Incidence and Epidemiology of Irritable Bowel Syndrome After a Large Waterborne Outbreak of Bacterial Dysentery

Background & Aims: Postinfectious irritable bowel syndrome (PI-IBS) is a common clinical phenomenon. To better define its incidence and epidemiology, a large cohort study was initiated after the contamination of a municipal water supply led to a large outbreak of acute Escherichia coli 0157:H7 and Campylobacter jejuni gastroenteritis. Methods: Local residents were invited to undergo structured assessments at research clinics established 2 years after the outbreak. Permanent adult residents with no prior history of inflammatory bowel disease or IBS were eligible. Standardized questionnaires defined past and current health. The cohort was divided into controls without gastroenteritis, subjects with clinically suspected gastroenteritis, and subjects with only self-reported gastroenteritis that could not be substantiated by another source. A modified Bowel Disease Questionnaire identified IBS according to Rome criteria. The incidence and epidemiology of PI-IBS was characterized. Risk factors were assessed using multiple logistic regression. Results: There were 2069 eligible study participants. Rome I criteria were met by 71 of 701 controls (10.1%) vs 249 of 904 subjects with self-reported gastroenteritis (27.5%) and 168 of 464 subjects with clinically suspected gastroenteritis (36.2%) (all comparisons, P < 001). Independent risk factors for PI-IBS included younger age, female sex, bloody stools, abdominal cramps, weight loss, and prolonged diarrhea. PI-IBS was more likely than sporadic IBS to show diarrhea-predominant features. Conclusions: PI-IBS is common after gastroenteritis from water contamination and often is diarrhea-predominant. Characteristics of the acute illness identify patients at increased risk for PI-IBS. Background & Aims: Postinfectious irritable bowel syndrome (PI-IBS) is a common clinical phenomenon. To better define its incidence and epidemiology, a large cohort study was initiated after the contamination of a municipal water supply led to a large outbreak of acute Escherichia coli 0157:H7 and Campylobacter jejuni gastroenteritis. Methods: Local residents were invited to undergo structured assessments at research clinics established 2 years after the outbreak. Permanent adult residents with no prior history of inflammatory bowel disease or IBS were eligible. Standardized questionnaires defined past and current health. The cohort was divided into controls without gastroenteritis, subjects with clinically suspected gastroenteritis, and subjects with only self-reported gastroenteritis that could not be substantiated by another source. A modified Bowel Disease Questionnaire identified IBS according to Rome criteria. The incidence and epidemiology of PI-IBS was characterized. Risk factors were assessed using multiple logistic regression. Results: There were 2069 eligible study participants. Rome I criteria were met by 71 of 701 controls (10.1%) vs 249 of 904 subjects with self-reported gastroenteritis (27.5%) and 168 of 464 subjects with clinically suspected gastroenteritis (36.2%) (all comparisons, P < 001). Independent risk factors for PI-IBS included younger age, female sex, bloody stools, abdominal cramps, weight loss, and prolonged diarrhea. PI-IBS was more likely than sporadic IBS to show diarrhea-predominant features. Conclusions: PI-IBS is common after gastroenteritis from water contamination and often is diarrhea-predominant. Characteristics of the acute illness identify patients at increased risk for PI-IBS. See CME Quiz on page 660. Between 5% and 30% of patients who suffer an acute episode of infectious gastroenteritis develop chronic gastrointestinal symptoms despite clearance of the inciting pathogen.1Borgaonkar M.R. Ford D.C. Marshall J.K. Churchill E. Collins S.M. The incidence of irritable bowel syndrome among community subjects with previous enteric infection.Dig Dis Sci. 2005; (e-pub ahead of print, Pub Med ID 16758307)Google Scholar, 2Gwee K.-A. Leong Y.