Irritable bowel syndrome (IBS) remains an enigma to many general physicians and even gastroenterologists. With its variable symptoms of abdominal pain, bloating, constipation, and diarrhea, occurring in the absence of obvious pathology, IBS was for years dismissed as being ‘all in the mind’, or simply blamed on an unhealthy diet. Not surprisingly, for a long time in Asia, there was little clinical and research interest in IBS. Two key developments have served to energize and excite a growing group of gastroenterologists and a diverse group of scientists in Asia. First, the development of pharmacological agents that can accurately and effectively target receptor molecules on intestinal tissues to modulate sensory, motor and secretory functions. Drugs such as alosetron, tegaserod and lubiprostone have given clinicians the prospect of efficacious treatment for many of the symptoms of IBS. Second, results from many studies begin to blur the division between functional and organic disorders. Beginning with post-infectious IBS, and followed in its wake by similar findings in other subsets of IBS, there is rapidly growing evidence of immune activation and altered fecal microbiota in its pathogenesis.1–9
Notwithstanding these exciting developments, many questions pertaining to the etiopathogenesis of IBS remain to be addressed. Asia, home to two-thirds of the world's population, with its tremendous diversity in diets, sociocultural norms, levels of sanitation and genetic polymorphism, provides opportunities to address many questions pertaining to IBS which cannot be investigated elsewhere in the world. The rapid socioeconomic development in the last 20 years has also created a transition in the health and environmental situation in Asia that opens a window for us to test some of the prevailing theories on the etiology of IBS. The prevalence of IBS among Asian communities appears to be on the rise. By the Rome II criteria, the prevalence of IBS in places like Singapore (8.6%) and Japan (9.8%) are comparable to Australia (6.9%) and Europe (9.6%), although not as high as in Canada and the UK (12%).10–14
In the 1960s, IBS was described as a disorder of civilization; based on a Western concept of civilization, reduced dietary fiber and psychological stress were pushed forward as important factors.15 There is little evidence to support the former, while the latter is a dominant factor in only a very small minority of IBS subjects. From a modern perspective, immunological challenge may prove to be a more profound factor. Will increasing hygiene lead to decreasing immune tolerance? Will increasing consumption of key Western dietary elements such as dairy and wheat grains pose antigenic challenge to the intestinal immune system?16 Singapore and Japan are two of the most affluent and Westernized societies with high levels of sanitation and consumption of Western diets. Are we forsaking dietary, psychosocial and physiological attributes in Asian societies that may be protective? To help us to uncover some clues, we will critically review the Asian epidemiology data with these perspectives in mind.
Sociodemographic profile of IBS in Asia
The female predominance reported in the West has not been observed in a number of Asian countries (Table 1). Notably in India, two major recent community studies reported higher prevalence of IBS in the male population. In Mumbai, male prevalence was 7.9% versus female prevalence of 6.9%, and in a pan-Indian study male prevalence was 4.3% versus female prevalence of 4.0%.17,18 Similarly, a higher prevalence in the male population has also been reported in Korea, while in Hong Kong and in Pakistan there was little difference between the male and female population.19–21 Even in several other Chinese populations where the prevalence was numerically higher in the female population, the differences were not statistically different, while the female to male ratios were generally lower than those reported from Western countries.10,22,23 Other than the Indian studies, all these studies employed the Rome II criteria. When comparing these Asian figures against studies from Spain, Australia and Canada using the Rome II criteria, clear differences emerge. Thus, in all these Westernized societies, the prevalence of IBS in women outnumbers that in men by approximately 2:1.12,24,25 The only big study from Asia to show convincingly a female predominance was from Japan where the reported prevalence of IBS by the Rome III criteria in the female population was 16% against a prevalence of 11% in the male population.11
