Content uploaded by Virender K Sharma
Author content
All content in this area was uploaded by Virender K Sharma
Content may be subject to copyright.
Reprint
Endoscopy 2010
October
Volume 42
Page 781–789
Endoscopic radiofrequency
ablation for Barrett’s
esophagus: 5-year outcomes
from aprospective
multicenter trial
D. E. Fleischer, B. F.Overholt, V.K.Sharma,
A. Reymunde, M. B. Kimmey, R. Chuttani,
K. J. Chang, R. Muthasamy, C. J. Lightdale,
N. Santiago, D. K. Pleskow, P. J. Dean,
K. K. Wang
Endoscopy
Official Organ of
the European Society of
Gastrointestinal Endoscopy
(ESGE) and
Affiliated Societies
www.thieme-connect.com/
ejournals
www.thieme.de/endoscopy
© 2010 by
Georg Thieme Verlag KG
Rüdigerstraße 14
D-70469 Stuttgart
ISSN 0013-726X
Reprint with the permission
of the publishers only
Endoscopy_Sonder_silbern_1c_u1u4:Muster_Sonder_silbern_1c_u1u4.qxd 30.09.2010 10:49 Seite 1
Endoscopic radiofrequency ablation for Barrett’s
esophagus: 5-year outcomes from a prospective
multicenter trial*
Authors D. E. Fleischer1, B. F. Overholt2, V. K. Sharma1, A. Reymunde3, M. B. Kimmey4, R. Chuttani5, K. J. Chang6, R. Muthasamy6,
C. J. Lightdale7, N. Santiago3, D. K. Pleskow5,P.J.Dean
8, K. K. Wang9
Institutions Institutions are listed at the end of article.
submitted 23 April 2010
accepted after revision
16 May 2010
Bibliography
DOI http://dx.doi.org/
10.1055/s-0030-1255779
Published online
20 September 2010
Endoscopy 2010; 42:
781–789 © Georg Thieme
Verlag KG Stuttgart · New York
ISSN 0013-726X
Corresponding author
D. E. Fleischer, MD
Mayo Clinic, Scottsdale
13400 East Shea Boulevard
Scottsdale, AZ 85259-5499
USA
Fax: +1-480-301-8673
Fleischer.David@mayo.edu
Original article 781
Introduction
!
Barrett’s esophagus occurs as a result of chronic
injury to the esophageal epithelium by reflux of
gastroduodenal contents associated with gastro-
esophageal reflux disease (GERD) [1, 2]. A diagno-
sis of Barrett’s esophagus is suspected upon dis-
covery of salmon-colored epithelium in the
esophagus and is confirmed by mucosal biopsy
that demonstrates intestinal epithelium contain-
ing goblet cells, referred to as intestinal metapla-
sia [3]. Barrett’s esophagus is categorized endo-
scopically by length and histologically according
to the absence or presence/severity of dysplastic
cellular changes: nondysplastic, low grade dys-
plasia (LGD), or high grade dysplasia (HGD). These
morphological categories represent surrogate
markers of increasing risk of developing esopha-
geal adenocarcinoma (EAC). Hence, medical so-
ciety guidelines have commonly recommended a
strategy of surveillance endoscopy with biopsy
for patients with nondysplastic and LGD Barrett’s
esophagus, to: (i) detect neoplastic progression
prior to EAC, or (ii) detect EAC at a treatable stage
[3–7].
A surveillance strategy for nondysplastic and LGD
Barrett’s esophagus has many recognized limita-
tions: biopsy sampling errors, lack of compliance
with surveillance protocols, cost, cost-utility con-
siderations, and failure to avert EAC in many
cases. Therefore, endoscopic therapies intended
to completely remove or ablate the nondysplastic
and LGD Barrett’s esophagus epithelium have
been evaluated as alternative strategies [8–13].
In a multicenter study, the AIM-II Trial, we have
evaluated endoscopic radiofrequency ablation
Background and study aims: The AIM-II Trial in-
cluded patients with nondysplastic Barrett’s
esophagus (NDBE) treated with radiofrequency
ablation (RFA). Complete eradication of NDBE
(complete response-intestinal metaplasia [CR-
IM]) was achieved in 98.4% of patients at 2.5
years. We report the proportion of patients de-
monstrating CR-IM at 5-year follow-up.
Patients and methods: Prospective, multicenter
US trial (NCT00489268). After endoscopic RFA of
NDBE up to 6 cm, patients with CR-IM at 2.5 years
were eligible for longer-term follow-up. At 5
years, we obtained four-quadrant biopsies from
every 1 cm of the original extent of Barrett’s
esophagus. All specimens were reviewed by one
expert gastrointestinal pathologist, followed by
focal RFA and repeat biopsy if NDBE was identi-
fied. Primary outcomes were (i) proportion of pa-
tients demonstrating CR-IM at 5-year biopsy, and
(ii) proportion of patients demonstrating CR-IM at
5-year biopsy or after the single-session focal RFA.
Results: Of 60 eligible patients, 50 consented to
participate. Of 1473 esophageal specimens ob-
tained at 5 years 85% contained lamina propria
or deeper tissue (per patient, mean 30 [13],
standard deviation [SD] 13). CR-IM was demon-
strated in 92% (46/ 50) of patients, while 8 % (4/
50) had focal NDBE; focal RFA converted all these
to CR-IM. There were no buried glands, dysplasia,
strictures, or serious adverse events. Kaplan-Mei-
er CR-IM survival analysis showed probability of
maintaining CR-IM for at least 4 years after first
durable CR-IM was 0.91 (95% confidence interval
[CI] 0.77–0.97) and mean duration of CR-IM was
4.22 years (standard error [SE] 0.12).
Conclusions: In patients with NDBE treated with
RFA, CR-IM was demonstrated in the majority of
patients (92%) at 5-year follow-up, biopsy depth
was adequate to detect recurrence, and all failures
(4 /4, 100 %) were converted to CR-IM with single-
session focal RFA.
* An oral presentation of this work was given at the Amer-
ican Society for Gastrointestinal Endoscopy (ASGE) Presi-
dential Plenary Session at Digestive Diseases Week
(DDW) 2010.
Fleischer D E. et al. Endoscopic RFA for Barrett’s esophagus: 5-year outcomes…Endoscopy 2010; 42: 781 –789
(RFA) for patients with nondysplastic Barrett’s esophagus
(NDBE), and have previously reported complete eradication of
NDBE (all esophageal biopsies negative for intestinal metaplasia,
termed complete response-intestinal metaplasia [CR-IM]) in
98.4% of patients at 2.5-year follow-up [14]. To assess the long-
er-term durability of complete reversion to a squamous epithe-
lium after ablation, we performed endoscopy and biopsy at 5-
year follow-up in patients from the trial who had shown CR-IM
at 2.5 years.
Patients and methods
!
