SEARCH

SEARCH BY CITATION

Keywords:

  • Asians;
  • Barrett’s oesophagus;
  • environment;
  • gastro-oesophageal reflux disease;
  • genes

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genetic influences on ethnic differences in Barrett’s oesophagus in Asians
  5. Environmental effects on Barrett’s oesophagus in Asians
  6. Conclusions and perspective
  7. Conflict of interest statement
  8. References

Abstract.  Rajendra S (University of New South Wales & Bankstown-Lidcombe Hospital, Sydney, NSW, Australia). Barrett’s oesophagus in Asians – are ethnic differences due to genes or the environment? (Review). J Intern Med 2011; 270: 421–427.

Ethnic differences in the prevalence of gastro-oesophageal reflux disease (GORD) and its complications, including Barrett’s oesophagus (BO), are well described in multiracial Asian patient populations. These findings together with familial aggregation of GORD symptoms and twin studies suggest the possibility of a genetic component to GORD. Nevertheless, environmental factors, e.g. Helicobacter pylori infection, abdominal adiposity and metabolic syndrome, could equally account for these differences. Indian (South Asian) race is a risk factor for Barrett’ oesophagus. This may be related to the Caucasian genetic make-up of Indians as opposed to an Oriental one as is the case of most other Asians. The HLA-B07 gene commonly found in South Asian and Caucasian populations, but not Orientals, may confer an increased risk for BO. Nevertheless, the high prevalence of H. pylori in South Asians and the consequent atrophic gastritis and hypochlorhydria may partially ameliorate this genetic predisposition to BO. The higher prevalence of obesity and the metabolic syndrome amongst certain Asiatic races may also contribute to the observed increased risk for BO. Future research should target the search for GORD/BO genes, ethnic differences in parietal cell mass and hiatal hernia, H. pylori colonization factors (e.g. MUC1 and MUC2) and adhesion molecules (BabA). Racial differences in lifestyle factors, i.e. abdominal adiposity, consumption of fruit and vegetables as well as smoking, should all be investigated as potential causes for this interethnic variation in GORD and BO. Nature or nurture, the clues are teasing and tantalizing and illustrate the complex relationship between the genetic make-up of man and the environment.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genetic influences on ethnic differences in Barrett’s oesophagus in Asians
  5. Environmental effects on Barrett’s oesophagus in Asians
  6. Conclusions and perspective
  7. Conflict of interest statement
  8. References

Barrett’s oesophagus (BO) is thought to develop as a result of chronic gastro-oesophageal reflux disease (GORD) and predisposes to oesophageal adenocarcinoma. The prevalence of BO in Asians is increasing as evidenced by hospital studies from Malaysia [1], South Korea [2], Japan [3], China[4] and India[5] (Table 1). Oesophageal adenocarcinoma is also on the rise in Singapore [6], Japan [7] and Iran (Figs 1 and 2) [8]. In Singapore, the age-standardized incidence rates (ASR) for squamous cell carcinoma decreased from 8.31 to 3.85 per 100 000 men (P = 0.017) and from 3.43 to 0.81 per 100 000 women (= 0.027). Conversely, the ASR for oesophageal adenocarcinoma rose from 0 to 0.54 per 100 000 men and from 0.03 to 0.13 per 100 000 women [6].

Table 1. Prevalence of endoscopic Barrett’s oesophagus (BO) in Asian countries
AuthorsCountryPatients (n)BO (n)BO (%)
Foaud 2009 [27]Egypt1000737.3
Nasseri-Moghaddam 2003 [28]Iran269103.7
Toruner 2004 [29]Turkey39571.50
Rajendra 2004 [1]Malaysia1985321.60
Ang TL 2005 [30]Singapore53391.7
Park JJ 2009 [31]Korea255362150.84
Azuma 2000 [3]Japan65040.60
Wong 2002 [32]China1660630.02
image

Figure 1. The age-standardized incidence rate of oesophageal cancer in men in Singapore.

Download figure to PowerPoint

image

Figure 2. Trends in age-adjusted incidence rate of oesophageal cancers in men by histological subtype in Japan.

Download figure to PowerPoint

The reasons include a true increase in the prevalence of GORD and its complications, increasing affluence and westernization of the Asian diet, rise in body mass index (BMI) including abdominal adiposity and smoking [9]. Heightened physician awareness, reduction in Helicobacter pylori infection, widespread availability of endoscopic equipment and an increase in trained medical/surgical gastroenterologists and thus greater patient referrals for gastrointestinal investigations may also account for this rise in BO and adenocarcinoma [9].