-L. Graham C. McKendrick M.W. Collins S.M. Walters S.J. Underwood J.E. Read N.W. The role of psychological and biological factors in post-infective gut dysfunction.Gut. 1999; 44: 400-406Crossref PubMed Scopus (705) Google Scholar, 3Neal K.R. Hebden J. Spiller R. Prevalence of gastrointestinal symptoms six months after bacterial gastroenteritis and risk factors for development of the irritable bowel syndrome postal survey of patients.BMJ. 1997; 314: 779-782Crossref PubMed Scopus (526) Google Scholar, 4McKendrick M.W. Read N.W. Irritable bowel syndrome post Salmonella infection.J Infect. 1994; 29: 1-3Abstract Full Text PDF PubMed Scopus (280) Google Scholar, 5Wang L.H. Fang X.C. Pan G.Z. Bacillary dysentery as a causative factor of irritable bowel syndrome and its pathogenesis.Gut. 2004; 53: 1096-1111Crossref PubMed Scopus (342) Google Scholar, 6James C. Thabane M. Borgaonkar M. Marshall J.K. Post-infectious irritable bowel syndrome is transient following a food-borne outbreak of acute gastroenteritis attributed to a viral pathogen.Gastroenterology. 2004; 126: A53Google Scholar, 7Ji S. Park H. Lee D. Song Y.K. Choi J.P. Lee S.I. Post-infectious irritable bowel syndrome in patients with Shigella infection.J Gastroenterol Hepatol. 2005; 20: 381-386Crossref PubMed Scopus (133) Google Scholar, 8Mearin F. Perez-Oliveras M. Perello A. Vinyet J. Ibanez A. Coderch J. Perona M. Dyspepsia and irritable bowel syndrome after a Salmonella gastroenteritis outbreak one-year follow-up cohort study.Gastroenterology. 2005; 129: 98-104Abstract Full Text Full Text PDF PubMed Scopus (353) Google Scholar This common clinical phenomenon of postinfectious irritable bowel syndrome (PI-IBS) was first described more than 5 decades ago.9Stewart G.T. Post-dysenteric colitis.BMJ. 1950; 1: 405-409Crossref PubMed Scopus (85) Google Scholar However, its pathogenesis remains poorly understood and no specific therapy has been identified. For the most part, previous studies of the epidemiology of PI-IBS have assessed relatively small population cohorts with limited power to define risks and factors that modify the effect.1Borgaonkar M.R. Ford D.C. Marshall J.K. Churchill E. Collins S.M. The incidence of irritable bowel syndrome among community subjects with previous enteric infection.Dig Dis Sci. 2005; (e-pub ahead of print, Pub Med ID 16758307)Google Scholar, 2Gwee K.-A. Leong Y.-L. Graham C. McKendrick M.W. Collins S.M. Walters S.J. Underwood J.E. Read N.W. The role of psychological and biological factors in post-infective gut dysfunction.Gut. 1999; 44: 400-406Crossref PubMed Scopus (705) Google Scholar, 3Neal K.R. Hebden J. Spiller R. Prevalence of gastrointestinal symptoms six months after bacterial gastroenteritis and risk factors for development of the irritable bowel syndrome postal survey of patients.BMJ. 1997; 314: 779-782Crossref PubMed Scopus (526) Google Scholar, 4McKendrick M.W. Read N.W. Irritable bowel syndrome post Salmonella infection.J Infect. 1994; 29: 1-3Abstract Full Text PDF PubMed Scopus (280) Google Scholar, 5Wang L.H. Fang X.C. Pan G.Z. Bacillary dysentery as a causative factor of irritable bowel syndrome and its pathogenesis.Gut. 2004; 53: 1096-1111Crossref PubMed Scopus (342) Google Scholar, 6James C. Thabane M. Borgaonkar M. Marshall J.K. Post-infectious irritable bowel syndrome is transient following a food-borne outbreak of acute gastroenteritis attributed to a viral pathogen.Gastroenterology. 2004; 126: A53Google Scholar, 7Ji S. Park H. Lee D. Song Y.K. Choi J.P. Lee S.I. Post-infectious irritable bowel syndrome in patients with Shigella infection.J Gastroenterol Hepatol. 2005; 20: 381-386Crossref PubMed Scopus (133) Google Scholar, 8Mearin F. Perez-Oliveras M. Perello A. Vinyet J. Ibanez A. Coderch J. Perona M. Dyspepsia and irritable bowel syndrome after a Salmonella gastroenteritis outbreak one-year follow-up cohort study.Gastroenterology. 