick AM, Vakil N. reveals new risk-stratification biomarker 69. Curvers WL, ten Kate FJ, Krishnadath
Screening and surveillance for Barrett panel for Barrett’s esophagus. Gastroen- KK, et al. Low-grade dysplasia in Barrett’s
esophagus in high-risk groups: a cost- terology 2012;143:927-35. esophagus: overdiagnosed and underes-
utility analysis. Ann Intern Med 2003;138: 57. Spechler SJ. Does Barrett’s esophagus timated. Am J Gastroenterol 2010;105:
176-86. regress after surgery (or proton pump in- 1523-30.
45. Shaheen NJ, Weinberg DS, Denberg hibitors)? Dig Dis 2014;32:156-63. 70. Wani S, Falk GW, Post J, et al. Risk
TD, Chou R, Qaseem A, Shekelle P. Upper 58. Kastelein F, Spaander MC, Steyerberg factors for progression of low-grade dys-
endoscopy for gastroesophageal reflux EW, et al. Proton pump inhibitors reduce plasia in patients with Barrett’s esopha-
disease: best practice advice from the the risk of neoplastic progression in pa- gus. Gastroenterology 2011;141:1179-86.
clinical guidelines committee of the Amer- tients with Barrett’s esophagus. Clin Gas- 71. Phoa KN, van Vilsteren FG, Weusten
ican College of Physicians. Ann Intern troenterol Hepatol 2013;11:382-8. BL, et al. Radiofrequency ablation vs endo-
Med 2012;157:808-16. 59. Huo X, Juergens S, Zhang X, et al. De- scopic surveillance for patients with Bar-
46. Chak A, Faulx A, Eng C, et al. Gastro- oxycholic acid causes DNA damage while rett esophagus and low-grade dysplasia:
esophageal reflux symptoms in patients inducing apoptotic resistance through a randomized clinical trial. JAMA 2014;
with adenocarcinoma of the esophagus or NF-κB activation in benign Barrett’s epi- 311:1209-17.
cardia. Cancer 2006;107:2160-6. thelial cells. Am J Physiol Gastrointest 72. Fleischer DE, Odze R, Overholt BF,
47. Dulai GS, Guha S, Kahn KL, Gorn- Liver Physiol 2011;301:G278-G286. et al. The case for endoscopic treatment
bein J, Weinstein WM. Preoperative preva- 60. Spechler SJ. Dysplasia in Barrett’s of non-dysplastic and low-grade dysplas-
lence of Barrett’s esophagus in esopha- esophagus: limitations of current man- tic Barrett’s esophagus. Dig Dis Sci 2010;
geal adenocarcinoma: a systematic review. agement strategies. Am J Gastroenterol 55:1918-31.
Gastroenterology 2002;122:26-33. 2005;100:927-35. 73. Anders M, Lucks Y, El-Masry MA, et al.
48. Bhat SK, McManus DT, Coleman HG, 61. Rastogi A, Puli S, El-Serag HB, Bansal Subsquamous extension of intestinal
et al. Oesophageal adenocarcinoma and A, Wani S, Sharma P. Incidence of esopha- metaplasia is detected in 98% of cases of
prior diagnosis of Barrett’s oesophagus: geal adenocarcinoma in patients with Bar- neoplastic Barrett’s esophagus. Clin Gas-
a population-based study. Gut 2014 April 3 rett’s esophagus and high-grade dyspla- troenterol Hepatol 2014;12:405-10.
(Epub ahead of print). sia: a meta-analysis. Gastrointest Endosc 74. Titi M, Overhiser A, Ulusarac O, et al.
49. Corley DA, Mehtani K, Quesenberry C, 2008;67:394-8. Development of subsquamous high-grade
Zhao W, de Boer J, Weiss NS. Impact of 62. Overholt BF, Lightdale CJ, Wang KK, dysplasia and adenocarcinoma after suc-
endoscopic surveillance on mortality from et al. Photodynamic therapy with porfimer cessful radiofrequency ablation of Bar-
Barrett’s esophagus-associated esophageal sodium for ablation of high-grade dyspla- rett’s esophagus. Gastroenterology 2012;
adenocarcinomas. Gastroenterology 2013; sia in Barrett’s esophagus: international, 143:564-6.
145:312-9. partially blinded, randomized phase III 75. Gray NA, Odze RD, Spechler SJ. Buried
50. Boerwinkel DF, Swager AF, Curvers trial. Gastrointest Endosc 2005;62:488-98. metaplasia after endoscopic ablation of
WL, Bergman JJ. The clinical consequences [Erratum, Gastrointest Endosc 2006;63: Barrett’s esophagus: a systematic review.
of advanced imaging techniques in Bar- 359.] Am J Gastroenterol 2011;106:1899-908.
rett’s esophagus. Gastroenterology 2014; 63. Shaheen NJ, Sharma P, Overholt BF, 76. Gupta M, Iyer PG, Lutzke L, et al. Re-
146:622-9. et al. Radiofrequency ablation in Barrett’s currence of esophageal intestinal meta-
51. Kastelein F, Biermann K, Steyerberg esophagus with dysplasia. N Engl J Med plasia after endoscopic mucosal resection
EW, et al. Aberrant p53 protein expression 2009;360:2277-88. and radiofrequency ablation of Barrett’s
is associated with an increased risk of 64. Dunbar KB. Endoscopic eradication esophagus: results from a US Multicenter
neoplastic progression in patients with therapy for mucosal neoplasia in Barrett’s Consortium. Gastroenterology 2013;145:
Barrett’s oesophagus. Gut 2013;62:1676- esophagus. Curr Opin Gastroenterol 2013; 79-86.
83. 29:446-53. 77. Hur C, Choi SE, Rubenstein JH, et al.
52. Fritcher EG, Brankley SM, Kipp BR, 65. Larghi A, Lightdale CJ, Memeo L, The cost effectiveness of radiofrequency
et al. A comparison of conventional cytol- Bhagat G, Okpara N, Rotterdam H. EUS ablation for Barrett’s esophagus. Gastro-
ogy, DNA ploidy analysis, and fluores- followed by EMR for staging of high-grade enterology 2012;143:567-75.
cence in situ hybridization for the detec- dysplasia and early cancer in Barrett’s Copyright © 2014 Massachusetts Medical Society.