The incidence of oesophageal adenocarcinoma has risen steeply in the last 35 years in most Western countries including the UK, Australia, Holland and the United States [10–14]. The number of new cases per 100 000 White men in 2000 was between 1 and 5. Amongst Caucasian males in the United States, the incidence of adenocarcinoma of the oesophagus rose more than 350% since the mid-1970s, surpassing squamous cell carcinoma in the 1990s [10, 15]. GORD, obesity, smoking and poor diet have been collectively attributed for 79% of oesophageal adenocarcinomas in the United States [16].

Genetic influences on ethnic differences in Barrett’s oesophagus in Asians

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genetic influences on ethnic differences in Barrett’s oesophagus in Asians
  5. Environmental effects on Barrett’s oesophagus in Asians
  6. Conclusions and perspective
  7. Conflict of interest statement
  8. References

Ethnic differences in the prevalence of heartburn, [17, 18] oesophagitis [19] and BO [1] are well documented in multiracial Asian patients (Table 2). These racial differences as well as familial aggregation of GORD symptoms [20], BO [21]/adenocarcinoma [22]and twin studies [23, 24] suggest the possibility of a genetic component to GORD and BO although race-specific environmental factors as documented above could also account for these differences (Table 3). Reflux symptoms were twice as common in Indians as compared to the Chinese in Singapore [18]. In neighbouring Malaysia, Indians have the highest prevalence of heartburn [17], oesophagitis [19] and BO [1]. The HLA-B7 gene (commonly found in South Asian and Caucasian populations but infrequently in Orientals) has been reported to be significantly positively associated with BO in this racial group but not the Malays or Chinese [25]. Interestingly, the HLA B7-positive patients with BO had a significantly higher family history of heartburn compared with their HLA-B7-negative counterparts. Not surprisingly, Indians are categorized as Caucasians in the world racial grouping as they share a common genetic make-up, whereas Malays and Chinese are grouped as East Asians [26]. The other major world populations being Africans, Pacific Islanders and Native Americans.

Table 2. Prevalence of Barrett’s oesophagus (BO) in patients of different racial groups examined in the same endoscopy units [33]
AuthorsCountryRacenBO (n)BO (%)
Hirota 1999 [34]USAWhite611406.5
Black20000.0
Rajendra 2004 [1]MalaysiaMalays50271.4
Chinese824101.2
Indians659152.3
Ford 2004 [35]UKWhite150636904.6
Asian5297450.8
Table 3. Positive and negative associations with Barrett’s oesophagus (BO)
Positive associations with BONegative associations with BO
Gastro-oesophageal reflux diseaseHelicobacter pylori
Genetics (Familial history, ? HLA B07)Fruit and vegetables
Race: Caucasian including Indian 
Male sex 
Smoking 
Obesity i.e. abdominal adiposity 
Hiatus hernia

It is well known that familial clustering of reflux symptoms is seen in relatives of Caucasian patients with reflux symptoms and increased oesophageal acid exposure in relatives of patients with BO [20]. Romero et al. [21] also reported that reflux symptoms are more common amongst relatives of patients with BO or oesophageal adenocarcinoma but not amongst relatives of patients with endoscopic oesophagitis. There is an increased frequency of positive family history (first- and second-degree relatives) for BO and adenocarcinoma amongst patients with these conditions (24%) compared to GORD without BO (5%) [22]. The above data pertain to Caucasians, and to date, there are no Asian studies on familial clustering of BO or adenocarcinoma.

A study of GORD symptoms in 8411 Swedish twin pairs over the age of 55 years found a casewise concordance for GORD of 31% amongst female MZ twins as compared with 21% in female DZ twins [23]. Heritability was estimated to account for 30% of the liability to GORD. A study of 1960 British twin pairs to determine the relative contribution of genetic and environmental influences revealed that casewise concordance were significantly higher for MZ than DZ twins (42% vs. 26%; P < 0.001) [24]. Multifactorial liability threshold modelling suggested that 43% of the variation in liability to GORD was because of multiple small genetic effects. The limited concordance rates suggest either Mendelian inheritance with a markedly reduced penetrance or a complex disorder with multiple genetic and environmental influences modifying disease phenotype. Unfortunately, there are no Asian twin studies in GORD/BO. Nevertheless, it has been shown that Asian (Chinese) patients with GORD have a similar pathogenesis for GORD as in Caucasians, i.e. increased oesophageal acid exposure mainly because of impaired oesophageal clearance by defective peristalsis [36]. Acid output though has been shown to be lower in East Asians as compared to Scottish patients as a result of a smaller parietal cell mass [37].

