Barrett's Esophagus: Prevalence and its Relationship With Dyspeptic Symptoms

[Guest guest]

From Journal of Gastroenterology & Hepatology

Barrett's Esophagus: Prevalence and its Relationship With Dyspeptic Symptoms

Posted 05/17/2004

Murat Toruner; Irfan Soykan; Arzu Ensari; Isinsu Kuzu; Cihan Yurdaydin; Ali

Özden

Abstract and Introduction

Abstract

Background and Aim: Barrett's metaplasia is a premalign condition which

plays a pivotal role in the development of esophageal adenocarcinoma. It is

considered a complication of chronic gastroesophageal reflux disease.

Although esophageal adenocarcinoma is an uncommon cancer, its incidence is

rapidly increasing. The aims of the present study were to determine the

prevalence of Barrett's metaplasia in outpatients referred for gastroscopy

for upper gastrointestinal symptoms, and to clarify the relationship between

Barrett's metaplasia and upper gastrointestinal symptoms.

Methods: Three-hundred and ninety-five consecutive dyspeptic patients, never

previously investigated, underwent gastroscopy and were enrolled into the

study.

Results: Barrett's metaplasia was detected in 29 patients (7.4%). The

age-specific prevalence of Barrett's metaplasia increased with age. In

multivariate analysis, Barrett's metaplasia was independently and positively

related to age, sex and duration of symptoms, but not with upper

gastrointestinal symptoms. In univariate analysis, Barrett's metaplasia was

significantly more common in patients with antral intestinal metaplasia

(24%) and presence of hiatal hernia (65.5%), compared with those with normal

endoscopic findings (6.2% and 39.2%, respectively, p = 0.001).

Conclusion: Symptoms do not predict Barrett's metaplasia. Barrett's

metaplasia is age-related and more common in patients with a longer duration

of symptoms, presence of hiatal hernia and antral intestinal metaplasia.

Introduction

Barrett's esophagus (BE) is a metaplastic change of the lining of the

esophagus with replacement of the normal squamous epithelium with columnar

epithelium containing goblet cells. It is the most severe histological

consequence of chronic gastroesophageal reflux and predisposes to the

development of adenocarcinoma of the esophagus.[1,2] As increases in the

incidence of esophageal adenocarcinoma have been reported in Western

countries such as the USA, New Zealand, Australia and western European

countries in recent years,[3-6] interest in the prevalence of BE and its

predicting factors has increased. Barrett's esophagus is usually discovered

during endoscopic examination of middle-aged and older adults whose mean age

at the time of diagnosis is approximately 55 years.[7] Series of surgical

resections of adenocarcinoma of the esophagus associated with BE reveal that

it is overwhelmingly predominate in white males.[8,9] The columnar

metaplasia in BE causes no symptoms. Thus, patients are seen initially for

symptoms of the associated gastroesophageal reflux disease (GERD), such as

heartburn and regurgitation. Endoscopically obvious BE can be seen in

approximately 10% of patients who have endoscopic examinations for symptoms

of GERD. However, approximately 25% of patients have no esophageal

symptoms.[10] Additionally, there are data suggesting that the duration of

reflux symptoms significantly correlates with the prevalence of BE.[11,12]

In this setting, BE is discovered during endoscopic examinations performed

for unrelated conditions.

The epidemiology of BE is incompletely described. In recent years, by

definition of short segment Barrett's esophagus (SSBE), an increase in the

prevalence of BE has been detected. In a study from the USA, BE was detected

in 25% of asymptomatic male veterans older than 50 years undergoing

screening sigmoidoscopy for colorectal cancer.[13] In another study from

Japan, BE was detected in 15.7% of individuals who underwent an upper

gastrointestinal endoscopy for the evaluation of BE.[14]

The aims of the present study were to investigate the prevalence of BE in a

group of patients who had undergone gastroscopy due to dyspeptic symptoms,

and to determine the relationship between BE and these symptoms.

Methods

The study population consisted of patients who were referred to the

endoscopy unit between March and July 2001. A total of 395 patients who were

admitted to the endoscopy unit were enrolled into the study. Patients who

had esophageal or gastric cancer, previous esophageal or gastric surgery,

gastrointestinal bleeding at the time of endoscopy, or coagulation defects

were excluded from the study.