2005; 129: 98-104Abstract Full Text Full Text PDF PubMed Scopus (353) Google Scholar To access larger populations, researchers have relied on large databases with inherent limitations.10Garcia Rodriguez L.A. Ruigomez A. Increased risk of irritable bowel syndrome after bacterial gastroenteritis cohort study.BMJ. 1999; 318: 565-566Crossref PubMed Google Scholar Furthermore, only a few studies have assessed the long-term natural history of PI-IBS.11Neal K.R. Barker L. Spiller R.C. Prognosis in post-infective irritable bowel syndrome a 6-year follow-up.Gut. 2002; 51: 410-413Crossref PubMed Scopus (222) Google Scholar Walkerton is a small rural town in Ontario, Canada. In May 2000, heavy rainfall washed livestock fecal residue from nearby farms into inadequately chlorinated drinking water supplied from a shallow well.12Anonymous. The investigative report of the Walkerton outbreak of waterborne gastroenteritis May-June 2000. Bruce-Grey-Owen Sound Health Unit, 2000:1–67.Google Scholar Contamination of the regional water supply with Escherichia coli 0157:H7, Campylobacter jejuni, and other pathogens led to a large outbreak of acute bacterial gastroenteritis that affected at least 2300 local residents, with 27 recognized cases of the hemolytic uremic syndrome and 7 deaths.12Anonymous. The investigative report of the Walkerton outbreak of waterborne gastroenteritis May-June 2000. Bruce-Grey-Owen Sound Health Unit, 2000:1–67.Google Scholar, 13O’Connor D.R. Report of the Walkerton inquiry part 1 the events of May 2000 and related issues. The Walkerton Inquiry, Toronto2002: 1-504Google Scholar Since then, many residents have described continued abdominal discomfort and altered bowel habit. A large cohort study was undertaken in this population to define the nature and epidemiology of PI-IBS. The Walkerton Health Study (WHS) was initiated 2 years after the outbreak to study the epidemiology and long-term health outcomes of the municipal water contamination that occurred in May 2000, and to facilitate local residents’ access to specialty clinical care. Some details of its methodology have been reported elsewhere.14Garg A.X. Moist L. Matsell D. Thiessen-Philbrook H.R. Haynes R.B. Suri S.S. et al.Risk of hypertension and reduced kidney function after acute gastroenteritis from bacteria-contaminated drinking water.CMAJ. 2005; 173: 261-268Crossref PubMed Scopus (26) Google Scholar, 15Garg A.X. MacNab J. Clark W. Ray J.H. Marshall J.K. Suri R.S. Devereaux P.J. Haynes R.B. Walkerton Health Study InvestigatorsLong-term health sequelae following E. coli and Campylobacter contamination of municipal water population sampling and assessing non-participation biases.Can J Public Health. 2005; 96: 125-130PubMed Google Scholar, 16Garg A.X. Marshall J.K. Salvadori M. Thiessen H.R. MacNab J. Suri R.S. Haynes R.B. Pope J. Clark W. Walkerton Health Study Investigators. A gradient of acute gastroenteritis to assess risk of long-term health sequelae after drinking bacteria contaminated water.J Clin Epidemiol. 2006; 59: 421-428Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar The WHS is directed by a multidisciplinary team of investigators and receives unrestricted funding from the Ontario provincial Ministry of Health and Long-Term Care. Soon after its inception, the WHS established a special multidisciplinary ambulatory clinic at the local Walkerton Hospital to conduct annual structured assessments of all affected and unaffected residents of Walkerton and its surrounding areas. Annual reassessment is planned to continue until 2009. At study entry, all WHS participants underwent a standardized interview that included a 20-minute structured survey questionnaire. The questionnaire addressed several domains including: (1) demographics, (2) municipal water exposure, (3) prior medical history, (4) details of any acute illness experienced during the outbreak, (5) current gastrointestinal symptoms, and (6) medical history since the outbreak. All WHS participants also were asked for written consent to link WHS data with ancillary health records, largely to verify self-reported information about premorbid health and acute illness during the outbreak. These records included the following: (1) individual responses to a survey of 899 Walkerton residents conducted by the local Bruce-Gray-Owen Sound Health Unit once the outbreak was recognized, (2) regional hospital laboratory data, (3) pharmacy records, (4) family physician charts, and (5) Walkerton Hospital medical records. All such data were abstracted by trained research assistants using standardized forms. WHS participants who fulfilled the following eligibility criteria were eligible for the study