Irritable bowel syndrome (IBS) encompasses a wide spectrum of gastrointestinal disorders that affect 12% to 20% of the population worldwide.[1, 2] In a subgroup of patients with IBS, a bout of acute diarrhea or gastroenteritis can be identified as an inciting event that results in the onset of IBS. This condition is known as postinfectious IBS (PI-IBS) and has distinctive clinical features.
The relative risk (RR) of acquiring PI-IBS after acute bacterial enteric infection has been reported to occur in 4 to 31% of cases. Multiple studies have provided evidence that travelers' diarrhea (TD), usually caused by one of a variety of bacterial enteropathogens, is also a risk factor for the acquisition of PI-IBS.[5, 6] International travel during the previous 6 months before onset of the diagnosis of IBS was explored in a general medical population. The study found that 16% of patients with PI-IBS and 8% of overall patients with IBS in a general medicine clinic had a history of international travel 6 months before onset of their chronic intestinal disorder.
We conducted a 6-month follow-up study in a cohort of 817 US students traveling to Mexico with self-described good health to analyze the association of TD with development of persistent abdominal symptoms (PAS) including IBS and to define the variables that predicted their occurrence.
Materials and Methods
- Top of page
- Materials and Methods
- Declaration of Interests
Participants were US students who traveled to the Mexican cities of Cuernavaca and Guadalajara for short-term (2–5 weeks) studies between June 2002 and January 2008. The subjects were consenting adults aged 18 years or older and adolescents ≥16 years who participated with parental consent. They were enrolled within 72 hours of arrival in Mexico and were required to stay for at least 5 days in the country to be included in the study. The exclusion criteria were history of unstable medical illness, preexisting self-identified persisting gastrointestinal symptoms, pregnancy, use of antibiotics in the past week, and lactose intolerance. The cohort was observed for the occurrence, duration, and severity of acute diarrhea during their stay in Mexico. Participants who had soft (would take the shape of a container) or watery (could be poured) stools with a frequency of three or more times in a 24-hour period were instructed to report to the clinic for stool collection and identification of enteropathogens.
Stool specimens were processed for enteric pathogens by methods previously published. Enterotoxigenic Escherichia coli (ETEC) was identified by colony hybridization and enteroaggregative E. coli was detected by HEp2 cell adherence assay. Identification of E. coli subtypes (enterotoxigenic E. coli, enteropathogenic E. coli, enteroaggregative E. coli, and enterohemorrhagic E. coli) was also conducted by polymerase chain reaction. Stool samples were examined for the presence of Cryptosporidium, Giardia, and Entamoeba by commercially available microplate ELISA Kits (Remel ProSpecT, Lenexa, KS, USA). Presence of Salmonella and Providencia ssp. was diagnosed by culture.
A questionnaire based on the Rome II criterion for the diagnosis of chronic gastrointestinal symptoms was administered to the participants on enrollment and 6 months after the return to the United States. Subjects were diagnosed with PAS on the basis of their responses to the questionnaire. PAS cases were categorized as functional abdominal disease (FAD) or IBS based on frequency and severity of symptom reporting (see definitions below). This study was approved by the University of Texas Health Science Center Committee for the Protection of Human Subjects (Houston, TX, USA).
PAS was defined as the presence of abdominal pain and discomfort with no obvious cause by history, with onset within the last 6 months. The symptoms reported were required to be present at least 20% of the time during the 3 months preceding the completion of the follow-up questionnaire. PAS was subclassified into IBS or FAD. IBS was defined as recurrent abdominal pain or discomfort associated with the following: (1) improvement of pain or discomfort with defecation; (2) onset associated with a change in frequency of stool; and (3) onset associated with a change in form (appearance) of stool.
FAD encompassed the presence of chronic gastrointestinal symptoms that did not otherwise meet the criteria for IBS. These included diarrhea, constipation, straining while passing hard and difficult-to-pass stools, bloating, distension of the abdomen or tightening of clothes after meals, passing mucus in stools, tenesmus, urgency, or a change in bowel habits.
