bs_bs_banner

doi:10.1111/jgh.12879

O R I G I N A L A RT I C L E

Surveillance of short-segment Barretts esophagus using

ultrathin transnasal endoscopy
Hiroko Sugimoto,* Takashi Kawai,* Sakiko Naito,* Kyosuke Yanagizawa,* Tetsuya Yamagishi,*
Masakatsu Fukuzawa, Kenji Yagi, Jun Matsubayashi, Toshitaka Nagao, Hirofumi Tomiyama,
Sumito Hoshino, Akihiko Tsuchida and Fuminori Moriyasu
*Endoscopy Center, Department of Gastroenterology and Hepatology, Department of Pathology and Department of Cardiovascular, Tokyo
Medical University Hospital, Tokyo, Japan, Department of Gastrointestinal and pediatric Surgery, Tokyo Medical University Hospital

Keywords
Barretts esophagus, transnasal endoscopy,
Narrow band imaging, close examination,
mucosal structure.
Correspondence
Dr Takashi Kawai, Endoscopy Center, Tokyo
Medical University Hospital, 6-7-1
Nishi-Shinjuku, Shinjuku-ku, Tokyo 160-0023,
Japan. Email: t-kawai@tokyo-med.ac.jp

Abstract
Background and Aim: Newly developed ultrathin transnasal endoscope, the GIFXP290N, makes possible a resolving power similar to the GIF-H260 at a distance of 3 mm.
We conducted surveillance of subjects with Barretts esophagus using this ultrathin
transnasal endoscopy. In Japan the lower margin of the lower esophageal palisade vessels
is defined the gastroesophageal junction in deep inspiration. We diagnose Barretts esophagus if columnar epithelium is present on the oral side of the gastroesophageal junction.
Methods and Results: Barretts esophagus was confirmed in 116 out of 135 subjects
(85.9%), with 17 cases of short-segment Barretts esophagus (SSBE) and 99 of ultra-shortsegment Barretts esophagus. Close observation of the Barretts esophagus mucosal structural pattern using narrow band imaging revealed 29 cases with an oval or round pattern,
29 with a long straight pattern, 47 with a villous pattern, 8 with a cerebriform pattern, and
6 with an irregular pattern according to Goda classification. Mucosal biopsies from all
subjects with SSBE are examined. Histological examination revealed intestinal metaplasia
in only eight subjects. We grouped the oval/round and long straight patterns as closed type,
and the villous, cerebriform, and irregular patterns as open type. Analysis of the relationship between these mucosal patterns and background factors revealed a significant correlation between intestinal metaplasia and the open-type pattern.
Conclusion: We consider this new ultrathin transnasal endoscopy to be a useful technique
for surveillance of Barretts esophagus, especially SSBE.

Introduction
Barretts esophagus is a condition in which chronic inflammation of
the lower esophagus associated with gastrointestinal reflux disease
causes replacement of squamous epithelium by columnar epithelium. The intestinal metaplasia of Barretts esophagus can further
lead to Barretts esophageal adenocarcinoma. The incidence of
Barretts esophagus and Barretts esophageal adenocarcinoma is
increasing, particularly in Western countries.1,2 Although there have
been no reports of definitive epidemiological studies of esophageal
adenocarcinoma in Japan, studies indicate a gradual increase in the
incidence of adenocarcinoma of the esophagogastric junction.3 In
comparison with Western countries, in Asian countries including
Japan the incidence of long-segment Barretts esophagus (LSBE),
with a circumferential segment of columnar epithelium greater than
3 cm in length, is lower, and short-segment Barretts esophagus
(SSBE) is more common.4 However, the malignant potential of this

SSBE is also high, and surveillance is recommended as for LSBE.5

In fact, Koike et al. reported that 24 out of 28 cases of Barretts
esophageal adenocarcinoma diagnosed between June 2001 and
September 2010 arose from SSBE.6 Endoscopic surveillance of
patients with Barretts esophagus has been reported to contribute to
early detection of esophageal adenocarcinoma and improved outcomes.7,8 However, the optimum method and interval for surveillance has yet to be clarified.
In recent years, transnasal endoscopy has become widely used
for screening of the upper gastrointestinal tract. Until now,
transnasal endoscopes have been inferior to transoral endoscopes
in terms of optical resolution. However, the recently developed
ultrathin transnasal endoscope, the GIF-XP290N, has a brighter
light source and uses an objective optical system that prevents any
reduction in contrast when the endoscope tip nears the area of
interest. This makes possible a resolving power similar to the
GIF-H260 at a distance of 3 mm.9 Furthermore, close observation

Journal of Gastroenterology and Hepatology 2015; 30 (Suppl. 1): 4145

2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd

41

Barretts esophagus and transnasal endoscopy

H Sugimoto et al.