This is a report of the 5-year outcomes of a prospective cohort
study (ClinicalTrials.gov identifier NCT00489268) conducted at
eight US centers between May 2004 and November 2009. Five of
the study centers were tertiary referral academic teaching hospi-
tals, while three were large community practices with the ex-
perience and staffing for conducting clinical trials. The institu-
tional review board at each site approved the study protocol and
the form provided for informed consent from patients. All the
study participants gave their informed consent.
Patients
In Phase I of this study (from study entry to 2.5-year follow-up),
eligibility criteria included age 18 –75 years and presence of
NDBE (2–6 cm). All patients underwent confirmatory endoscopy
with biopsies from the esophageal body within 6 months of en-
rollment for corroboration that the morphological grade of Bar-
rett’s esophagus was not worse than nondysplastic. Exclusion
criteria were: esophageal stricture or varices, active esophagitis,
previous ablation, previous endoscopic resection, previous radia-
tion therapy to the esophagus, any history of esophageal dyspla-
sia or malignancy, or presence of an implanted electrical device.
Patients were provided with antisecretory medication; oral
esomeprazole (AstraZeneca LP, Wilmington, Delaware, USA) was
used during the first 2.5 years of the study, at 40 mg per day with
dose escalation to 40 mg twice a day for 1 month after ablation.
In Phase II of this study (2.5 years to 5 years), attempts were
made to contact all 60 patients who met eligibility criteria for
the 5-year biopsy. Antisecretory medication type and dosage
were decided at the discretion of the investigator. Patients were
provided with oral esomeprazole 40 mg twice per day for 2
months prior to the 5-year visit to minimize inflammation at
the time of endoscopy with biopsy.
Study devices
In Phase I of the study, endoscopic circumferential RFA was per-
formed with the HALO360 system (BÂRRX Medical, Inc., Sunny-
vale, California, USA) comprising a sizing balloon, a balloon-based
ablation catheter and an energy generator. Endoscopic focal RFA
was done using the HALO90 system, comprising an electrode ar-
ray fitted to the distal end of an endoscope.
Patient flow
At the beginning of the study (Phase I), we enrolled 70 patients
who met all the inclusion criteria for the study and to whom
none of the exclusion criteria applied (
●
"Fig. 1).
After circumferential RFA, a complete response for intestinal me-
taplasia (CR-IM), defined as all esophageal biopsies being nega-
tive for intestinal metaplasia, was seen in 48 of 69 patients
(70%) at 1-year follow-up. After 1-year follow-up, patients with
persistent Barrett’s esophagus or an irregular squamocolumnar
junction received focal RFA. As a result, CR-IM was found in 60
of 61 patients (98 %) at 2.5-year follow-up. In the present trial
(phase II), patients with CR-IM at the 2.5-year follow-up were eli-
gible for 5-year endoscopic biopsy to assess the durability of re-
version to a squamous epithelium.●
"Fig. 2 shows a series of en-
doscopic images from representative patients at various time
points in the study.
5-year endoscopy with biopsy visit
Radial Jaw-4 2.8-mm forceps (Boston Scientific Corp, Natick,
Massachusetts, USA) were provided at each site for obtaining
biopsies at the 5-year visit. Investigators were permitted, at their
discretion, to use other large-jaw forceps including the Olympus
Endojaw 220 FE 2.8-mm forceps and the Olympus FB-13K 3.7-
mm forceps (Olympus America, Center Valley, Pennsylvania,
USA). Biopsy specimens were obtained using these types of biop-
sy forceps, with tissue samples taken from four quadrants per
Fig. 1 AIM-II Trial: enrollment and follow-up of all patients from inception
of trial. Among the 60 patients eligible for the 5-year follow-up phase based
on complete response of intestinal metaplasia (CR-IM) status at 2.5 years,
50 (83 %) patients elected to participate. RFA, radiofrequency ablation.
Original article782
Fleischer D E. et al. Endoscopic RFA for Barrett’s esophagus: 5-year outcomes…Endoscopy 2010; 42: 781 –789
level, beginning at the top of the gastric folds and moving proxi-
mally in 1-cm increments to encompass the entire baseline ex-
tent of Barrett’s esophagus. Additionally, directed biopsy speci-
mens were obtained from any areas that appeared abnormal.
Biopsies distal to the top of the gastric folds were not mandated
in this trial nor were they included in the analysis if obtained. All
biopsy specimens from one level or one focal area were submit-
ted in one jar containing formalin and labelled with the location
of the biopsy as well as study subject identifier codes.
The study protocol did not mandate a specific type of endoscopic
technique or type of endoscopic equipment for the 5-year endos-
copy with biopsy visit. Investigators indicated on each case re-
port form whether or not the following techniques were used:
(i) Lugol chromoendoscopy, (ii) high-definition endoscopy, or
(iii) electronic imaging, such as narrow band imaging (NBI).
Central pathology interpretation and processing
Study specimens were sent in a standardized kit to a central pa-
thology laboratory (Gastrointestinal Pathology, LLC, Memphis,
Tennessee, USA). The formalin-fixed specimens from each con-
tainer were embedded in paraffin (one block per jar), affixed to
slides, and stained with hematoxylin and eosin. One slide repre-
sented each level or focal area sampled.
A board-certified pathologist (P.J.D.) specializing in gastrointesti-
nal pathology evaluated each specimen on each slide and cate-
gorized each according to tissue type (intestinal metaplasia ver-
sus squamous), dysplasia or cancer in intestinal metaplasia if
present, and biopsy depth. Depth was defined as the deepest tis-
sue layer present in each specimen: epithelium, lamina propria,
muscularis mucosae, or submucosa. Biopsy specimens were con-
sidered to contain subepithelial structures if they contained la-
mina propria, muscularis mucosae, or submucosa. Each biopsy
specimen was also evaluated for the presence of buried glandular
mucosa, defined as “any specialized columnar epithelium cov-
ered by a layer of squamous epithelium with no communication
with the surface”[15]. Finally, each tissue block was evaluated for
the presence of inf lammation. All data were entered into a stand-
ardized pathology case report form.
Salvage focal ablation
Patients in whom intestinal metaplasia was detected at the 5-
year follow-up underwent endoscopy with focal RFA 1 month la-
ter, followed by repeat biopsy 2 months after RFA to assess histo-
logical response. Treatment settings were 12 J/cm2and 40 W/
cm2. Areas positive for intestinal metaplasia at 5-year biopsy
were targeted, along with the region of the top of the gastric
folds. Each location was treated twice successively, followed by
cleaning of the coagulum from the treated area and electrode, fol-
lowed by treatment of each location twice again.
Outcome measures
The primary study outcomes were defined a priori as: (i) the pro-
portion of patients with CR-IM at the 5-year biopsy visit, and (ii)
the proportion of patients demonstrating CR-IM at the 5-year
biopsy visit or at the biopsy visit after single-session salvage focal
RFA.
Secondary outcomes were defined a priori as: (i) proportion of 5-
year failures converted to CR-IM after single-session salvage focal
RFA; (ii) biopsy depth; (iii) prevalence of buried glandular muco-
sa; (iv) prevalence of dysplasia; (v) Kaplan-Meier CR-IM survival
analysis; and (vi) adverse events.