Environmental effects on Barrett’s oesophagus in Asians

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genetic influences on ethnic differences in Barrett’s oesophagus in Asians
  5. Environmental effects on Barrett’s oesophagus in Asians
  6. Conclusions and perspective
  7. Conflict of interest statement
  8. References

Helicobacter pylori infection and Barrett’s oesophagus

Helicobacter pylori is known to be more prevalent in Asia than in the West, and most Asian studies have shown a protective effect of this bacterium for BO [38, 39]. The declining rate of H. pylori infection and hence pangastritis and corpus atrophy has been postulated as a cause of increased GORD and complications in Asians [40]. The author (SR) estimated the strength of the association between H. pylori, ethnicity and GORD spectrum, including BO, in 188 multiethnic Asian patients presenting for endoscopy in a tertiary Malaysian referral centre. Indians had the highest prevalence of H. pylori (45/60, 75%) as compared to the Chinese (42/72, 58.3%) and Malays (11/56, 19.6%) Presence of H. pylori was associated with a reduced severity of GORD spectrum including BO, in Asians, and this was especially so in Indians (OR 0.56; CI 95% 0.31–0.99; = 0.048). Not surprisingly, Indians had a significantly greater propensity to develop corpus atrophy than the Chinese or Malays. When BO was present, the length of abnormality was 45% shorter in the presence of H. pylori compared with uninfected patients with BO [40].

Ho et al. [41] has also asserted that the diminishing prevalence of H. pylori infection maybe an aetiological factor in the increasing incidence of reflux disease and its complications in Singapore. In Japan, Abe et al. studied the H. pylori status in 112 patients with reflux oesophagitis, BO and controls. Helicobacter pylori infection in the oesophagitis group (24.1%) was significantly lower than the control group (71.2%). In patients with short-segment BO, H. pylori prevalence was 18.7% and in long-segment BO (LSBO), 0% [42]. More recently, the same author compared serum pepsinogen (surrogate marker of corpus atrophy) and Helicobacter pylori status between BO patients and age- and sex-matched controls. Helicobacter Pylori status was 11% (4/36) in LSBO versus 74% (80/108) in controls. In addition, serum pepsinogen levels were significantly higher in BO than in controls [43]. In a Taiwanese study of 1622 patients, there was a significantly lower rate of H. pylori infection in patients with reflux oesophagitis compared to those with a normal oesophagus (33% vs. 67.5%) [44]. There were 10 patients with BO in their study population of whom only one was H. pylori positive. In Hong Kong, it is thought that cagA H. pylori results in a lower prevalence and milder GORD in the Hong Kong Chinese population [45]. Negative studies on BO and H. pylori in Asians are rare. One such example is that from China by Zhang et al. [46] In a study of 391 patients with GORD, oesophagitis and BO, the authors found a protective role of H pylori infection to GORD but not BO.

Two systematic reviews have found an inverse association of GORD with the prevalence of H. pylori infection, being 10% lower in GORD patients than in controls especially in studies from the Far East [38, 39]. This may be related to the higher prevalence rate of this infection amongst Asians, and thus, smaller sample sizes required to detect a difference and/or other causes yet to be elucidated.

Intriguingly, Indians in Malaysia and who have the highest prevalence of H. pylori are also more likely to be diagnosed with BO than the Malays or the Chinese indicating a complex interplay of genetic and environmental factors [1, 40]. It has been hypothesized that the high prevalence of H. pylori infection amongst Indians and the consequent atrophic gastritis and hypochlorhydria may at least partially ameliorate the genetic predisposition (HLA-B7) to BO [40].

Abdominal adiposity and Barrett’s oesophagus

Abdominal adiposity and obesity increase the risk of Barrett’s metaplasia and oesophageal adenocarcinoma, respectively [47, 48]. Abdominal obesity might increase the risk of BO and oesophageal adenocarcinoma by promoting reflux through increasing intragastric pressure [49]. Kendall et al. [50] reported that high serum leptin (a visceral fat hormone) was associated with an increased risk of BO, particularly amongst men. This hormone might promote carcinogenesis by mitogenic and angiogenic means. Obesity decreases serum levels of adiponectin, which is a hormone associated with decreased proliferation. The end result is increased proliferation that favours carcinogenesis.

A cross-sectional study involving 111, 639 individuals of whom 11% had heartburn (White = 78%, Black = 13% and Asian = 4%) evaluated the link between obesity (BMI and abdominal diameter) and GORD. There was an association between BMI (mostly abdominal diameter) and GORD in Caucasians but none in Asians or Blacks. The flaw in this study was to use Caucasian overweight and obese BMI cut-offs for Asians and Blacks [51]. Two much smaller studies involving exclusively multiethnic Asian patients found no difference in BMI between those with GORD/BO and controls [25, 40]. Again, the above-mentioned studies suffered with a similar drawback of using Caucasian cut-offs for the overweight and obese categories. Moreover, abdominal girth measurement was not undertaken. In a cross-sectional study of 1495 Japanese subjects undergoing health check-ups, it was found that 127 (8%) had reflux oesophagitis. Multivariate logistic regression analyses revealed that waist circumference (measured in cm at the mid-point between the last rib and the iliac crest) and hiatus hernia were associated with reflux oesophagitis in men [52]. In a case–control study of 3539 Koreans with reflux oesophagitis, waist circumference (OR = 1.47) and triglyceride concentration (OR = 1.20) independently increased the risk for this condition [53].