cohort: (1) at least 16 years of age at the time of the outbreak, (2) no diagnosis of IBS or inflammatory bowel disease before the outbreak, and (3) permanent residency in Walkerton (as identified by postal code) at the time of the outbreak. We divided all participants into 3 groups to characterize acute gastroenteritis experienced at the time of the outbreak: (1) those who reported no acute illness during the outbreak (controls), (2) those who reported an acute illness that could not be substantiated by prior health records (self-reported gastroenteritis), and (3) those with symptoms whose illness was substantiated by review of health records (clinically suspected gastroenteritis). Gastroenteritis was substantiated by any of the following: (1) documented health care contact for acute gastrointestinal symptoms during the outbreak, (2) bloody diarrhea in May 2000 reported in health records or in responses to the Bruce-Gray-Owen Sound Health Unit survey, (3) diarrhea lasting at least 3 days with more than 3 stools per day reported on the Bruce-Gray-Owen Sound Health Unit survey, or (4) a documented positive stool culture. In a previous analysis, the ability to confirm gastroenteritis was found to correlate with severity of the acute illness, and with preliminary measures of chronic health outcome.16Garg A.X. Marshall J.K. Salvadori M. Thiessen H.R. MacNab J. Suri R.S. Haynes R.B. Pope J. Clark W. Walkerton Health Study Investigators. A gradient of acute gastroenteritis to assess risk of long-term health sequelae after drinking bacteria contaminated water.J Clin Epidemiol. 2006; 59: 421-428Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar A modified version of Talley’s Bowel Disease Questionnaire17Talley N.J. Phillips S.F. Melton J. Wiltgen C. Zinsmeister A.R. A patient questionnaire to identify bowel disease.Ann Intern Med. 1989; 111: 671-674Crossref PubMed Scopus (426) Google Scholar was administered to all subjects at least 16 years of age and residents of Walkerton in May 2000. Rome I diagnostic criteria were applied to these responses to identify participants with IBS.18Thompson W.G. Creed F.H. Drossman D.A. et al.Functional bowel disorders and functional abdominal pain.Gastroenterol Int. 1992; 5: 75-91Google Scholar The prevalence of IBS was estimated in each cohort stratum as a proportion with a 95% confidence interval. The association between acute gastroenteritis (clinically suspected and/or self-reported) and a diagnosis of IBS was tested using χ2 analysis adjusted for multiple comparisons using Bonferroni correction. Relative risks were calculated using controls as the referent group. Among subjects who experienced acute gastroenteritis, risk factors for developing IBS were tested using univariate and multivariate logistic regression. Risk factors considered in this analysis included age, sex, exposure status, specific features of illness during outbreak, self-reported premorbid anxiety or depression, antibiotic use, and direct contact with livestock. A similar approach was used previously to test associations between acute gastroenteritis and long-term renal dysfunction in the same population.14Garg A.X. Moist L. Matsell D. Thiessen-Philbrook H.R. Haynes R.B. Suri S.S. et al.Risk of hypertension and reduced kidney function after acute gastroenteritis from bacteria-contaminated drinking water.CMAJ. 2005; 173: 261-268Crossref PubMed Scopus (26) Google Scholar All analyses were conducted using SPSS software (version 13.0; SPSS Inc., Chicago, IL). The study protocol received full approval from both the Hamilton Health Sciences/McMaster University Research Ethics Board (Hamilton, Ontario) and the University of Western Ontario’s Office of Research Ethics (London, Ontario). The WHS enrolled a total of 4315 subjects; 3959 entered 2 years after the outbreak and 356 entered 3 years after the outbreak. A previous analysis of participation bias in this cohort confirmed that the demographics of the study sample were similar to the affected population, and that potential participation biases (also called response bias or selection bias) would not lead to overestimates of risk between acute bacterial gastroenteritis and long-term health sequelae.15Garg A.X. MacNab J. Clark W. Ray J.H. Marshall J.K. Suri R.S. Devereaux P.J. Haynes R.B. Walkerton Health Study InvestigatorsLong-term health sequelae following E. coli and Campylobacter contamination of municipal water population sampling and assessing non-participation biases.Can J Public Health. 