All adverse events and serious adverse events were recorded on a
standardized case report form. Specifically, a stricture in this
study was defined a priori as any narrowing of the esophageal lu-
men in the area of treatment causing symptoms or requiring di-
lation. All case repor t forms at each site were monitored through-
out the study period and queries issued in cases of noncompli-
ance.
Statistical analysis
Data analysis was performed using SAS software, version 9 (SAS
Institute, Cary, North Carolina, USA). The study population for the
primary analysis included all eligible patients who gave their in-
formed consent for the 5-year follow-up. Baseline patient data for
the eligible patient group were assessed to detect differences be-
tween participants and decliners. Biopsy sample characteristics
were compared for patients with CR-IM at 5 years vs. those with
failure at 5 years, as well as for those with failure at 5 years vs.
post-salvage RFA after 5 years. Fisher’s exact test was used to
compare categorical variables. For continuous variables, a ttest
was used for variables which did not display non-normality for
either treatment group (Shapiro-Wilk Pvalue > 0.05) and the
Mann-Whitney test was used otherwise. All tests were two-sided
and Pvalues less than 0.05 were considered statistically signifi-
cant.
In participants, the durability of CR-IM was assessed using the
Kaplan-Meier survival curve with 95 % confidence intervals (CIs)
generated using Greenwood’s formula for computing standard
errors. In this analysis, “time-zero”(start of CR-IM period for the
survival analysis calculation) was the first durable CR-IM after
enrollment (defined as either the 2.5-year time point or an earli-
er time point at which the first of two consecutive study biopsy
sessions demonstrated CR-IM [6 or 12 months]). Sinceall partici-
pants in this phase of the study had CR-IM at 2.5 years, time-zero
was therefore either the 6-month, 12-month, or 2.5-year follow-
Fig. 2 Endoscopic images from representative study patients. aA 6-cm length of Barrett’s esophagus without dysplasia at baseline. bAppearance immedi-
ately following circumferential ablation. cFocal radiofrequency device mounted on the tip of the gastroscope, with ablation zone seen distally. Ablation was
done for residual intestinal metaplasia after 1-year follow-up. dEndoscopy demonstrating no Barrett’s esophagus at 2.5 years with all biopsies negative for
intestinal metaplasia. eRetroflexed view of gastroesophageal junction in patient at 5-year follow-up, showing neosquamous epithelium extending to the car-
dia. The esophagus was normal on endoscopy, with all biopsies negative for intestinal metaplasia.
Original article 783
Fleischer D E. et al. Endoscopic RFA for Barrett’s esophagus: 5-year outcomes…Endoscopy 2010; 42: 781 –789
up visit (endoscopy with biopsy) for all patients. For the purpose
of this survival analysis, where patients were found to have re-
current intestinal metaplasia after 2.5 years it was assumed that
for half the elapsed time between endoscopies their status had
been CR-IM and for the other half intestinal metaplasia had
been present.
Results
!
For this 5-year follow-up of the AIM-II trial there were 60 eligible
patients from eight US centers, of whom 50 (83 %) were willing to
participate and gave signed consent for the extended follow-up
described in the approved protocol (
●
"Fig. 1). The baseline pa-
tient characteristics of the eligible (n = 60), participant (n = 50),
and declining (n = 10) subgroups are shown in ●
"Table 1. Com-
paring baseline characteristics of participants (n = 50) versus de-
cliners (n = 10), only entr y age was different between the groups,
with participants being younger than decliners (P< 0.05).
At the 5-year biopsy visit, the mean (SD) number of biopsies ob-
tained per patient was 30 (13) comprising a total of 1473 biopsies
(
●
"Table 2).
Regarding the first primary outcome, 92% (46 / 50) of patients
showed CR-IM at the 5-year biopsy visit. Four patients (8%) had
intestinal metaplasia at 5 years (
●
"Fig. 3).
In these four patients, a mean (SD) of 32 (17) esophageal speci-
mens were obtained from each patient; one patient had three
specimens positive for intestinal metaplasia and three patients
each had one such specimen. Five of the 6 specimens positive for
intestinal metaplasia were located within 1 cm of the squamoco-
lumnar junction near the top of the gastric folds. These four pa-
tients received a single session of focal RFA 1 month after biopsy
and all (4/ 4, 100%) were found to have CR-IM status upon subse-
quent biopsy at 2 months after RFA. Therefore, regarding the sec-
Table 1 Radiofrequency ablation of nondysplastic Barrett’s esophagus. Patient demographic data at 5-year follow-up.
Parameter All those eligible
for 5-year follow-up
Participating patients Patients declining
participation
Number of patients 60 50 10
Gender, male/female 44 / 16 37 / 13 7 / 3
Ethnicity
White
Black
Hispanic/Latino
42 (70%)
2(3%)
16 (27%)
36 (72 %)
1(2%)
13 (26 %)
6 (60.0 %)
1 (10.0 %)
3 (30.0 %)
Baseline data
Age, years
Mean (SD)
Range
Bodyweight,mean(SD),lbs
History of GERD, n (%)
Barrett esophagus length, mean (SD), cm
Hiatal hernia
Present/Absent, n/n
Length,mean(SD),cm
55.8 (11.1)
26–79
177.8 (38.8)
60 (100%)
3.2 (1.3)
50/10
2.5 (1.1)
54.3 (10.8)*
26 –72
177.6 (38.0)
50 (100 %)
3.1 (1.3)
43 / 7
2.6 (1.2)
63.6 (9.8)*
48 –79
179.0 (44.8)
10 (100.0 %)
3.7 (1.4)
7/3
1.9 (0.7)
RFA treatments in prior study phase, mean (SD), n
Tot al
Circumferential
Focal
3.4 (1.0)
1.6 (0.5)
1.8 (0.7)
3.4 (0.9)
1.5 (0.5)
1.9 (0.7)
3.3 (1.1)
1.6 (0.5)
1.7 (0.7)
SD, standard deviation; GERD, gastroesophageal reflux disease
*P< 0.05, patients who participated vs. those who declined participation
Table 2 Biopsy characteristics at 5-year follow -up after radiofrequency ablation (RFA) of nondysplastic Barrett’s esophagus. Data are shown for all par ticipants, for
those with persisting complete response of intestinal metaplasia (CR-IM), for those with treatment failure, and following salvage RFA for those with treatment
failure.