Figure 1 Procedure of observation for Barretts esophagus. We then confirm the lower margin of the lower esophageal palisade vessels, the
gastroesophageal junction as defined in Japan (left). If Barretts esophagus is detected (center), we switch over to NBI, making a close observation
of the mucosal structure at a distance of 3 mm (right). NBI, narrow band imaging; WLI, white light imaging.

Table 1

Characteristics of subjects

Subjects
Atrophic gastritis
Gastric polyps
Gastric ulcer
Duodenal ulcer
Gastric cancer (post-ESD)
Esophageal cancer (post-ESD)
Gastric submucosal tumor

71 cases
24 cases
15 cases
7 cases
14 cases
2 cases
2 cases

ESD, endoscopic submucosal dissection.

using narrow band imaging (NBI) enables us to delineate the

mucosal structure, useful in the endoscopic diagnosis (in particular
qualitative diagnosis) of lesions.
In this paper, we present the results of endoscopic surveillance
of Barretts esophagus using the GIF-XP290N.

Methods
This prospective continuous study was conducted at the Tokyo
Medical University Hospital Endoscopy Center between August
2012 and May 2013. The subjects were 135 patients who underwent upper gastrointestinal screening using thin transnasal endoscopy. Their average age was 63.5 9.7 years, with a male : female
ratio of 2.5:1. All examinations were conducted using both white
light imaging (WLI) and close examination using NBI. Subject
diagnoses are given in Table 1. The following subject background
factors were collated: age, gender, history of smoking and alcohol
intake, Helicobacter pylori (H. pylori) infection (including history
of eradication therapy), and medication (suppressors of gastric
acid secretion).
First, we examine from the upper esophagus to the gastroesophageal junction using WLI. We also check for the presence
of a hiatus hernia. We then confirm the lower margin of the lower
esophageal palisade vessels, the gastroesophageal junction as
defined in Japan (Fig. 1 left),10 with the diaphragm lowered in deep
inspiration. We diagnose Barretts esophagus if columnar epithelium is present on the oral side of the gastroesophageal junction. If
Barretts esophagus is detected (Fig. 1 center), we switch over to
NBI, making a close observation of the mucosal structure at a
42

distance of 3 mm (Fig. 1 right). Barretts esophagus is classified as

LSBE (> 30 mm), SSBE (< 30 mm, > 10 mm), or ultra-shortsegment Barretts esophagus (USSBE) (< 10 mm). We classify the
mucosal pattern as oval or round pattern (Fig. 2a), long straight
pattern (Fig. 2b), villous pattern (Fig. 2c), cerebriform pattern
(Fig. 2d), or irregular pattern (Fig. 2e) according to Goda classification.11 Mucosal biopsies from all subjects with SSBE are examined histologically for the presence of metaplasia. We measure the
length of segments of Barretts esophagus using biopsy forceps.
We also check reflux esophagitis (grades A, B, C, D) according to
Los Angeles Classification.12
We diagnose H. pylori infection using either the rapid urease
test, urea breath test, histology, or serum IgG H. pylori antibodies.
We excluded subjects aged < 20 years and those with a history
of gastrointestinal surgery, including the esophagus or stomach.
Subjects taking inhibitors of gastric acid secretion, such as proton
pump inhibitors, were also excluded.
The study protocol was approved by Tokyo Medical University
Hospital Clinical Ethics Committee on Human Experiments, in
accordance with the Helsinki Declaration of 1975 as revised in
1983.
Endoscopy system. Endoscopic examinations are performed on unsedated subjects. We use the Olympus GIF-XP290N
transnasal endoscopy (Olympus Medical System, Tokyo, Japan;
outer diameter at the distal end 5.0 mm) and endoscope (Evis
Lucera Elite System; Olympus Medical System). Premedication
and anesthesia to the nasal cavity are performed as previously
described.13 Antispasmodics, such as scopolamine butyl-bromide,
are not administered as premedication.
Histological examination. Biopsy specimens are fixed in
formalin, then embedded in paraffin, and 4-m slices were made.
These specimens were examined for the presence of intestinal
metaplasia.
Statistical analysis. Statistical analyses were performed
using the analysis software SPSS 16.0J for Windows (SPSS:
SPSSR22.0J, IBM, New York, USA). For subject background
factors, age was analyzed using the unpaired t-test, whereas other

Journal of Gastroenterology and Hepatology 2015; 30 (Suppl. 1): 4145

2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd

H Sugimoto et al.