Table 1. Prevalence of irritable bowel syndrome in male and female population
|Hong Kong||Rome II||6.6%||6.5%||Kwan20|
Across Asia, the prevalence of IBS is higher in the younger age groups. Applying the Rome II criteria, IBS was significantly more prevalent in those below 50 years of age than those 50 years and older, in Singapore (9.7% vs 5.8%) and in Taiwan (38% vs 18%).10,23 In Hong Kong, the average age of IBS subjects was younger than non-IBS subjects by 5 years (40 vs 45 years).20 In Korea and Japan, subjects below 30 years of age had the highest prevalence.11,19 In South China, IBS was only marginally more prevalent in subjects below 50 years of age.22
In studies from multiracial societies, like Singapore and Malaysia, the prevalence of IBS appears to be similar in Chinese, Indians and Malays. Using a race-stratified disproportionate random sampling of the Singapore population, Ho et al. reported IBS (Manning criteria) prevalence of 2.1% in Chinese, 3.4% in Malays and 2.6% in Indians.26 In a later study by Gwee, the prevalence of IBS by Rome II criteria was Chinese 8.7%, Indian 10% and Malay 6.8%.10 In Malaysia, a study of medical students reported prevalence of IBS by Rome I criteria of Chinese 16%, Indian 15% and Malay 16%, while a community study using race-stratified disproportionate random sampling, reported IBS prevalence by Rome II criteria of Chinese 18%, Indian 17% and Malay 12%.27,28
There appears to be a trend in the largely urban populations of Beijing, Guangzhou, Hong Kong and Singapore for the better educated and more affluent to have a higher prevalence of IBS.10,22,29 In Pakistan, however, IBS was more common in subjects with lower levels of education, and was more common in laborers than in those with clerical and professional occupations.21 Similarly, a study from the USA has suggested that functional bowel disorders occurred more frequently in the lower household income group.30 Looking across different Asian countries, there was also a trend for higher consultation rates in the more affluent city states (Table 2). In Singapore, Ho et al. recorded the highest rate of medical consultation for IBS at 84%, while consultation rates in Hong Kong and Taipei were 57% and 55%, respectively.23,26,31 In the Taipei study, possible reasons for the high consultation rate included the participation of subjects in a paid physical check-up program, the comprehensive national health insurance system, and the relative affluence of Taipei residents. Of note also, was that these participants generally had a high level of education, with over 80% having at least 12 years of education. On the other hand, more modest consultation rates were reported in Malaysia (43%), Pakistan (39%), Bangladesh (35%) and India (34%).18,21,28,32 In Singapore, Taiwan and Bangladesh, sex did not predict consultation behavior.23,26,32 Female sex was an independent risk factor in one study from Hong Kong, but was not in another study after controlling for anxiety which proved to be the only independent predictor in this study.20,31 In Pakistan and India, more men than women presented for hospital investigations even though the prevalence in the community were almost equal.18,21 In Pakistan, a high psychological distress score was a strong predictor of IBS in men but not in women, while in India there appeared to be more consulters in the higher socioeconomic classes.18,21
Table 2. Rates of consultation for irritable bowel syndrome (IBS)
Looking at the type of symptoms, incomplete evacuation and passage of mucus appears to be common risk factors for consultation in Singapore, Taipei, Hong Kong and Bangladesh.20,23,26,32 In Taipei, consulters had a greater number of symptoms than non-consulters, while a study from Tokyo suggests that an increase in symptoms preceded the decision to consult.23,33 Several other studies from Japan indicate that psychosocial factors do not predict consultation for IBS, and the results of one unusual study were interpreted as indicating that while psychological factors do not determine consultation, a parental history of IBS may contribute to psychological distress in their children who developed IBS.34
Taking into account the immensity and diversity of Asia, the rest of the review will be organized along ethno-geographical lines.