Parameter 5-year biopsy Following salvage RFA
All participants CR-IM Treatment failures
Number of patients 50 46 4 4
Total biopsy specimens, n 1473 1347 126 112
Specimens per patient, mean (SD), n 30 (13) 29 (13) 32 (17) 28 (17)
Specimens with lamina propria or deeper
Tot al , n
Proportion, %
Number per patient, mean (SD), n
Proportion per patient, mean (SD), %
1255
85 %
25 (10)
87 % (11 %)
1145
85 %
25 (10)
87 % (11 %)
110
87 %
28 (15)
88 % (5 %)
99
88%
24 (11)
94 % (13%)
Specimens containingburied glandular mucosa, n 0 0 0 0
Specimens with dysplasia or cancer 0 0 0 0
No significant differences for any variables
Original article784
Fleischer D E. et al. Endoscopic RFA for Barrett’s esophagus: 5-year outcomes…Endoscopy 2010; 42: 781 –789
ond primary outcome, all patients (50 /50, 100 %) were in CR-IM
at either the 5-year biopsy visit or after a single salvage focal ab-
lation after the 5-year biopsy visit.
In the biopsies obtained at the 5-year visit, subepithelial struc-
tures were present in 85 % of the specimens (
●
"Fig. 4).
In the four patients with intestinal metaplasia at 5-year biopsy,
the mean (SD) number of biopsies per patient obtained 2 months
after salvage RFA was 28 (17) (total 112 biopsies), with subepi-
thelium present in 88 % of these specimens. No evidence of bur-
ied glandular mucosa or dysplasia was detected on any 5-year
biopsy or post-salvage biopsy.
At the 5-year visit, the Radial Jaw-4 2.8-mm biopsy forceps was
used in 72% of patients (36 / 50), the Olympus Endojaw 220-FE
2.8-mm forceps in 26% (13/ 50), and the Olympus FB-13K 3.7-
mm device in 2% (1 /50). There was no difference in biopsy speci-
men depth according to type of biopsy forceps used.
At the 5-year follow-up biopsies investigators used high-defini-
tion endoscopy in 35/50 of patients (70 %) and electronic ima-
ging (such as NBI) in 17 / 50 (34 %). None used Lugol chromoen-
doscopy. High-definition endoscopy was used in 32 / 46 (70 %) of
the CR-IM patients and 3 / 4 (75%) of those with intestinal meta-
plasia. Corresponding values for electronic imaging were 16 / 46
(35%) and 1 /4 (25 %).
The Kaplan-Meier survival curve for the secondary outcome of
CR-IM survival is shown in●
"Fig. 5.
All par ticipants (n = 50) are represented in this analysis. The
probability of maintaining CR-IM for at least 4 years after the f irst
durable CR-IM was 0.91 (95 % confidence interval [CI] 0.77 –0.97).
The mean (standard error [SE]) duration of CR-IM in participants
was 4.22 (0.12) years. Included in this survival analysis is one pa-
tient with CR-IM status at 5 years, but who had an off-protocol
biopsy between the 2.5- and 5-year study visits that detected
nondysplastic intestinal metaplasia in one specimen; this was
followed by focal ablation. Although the biopsy was obtained out-
side of the study protocol and the histological findings were not
reviewed by the study pathologist, treatment is considered to
have failed in the survival analysis. However it is not considered
to have failed with regard to the study’s primary end point (CR-
IM at 5 years). Of note, amongst the remaining 49 patients, 33
also had a nonprotocol endoscopy with biopsy procedures as
part of standard care during the interval between 2.5 and 5 years
with none having intestinal metaplasia in esophageal biopsies
and none undergoing ablation in the esophagus.
At the 5-year biopsy visit, 45 patients reported taking oral
esomeprazole 40 mg twice per day, 3 patients oral omeprazole
40 mg twice per day, 1 patient oral lansoprazole 30 mg per day,
and 1 patient no antisecretory medication. Of the 50 patients,
three (6 %) had either grade A (n = 1) or grade B (n = 2) erosive
esophagitis (Los Angeles Classification) at the 5-year visit, while
the remainder had no signs of erosive esophagitis. All three pa-
tients with erosive esophagitis were taking oral esomeprazole
40 mg twice per day. None of the four patients subsequently no-
ted to have intestinal metaplasia at 5-year biopsy had erosive
esophagitis during that visit.
There were no differences in baseline demographic data or Bar-
rett’s esophagus characteristics between patients with CR-IM
status or treatment failure at 5 years. In addition, there was no
difference between these groups in the mean (SD) number of
biopsies collected (29 [13] vs. 32 [17]) or in the percentage per
patient of biopsy specimens containing subepithelium (87 % vs.
88%). Similarly in the treatment failure group, comparing the 5-
year biopsy visit with the follow-up biopsy visit 2 months after
salvage focal RFA, there was no difference between the mean
number of biopsies per patient (32 [17] vs. 28 [17]) or the propor-
tion per patient of biopsies containing subepithelium (88 % vs.
94%) (
●
"Table 2).
No strictures were noted at any follow-up endoscopy. One pa-
tient reported a globus sensation 1 week after salvage focal abla-
tion, which resolved without intervention. There were no other
adverse events.
Discussion
!
Esophageal mucosa demonstrating glandular epithelium with
goblet cells defines the entity Barrett’s esophagus [3]. The mor-
phological categories in Barrett’s esophagus of nondysplastic tis-
sue, LGD, and HGD are surrogate markers of increasing risk for in-
cident EAC, although genetic alterations enabling neoplastic
behavior and progression are present in NDBE prior to develop-
ment of a dysplasia phenotype [16 –20]. While nondysplastic tis-
sue is associated with the lowest risk for EAC of the nondysplas-
tic/LGD/HGD categories (0.5 %–0.6 % incidence per patient year of
follow-up; 5% –8% lifetime risk), its detection prompts the initia-
tion of a surveillance endoscopy and biopsy regimen intended to
detect neoplastic progression [21–24]. Recognized limitations of
Fig. 3 Photomicrographs of esophageal biopsy specimens obtained from a patient with intestinal metaplasia at 5-year follow-up af ter radiofrequency ablation
(RFA), i. e. treatment failure with regard to primary outcome (both images hematoxylin and eosin [H&E]; original magnification × 200). aIntestinal metaplasia
seen in a biopsy obtained in the distal esophagus at the top of the gastric folds. Biopsy depth extends to the muscularis mucosae and intestinal metaplasia is
present at the sur face layer. bEsophageal biopsy 2 months after single-session focal RFA at 5 years for the residual intestinal metaplasia. Neosquamous epi-
thelium is completely re-established and there was no evidence of intestinal metaplasia in any of 30 biopsies. The biopsy depth extends to the muscularis
mucosae and shows completely squamous epithelium.
Original article 785
Fleischer D E. et al. Endoscopic RFA for Barrett’s esophagus: 5-year outcomes…Endoscopy 2010; 42: 781 –789
a surveillance strategy include biopsy sampling error, lack of
compliance with surveillance protocols, cost, cost-utility issues,
and failure to avert EAC in many cases [8–13]. An ideal manage-
ment paradigm for an nondysplastic population in the future
might be to risk-stratify patients by assaying for a genotype asso-
ciated with propensity for neoplastic progression, and then to
eradicate the NDBE in those patients at highest risk, with surveil-
lance or no action in those patients at lower or zero risk. The abil-
ity to risk-stratify a Barrett’s esophagus population according to
genotype has proven elusive, and is not yet possible.