Barretts esophagus and transnasal endoscopy

Table 2 Correlations between the presence of Barretts esophagus

and subject background factors

Age
Male : female
Height
Weight
Tobacco
Alcohol
Hiatus hernia
H. pylori infection

BE present
n = 116

BE absent
n = 19

P-value

65.8 10.4
62:54
160.5
58.1
27
48
84
17

23.3
41.4
72.4
14.6

66.0 12.5
9:10
160.8
60.4
4
8
3
5

21.1
42.1
15.7
26.3

0.987
0.649
0.914
0.457
0.803
0.976
0.000
0.205

BE, Barretts esophagus.

Table 3 Correlation between mucosal pattern of Barretts esophagus

and subject background factors

Age
Male : female
Height
Weight
Tobacco
Alcohol
Hiatus hernia
H. pylori infection
Intestinal metaplasia

Closed type
n = 54

Open type
n = 62

P-value

65.4 10.8
27:27
160.3
57.7
13
24
28
13
0

24.1
44.4
51.8
24.1
0

66.3 10.2
35:27
160.9
58.5
14
24
56
4
8

22.5
38.7
90.3
6.5
12.5

0.710
0.491
0.710
0.743
0.729
0.547
0.000
0.006
0.006

Closed type: oval/round pattern and long straight pattern. Open type:
villous pattern, cerebriform pattern, and irregular pattern.

background factors, as well as the presence of H. pylori infection

and hiatus hernia, were analyzed using the 2 test or Fishers exact
test. A level of significance < 5% derived using these tests was
considered statistically significant.

Results

Figure 2 Picture and schema of Barretts esophagus was divided into

five patterns using close observation of GIF-XP290N: (a) oval or round
pattern, (b) long straight pattern, (c) villous pattern, (d) cerebriform
pattern, and (e) irregular pattern.

Barretts esophagus was confirmed in 116 subjects (85.9%). All

were < 30 mm in length, with 18 cases of SSBE and 98 of USSBE.
Only 6 cases of reflux esophagitis (LA-A:3, LA-B:1, LA-C:2)
were recognized in BE. The mucosal pattern was oval or round
pattern in 29 cases, long straight pattern in 29, villous pattern in
47, cerebriform pattern in 8, and irregular pattern in 6. Histological
examination revealed intestinal metaplasia in eight subjects, three
of whom showed a villous pattern, two a cerebriform pattern, and
three an irregular pattern.
There were 22 subjects positive and 28 negative for H. pylori
infection, with 85 having successfully undergone eradication
therapy.
Analysis of the relationship between the presence of Barretts
esophagus and background factors revealed a significant correlation with the presence of hiatus hernia, but no other significant
correlations (Table 2). Using the simplified classification of
mucosal patterns proposed by Yuki et al., we grouped the oval/
round and long straight patterns as closed type, and the villous,

Journal of Gastroenterology and Hepatology 2015; 30 (Suppl. 1): 4145

2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd

43

Barretts esophagus and transnasal endoscopy

H Sugimoto et al.

50
45
40
35
30
25
20
15
10
5
0

cerebriform, and irregular patterns as open type.14 Analysis of the

relationship between these mucosal patterns and background
factors revealed a significant correlation between hiatus hernia,
intestinal metaplasia, and H. pylori infection, and the open-type
pattern (Table 3).
The longest orientation of Barretts esophagus was the 7 oclock
position, seen in 49 cases, followed by the 1 oclock position in 35
cases (Fig. 3). The longest orientation of SSBE was 1 oclock in
eight cases, followed by the 7 oclock in five cases; on the other
hand, the longest orientation of USSBE was 7 oclock in 44 cases,
followed by the 1 oclock in 27 cases.