2005; 96: 125-130PubMed Google Scholar Of the 4315 WHS participants, 2246 (52.1%) were excluded from the study cohort at risk of IBS. Specifically, 1268 were not permanent residents of the Walkerton postal code during the outbreak, 1180 were younger than 16 years of age at the time of the outbreak, and 247 reported a prior diagnosis of IBS or inflammatory bowel disease. Among the 2069 eligible study cohort participants, 921 (44.5%) were male and the mean age was 46.6 years (Table 1). All but 6 participants reported having consumed Walkerton tap water before the outbreak. A total of 464 participants (22.4%) fulfilled criteria for clinically suspected gastroenteritis, 904 (43.7%) had self-reported gastroenteritis, and 701 (33.9%) were classified as controls.Table 1Characteristics of Eligible Study CohortControls (N = 701)Self-reported gastroenteritis (N = 904)Clinically suspected gastroenteritis (N = 464)All cases of gastroenteritis (N = 1368)Female, N (%)388 (55.3)493 (54.5)267 (57.5)760 (55.6)Caucasian, N (%)692 (98.7)900 (99.6)459 (98.9)1359 (99.6)Age at time of outbreak, y: Mean (SD)49.4 (17.5)45.2 (17.0)45.1 (17.4)45.2 (17.1)Self-reported symptoms during outbreak, N (%) Duration of diarrhea Not reported–94 (10.4)37 (8.0)131 (9.6) 0–1 day–67 (7.4)19 (4.1)86 (6.3) 2–3 days–238 (26.3)75 (16.2)313 (22.9) 4–5 days–141 (15.6)75 (16.2)216 (15.8) 6–7 days–132 (14.6)77 (16.6)209 (15.3) >7 days–232 (25.7)181 (39.0)413 (30.2) Maximum no. of loose stools/day Not reported–133 (14.7)67 (14.4)200 (14.6) 1–3 loose stools/day–162 (17.9)40 (8.6)202 (14.8) 4–6 loose stools/day–305 (33.7)112 (24.1)417 (30.5) 7–10 loose stools/day–168 (18.6)103 (22.2)271 (19.8) > 10 loose stools/day–136 (15.0)142 (30.6)278 (20.3) Bloody diarrhea–148 (16.4)156 (33.6)304 (22.2) Abdominal pain–776 (85.8)420 (90.5)1196 (87.4) Weight loss > 10 pounds–145 (16.0)150 (32.3)295 (21.6) Fever–280 (31.0)221 (47.6)501 (36.6)Health care use during outbreak, N (%) Family physician visit–11 (12.3)118 (25.4)229 (16.7) Emergency room visit or hospital admission–74 (8.2)227 (48.9)301 (22.0)Stool cultures during outbreak, N (%) Negative3 (4.3)1 (.1)127 (27.4)128 (9.4) E coli 0157:H7 alone0 (.0)0 (.0)21 (4.5)21 (1.5) Campylobacter alone0 (.0)0 (.0)35 (7.5)35 (2.6) Both E coli 0157:H7 and Campylobacter0 (.0)0 (.0)3 (0.6)3 (.0) Open table in a new tab Because of constraints on local resources during the initial outbreak, only a minority of local residents were encouraged to submit stool specimens for culture. Among the 2069 eligible study cohort participants, 35 had documented stool cultures positive for Campylobacter species, 21 had cultures positive for E coli 0157:H7, and 3 had cultures positive for both (Table 1). Overall, 488 subjects (23.6%) met Rome I criteria for IBS at their first WHS visit in 2002 or 2003. This included 71 of 701 controls (10.1%) vs 249 of 904 subjects with self-reported gastroenteritis (27.5%; P < .001 vs controls) and 168 of 464 subjects with clinically suspected gastroenteritis (36.2%; P < .001 vs controls and vs self-reported gastroenteritis). The odds ratio for developing IBS with clinically suspected gastroenteritis vs controls was 4.8 (95% confidence interval, 3.4–6.8) (Figure 1). Among participants who developed acute gastroenteritis during the outbreak, risk factors for subsequent IBS were assessed by multiple logistic regression analysis. Independent predictors of IBS included younger age, female sex, and 4 features of the acute enteric illness: duration of diarrhea (significant if >7 days), presence of blood in the stools, abdominal cramps, and weight loss of at least 10 pounds (Table 2).Table 2Multiple Logistic Regression Analysis to Identify Independent Predictors of IBS After Acute Gastroenteritis (Either Self-Reported or Clinically Suspected)Variableβ coefficient (standard error)P valueOdds ratio with 95% confidence intervalAge, y−.009 (.004).029−.991 (.983–.999)Female sex.383 (.141).0061.466 (1.116–1.928)Features of acute enteric illness Duration of diarrhea, days 0–1Reference 2–3.295 (.355).4061.344 (.670–2.696) 4–5.541 (.363).1371.717 (.843–3.499) 6–7.409 (.366).2641.505 (.735–3.084) >7.862 (.348).0132.369 (1.198–4.686) Bloody stools.506 (.155).0011.658 (1.225–2.245) Abdominal cramps1.467 (.362).0004.337 (2.133–8.816) Weight loss > 10 pounds.601 (.162).0001.824 (1.329–2.504) Constant−2.698NOTE. Covariates were included in the multivariable model if P < .10 in the univariate analysis. Statistically nonsignificant covariates included in the multivariable model but not shown include peak stool frequency, fever, and premorbid anxiety or depression. Open table in a new tab NOTE. Covariates were included in the multivariable model if P < .10 in the univariate analysis. Statistically nonsignificant covariates included in the multivariable model but not shown include peak stool frequency, fever, and premorbid anxiety or depression. Individual Bowel Disease Questionnaire responses were compared among IBS subjects with and without a history of acute gastroenteritis (either self-reported or clinically suspected). Those who had IBS after gastroenteritis (PI-IBS) were more likely than those with IBS but no gastroenteritis (sporadic IBS) to report increased stool frequency (72.2% vs 60.6%, P = .047) and looser bowel movements (88.5% vs 77.5%, P = .011) with the onset of abdominal discomfort. Those with PI-IBS were also more likely than those with sporadic IBS to report increases in stool frequency (50.8% vs 36.6%, P = .027), watery stools (60.7% vs 39.4%, P < 001), or urgency (81.5% vs 64.8%, P < .001) at least 25% of the time. Overall, these observations suggest that PI-IBS is more likely than sporadic IBS to express a diarrhea-predominant phenotype. The results of this study confirm a strong and highly significant association between acute bacterial dysentery and subsequent IBS symptoms. After an interval of at least 2 years, the relative risk of developing IBS among Walkerton residents who experienced bacterial gastroenteritis from water contamination was increased more than 3-fold (odds ratio, 4.8; 95% confidence interval, 3.4–6.8), with an absolute increase of 26.1%. Younger age, female sex, and several characteristics of the acute enteric illness were identified as independent predictors of IBS. This study has a number of methodologic advantages over previous reports, most notably its access to a large, well-defined, at-risk cohort with a simultaneous and well-characterized acute enteric illness. Follow-up evaluations of other smaller cohorts with bacterial gastroenteritis have estimated that 5%–32% of patients go on to develop PI-IBS.1Borgaonkar M.R. Ford D.C. Marshall J.K. Churchill E. Collins S.M. The incidence of irritable bowel syndrome among community subjects with previous enteric infection.Dig Dis Sci. 2005; (e-pub ahead of print, Pub Med ID 16758307)Google Scholar, 2Gwee K.-A. Leong Y.-L. Graham C. McKendrick M.W. Collins S.M. Walters S.J. Underwood J.E. Read N.W. The role of psychological and biological factors in post-infective gut dysfunction.Gut. 1999; 44: 400-406Crossref PubMed Scopus (705) Google Scholar, 3Neal K.R. Hebden J. Spiller R. Prevalence of gastrointestinal symptoms six months after bacterial gastroenteritis and risk factors for development of the irritable bowel syndrome postal survey of patients.BMJ. 1997; 314: 779-782Crossref PubMed Scopus (526) Google Scholar, 4McKendrick M.W. Read N.W. Irritable bowel syndrome post Salmonella infection.J Infect. 1994; 29: 1-3Abstract Full Text PDF PubMed Scopus (280) Google Scholar, 5Wang L.H. Fang X.C. Pan G.Z. Bacillary dysentery as a causative factor of irritable bowel syndrome and its pathogenesis.Gut. 2004; 53: 1096-1111Crossref PubMed Scopus (342) Google Scholar, 6James C. Thabane M. Borgaonkar M. Marshall J.K. Post-infectious irritable bowel syndrome is transient following a food-borne outbreak of acute gastroenteritis attributed to a viral pathogen.Gastroenterology. 