More, however, is now known regarding the safety and short- to
intermediate-term efficacy of a number of endoscopic techniques
used to eradicate the Barrett’s esophagus epithelium. We pre-
viously reported the 1- and 2.5-year safety and efficacy outcomes
of the AIM-II trial, which applied endoscopic RFA in patients with
up to 6 cm of NDBE. Circumferential ablation achieved CR-IM in
70% of patients at 1-year follow-up [25]. Thereafter, focal abla-
tion of residual intestinal metaplasia achieved CR-IM in 98% of
these patients at 2.5-year follow-up [14]. There were no stric-
tures, neoplastic progression, buried glands, or serious adverse
events. Others studies of RFA applied in nondysplastic, LGD, and
HGD Barrett’s esophagus populations, including a randomized,
sham-controlled multicenter trial, report similarly high rates of
CR-IM (or complete response for dysplasia) with acceptable safe-
ty profiles [26 –40].
Since Barrett’s esophagus is a chronic disease state with long-
term implications for neoplastic progression, information on
longer-term outcomes regarding the durability of CR-IM after
RFA are essential to assess the utility of this strategy. In this
long-term follow-up of patients from the AIM-II trial, we con-
ducted endoscopy and biopsy at 5-year follow-up and found
that in 92% of patients (46 / 50) CR-IM was maintained. We uti-
lized the more objective measurement of histological findings
rather than endoscopic observation. A Kaplan-Meier survival
curve showed the probability of maintaining CR-IM for at least 4
years after first durable CR-IM was 0.91 (95 %CI 0.77 –0.97), while
the mean (SE) duration of CR-IM was 4.22 (0.12) years. Recurrent
NDBE was identified in four patients at 5 years (8%) in 6/128
esophageal specimens, most of which were located within 1 cm
of the top of the gastric folds. These patients underwent single-
session focal RFA and all showed CR-IM thereafter. No dysplasia,
stricture, serious adverse event, or buried glandular mucosa was
noted. Assessment of the depth of esophageal biopsy specimens
showed that 85% of the 1473 esophageal biopsy specimens con-
tained lamina propria or deeper structures, a depth considered
adequate for detection of buried glandular mucosa. Our 5-year
data with RFA can be compared with long-term outcomes of
other ablative modalities. A technical review by Wani et al. re-
ported recurrence rates of up to 68 % in patients with NDBE treat-
ed with argon plasma coagulation (APC) [41,42] . While long-
term data for multipolar electrocoagulation (MPEC) are limited,
one case series described recurrence at 2-year follow-up in 27 %
Fig. 4 Photomicro-
graphs of esophageal
biopsy specimens
obtained from a patient
deemed to have com-
plete response-intes-
tinal metaplasia (CR-IM)
at 5 years (hematoxylin
and eosin [H&E]; origi-
nal magnification
× 200). Depth of speci-
men categorized as:
afull epithelium;
blamina propria;
cmuscularis mucosae;
dsubmucosa.
Fig. 5 Kaplan-Meier survival curve for duration of complete response of
intestinal metaplasia (CR-IM). All par ticipants (n = 50) are represented in
this analysis. The probability of maintaining CR-IM for at least 4 years after
first durable CR-IM is 0.91 (95 %CI 0.77 –0.97). The mean duration of CR-IM
in participants was 4.22 years (standard error [SE] 012).
Original article786
Fleischer D E. et al. Endoscopic RFA for Barrett’s esophagus: 5-year outcomes…Endoscopy 2010; 42: 781 –789
of patients with baseline nondysplastic or LGD Barrett’s esopha-
gus [43]. In the PHOBAR trial, which evaluated the safety, efficacy
and durability of photodynamic therapy in patients with HGD in
Barrett’s esophagus, there was a 48 % probability of maintaining
complete eradication of HGD at 5 years [44].
Several cost-utility models indicate that endoscopic ablation is a
cost-effective strategy for Barrett’s esophagus with nondysplastic
tissue, LGD, and HGD. The cost-utility evidence is strongest for
the highest risk lesions, i. e., LGD/HGD, with ablation being pre-
ferred or dominant to other strategies. Specifically regarding no
dysplasia, however, Inadomi et al. reported that ablation is the
preferred strategy for Barrett’s with no dysplasia if a permanent
CR-IM is achieved in 40% of patients and surveillance discontin-
ued after CR-IM. This model’s threshold for cost–effectiveness re-
presents a lower CR-IM (40%) than reported in published trials of
RFA, yet sets a high threshold for the durability of CR-IM and re-
quires cessation of surveillance in CR-IM patients, a step that may
not be acceptable to patients or physicians. However, the next
most preferred strategy from this model (over that of surveil-
lance) is ablation in which CR-IM is achieved in 40 % and surveil-
lance is, in fact, continued in all patients [45]. In a separate model,
Das et al. found that, if CR-IM was achieved in 50% of patients,
RFA yielded the most quality-adjusted life-years (QALYs) when
compared with strategies of no intervention (natural history)
and management following current American College of Gastro-
enterology surveillance guidelines [46]. Both of these models in-
dicate that durability of CR-IM after ablation of NDBE has an im-
portant influence on the cost-utility of the strategy. The persist-
ence of CR-IM in 92% of NDBE patients at 5-year follow-up in the
present study, as well as the successful re-establishment of CR-IM
in all patients with treatment failure by single-session focal RFA
after 5 years, is promising in that it permits us to apply these
models for the first time in clinical practice and to consider RFA
as a viable alternative to surveillance alone for NDBE.
As Barrett’s esophagus is a surrogate marker identifying risk for
neoplasia progression and EAC incidence, therapeutic manage-
ment strategies should be designed to avert these outcomes. In a
meta-analysis, Wani et al. reported that the incidence of EAC in
an observed NDBE population was 0.598 % per patient per year
of follow-up, while the incidence of EAC in ablated NDBE popula-
tions (regardless of CR-IM outcome) was 0.163% [21]. In this a-
nalysis, ablation affords an absolute risk reduction (ARR) of
0.435% per patient per year of follow-up in patients with NDBE,
demonstrating that ablation does avert EAC for this population.
Using these data for NDBE, the number needed to treat (NNT) to
avoid one EAC over a 1-year period is 230 (1/ARR), while the NNT
to avoid one EAC over 5 years is 46. Sharma et al. found that non-
dysplastic tissue progresses to HGD at a rate of 0.9 % per patient
per year of follow-up [22]. Using the natural history progression
rates from Wani et al. and Sharma et al. for NDBE to EAC (0.6%)
and HGD (0.9%), we might have expected seven cases of EAC or
HGD in this trial over the 5-year study period (if the patients
were observed and not treated), rather than the absence of HGD
or EAC cases that was noted.