Chinese people in China, Taiwan, Hong Kong, Singapore
By the Manning criteria, the prevalence of IBS in Chinese communities has been reported as 7.3% in Beijing, 11% in Singapore and 12% in Guangzhou. By the Rome II criteria, the prevalence of IBS appeared to be lower, with 5.7% in Guangzhou, 6.6% in Hong Kong and 8.6% in Singapore.10,20,22 There appeared to be problems of interpretation in two studies. The prevalence of IBS in Beijing fell from 7.3% by the Manning criteria to 0.82% by the Rome I criteria.29 A similar discrepancy was seen in a Hong Kong study where the prevalence of IBS ‘dropped’ from 21% and 13% by Manning criteria to 3.8% and 3.6% by Rome II criteria for women and men, respectively.35 Although in the Singapore study the prevalence of IBS was lower with each succeeding criteria— Manning 8.6%, Rome I 10% and Rome II 11%—there was, nevertheless, 96% overall agreement with a κ 0.77 between Manning and Rome II criteria.10 Similarly, a study from Taipei reported good agreement between Rome I and Rome II criteria (89% agreement, κ 0.73).23
It remains possible that IBS may be underdiagnosed and underestimated among Chinese communities. In Hong Kong, it was reported that only 21% of patients with IBS criteria seen by their medical practitioners had received this diagnosis.20 In Singapore, IBS made up 17% of new cases referred to a gastroenterology center in Singapore, more than even peptic ulcer and hepatitis B, and yet, Ho reported a prevalence of only 2.3% in a community study.26,36 In his study, Ho had classified all subjects with upper abdominal symptoms as dyspepsia.26 There appears to be a high frequency of upper abdominal pain and meal-related symptoms (such as bloating) among IBS subjects in Chinese communities. In a later Singapore study, Gwee observed that more than half of community subjects with IBS reported pain in the upper abdomen, and 62% also reported bloating (Table 3). Further, bloating was also found to be an independent predictor of consultation for IBS.10,26 In a study of patients with upper abdominal symptoms referred for gastroscopy by their general practitioners, and subsequently treated with a proton pump inhibitor, 28% were found to have IBS.39 In a study of 481 patients in Taipei who were initially classified as suffering from functional dyspepsia (FD) by the Rome criteria, more than half had upper abdominal pain that was exclusively relieved by defecation or associated with changed bowel pattern.40 When these patients with IBS criteria were excluded, the prevalence of FD fell from 24% to 12%. In Hong Kong and mainland China, 60–70% of IBS patients had concomitant dyspeptic symptoms.31,38 Furthermore, bloating was a more common complaint than abdominal pain with 68% of IBS consulters in Hong Kong reporting bloating and only 32% reporting moderate to severe abdominal pain, while in a Chinese hospital 58% reported abdominal pain against 63% with bloating (Table 3).20,38 General sociological observations suggest that the Chinese place a great importance on the enjoyment of food. Is it possible, then, that in Chinese societies the impact of IBS on the enjoyment of food is viewed as more important than its impact on stool frequency?
Table 3. Prevalence of abdominal pain and bloating in irritable bowel syndrome populations
|Bangladesh community, non-consulters||79%||45%||Masud32|
|Bangladesh community, consulters||84%||49%||Masud32|
|Hong Kong community, non-consulters||11%||83%||Kwan20|
|Hong Kong community, consulters||32%||68%||Kwan20|
|Taiwan, annual medical, non-consulters||100%||26%||Lu23|
|Taiwan, annual medical, consulters||100%||33%||Lu23|
|Korea community, consulters||50%||35%||Han19|
The Rome criteria concentrate on the relationship of symptoms to changes in stool frequency and consistency, and the relief of pain or discomfort with defecation. They do not, however, take into account the relationship to a meal. A study from Sweden has highlighted this weakness of the Rome criteria; even though 50% of patients felt that defecation relieved their pain, daily symptom recording actually demonstrated that the pain in IBS patients had a stronger temporal relationship to eating than to defecation.41 The failure of the Rome criteria to recognize the association of IBS symptoms with meals could potentially contribute to a misdiagnosis of IBS as functional dyspepsia in Asians. Furthermore, IBS subjects in Asia do not appear to be as bothered about their stool frequency. In Singapore, 77% of IBS subjects perceived their bowel habits to be normal, and the sensation of incomplete evacuation, rather than stool frequency, was a predictor of health-care utilization.10,26 Similarly in Taiwan, no differences were found in stool frequency between consulters and non-consulters, whereas, more consulters complained of incomplete evacuation and straining.23
This weakness of the early Rome criteria could potentially contribute to a misdiagnosis of IBS as dyspepsia, and therein to inappropriate surgery. In a study from Taiwan, Lu et al. found that IBS patients were more likely to have had cholecystectomy.23 Another interesting finding was that women who had consulted for dyspepsia had almost a threefold greater chance of a hysterectomy than non-consulters.40 Could it be that these Taiwanese women were dissatisfied with the treatment they received from their doctors for ‘dyspepsia’ and had hoped that a surgical quick fix by their gynecologist would cure them of their bloating? Thus, we have to question the relevance of the Rome criteria for Asia.