2004; 126: A53Google Scholar Indeed, a British primary care database study identified acute bacterial gastroenteritis as the strongest risk factor identified to date for the development of IBS (adjusted relative risk, 11.9).10Garcia Rodriguez L.A. Ruigomez A. Increased risk of irritable bowel syndrome after bacterial gastroenteritis cohort study.BMJ. 1999; 318: 565-566Crossref PubMed Google Scholar Risk factors for PI-IBS identified previously have included female sex, psychologic profile, and the severity of the acute illness.2Gwee K.-A. Leong Y.-L. Graham C. McKendrick M.W. Collins S.M. Walters S.J. Underwood J.E. Read N.W. The role of psychological and biological factors in post-infective gut dysfunction.Gut. 1999; 44: 400-406Crossref PubMed Scopus (705) Google Scholar, 3Neal K.R. Hebden J. Spiller R. Prevalence of gastrointestinal symptoms six months after bacterial gastroenteritis and risk factors for development of the irritable bowel syndrome postal survey of patients.BMJ. 1997; 314: 779-782Crossref PubMed Scopus (526) Google Scholar Microbial virulence factors also have been described.6James C. Thabane M. Borgaonkar M. Marshall J.K. Post-infectious irritable bowel syndrome is transient following a food-borne outbreak of acute gastroenteritis attributed to a viral pathogen.Gastroenterology. 2004; 126: A53Google Scholar, 19Thornley J.P. Jenkins D. Neal K. et al.Relationship of Campylobacter toxigenicity in vitro to the development of post-infectious irritable bowel syndrome.J Infect Dis. 2001; 184: 606-609Crossref PubMed Scopus (133) Google Scholar Because our study began data collection 2 years after the outbreak we could not generate premorbid psychologic profiles of our participants. However, results of our multiple logistic regression analysis are otherwise consistent with these reports. Participation of Walkerton residents in the study cohort was voluntary, in response to an intense local media campaign. We enrolled 3047 residents of the geographic area defined by the Walkerton postal code, including 2756 residents of the Walkerton town (55% of its estimated population).15Garg A.X. MacNab J. Clark W. Ray J.H. Marshall J.K. Suri R.S. Devereaux P.J. Haynes R.B. Walkerton Health Study InvestigatorsLong-term health sequelae following E. coli and Campylobacter contamination of municipal water population sampling and assessing non-participation biases.Can J Public Health. 2005; 96: 125-130PubMed Google Scholar We restricted our study cohort to Walkerton residents to better assess participation bias. We compared our study cohort with the local population using local census data and public health records to assess sample representativeness.15Garg A.X. MacNab J. Clark W. Ray J.H. Marshall J.K. Suri R.S. Devereaux P.J. Haynes R.B. Walkerton Health Study InvestigatorsLong-term health sequelae following E. coli and Campylobacter contamination of municipal water population sampling and assessing non-participation biases.Can J Public Health. 2005; 96: 125-130PubMed Google Scholar Briefly, the age and sex distribution of WHS participants was found to be similar to that of the local population apart from a slight overrepresentation of women (55% of sample vs 52% of population) and a slight underrepresentation of young adults and the very elderly. Furthermore, participants who joined in the second year of the study were similar to those enrolled in its first year, but were less likely to have been unwell during the outbreak. Although participation bias cannot be excluded, this analysis concluded that the study sample was sufficiently representative of the target population to generate relatively unbiased estimates of disease association. For various reasons, including constraints placed on local resources by the magnitude of the outbreak and delays in funding and establishing the research clinic, subjects were not enrolled until 2 years after the outbreak. Accordingly, recall of premorbid illness and details of any acute illness experienced during the outbreak could have been compromised. Accordingly, we sought to confirm many of these data by linking WHS responses to local health records and to responses to questionnaires administered by local health authorities during the outbreak. It is of interest that the observed relationship between acute gastroenteritis and IBS was strongest among subjects whose acute illness was substantiated by at