Acute TD was defined as the passage of ≥3 unformed stools within a 24-hour period plus one or more signs or symptoms of enteric infection (abdominal pain, cramping, excessive gas, nausea, vomiting, fever, urgency, bloody or mucous stool, or tenesmus).
Participants taking antidiarrheal prophylaxis, dropouts, and nonrespondents to the questionnaire were excluded from the analysis. We used STATA SE version 9 (Stata Corp., College Station, TX, USA) to perform the statistical analysis. The chi-square test of association and Fisher's exact test were used to determine statistically significant association of the sociodemographic factors, and the clinical and microbiology data with the study outcomes of PAS, FAD, and IBS. The p values reported were for two-sided tests. The t-test was used to compare the means of continuous variables. RR was calculated wherever applicable.
- Top of page
- Materials and Methods
- Declaration of Interests
Between 2002 and 2008, we obtained clinical histories from a total of 817 US adults studying in Mexico for 2 to 5 weeks and again 6 months after leaving Mexico. The sociodemographic characteristics of the study group are presented in Table 1. The population represented young adults, with a preponderance of women, spending an average of just over 3 weeks in Mexico, most often during the spring and summer seasons.
Table 1. Epidemiologic characteristics of the study population
|Characteristic||N (%), N = 817|
|African Americans|| 12 (1.5)|
|Asians|| 2 (0.2)|
|Others|| 3 (0.3)|
|Season of travel|
|September to February||110 (13)|
|March to August||797 (87)|
|Age (years), mean (SD)||32.0 ± 14.4|
|Length of stay (days), mean (SD)||23.3 ± 10.7|
In Table 2 the presence (or absence) of chronic abdominal complaints are recorded for those found to be suffering (or not) from defined gastrointestinal syndromes 6 months after leaving Mexico. Subjects meeting the criteria for PAS (either FAD or IBS) more frequently complained of gastrointestinal symptoms compared with subjects not fitting the diagnosis of PAS. The overall incidence of PI-IBS in subjects at 6-month follow-up was 4%. The incidence of PI-IBS without TD while in Mexico was 3.2% compared with 5.7% in subjects who had TD while in Mexico. Thus, 3% of individuals developed PI-IBS as a result of TD.
Table 2. Chronic abdominal symptoms among US residents present 6 months after leaving Mexico
|Symptoms||Healthy at follow-up (n = 589)||FAD at follow-up (n = 196)||Healthy versus FAD at follow-up||IBS at follow-up (n = 32)||Healthy versus IBS at follow-up||PAS at follow-up (FAD + IBS) (n = 228)||Healthy versus PAS at follow-up|
| ||N (%)||N (%)||p value||N (%)||p value||N (%)||p value|
|Abdominal pain||144 (24)||115 (60)||<0.001||31 (100)||<0.001||146 (66)||<0.001|
|Change in bowel habits||99 (17)||79 (41)||<0.001||31 (100)||<0.001||110 (50)||<0.001|
|Loose stools||148 (25)||134 (70)||<0.001||31 (100)||<0.001||165 (74)||<0.001|
|Straining||99 (17)||70 (37)||<0.001||18 (58)||<0.001||88 (40)||<0.001|
|Urgency||105 (18)||93 (49)||<0.001||25 (81)||<0.001||118 (53)||<0.001|
|Tenesmus||115 (19)||93 (49)||<0.001||23 (74)||<0.001||116 (52)||<0.001|
|Mucus in stools||95 (16)||48 (25)||0.01||17 (55)||<0.001||65 (29)||0.01|
|Bloating after eating||109 (18)||119 (62)||<0.001||17 (55)||<0.001||146 (66)||<0.001|
|Loosen clothes after eating||90 (15)||60 (31)||<0.001||16 (52)||<0.001||76 (34)||<0.001|
Men and women experienced PAS and IBS with approximately equal frequency. Hispanics and non-Hispanics were equally affected (Table 3). The timing of travel (spring and summer vs fall and winter) did not influence the occurrence of persistent abdominal complaints. Two variables correlating with increasing rates of persistent symptoms were younger age (for FAD or IBS) and longer length of stay in Mexico (for FAD) as shown in Table 3. When compared with the reference age strata of travelers aged 16 to 24 years, travelers in the 25 to 44 years had a RR reduction of 20% in PAS. This RR reduction was decreased to 40% for participants in the ≥45 years of age group.