All patients entering the study were asked to complete a symptom

questionnaire concerning the presence of heartburn, regurgitation,

epigastric pain, nausea and vomiting. All endoscopic examinations were

performed by one of the investigators (MT) using a standard video upper

endoscope (Fujinon EG-410 HR, Tokyo, Japan). During endoscopic examination,

the appearance of the esophagogastric junction was carefully inspected (EGJ;

defined as the junction of the proximal gastric folds and the tubular

esophagus). The squamocolumnar junction was also identified as the point

where the squamous mucosa joined the salmon-color columnar mucosa. All

patients were classified as having either a normal appearing EGJ, a

long-segment BE (LSBE; defined as the presence of intestinal metaplasia

extending proximally from the EGJ and measuring 3 cm or more in length), or

a short-segment BE (SSBE; defined as segments of intestinal metaplasia

emanating from the EGJ and measuring less than 3 cm). The diagnosis of

specialized intestinal metaplasia of the esophagogastric junction (SIM-EGJ)

was defined by the presence of a normal appearing squamocolumnar junction,

devoid of any tongues of pink columnar lined epithelium above the

endoscopically defined EGJ, but associated with specialized intestinal

metaplasia containing goblet cells on biopsy specimens. Cardiac type

metaplasia was defined as columnar metaplasia with glands containing mucous

cells.[13,15,16] Hiatal hernia was diagnosed when the length between the EGJ

and diaphragmatic pinch was >2 cm. At endoscopy, esophagitis, if present,

was noted and graded according to the Los Angeles classification as follows:

grade A, one (or more) mucosal break no longer than 5 mm that does not

extend between the tops of two mucosal folds; grade B, one (or more) mucosal

break more than 5 mm long that does not extend between the tops of two

mucosal folds; grade C, one (or more) mucosal break that is continuous

between the tops of two or more mucosal folds but which involves less than

75% of the circumference; grade D, one (or more) mucosal break which

involves at least 75% of the esophageal circumference.[17] The following

additional data were gathered for each patient: age; sex; alcohol (current

alcohol users); tobacco (current cigarette smokers); medication use and;

histological findings in the antrum, including presence of Helicobacter

pylori infection.

When a normal appearing EGJ was encountered, at least six mucosal biopsy

specimens were taken for histological assessment. Four biopsy specimens were

obtained from the squamocolumnar junction, with one cup of the biopsy

forceps positioned on squamous epithelium, and the other cup on the columnar

epithelium. Two were acquired from the gastric antrum. If BE was suspected,

additional biopsy specimens were obtained to sample any mucosal tongues of

glandular epithelium arising from the EGJ. These biopsy specimens were

obtained from all possible quadrants and at every 1 cm up the length of the

columnar epithelium. All biopsy specimens were immediately submerged in a

10% formalin solution and later stained with hematoxylin and eosin and with

Alcian blue staining at a pH of 2.5. The evaluation of these biopsy

specimens was performed by two qualified pathologists (AE and IK).

The present study was approved by the Institutional Review Board of Ankara

University Medical School and all patients signed informed consent forms

beforehand.

Statistical Analysis

Analysis was performed using the SPSS software package, version 7.5.2 (SPSS,

Chicago, IL, USA). The chi-squared test was used to compare categorical

data. Continuous variables were analyzed using Student's t-test. To test for

a potential relationship between BE and upper gastrointestinal symptoms,

logistic regression analysis was performed with SPSS. The level of

significance was set at p < 0.05.

Results

Median age of the patients enrolled in the study (n= 395, 224 women) was 47

years (range: 19-48 years). Demographic and clinical characteristics of

patients are shown in Table 1. Gastrointestinal symptoms such as heartburn,

epigastric pain, vomiting and nausea were seen predominantly in female

patients. Smoking and alcohol consumption were observed more frequently in

male patients. Symptom duration did not differ between the sexes.

Helicobacter pylori positivity was 62.5% and there was no difference between

males (63.1%) and females (62.1%).

Endoscopic findings from the patients are described in Table 2. Endoscopic

examination revealed squamocolumnar junction (SCJ) irregularity and/or

Barrett's segments, islands and tongues in 129 patients (32.7%),

predominantly in males (38.1%vs 28.6% females). Endoscopic esophagitis was

observed in 61 (15.4%) patients (36 grade A and 25 grade . Esophagitis was

detected more frequently in male patients (male 38 [22.2%]vs female 23

[10.3%]; p < 0.05). Hiatus hernia frequency did not differ between the sexes

(male 77 [45.6%]vs female 81 [36.6%]; p = not significant [NS]).