With regards to possible etiological factors, a history of dysentery has been reported in several studies from China to be a significant independent risk factor (Beijing odds ratio [OR] = 3.0, Guangzhou OR = 1.63.22,29 In a prospective cohort study in a major hospital in Beijing, 8% of patients presenting with Shigella dysentery developed IBS, whereas only 0.8% of family members not exposed to dysentery developed IBS.42 As with the Western studies, a longer duration of diarrhea was a significant risk factor.42,43 However, unlike the studies from the UK, 77% of women developed IBS, compared to only 36% of men; in China, similar risks were observed for men and women.42,43
The prevalence of IBS in Korea appears to be lower than those reported in Chinese communities. In a community study of 1066 (535 male, 531 female) persons in Seoul, the prevalence of IBS by Rome II criteria was estimated to be only 6.6% with a prevalence in male of 7.1% vs 6.0% in female subjects.19 In another study from Seoul involving 1717 young university students, the prevalence was just as low, only 5.7%.44 In a study from a small inland city, the prevalence, also by Rome II criteria, was only 2.2% with no significant difference between the female (2.6%) and male (1.8%) populations.45 In contrast, the same study reported a high prevalence of uninvestigated dyspepsia (12%), the majority (70%) of whom had a dysmotility-like dyspepsia. Given the observations in the Chinese populations of China, Taiwan, Hong Kong and Singapore, where there was a substantial overlap between IBS and dyspepsia, and a high frequency of meal-related symptoms mistaken for dyspepsia among IBS subjects, it is questioned whether a substantial number of IBS subjects in Korea may have been misclassified as dyspepsia. Surprisingly, in a rural Korean community, the prevalence was as high as 8.6% when applying the Rome I criteria.46
Seoul is the only Asian center besides Beijing that has reported the development of post-infectious IBS. In December 2001, 181 health-care workers in a major hospital were involved in an outbreak of Shigella dysentery.47 In this study, the OR of developing IBS was calculated to be 2.9 at 12 months; as with the Beijing study, the length of diarrhea during the acute illness was an independent risk factor, but the risk of developing post-infectious IBS was the same for men as for women.47 In another study from South Korea, routine colonoscopy was performed as part of a general health screening.48 Random colonic biopsies taken from 42 patients with IBS by Rome II criteria revealed significantly more non-specific inflammatory findings (mucosal hyperplasia, lymphocyte aggregation and eosinophilia) than those taken from asymptomatic subjects.
In Japan, the reported prevalence of IBS varied from 6.1% to as high as 31% depending on the sample size, study population and the criteria used. In the largest and most recent study, which involved 10 000 subjects completing an Internet survey, the prevalence of IBS was 9.8% by Rome II and 13% by Rome III criteria; there was a higher prevalence in women than in men (16% vs 11%).11 The highest prevalence was in the 20–29-year age group (women 22%, men 14%), decreasing progressively with increasing age. In a survey of 2495 newly enrolled undergraduates between 17 and 22 years of age, the prevalence of IBS by Rome II was 11%, with a significantly higher prevalence of in women (16%) than in men (7.7%).49 No information was provided in these two latest studies on consultation rates.