In Singapore, Indians also have the highest mean BMI as compared to the other races [54]. Moreover, the metabolic syndrome is most prevalent amongst Indians as compared to the Chinese or Malays [55]. It is therefore tempting to speculate that abdominal adiposity/metabolic syndrome could play a role in explaining ethnic differences in BO amongst Asians.

Hiatus hernia

There is a strong association of BO with the presence of a hiatus hernia in both Asians and Caucasians [31, 40, 56, 57]. In the West, more than 90% of patients with BO have a hiatal hernia, whereas in Japan 70% of BO individuals aged above 70 had one [58]. In Malaysia, 52.7% of patients with BO (both short and long segment) were found to have a hiatus hernia, and in Korea, it was 13% [31, 40]. Unpublished data by the author (SR) reveal ethnic differences in the relative contribution of a hiatal hernia to the risk and severity of GORD and BO amongst Malay, Chinese and Indian patients [40]. Epidemiological studies have suggested that hiatus hernia was a lifestyle-related factor in the development of GORD [59]. Recently though, collagen type III alpha (COL3A1) on chromosome 2 has been identified as a disease-associated gene in both paediatric (Australian) and adult (Swedish) GORD. Moreover, it has been genetically associated with hiatus hernia in adult males [60]. Interestingly, a study of familial hiatal hernia in a large five-generation family revealed true autosomal dominant inheritance suggesting further genetic basis for the development of this condition [61].

The presence of a hiatus hernia may increase reflux and impair oesophageal clearance by lowering lower oesophageal sphincter pressure, trapping refluxate in the sac and interfering with the antireflux mechanism of the crural diaphragm [62, 63].

Smoking

There is a moderate risk for current smokers to develop BO, although this has not been a universal finding [64]. In Asians with BO, smoking has been identified as a risk factor in Korea and Malaysia [40, 65]. Again, there is no data on interethnic differences on this lifestyle habit amongst Asians.

Diet

Apparently, curry promotes reflux [66]. Anecdotally, most of us know it is true. Nevertheless, there are no data to support the theory that ethnic differences in curry or chilli consumption could explain the racial differences in BO. In the author’s opinion, much of South-east Asian cuisine is a fusion of Indian, Malay and Chinese food. As such, it would make it difficult to conduct studies investigating the relationship if any between spice/chilli consumption and the development of BO in the different races in that region. Nevertheless, it has been shown that the risk of BO is moderately decreased with increasing intake of fruit and vegetables [67, 68].

Conclusions and perspective

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genetic influences on ethnic differences in Barrett’s oesophagus in Asians
  5. Environmental effects on Barrett’s oesophagus in Asians
  6. Conclusions and perspective
  7. Conflict of interest statement
  8. References

In the author’s opinion, BO is being increasingly diagnosed in Asia as endoscopists are more aware of this condition, a true increase in this condition as a result of increasing prevalence of GORD, abdominal adiposity, widespread eradication of H. pylori and increasing westernization of the Asian diet. The growing economic strength of many Asian countries has enabled more to be spent on health care and thus increased availability of endoscopic equipment, including chromoendoscopy. The growing research interest in BO amongst Asian gastroenterologists and scientists should also result in a better understanding of the reasons for the increasing prevalence of BO in that region.

Future research should target the search for GORD genes, ethnic differences in parietal cell mass, hiatal hernia, H. pylori colonization factors like MUC1 and MUC2, Lewis antigens, BabA (H. pylori adhesion) and host receptors. Inflammatory mediators including IL-1β and tumour necrosis factor α should also be explored to further elucidate the reasons for the observed interethnic variations in BO. Racial differences in lifestyle factors like abdominal adiposity, consumption of fruit and vegetables as well as smoking and possibly alcohol intake should all be investigated as potential causes/risk factors.

‘Nature or nurture; the clues are teasing and tantalising and illustrate the complex relationship between the genetic makeup of man, migration, the environment and the emergence of disease. This is a story set to run!’ (K.D. Bardhan, Personal communication).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Genetic influences on ethnic differences in Barrett’s oesophagus in Asians
  5. Environmental effects on Barrett’s oesophagus in Asians
  6. Conclusions and perspective
  7. Conflict of interest statement
  8. References