Barrett's metaplasia was histologically confirmed in 29 (7.4%) patients

(14.8% of patients with irregular Z-line and Barrett's tongues, 27.7% of

patients with Barrett's islands and 71.4% of patients with long segment

Barrett's epithelium were confirmed histologically). Two of these 29

Barrett's metaplasia patients were diagnosed as having low-grade dysplasia.

After histopathological examination of the biopsy specimens, patients

enrolled into the study were divided into four groups: normal (n = 306,

77.6%); cardiac type metaplasia (CM; n = 31, 7.9%); specialized intestinal

metaplasia at squamocolumnar junction (SIM-SCJ; n = 28, 7.1%) and; Barrett's

metaplasia (BM; n = 29, 7.4%).

Patients with BM were more likely to be male (55.2%vs 44.8% females),

whereas patients in the normal (43.8%, n = 134 vs 56.2%, n = 172), CM (29%,

n = 9 vs 71%, n = 22) and SIM-SCJ (42.9%, n = 12 vs 57.1%, n = 16) groups

were predominantly female. The distribution of H. pylori infection in each

group according to sex was as follows: CM, H. pylori (+) n = 22 (14 women),

H. pylori (-) n = 9 (7 women); SIM-SCJ, H. pylori (+) n = 16 (9 women), H.

pylori (-) n = 12 (7 women) (P = NS). Although gastrointestinal symptoms

such as heartburn, epigastric pain, vomiting and nausea were seen more

frequently in female patients and male patients had more social habits such

as smoking and alcohol consumption, there was no significant difference

between these four groups.

Symptom duration was longer in the BM group than in the other three groups

(normal, CM and SIM-SCJ; 92 [2-360]vs 36 [0.2-480] and 48 [1-360] months,

respectively, p < 0.05).

Hiatal hernia was seen more frequently in the BM group than in the other

three groups (normal, CM and SIM-SCJ; 65.5%vs 39.2%, 35.5% and 30.8,

respectively [P< 0.05]). Hiatus hernia was found in 65.5% (19/29) of

patients with BE and only 35.2% (129/366) of patients without BE (P< 0.01).

Intestinal metaplasia in the antrum was observed more in the CM, SIM-SCJ and

BM groups than in the normal group (32.3%, 17.9%, 24.1%vs 6.2%,

respectively, p < 0.05). Helicobacter pylori positivity did not differ with

presence of Barrett's metaplasia and endoscopic esophagitis. Clinical and

endoscopic characteristics of the patients are summarized in Table 3.

Regarding age, the incidence of Barrett's metaplasia increased with age

(Fig. 1).

Discussion

Barrett's metaplasia has been accepted as a premalign lesion of the

esophagus which can lead to the development of adenocarcinoma of the

esophagus. Therefore, investigations in this area have increased in recent

years. However, little is known about its epidemiology and the predictors of

BE. In the present study, we found that 7.4% of patients who had undergone

upper gastrointestinal (GI) endoscopy for different reasons had Barrett's

metaplasia. Differences between the reported prevalence of BE have been

observed among studies from different countries. In Germany, prevalence of

BE was reported to be 4.2%.[18] In Japan, where esophageal adenocarcinoma is

accounting for <5% of esophageal cancers, BE prevalence is 0.3-0.6%. As

esophageal adenocarcinoma is increasing worldwide, in Turkey the ratio of

esophageal cancer is 19.2% among all GI cancers, which is higher than that

seen in other Balkan countries. The majority of patients, especially in

eastern regions of Turkey, have squamous cell esophageal carcinoma because

of the common consumption of hot beverages in this region. Esophageal

adenocarcinoma accounts for approximately 10% of esophageal cancers in

Turkey.[19] Adenocarcinoma of the esophagus and BE are not common in Africa

and the Middle-East. However, BM prevalence was reported to be 8.4% in

Jordan.[18] In another study from Turkey, Dincer et al. reported a very low

prevalence (1.7%) of BE in their study population. Although the number of

patients studied was quite small (n = 60), they concluded that this finding

may be related to the relatively younger mean age of patients.[20]

We also found that BE is seen predominantly in older age groups and males.