least one other source. In part, this gradient could reflect inaccurate recall of acute symptoms in those with unconfirmed illness. However, it is more likely that subjects with more severe acute illness were more likely to seek health care and/or submit stool specimens for culture. Accordingly, confirmation of illness is, in part, a marker of illness severity. The prevalence of IBS in our participants with clinically suspected gastroenteritis is among the highest ever reported. Despite the size of the study cohort, all participants were assessed for diagnoses of IBS by direct interview using a standardized questionnaire. We excluded subjects with premorbid IBS from the study cohort subjects by asking about prior clinical diagnoses. It would have been preferable, but infeasible, to have administered our standardized questionnaire to subjects before the outbreak. Accordingly, we recognize that some of the subjects we diagnosed with IBS with onset after acute gastroenteritis may have had IBS before the outbreak. PI-IBS is the only functional disorder whose onset is predictable and associated with a clear trigger. Hence, by defining clinical and genetic risk factors for PI-IBS and a better understanding of its natural history, one could select high-risk individuals for intervention to abort or attenuate its course. Although the pathogenesis of PI-IBS remains uncertain, animal models and human observation suggest that subclinical inflammation and altered intestinal permeability are important.20Marshall J.K. Thabane M. Garg A.X. Clark W. Meddings J. Collins S.M. Intestinal permeability in patients with irritable bowel syndrome after a waterborne outbreak of acute gastroenteritis in Walkerton, Ontario.Aliment Pharmacol Ther. 2004; 20: 1317-1322Crossref PubMed Scopus (235) Google Scholar, 21Spiller R.C. Jenkins D. Thornley J.P. Hebden J.M. Wright T. Skinner M. Neal K.R. Increased rectal mucosal enteroendocrine cells, T lymphocytes, and increased gut permeability following acute Campylobacter enteritis and in post-dysenteric irritable bowel syndrome.Gut. 2000; 47: 804-811Crossref PubMed Scopus (999) Google Scholar, 22Barbara G. De Giorgio R. Deng Y. Vallance B. Blennerhassett P. Collins S.M. Role of immunologic factors and cyclooxygenase 2 in persistent postinfective enteric muscle dysfunction in mice.Gastroenterology. 2001; 120: 1729-1736Abstract Full Text Full Text PDF PubMed Scopus (88) Google Scholar, 23Collins S.M. A case for an immunological basis for irritable bowel syndrome.Gastroenterology. 2002; 122: 2078-2080Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar A clinical trial of oral prednisolone to prevent PI-IBS was disappointingly negative.24Dunlop S.P. Jenkins D. Neal K.R. et al.Randomized double-blind placebo-controlled trial of prednisolone in post-infectious irritable bowel syndrome.Aliment Pharmacol Ther. 2003; 18: 77-84Crossref PubMed Scopus (154) Google Scholar However, the concept of anti-inflammatory therapy soon after inoculation remains promising. In summary, results of this study confirm a strong and significant relationship between acute enteric infection and subsequent IBS symptoms. Long-term follow-up evaluation of this cohort is underway to clarify the natural history and sequelae of PI-IBS in this population. The authors thank Ms. Arlene Richards and all staff and students who assisted with data collection, and Ms. Heather Thiessen-Philbrook who performed previous statistical analyses cited in the current publication. Continuing Medical Education Exam 2: August 2006GastroenterologyVol. 131Issue 2Preview Full-Text PDF This Month in GastroenterologyGastroenterologyVol. 131Issue 2PreviewIn view of the high prevalence of functional gastrointestinal disorders, clinicians who are confronted with gastrointestinal symptoms in daily practice need to decide whether or not they need to pursue additional testing, or whether empirical therapy can be considered as a first approach. For lower gastrointestinal symptoms, it has now been established that the presence of the Rome criteria for the irritable bowel syndrome, in the absence of alarm features, has a very high predictive value for diagnosing irritable bowel syndrome. Full-Text PDF