In an earlier survey of 4000 community subjects using a modified Rome I criteria, the estimated prevalence was 6.1%, with a female predominance of 7.8% to male 4.5%.50 In this study, the consultation rate was 25%. In a study of 437 employees undergoing annual health screening, the prevalence of IBS (Rome II criteria) was 14% with the prevalence in men coming close to that in women (13% vs 16%).34 In this study, a total of 67 (59%) subjects had consulted doctors for their IBS symptom, with no significant difference in female to male ratios for consulters and non-consulters.
Not surprisingly, in another study involving 633 medical outpatients, a much higher prevalence of 31% was found compared with the studies involving unselected community subjects.51 In this hospital-based study, no significant difference was found between male and female prevalence of IBS (28% vs 34%) It appears then, that in the Japanese population IBS is more common in women than in men, but in health-care facilities there is increased male representation. This would suggest that Japanese men may either have more access to health care or may be more likely to seek medical attention when they have IBS.
The studies from Japan have also sought to ascertain the contribution of stress and other psychological factors to the development of IBS, and in particular to consultation for IBS. In Kumano's study, significantly more IBS than non-IBS subjects met criteria for agoraphobia and panic disorder, but no significant differences were found between consulters and non-consulters, or between men and women.50 In Kanazawa's study, IBS subjects reported more anxiety, depression and perceived stress than asymptomatic controls, but again no differences were found between consulters and non-consulters, or between male and female subjects.34 In this study, a family history of bowel problems was found to be a risk factor for IBS, independent of psychological factors and consultation behavior.
In Shinozaki's study, perceived stress was present in 55% of patients with IBS by Rome II criteria.51 In this study, patients who failed to meet the strict Rome II criteria were classified as having functional bowel disorders; these patients had similar rates of perceived stress (32%) as patients without bowel symptoms (29%). This study demonstrates that the perception of stress as a risk factor for IBS may be influenced by the selection of patients with more severe symptoms by the Rome II criteria. In a study comparing subjects with IBS by the Rome II criteria, that demand a higher frequency of symptoms, with IBS by the more ‘lenient’ Rome III criteria, it was observed that the former criteria selected patients with greater psychosocial problems and higher consultation rates compared with the latter.52,53 Furthermore, the role of perceived stress could have been overestimated in these Japanese studies, because participating patients were recruited from the clinics of psychosomatic physicians with a special interest in IBS, rather than from general gastroenterology clinics. In particular, the high prevalence of IBS in the Shinozaki study compared with the other much larger community studies, points to a selection bias.11,34,49–51
While these studies indicate that IBS subjects may have more psychological disturbances than non-IBS controls, they do not tell us what drives consultation behavior in Japan as no differences were reported between consulters and non-consulters. A more informative way to address this question was suggested by a recent study that prospectively followed up subjects with IBS criteria who had not sought medical attention. Of 105 subjects who were interviewed at 3-month intervals for 3 years, 37 (35%) became patients during this period, while at the same time, 26 subjects (25%) lost their IBS symptoms.33 This study found that the longer a person had IBS symptoms, the greater the chance of becoming an IBS patient, while those who had had short periods of symptoms were more likely to eventually lose them without seeking medical care (i.e. without becoming a ‘patient’). There was also an exacerbation of symptoms just before a decision to seek medical attention.
The above interesting observations indicate that we should test the possibility that an earlier and more aggressive approach to treatment of symptoms may lead to better resolution of IBS, while at the same time call into question the role of psychological factors in consultation behavior in Asian communities. In an early study from Thailand, it was also reported that rural Thais with IBS rarely attributed their abdominal pain or changes in bowel habit to stress (1.8% and 3.5%, respectively).54 While urban Thais in Bangkok were more likely to attribute their symptoms to stress, this was still substantially below that of a US population (abdominal pain 16% vs 54%; changes in bowel pattern 21% vs 71%). Thus, we should be wary of extrapolating the results of psychological studies from the West to our Asian societies.