These results are in agreement with previous studies that reported a higher

prevalence in older groups of patients and males.[21] Although 57% of our

patients were female, this difference did not reach statistical

significance. Vontilainen et al. found that the male to female ratio of BE

patients was 1.[22] Hirota et al. have also reported an equal distribution

of BE between males and females.[23]

The diagnosis of SSBE can be missed if endoscopic landmarks are not

carefully determined and there is lack of targeting of biopsy samples. A

recent study by Padda et al. showed that histological confirmation of SSBE

and LSBE was 38.4% and 75%, respectively.[24] Our results for LSBE were

similar but we found a lower confirmation rate for SSBE. The sensitivity and

positive predictive values of standard upper endoscopy for diagnosing BE

have been reported as 82% and 34%, respectively.[25] This may be secondary

to the patchy and mosaic presence of intestinal metaplasia in the distal

esophagus. Hiatus hernia is known to be more common in patients with

Barrett's metaplasia than in healthy controls. A recent study by Cameron

demonstrated that 96% of his patients with BE harbored a hiatus hernia.[26]

Additionally, the hiatus sac tended to be longer and wider in patients with

BE. In our study, hiatus hernia was found in 65.6% (n = 19) of patients with

BE and in only 35.2% (n = 129) of patients without BE (P < 0.01). It has

been known that hiatus hernia presence may lead to GERD which can cause

BE.[27] Although esophagitis is one of the important factors in the

development of BE, in our study the rate of endoscopic esophagitis was

almost the same between patients with BE and the normal group. Results of

previous studies investigating the association between esophagitis and BE

are controversial. While some previous studies found no correlation between

esophagitis and BE,[28,29] Voutilainen et al. demonstrated that junctional

specialized columnar epithelium was associated with erosive esophagitis,

suggesting that specialized columnar epithelium may be a result of GERD.[23]

In our study, we found no association between BE presence with

gastroesophageal reflux and/or other dyspeptic symptoms. In most of the

published studies there are some controversial reports discussing this

possible association. Spechler et al. and Nandurkar et al. showed that the

prevalence of the SSBE has not been linked with gastroesophageal reflux

symptoms.[29,30] However, Pereira et al. and ston et al. found an

association between prevalence of SSBE and gastroesophageal reflux

symptoms.[31,32] The possible reason for lack of this kind of association in

our study may be a result of the small number of LSBE in our study

population. However, we found a significant correlation between the symptom

duration and the presence of BE, which demonstrates the longer duration of

acid/bile reflux to esophagus is a risk factor for developing BE.

The role of H. pylori infection in BE development is still controversial.

Some investigators have reported that H. pylori infection plays a protective

role in GERD and BE development.[33] This protective role is explained by

reduced gastric acid secretion after H. pylori infection. Some recent

reports have indicated that not all H. pylori infections, but infections

with Cag A positive strains, protect against GERD. However, there are some

reports indicating no relationship between H. pylori infection and BE

prevalence.[34-36] We found a lower incidence of H. pylori infection in

patients with esophagitis than without esophagitis, but this difference did

not reach any significance. In our study there was no correlation between H.

pylori infection and BE prevalence. Hackelsberger et al. found that in

patients with endoscopic features suggestive or typical of BE, 34.3% had

intestinal metaplasia at the SCJ which was associated with male sex,

esophagitis and reflux symptoms, but not with H. pylori or gastric

intestinal metaplasia.[37] Conversely, Oberg et al. investigated 229

patients with symptoms suggestive of foregut disease by means of esophageal

manometry, 24-h pH monitoring and upper gastrointestinal endoscopy with

biopsy specimens obtained from the gastric antrum, the GEJ and the distal

esophagus. They concluded that H. pylori plays no role in the pathogenesis

of GERD or its complications.[38] In our study, the rate of antral

intestinal metaplasia in the BE group was higher than that in the normal

group. However, antral H. pylori infection in the BE group and normal group

was almost the same. This discrepancy between the two groups suggests that

there are some factors other than H. pylori infection, such as long standing

chronic gastritis (autoimmune, type A chronic atrophic gastritis) and

probably other environmental factors that affect the development of

intestinal metaplasia.[39]

In conclusion, all patients with GERD/dyspeptic symptoms and a longer

duration of symptoms should undergo upper gastrointestinal endoscopy for

screening of BE, and those with observed hiatus hernia should be

investigated for the presence of BE. We believe that the present study will

shed new light on the epidemiology and geoprevalence of BE around the world.

Dr Murat Toruner, Ankara University Medical School, Ibni Sina Hospital,

Division of Gastroenterology, Sihhiye, 06100, Ankara, Turkey. Email:

toruner@...