Buried glandular mucosa is an important consideration in the
management of Barrett’s esophagus. This entity is present in a
significant proportion of ablation-naïve Barrett’s esophagus pa-
tients (25% –39 %), as well as in post-PDT (51%) and post-APC
(44%) patients [26, 42, 47, 48]. In the present study, we collected
1473 esophageal biopsies at 5-year follow-up and 3930 biopsies
at earlier visits and found no evidence of buried glandular muco-
sa. Others have reported similarly low rates of this finding after
RFA. Shaheen et al. found that 25% of LGD and HGD patients had
buried glands prior to RFA, while at 1-year buried glands were
present in 5% of RFA patients and 40% of sham patients [26]. In
our study, and in the randomized controlled trial of Shaheen et
al., post-RFA biopsy specimens contained lamina propria or dee-
per structures in 85 % and 79 % of cases, respectively, suggesting
that biopsy depth after RFA is adequate for detection of buried
glandular mucosa and that occult buried glandular mucosa after
RFA is unlikely [49].
Strengths of this study include the prospective, multicenter de-
sign, long duration of follow-up, large number of biopsy speci-
mens obtained, standardized biopsy sampling protocol, use of
large-jaw biopsy forceps, use of a single expert gastrointestinal
pathologist for histological interpretation, objective histological
outcomes, and use of complete histological response as the pri-
mary outcome.
This study has some possible limitations:
"Regarding patient attrition, the original patient group com-
prised 70 patients, whereas 69 and 61 patients were available
for evaluation at 1 and 2.5 years, respectively. There were 60
eligible patients for the present study (those with CR-IM at 2.5
years), while 50 (83%) provided consent and participated.
Nonparticipants were older than participants. If, because of
age or other unmeasured factors, nonparticipants were three
times more likely to have treatment failure (a very conserva-
tive estimate given that age was not associated with response
to therapy in this study), our 5-year CR-IM outcome would
drop nominally from 92% (46 /50) to 82 % (49/ 60). Therefore,
it is unlikely that attrition significantly impacted our primary
findings.
"One patient with CR-IM at 5 years had an off-protocol biopsy
between the 2.5-year and 5-year study visits that showed
nondysplastic intestinal metaplasia in one specimen obtained
from the region of the gastric folds and the patient was treat-
ed with focal ablation. Although the biopsy was obtained out-
side of the protocol and not reviewed histologically by the
study pathologist, to present the data transparently and as
conservatively as possible, this patient was considered to have
treatment failure in the Kaplan-Meier CR-IM survival analysis.
No other patients underwent ablation between 2.5 and 5
years. If we considered this single patient as having treatment
failure for the primary analysis, the 5-year CR-IM would drop
nominally from 92% (46 / 50) to 90% (45 /50).
"Another possible limitation of this study relates to the Kaplan-
Meier CR-IM durability analysis and our methodology for as-
signment of a “time zero”for durable CR-IM for each patient.
After the 1-year follow-up, focal RFA was applied at 2-month
intervals (maximum of 3 sessions) in patients with intestinal
metaplasia (1-year failure) as well as in patients with an ir-
regular squamocolumnar junction (without confirmation of
intestinal metaplasia) with the aim of optimizing efficacy
outcomes at the 2.5-year follow-up. If some of the focal RFA
sessions in patients without confirmation of intestinal meta-
plasia actually ablated minute residual foci of intestinal meta-
plasia, then it is possible that our durability analysis overesti-
mated the true duration of CR-IM in these patients by setting
the “time zero”point too early. While we do not believe that
this would significantly affect the durability curve, an optimal
study design might have required histological confirmation of
intestinal metaplasia prior to any focal ablation after the
1-year follow-up.
Original article 787
Fleischer D E. et al. Endoscopic RFA for Barrett’s esophagus: 5-year outcomes…Endoscopy 2010; 42: 781 –789
"The study had no concurrent control arm, although the objec-
tive histological end points mitigate the likelihood of bias due
to lack of controls at 5 years.
"There was no standardized post 2.5-year antisecretory medi-
cation regimen. Up to 2.5 years, all patients were provided
with oral esomeprazole 40 mg per day (with escalation to
twice per day for 1 month post-RFA). The inability to assess
adequacy of acid suppression and compliance with medica-
tion during the post 2.5-year period, however, limits our abil-
ity to draw conclusions about the role of these factors in dis-
ease recurrence or persistent cure. However, given the high
degree of maintenance of CR-IM at 5 years, despite the het-
erogeneity of antisecretory regimens in the post 2.5-year fol-
low-up period, it is unlikely that the composition of the anti-
secretory regimen is an important predictor of CR-IM dur-
ability.
"There is an inherent lack of precision in identifying the precise
location of the distal terminus of the esophagus and in accu-
rately distinguishing this from the proximal extent of the
stomach. This may be important regarding accurate assess-
ment of the presence or absence of intestinal metaplasia in
the esophagus after ablative therapy. Our biopsy methodology
in this trial used the top of the gastric folds as the landmark for
the distal terminus of the esophagus, and we obtained biop-
sies from this location (and proximally at 1-cm inter vals) in all
patients at all follow-up visits. It is possible that this metho-
dology could: (i) overestimate the presence of intestinal me-
taplasia if these distal biopsies were actually from the un-
treated gastric cardia and if they contained intestinal meta-
plasia (a common finding); or (ii) underestimate the presence
of intestinal metaplasia if the location of the gastric folds was
located too proximally, as we would miss sampling a portion
of the distal, previously treated esophagus. The importance of
thoroughly assessing the distal esophagus is emphasized by
our finding that 5 /6 biopsy specimens positive for intestinal
metaplasia in the patients with treatment failure were located
within 1 cm of the top of the gastric folds.
"The interval of 2 months from salvage RFA to subsequent
biopsy to assess CR-IM after salvage was short. It is possible
that after salvage RFA occult intestinal metaplasia was present
that would have been detected after additional time or with
further biopsy sessions. However, this only affected four pa-
tients (8 %) and the salvage outcome was not used for calcula-
tion of the main study end point, that of CR-IM at 5 years.
"A final possible limitation is that biopsy forceps were not
standardized for all cases, with the Radial Jaw-4 2.8-mm for-
ceps employed in two-thirds of the biopsy visits and the En-
dojaw 220 FE 2.8-mm or Olympus FB-13K 3.7-mm forceps
used in one-third. However biopsy depth did not differ ac-
cording to forceps type, so this is unlikely to be a confounding
variable.
In summary, this is the first report of 5-year CR-IM outcomes
related to RFA for Barrett’s esophagus. In our patients from the
AIM-II Trial who had NDBE treated previously with RFA, CR-IM
persisted in the majority of patients (92 %) at 5-year follow-up,
the biopsy depth was adequate to detect recurrence, and all treat-
ment failures were converted to CR-IM with single-session focal
RFA (4/ 4, 100%). Kaplan-Meier survival analysis of CR-IM dur-
ability demonstrated that the probability of maintaining CR-IM
for at least 4 years is 0.91 (95 %CI 0.77 –0.97) and mean (SE) dura-
tion of CR-IM is 4.22 (0.12) years. There were no buried glands,
dysplasia, strictures or serious adverse events. These long-term
data have important implications for the clinical management of
patients with NDBE and perhaps also those with LGD/HGD. If ear-
ly recurrence of Barrett’s esophagus had been common in these
patients, the value of this therapeutic intervention for Barrett’s
esophagus (specifically, nondysplastic) would be called into
doubt. Our report of CR-IM at 5 years in the context of the many
other studies reporting favorable outcomes related to the safety,
efficacy, cost-utility, and reduction in neoplastic progression lend
further credence for a role of RFA in the treatment of NDBE.