Table 3. Distribution of selected characteristics among the North American travelers to Mexico, N (%)
|Variable||Healthy at follow-up (n = 589)||FAD at follow-up (n = 196)||FAD versus healthy, RR (CI), p||IBS at follow-up (n = 32)||IBS versus healthy, RR (CI), p||PAS at follow-up (FAD + IBS) (n = 228)||PAS versus healthy RR (CI), p|
|Male (n = 227)||170 (29)|| 49 (25)||Ref|| 8 (25)||Ref|| 57 (25)||Ref|
|Female (n = 590)||419 (71)||147 (75)||1.2 (0.9–1.5), 0.3||24 (75)||1.2 (0.6–2.6), 0.8||171 (75)||1.2 (0.9–1.5), 0.3|
|Hispanics (n = 88)|| 66 (11)|| 21 (11)||Ref||1 (3)||Ref|| 22 (10)||Ref|
|Non-Hispanics (n = 729)||523 (89)||175 (89)||1.0 (0.7–1.6), 0.8||31 (97)||3.8 (0.7–22.0), 0.2||206 (90)||1.1 (0.8–1.7), 0.6|
|Age (years) mean (SD)||33 ± 0.6||28 ± 0.8||<0.001||29 ± 2.1||0.12||28 ± 0.8||<0.001|
|Age (years)|| || || || || || || |
|<25||276 (66)||126 (30)||Ref||18 (4)||Ref||144 (34)||Ref|
|25–44||150 (75)|| 43 (22)||0.6 (0.4–0.9), 0.02||8 (4)||0.8 (0.3–1.9), 0.64|| 51 (25)||0.7 (0.4–0.9), 0.03|
|≥45||163 (84)|| 27 (14)||0.4 (0.3–0.6), <0.001||6 (4)||0.6 (0.2–1.4), 0.23|| 33 (17)||0.4 (0.2–0.6), <0.001|
|Length of stay (days), mean (SD)||23 ± 0.5||26 ± 0.7||0.001||23 ± 1.8||0.9||25 ± 0.7||0.004|
|Season of travel|| || || || || || || |
|September to February (n = 110)|| 81 (14)|| 26 (13)||Ref||3 (9)||Ref|| 29 (13)||Ref|
|March to August (n = 707)||508 (86)||170 (87)||1.0 (0.7–1.5), 0.9||29 (91)||1.5 (0.5–4.6), 0.6||199 (87)||1.1 (0.8–1.5), 0.7|
Diarrhea more commonly occurred while in Mexico for those later diagnosed with FAD, 101/196 (52%), RR = 1.5 [confidence interval (CI) 1.2–1.8; p = 0.001]; IBS, 20/32 (63%), RR = 2.5 (CI 1.2–5.0; p = 0.007); and PAS, 121/228 (53%), RR = 1.5 (CI 1.2–1.8; p < 0.001), compared with those who were healthy when later surveyed, where diarrhea was reported to have occurred in Mexico in 227/589 (39%) (see Table 4). The number of unformed stools passed per day while in Mexico was increased for subjects found later to have FAD (p = 0.008) and PAS (FAD plus IBS subjects combined) (p = 0.004). The number of watery stools passed per day while in Mexico was not related to PAS. Development of diarrhea in Mexico and the occurrence of multiple bouts of illness were greater in subjects later diagnosed with FAD (RR; 95% CI: 2.3; 1.6–3.0; p < 0.001) and PAS (RR; 95% CI: 2.2; 1.6–2.8; p <0.001) (data not shown in tables).