South Asia (India, Bangladesh, Pakistan)
Early studies from India were the first to suggest a possibly higher prevalence of IBS in men, with as much as a two- to fourfold predominance compared with women.55–57 As these studies were based on IBS patients attending specialist clinics, this male predominance was simply attributed to greater access to health care. On the other hand, there has been one study of patients attending a psychiatry clinic where it was reported that 75% of 55 IBS patients were female.37 These early Indian studies also found that more than half of patients complained of upper abdominal pain, whereas in Western series the majority of patients presented with lower abdominal pain, and only approximately one-quarter to one-third would complain of pain in the upper abdomen.55–60
Recently, the Indian Society of Gastroenterology (ISG) conducted a study involving close to 3000 IBS patients and 4500 community subjects drawn from 18 centers.18 This study is unique in that the ISG adopted an entirely clinically based diagnosis of IBS rather than any of the Rome criteria. Their findings largely validate the findings of the early Indian studies. The estimated prevalence was 4.2%, comparable in male (4.3%) and female (4.0%) subjects. The consultation behavior of the community group where 33% of men and 38% of women had consulted doctors in the preceding 12 months, infer that this greater male prevalence may not be just due to more health care seeking by men in India. The ISG study also found that 49% of patients with IBS reported epigastric pain, and that 70% complained of abdominal fullness rather than pain (Table 3).18
Studies in other parts of the Indian subcontinent have reported similar findings. In a community-based study of 2549 adults in Mumbai, the sex-adjusted prevalence of IBS in men was 7.9% compared with 6.9% in women.17 In a population-based survey from Pakistan, 13% of women and men met Rome II criteria of IBS.21 It did appear that men had greater access to health care, with more men receiving investigations than women (54% vs 27%).21
In a Bangladeshi rural community comprised of farmers and housewives, the prevalence of IBS by Rome I criteria was 8.5%, 11% in women and 5.8% in men.32 More than half of the IBS subjects also reported pain in the upper abdomen, and in the majority of subjects (64%) the bowel pattern could not be classified into either constipation or diarrhea.32
Similarly, in the pan-India study by Ghoshal, the bowel pattern in the majority of Indian subjects could not be classified based on stool frequency.18 In contrast to the West, where stool frequency between 3/week to 3/day is considered normal, in the Indian population 99% of subjects had stool frequency of 1 or more a day, while among IBS patients, the median stool frequency was twice a day, regardless of whether they had constipation or diarrhea predominant bowel habits.18 Of 1404 subjects who complained of constipation, only 507 (36%) had stool frequency of less than three times a week; of 1252 subjects who complained of diarrhea, only 50 (4%) had stool frequency of more than three times a day. Possible factors that may account for the high stool frequency among Indian subjects and patients with IBS include a relatively high mean dietary fiber intake (IBS patients and healthy subjects both 52 g/day), and faster colonic transit time (mean 18 h in a North Indian population vs 35 h in a Western population using the same method).61–64 Similarly, the mean colonic transit times of healthy Chinese adults in Hong Kong were shorter than in the USA (men 18 h vs 33 h, women 39 h vs 47 h).65,66
Another aspect which may pose difficulties for assessment of patients from India is in the diagnosis of small intestinal bacterial overgrowth (SIBO), which has been implicated in some IBS patients. Two studies from India estimate the prevalence of SIBO to approximate 11%.67,68 A condition known as tropical enteropathy, in which intestinal villi are shortened in the presence of bacterial colonization of the small intestine, may be present in as many as 50% of healthy asymptomatic Indians.69–71 This would mean that presence of SIBO may just be an innocent bystander. Furthermore, several studies from India suggest that past exposure to amoeba rather than predispose to IBS, could even be protective.72–74