Discussion
We used ultrathin transnasal endoscopy for surveillance of Barretts esophagus for the following four reasons: (i) reduced
patient discomfort allows us to observe the gastroesophageal
junction thoroughly and without haste; (ii) because the patient
is awake, we can identify the lower esophageal palisade vessels
in deep inspiration (under sedation, we cannot examine the lower
esophagus in deep inspiration, making identification of the lower
esophageal palisade vessels difficult); (iii) close examination
allows us to delineate the Barretts esophagus mucosal pattern;
and (iv) we can take biopsy specimens.
In this study, the incidence of Barretts esophagus was 85.9%,
considerably greater than that of 20.9% reported by Kawano
et al.15 However, none were LSBE, and 85.3% were USSBE, indicating that thorough examination of the gastroesophageal junction
using an ultrathin transnasal endoscope enabled the detection of
small segments of Barretts esophagus, leading to the high incidence in this study. Ishimura et al. have also identified inconsistencies between endoscopists in the diagnosis of Barretts
esophagus.16
We use the classification of Goda et al. to classify the mucosal
structural pattern of Barretts esophagus into five patterns.11 Of
these patterns, correlations have been reported between the villous
and cerebriform patterns with intestinal metaplasia and Barretts
esophageal adenocarcinoma. Apart from Goda et al., there have
44

10

11

12

Figure 3 Number of orientation of Barretts

esophagus.

been a number of reports of mucosal and vascular classifications of

Barretts esophagus,1719 although these have all used high-vision or
magnifying endoscopes. These endoscopes are larger in caliber,
requiring sedation, and are unsuited to screening endoscopy.
Accordingly, we used a newly developed ultrathin transnasal endoscope, allowing close observation of the mucosal structure, and
investigated the incidence of Barretts esophagus, in particular
SSBE/USSBE, and the usefulness of classification of mucosal
structural patterns. Yuki et al. classified mucosal patterns into an
open type and closed type, reporting dysplasia was more common
with the open type.14 We similarly grouped patterns into closed type
(oval/round and long straight patterns) and open type (villous,
cerebriform, and irregular patterns), and found that histological
intestinal metaplasia was more common with the open type. The
presence of intestinal metaplasia is reported to be associated with
malignancy, and careful follow-up using ultrathin transnasal endoscopy is required in patients with open-type Barretts esophagus.
Our findings of the orientation of segments of Barretts esophagus and the multidirectionality were consistent in previous reports
that Barretts esophagus and adenocarcinoma are most commonly
located on the right esophageal wall.20,21
Concerning background factors, Fujiwara et al. and Amano
et al. reported advanced age, male gender, smoking, H. pylori
infection, presence of hiatus hernia, and elevated body mass index
as significant risk factors for gastroesophageal reflux disease
(GERD), but only hiatus hernia for SSBE.22,23 Although the risk
factors for LSBE are similar to those for GERD, all cases of
Barretts esophagus in this study were SSBE or USSBE, accounting for our identification of only hiatus hernia as a significant risk
factor.
In conclusion, we believe that transnasal endoscopy will be a
useful method for surveillance of SSBE/USSBE, common conditions in Japanese people.

References
1 Powell J, McConkey CC. The rising trend in oesophageal
adenocarcinoma and gastric cardia. Eur. J. Cancer Prev. 1992; 1:
2659.

Journal of Gastroenterology and Hepatology 2015; 30 (Suppl. 1): 4145

2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd

H Sugimoto et al.