Table 4. Association of diarrhea in Mexico with persistent abdominal symptoms among US travelers 6 months after return
|Characteristics of travelers' diarrhea||Healthy at follow-up (n = 589), N (%)||FAD at follow-up (n = 196), N (%)||FAD versus healthy RR (95% CI), p value||IBS at follow-up (n = 32), N (%)||IBS versus healthy RR (95% CI), p value||PAS at follow-up (FAD + IBS) (n = 228), N (%)||PAS versus healthy RR (95% CI), p value|
|Diarrhea status in Mexico|| |
|Travelers' diarrhea|| || || || || || || |
|No (n = 469)||362 (61)||95 (48)|| ||12 (38)|| ||107 (47)|| |
|Yes (n = 348)||227 (39)||101 (52)||1.5 (1.2–1.8), 0.001||20 (63)||2.5 (1.2–5.0), 0.007||121 (53)||1.5 (1.2–1.8), <0.001|
|Unformed stools per day of stay in Mexico||0.36 ± 0.01||0.49 ± 0.04||0.008||0.48 ± 0.07||0.08||0.49 ± 0.04||0.004|
|Adjusted watery stools per day of stay in Mexico||0.14 ± 0.01||0.18 ± 0.02||0.14||0.16 ± 0.03||0.60||0.18 ± 0.02||0.13|
As shown in Table 5, statistically significant associations were found between diarrhea due to heat-labile enterotoxin (LT)-producing ETEC (p = 0.05) and Providencia ssp. (p = 0.03). An association of development of PAS was not seen for TD due to other defined pathogens. Shigella spp. and Campylobacter spp. were not identified in subjects with TD acquired in Mexico.
Table 5. Enteric pathogens associated with the development of PAS
|Enteric pathogen identified||Healthy at follow-up, N (%)||FAD at follow-up, N (%)||FAD versus healthy, p value||IBS at follow-up, N (%)||IBS versus healthy, p value||PAS at follow-up (FAD + IBS), N (%)||PAS versus healthy, p value|
|Enteroaggregative Escherichia coli (EAEC)||64/171 (37) ||24/68 (35)||0.8||3/10 (30)||0.7||27/78 (35)||0.8|
|Enterotoxigenic E. coli (ETEC)||57/169 (34) ||28/64 (44)||0.2||2/10 (20)||0.4||30/74 (41)||0.3|
|ETEC-LT||18/169 (11) ||13/64 (20)||0.05||2/10 (20)||0.6||15/74 (20)||0.06|
|ETEC-ST||13/169 (8) || 9/64 (14)||0.1||0/10 (0)||NA|| 9/74 (12)||0.2|
|ETEC-ST/LT||26/169 (15)|| 6/64 (9)||0.5||0/10 (0)||NA|| 6/74 (8)||0.2|
|Enteropathogenic E. coli (EPEC)||32/88 (36)|| 13/37 (35)||1.0|| 3/6 (50)||0.7||16/43 (37)||1.0|
|Enterohemorrhagic E. coli (EHEC)||14/88 (16)|| 5/37 (14)||1.0|| 1/6 (17)||1.0|| 6/43 (14)||1.0|
|Salmonella ssp.||3/143 (2) || 3/70 (4)||0.39||1/11 (10)||0.24||4/81 (5)||0.21|
|Providencia ssp.||0/131 (0) || 3/62 (5)||0.03|| 0/8 (0)||NA||3/70 (4)||0.04|
|Cryptosporidium spp.||6/146 (4) || 6/65 (9)||0.2|| 1/10 (10)||0.4||7/75 (9)||0.1|
- Top of page
- Materials and Methods
- Declaration of Interests
Based on a questionnaire completed 6 months after leaving Mexico, chronic abdominal complaints were commonly found in a large group of young adults traveling to Mexico from the United States. An earlier evaluation of a similar cohort of students reported an 18% frequency of persistent diarrhea 6 months after leaving Mexico, which is consistent with the present findings. In this study, we found a 4% incidence of PI-IBS at 6-month follow-up of previously healthy travelers who did not report abdominal complaints at study enrollment. As reported in other studies, diarrhea-predominant disease was the most frequently observed subtype of PI-IBS that followed enteric infection.[3, 11]
In this study, TD was a significant risk factor for the development of PAS including PI-IBS. The risk of developing FAD and IBS among subjects who reported TD while at Mexico was 1.5 to 2.5 times greater compared with subjects who were healthy during their earlier travel. A literature review of studies on PI-IBS identified a sevenfold increase in the risk of acquiring IBS after a bout of acute infectious diarrhea. A 1-year follow-up of children following water contamination by E. coli 0157:H7 and Campylobacter species revealed a 10.5% incidence of PI-IBS among those exposed compared with 2.5% among controls.35 An 8-year follow-up of adults in the same study found that 15.4% of those who had acute gastroenteritis in the epidemic continued to report symptoms consistent with IBS.36