Acknowledgments
!
This research was supported at each research institution by a
study grant from BÂRRX Medical Inc., Sunnyvale, California,
USA. This research was also supported by the Investigator-Spon-
sored Study Program of AstraZeneca (AstraZeneca LP, Wilming-
ton, Delaware, USA).
Competing interests: D.E.F., B.F.O., V.K.S., R.C., R.M., C.J.L., and D.K.
P. report having received professional lecture fees from BÂRRX
Medical. K.J.C. reports a minor equity ownership in, a royalty
agreement with, and receipt of professional lecture fees from
BÂRRX Medical. A.R., M.B.K., N.S., P.J.D., and K.K.W. report no con-
flict of interest and no disclosures
Institutions
1Mayo Clinic, Scottsdale, Arizona, USA
2Gastrointestinal Associates, Knoxville, Tennessee, USA
3Ponce Gastroenterology, Ponce, Puerto Rico, USA
4Tacoma Digestive Disease Center, Tacoma, Washington, USA
5Beth Israel Deaconess Medical Center, Boston, Massachusetts, USA
6University California, Irvine, California, USA
7Columbia Presbyterian Medical Center, New York, USA
8GI Pathology, Memphis, Tennessee, USA
9Mayo Clinic, Rochester, Minnesota, USA
References
1Haggitt RC. Barrettʼs esophagus: pathogenesis, dysplasia, and adeno-
carcinoma. Hum Pathol 1994; 25: 982 –993
2Winters C Jr., Spurling TJ, Chobanian SJ et al. Barrett’s esophagus: a pre-
valent, occult complication of gastroesophageal reflu x disease. Gastro-
enterology 1987; 92: 118 –124
3Wang KK, Sampliner RE. Updated Guidelines 2008 for the diagnosis,
surveillance and therapy of barrett’s esophagus. Am J Gastroenterol
2008; 103: 788–797
4Sharma P, McQuaid K, Dent J et al. A critical review of the diagnosis and
management of Barrett’s esophagus: the AGA Chicago Workshop. Gas-
troenterology 2004; 127: 310 –330
5Watson A, Heading RC, Shepherd NA. Guidelines for the diagnosis and
management of Barrett’s columnar-lined oesophagus: a report of the
Working Party of the British Society of Gastroenterolog y. London: Br it-
ish Society of Gastroenterology, 2005
6Boyer J, Laugier R, Chemali M et al. French Society of Digestive Endosco-
py SFED guideline: monitoring of patients with Barrett’s esophagus.
Endoscopy 2007; 39: 840–842
7Society for Surgery of the Alimentar y Tract. SSAT patient care guide-
lines. Management of Barrett’s esophagus. J Gastrointest Surg 2007;
11: 1213–1215
8Inadomi JM. Surveillance in Barrett’s esophagus: a failed premise. Keio
J Med 2009; 58: 12 –18
9Shaheen NJ, Provenzale D, Sandler RS. Upper endoscopy as a screening
and surveillance tool in esophageal adenocarcinoma: a review of the
evidence. Am J Gastroenterol 2002; 97: 1319 –1327
10 Montgomery E, Bronner MP, Goldblum JR et al. Reproducibility of the di-
agnosis of dysplasia in Barrett esophagus: a reaffirmation. Hum Pathol
2001; 32: 368–378
Original article788
Fleischer D E. et al. Endoscopic RFA for Barrett’s esophagus: 5-year outcomes…Endoscopy 2010; 42: 781 –789
11 Reid BJ, Haggitt RC, Rubin EC et al. Observer variation in the diagnosis of
dysplasia in Barrett’s esophagus. Human Pathol 1988; 19: 166–178
12 Alikhan M, Rex D, Khan A et al. Variable pathologic interpretation of co-
lumnar lined esophagus by general pathologists in community prac-
tice. Gastrointest Endosc 1999; 50: 23–26
13 Abrams JA, Kapel RC, Lindberg GM et al. Adherence to biopsy guidelines
for Barrett’s esophagus surveillance in the community setting in the
United States. Clin Gastroenterol Hepatol 2009; 7: 736–742
14 Fleischer DE, Overholt BF, Sharma VK et al. Endoscopic ablation of Bar-
rett’s esophagus: a multicenter study with 2.5-year follow-up. Gastro-
intest Endosc 2008; 68: 867 –876
15 Dulai GS, Jensen DM, Cortina G et al. Randomized trial of argon plasma
coagulation vs. multipolar electrocoagulation for ablation of Barrett’s
esophagus. Gastrointest Endosc 2005; 61: 232 –240
16 Fitzgerald RC. Complex diseases in gastroenterology and hepatology:
GERD, Barrett’s, and esophageal adenocarcinoma. Clin Gastroenterol
Hepatol 2005; 3: 529–537
17 Rabinovitch PS, Longton G, Blount PL et al. Predictors of progression in
Barrett’s esophagus III: baseline flow cytometric variables. Am J Gas-
troenterol 2001; 96: 3071 –3083
18 Reid BJ, Levine DS, Longton G et al . Predictors of progression to cancer in
Barrett’s esophagus: baseline histology and flow cytometry identify
low- and high-risk patient subsets. Am J Gastroenterol 2000; 95:
1669–1676
19 Wijnhoven BP, Hussey DJ, Watson DI et al. MicroRNA profiling of Bar-
rett’s oesophagus and oesophageal adenocarcinoma. Br J Surg 2010;
97: 853–861
20 Huang Q, Hardie LJ. Biomarkers in Barrett’s oesophagus. Biochem Soc
Trans 2010; 38: 343–347
21 Wani S, Puli SR, Shaheen NJ et al. Esophageal adenocarcinoma in Bar-
rett’s esophagus after endoscopic ablative therapy: a meta-analysis
and systematic review. Am J Gastroenterol 2009; 104: 502 –513
22 Sharma P, Falk GW, Weston AP et al. Dysplasia and cancer in a large
multicenter cohort of patients with Barrett’s esophagus. Clin Gastro-
enterol Hepatol 2006; 4: 566–572
23 Shaheen NJ, Crosby MA, Bozymski EM, Sandler RS. Is there publication
bias in the report ing of cancer risk in Barrett’s esophagus? Gastroente-
rology 2000; 119: 333 –338
24 Anderson LA, Mur ray LJ, Murphy SJ et al. Mortality in Barrett’s oesoph-
agus: results from a population based study. Gut 2003; 52: 1081 –1084
25 Sharma VK, Wang KK, Overholt BF et al. Balloon-based, circumferential,
endoscopic radiofrequency ablation of Barrett’s esophagus: 1-year fol-