2 Powell J, McConkey CC, Gillison EW, Spychal RT. Continuing

rising trend in oesophageal adenocarcinoma. Int. J. Cancer 2002;
102: 4227.
3 Kusano C, Gotoda T, Khor CJ et al. Changing trends in the
proportion of adenocarcinoma of the esophagogastric junction in a
large tertiary referral center in Japan. J. Gastroenterol. Hepatol.
2008; 23: 16625.
4 Amano Y, Kinoshita Y. Barrett esophagus: perspectives on its
diagnosis and management in Asian populations. Gastroenterol.
Hepatol. 2008; 4: 4553.
5 Sharma P, Morales TG, Bhattacharyya A, Garewal HS, Sampliner
RE. Dysplasia in short-segment Barretts esophagus: a prospective
3-year follow-up. Am. J. Gastroenterol. 1997; 92: 20126.
6 Koike T, Abe Y, Iijima K et al. Diagnosis of Barretts esophageal
cancer by conventional endoscopic examination (in Japanese).
Stomach and Intestine 2011; 46: 180014.
7 Corley DA, Levin TR, Habel LA, Weiss NS, Buffler PA.
Surveillance and survival in Barretts adenocarcinomas: a
population-based study. Gastroenterology 2002; 122: 63340.
8 Cooper GS, Kou TD, Chak A. Receipt of previous diagnoses and
endoscopy and outcome from esophageal adenocarcinoma: a
population-based study with temporal trends. Am. J. Gastroenterol.
2009; 104: 135662. DOI: 10.1038/ajg.2009.159.
9 Kawai T, Fukuzawa M, Gotoda T. Evolution of ultra thin endoscope.
Dig. Endosc. 2013; 25: 46.
10 The Japan Esophageal Society, ed. Japanese Classification of
Esophageal Cancer (The 10th Edition/Revised Version). Tokyo:
Kanehara Shuppan, 2008; 402.
11 Goda K, Tajiri H, Ikegami M, Urashima M, Nakayoshi T, Kaise M.
Usefulness of magnifying endoscopy with narrow band imaging for
the detection of specialized intestinal metaplasia in columnar-lined
esophagus and Barretts adenocarcinoma. Gastrointest. Endosc.
2007; 65: 3646.
12 Armstrong D, Bennett JR, Blum AL et al. The endoscopic
assessment of esophagitis: a progress report on observer agreement.
Gastroenterology 1996; 111: 8592.
13 Kawai T, Miyazaki I, Yagi K et al. Comparison of the effects on
cardiopulmonary function of ultrathin transnasal versus normal

Barretts esophagus and transnasal endoscopy

14

15

16

17

18

19

20

21

22

23

diameter transoral esophagogastroduodenoscopy in Japan.

Hepatogastroenterology 2007; 54: 7704.
Yuki T, Amano Y, Kushiyama Y et al. Evaluation of modified
crystal violet chromoendoscopy procedure using new mucosal pit
pattern classification for detection of Barretts dysplastic lesions. Dig
Liv Dis. 2006; 38: 296300.
GERD Society Study Committee, Kawano T, Kouzu T, Ohara S,
Kusano M. The prevalence of Barretts mucosa in the Japanese (in
Japanese). Gastroenterol Endosc 2005; 47: 95161.
Ishimura N, Amano Y, Kinoshita Y. Endoscopic definition of
esophagogastric junction for diagnosis of Barretts esophagus:
importance of systematic education and training. Dig. Endosc. 2009;
21: 21318.
Sharma P, Bansal A, Mathur S et al. The utility of a novel narrow
band imaging endoscopy system in patients with Barretts
esophagus. Gastrointest. Endosc. 2006; 64: 16775.
Curvers WL, Bohmer CJ, Mallant-Hent RC et al. Mucosal
morphology in Barretts esophagus: interobserver agreement
and role of narrow band imaging. Endoscopy 2008; 40:
799805.
Singh R, Anagnostopoulos GK, Yao K et al. Narrow-band imaging
with magnification in Barretts esophagus: validation of a simplified
grading system of mucosal morphology patterns against histology.
Endoscopy 2008; 40: 45763.
Okita K, Amano Y, Takahashi Y et al. Barretts esophagus in
Japanese patients: its prevalence, form, and elongation. J.
Gastroenterol. 2008; 43: 92834.
Moriyama N, Amano Y, Okita K, Mishima Y, Ishihara S, Kinoshita
Y. Localization of early-stage dysplastic Barretts lesions in patients
with short-segment Barretts esophagus. Am. J. Gastroenterol. 2006;
101: 26667.
Fujiwara Y, Higuchi K, Shiba M et al. Differences in clinical
characteristics between patients with endoscopy-negative reflux
disease and erosive esophagitis in Japan. Am. J. Gastroenterol. 2005;
100: 7548.
Amano K, Adachi K, Katsube T et al. Role of hiatus hernia and
gastric mucosal atrophy in the development of reflux esophagitis in
the elderly. J. Gastroenterol. Hepatol. 2001; 16: 1326.

Journal of Gastroenterology and Hepatology 2015; 30 (Suppl. 1): 4145

2015 Journal of Gastroenterology and Hepatology Foundation and Wiley Publishing Asia Pty Ltd

45