Other studies have noted an increased risk of PI-IBS development for younger age groups, females, subjects with preexistent anxiety and depression,[3, 13] a longer duration of enteric symptoms with the inciting infection, Campylobacter enteritis, salmonellosis, and shigellosis.[17, 18] We hypothesize that infections of the gut by invasive bacterial enteric pathogens produce intestinal inflammation that appears to be an important event predisposing to PAS[19-22]. Small bowel motility changes can trigger the overgrowth of abnormal bacterial flora causing small intestinal bacterial overgrowth, which is found in idiopathic forms of IBS.[23-25] Other findings in subjects with IBS include mast cell activation and release of mast cell intermediaries that have physiologic effects in the gut[21, 26, 27] and alteration of serotonin metabolism with release of 5-hydroxytryptamine.[28, 29]
No significant difference in gender distribution was noted among IBS and FAD subsets in our population, a possible reason being the exclusion of subjects with chronic abdominal symptoms at baseline and the strong association of PAS with infectious diarrhea, which affects men and women equally. The risk of developing PAS was higher among younger subjects that may be attributed to the age-related lifestyle factors[30, 31] such as a more adventurous dietary intake or a higher volume of potentially contaminated food being consumed. Another possible explanation for decreasing PAS incidence with age is that the large bowel lymphocyte and mast cell numbers decrease with age, which reduces the inflammatory response or preexisting immunity to enteropathogens.
The isolation rates of Salmonella ssp., Shigella spp., and Campylobacter spp. were either low or absent. We found a significant association of LT-producing ETEC and Providencia ssp. with FAD. A nonsignificant trend toward higher frequency of Salmonella infection among PAS cases compared with healthy subjects was observed. On the basis of this study and a previous one by our group carried out in Mexico, where the predominant cause of TD is ETEC, we believe that ETEC should be considered as a cause of PI-IBS.
Although the Rome II criteria are acceptable for epidemiologic studies of IBS, this study has several limitations. First, a rigorous assessment for gastrointestinal disease and enteric pathogens was not done at baseline. The assessment for PI-IBS was done by a mailed questionnaire and subjects were not evaluated in person by a study physician or for the presence of enteropathogens at 6-month follow-up. Other limitations are that the questionnaire did not seek for symptoms that were atypical of IBS (ie, pain interfering with sleep, bloody stools, weight loss, fever, and nocturnal diarrhea) that could be “red flags” suggestive of other pathology. Also, we did not inquire for the presence of chronic upper gastrointestinal symptoms. By doing so, we could have explored the possibility of the postinfectious onset of gastroesophageal reflux disease following norovirus infection as has been suggested by others.
In conclusion, we found that acute TD was a risk factor for the development of FAD and IBS. Further studies should be targeted at unraveling the interactions between host, environment, and pathogens, as well as to identify prognostic factors in a larger population base to facilitate a better understanding of the natural history of PAS among travelers. Studies are needed to determine if it is possible to prevent PAS development in international travelers with prophylactic antibiotics or vaccines. Before these study results are available, we believe that greater emphasis should be placed on the prevention and early treatment of acute enteric infections to prevent morbidity and possibly reduce the occurrence of PAS.