low-up of 100 patients. Gastrointest Endosc 2007; 65: 185 –195
26 Shaheen NJ, Sharma P, Overholt BF et al. Radiofrequency ablation in Bar-
rett’s esophagus with dysplasia. N Engl J Med 2009; 360: 2277 –2288
27 Pouw RE, Wirths K, Eisendrath P et al. Efficacy of radiofrequency abla-
tion combined with endoscopic resection for Barrett’s esophagus with
early neoplasia. Clin Gastroenterol Hepatol 2010; 8: 23 –29
28 Lyday WD, Corbett FS, Kuperman DA et al. Radiofrequency ablation of
Barrett’s esophagus: outcomes of 429 patients from a multicenter
community practice registry. Endoscopy 2010; 42: 272 –278
29 dos Santos RS, Bizekis C, Ebright M et al. Radiofrequency ablation for
Barrett’s esophagus and low-grade dysplasia in combination with an
antireflux procedure: a new paradigm. J Thorac Cardiovasc Surg
2010; 139: 713–716
30 Velanovich V. Endoscopic endoluminal radiofrequency ablation of Bar-
rett’s esophagus: initial results and lessons learned. Surg Endosc 2009;
23: 2175–2180
31 Sharma VK, Kim HJ, Das A et al. Circumferential and focal ablation of
Barrett’s esophagus containing dysplasia. Am J Gastroenterol 2009;
104: 310–317
32 Eldaif SM, Lin E, Singh KA et al. Radiofrequency ablation of Barrett’s
esophagus: short-term results. Ann Thorac Surg 2009; 87: 405 –410
33 Pouw RE, Gondrie JJ, Sondermeijer CM et al. Eradication of Barrett
esophagus with early neoplasia by radiofrequency ablation, with or
without endoscopic resection. J Gastrointest Surg 2008; 12: 1627 –
1636
34 Ganz RA, Overholt BF, Sharma VK et al. Circumferential ablation of Bar-
rett’s esophagus that contains high-grade dysplasia: a U.S. multi center
registry. Gastrointest Endosc 2008; 68: 35 –40
35 Gondrie JJ, Pouw RE, Sondermeijer CM et al. Stepwise circumferential
and focal ablation of Barrett’s esophagus with high-grade dysplasia:
results of the first prospective series of 11 patients. Endoscopy 2008;
40: 359–369
36 Gondrie JJ, Pouw RE, Sondermeijer CM et al. Effective treatment of early
Barrett’s neoplasia with stepwise circumferential and focal ablation
using the HALO system. Endoscopy 2008; 40: 370–379
37 Sharma VK, Kim HJ, Das A et al. A prospective pilot trial of ablation of
Barrett’s esophagus with low-grade dysplasia using stepwise circum-
ferential and focal ablation (HALO system). Endoscopy 2008; 40:
380–387
38 Hernandez JC, Reicher S, Chung D et al. Pilot series of radiofrequency ab-
lation of Barrett’s esophagus with or without neoplasia. Endoscopy
2008; 40: 388–392
39 Roorda AK, Marcus SN, Triadafilopoulos G. Early experience with radio-
frequency energy ablation for Barrett’s esophagus with and without
dysplasia. Dis Esophagus 2007; 20: 516 –522
40 Hubbard N, Velanovich V. Endoscopic endolumenal radiofrequency ab-
lation of Barrett’s esophagus in patients with fundoplication. Surg En-
dosc 2007; 21: 625–628
41 Wani S, Sayana H, Sharma P. Endoscopic eradication of Barrett’s esoph-
agus. Gastrointest Endosc 2010; 71: 147 –166
42 Basu KK, Pick B, Bale R et al. Efficacy and one year follow up of argon
plasma coagulation therapy for ablation of Barrett’s oesophagus: fac-
tors determining persistence and recurrence of Barrett’s epithelium.
Gut 2002; 51: 776–780
43 Sharma P, Bhattacharyya A, Garewal HS et al. Durability of new squa-
mous epithelium after endoscopic reversal of Barrett’s esophagus.
Gastrointest Endosc 1999; 50: 159 –164
44 Overholt BF, Wang KK, Burdick JS et al. Five-year safety and efficacy of
photodynamic therapy with Photofrin in Barrett’s high-grade dyspla-
sia. Gastrointest Endosc 2007; 66: 460–468
45 Inadomi JM, Somsouk M, Madanick RD et al. A cost-utility analysis of
ablative therapy for Barrett’s esophagus. Gastroenterology 2009; 136:
2101–2114
46 Das A, Wells C, Kim HJ et al. An economic analysis of endoscopic abla-
tive therapy for management of nondysplastic Barrett’s esophagus. En-
doscopy 2009; 41: 400–408
47 Ban S, Mino M, Nishioka NS et al. Histopathologic aspects of photody-
namic therapy for dysplasia and early adenocarcinoma arising in Bar-
rett’s esophagus. Am J Surg Pathol 2004; 28: 1466–1473
48 Sharma P, Morales TG, Bhattacharyya A et al. Squamous islands in Bar-
rett’s esophagus: what lies underneath? Am J Gastroenterol 1998; 93:
332–335
49 Shaheen NJ, Goldblum JR, Sampliner RE et al. Are biopsies after ablation
for dysplastic Barrett’s esophagus of adequate depth to detect glandu-
lar mucosa beneath the neosquamous epithelium? Comparative histo-
pathological outcomes from a randomized, sham-controlled trial (AIM
Dysplasia Trial). Gastroenterology 2008; 134: A–724
Original article 789
Fleischer D E. et al. Endoscopic RFA for Barrett’s esophagus: 5-year outcomes…Endoscopy 2010; 42: 781 –789
Reprint
Endoscopy 2010
October
Volume 42
Page 781–789
Endoscopic radiofrequency
ablation for Barrett’s
esophagus: 5-year outcomes
from aprospective
multicenter trial
D. E. Fleischer, B. F.Overholt, V.K. Sharma,
A. Reymunde, M. B. Kimmey, R. Chuttani,
K. J. Chang, R. Muthasamy, C. J. Lightdale,
N. Santiago, D. K. Pleskow, P. J. Dean,
K. K. Wang
Endoscopy
Official Organ of
the European Society of
Gastrointestinal Endoscopy
(ESGE) and
Affiliated Societies
www.thieme-connect.com/
ejournals
www.thieme.de/endoscopy
© 2010 by
Georg Thieme Verlag KG
Rüdigerstraße 14
D-70469 Stuttgart
ISSN 0013-726X
Reprint with the permission
of the publishers only
Endoscopy_Sonder_silbern_1c_u1u4:Muster_Sonder_silbern_1c_u1u4.qxd 30.09.2010 10:49 Seite 1
