Abstract

Introduction

Gastrointestinal (GI) infections are globally prevalent and cause significant morbidity and mortality. Chronic GI disorders including celiac disease1, inflammatory bowel disease (IBD)2 and irritable bowel syndrome (IBS)3 have been causally associated with infectious gastroenteritis (IGE). Since the first description of post-infectious IBS (PI-IBS) provided by Chaudhary and Truelove in 1962 among patients with proven or presumptive episode of bacterial or amoebic dysentery4, numerous epidemiological studies have confirmed the association between acute IGE with different pathogens and development of IBS and other functional GI disorders (FGIDs)3. In addition to outbreaks in community or hospitalized settings, acute IGE sustained during a travel5 or military recruitment6 has also been associated with development of PI-IBS. Studies in different settings and with different pathogens have yielded a wide range (4-36%) of individuals developing PI-IBS after an episode of IGE3. Bacterial IGE seems to be more closely related to the development of long-term PI-IBS phenotype7 as compared to viral IGE8. Long-term follow up studies suggest that bacterial pathogen related PI-IBS phenotype may persist for up to 8-10 years after the acute IGE is over9,10. Epidemiological studies have identified female gender, age <60 yr, smoking, enteritis severity, and pre-enteritis psychological stress as risk factors for development of PI-IBS11. Existing literature supports the role of enterochromaffin (EC) cell hyperplasia, altered barrier function, immune dysregulation and potentially mast cell activation in the pathophysiology of PI-IBS. However, a number of prior studies have been conducted after epidemics often involving >1 pathogen or in highly selected tertiary referral patients. Thus, there is limited insight into unique pathogen-specific mechanisms from isolated, community-based gastroenteritis cases, and into physiological and clinical sequelae based on the pathogen and host characteristics.

Epidemiology

Gastrointestinal infections with isolated (bacterial, viral), mixed (> 1 bacteria, bacteria and a virus), and unspecified (travelers’ diarrhoea) pathogens have all been associated with the development of PI-IBS. Most of the larger cohorts are from outbreaks involving bacterial pathogens12. However, in several studies involving outbreaks, the information on the pathogen involved is often not available in all subjects and is only assumed. Although, these studies have shown a wide range (4-36%) of individuals developing PI-IBS, a meta-analysis revealed median prevalence of 10 per cent (compared to 1.2% in controls)13. This also showed a relative risk ranging between 2-12 and a pooled odds ratio of 7.3 for development of PI-IBS. The Table provides a comprehensive list of PI-IBS epidemiological studies according to the pathogen involved. An outbreak in Walkerton, Canada in 2000 caused by contamination of municipal water supply from livestock faecal material resulted in acute gastrointestinal illness in over half of the town's 4800 residents has provided significant epidemiological information about PI-IBS12. Although, pathogen information was not available in all cases, mixed bacterial infection with Campylobacter jejuni and Escherichia coli O157:H7 was found be responsible for most of the cases. Two years post-infection, a PI-IBS rate of 36 per cent among patients that had IGE during the outbreak was seen12.

Table

Percentage of individuals developing post-infections - irritable bowel syndrome (PI-IBS) with different enteritides

Open in a separate window

Risk factors

Clinical features

In the absence of a biomarker or a universally accepted definition for PI-IBS, the Rome criteria43 for defining IBS are usually used to define PI-IBS. This requires presence of recurrent abdominal pain or discomfort at least three days/month in the last three months associated with two or more of the following: (i) improvement with defaecation, (ii) onset associated with a change in frequency of stool, and (iii) onset associated with a change in form (appearance) of stool. These historical characteristics must be fulfilled for the last three months with symptom onset at least six months prior to diagnosis. To characterize as PI-IBS, the individual must not meet criterion for IBS prior to the infection and the acute IGE must be characterized either by presence of positive stool cultures or presence of ≥2 of the following: fever, vomiting, or diarrhoea. The stool cultures are often not available since a number of individuals develop acute IGE as travelers’ diarrhoea. The PI-IBS phenotype is most commonly IBS-D44 and IBS subtype phenotypic clusters have been found to remain stable over time45. Previous treatment for anxiety or depression was reported less frequently in PI-IBS than non-PI-IBS (26 vs 54%).

Natural history

The Walkerton outbreak cohort investigators reported the 8 year follow up data and the prevalence of IBS dropped to 15.4 per cent after 8 years, still remaining significantly increased compared with controls without history of acute IGE (OR 3.12; 95% CI 1.99 to 5.04)9. This suggests that PI-IBS phenotype persists for a prolonged duration. Another study showed that symptoms persist for upto 10 years following the initial infection10. A recent analysis of the defense database showed an increased risk of dyspepsia, constipation, and gastroesophageal reflux disease among military recruits who had suffered from IGE from norovirus46. In another food-borne outbreak study of norovirus, 23.6 per cent reported symptoms consistent with PI-IBS at three months vs 3.4 per cent who remained well (odds ratio: 6.9). However, at 6, 12, and 24 months, the prevalence of IBS was similar among exposed versus non-exposed individuals8. Vomiting during the IGE episode independently predicted risk of PI-IBS at three months8. From a recent viral outbreak in Italy, 13 per cent of IGE patients were found to have PI-IBS at 12 months, and most of these were mixed-IBS (IBS-M). This study also showed that the presence and severity of symptoms progressively diminished at 12 month follow up as compared to the 6 month survey22. Both these studies suggest that post-viral IBS may be a more transient phenomenon as compared to bacterial PI-IBS. This might be related to a lesser degree of initial epithelial damage and host inflammatory response caused by viral as compared to bacterial pathogens.

Mechanisms

Treatment

The first step in the management should be to illicit the history of IGE preceding the onset of symptoms. This finding can often be a subtle on history and patients may just endorse acute onset of symptoms during or after a travel without recall of classical symptoms of acute IGE. A stool culture is unlikely to be available in most cases. Although PI-IBS is characterized by acute onset of symptoms after a GI infection, many patients provide a history of respiratory illness preceding the onset of IBS. Although, this cannot be typically characterized as PI-IBS, it can be considered to be in the same spectrum of FGIDs. Once PI-IBS is clinically suspected, it is often helpful to educate the patient about the role of gut infections in development of IBS symptoms and a suspected causal link can be helpful in “legitimizing” the disorder. It is also helpful to reassure since the PI-IBS is likely to improve over time than deteriorate, especially if a viral pathogen is suspected to be involved.

Subsequently, the overall clinical management is driven by nature and severity of the symptoms and is not different from management of IBS, details of which can be found elsewhere58. A few clinical trials specific to PI-IBS have been performed. In a randomized, double-blind, placebo controlled trial of prednisolone at a dose of 30 mg/day for three weeks, lamina propria T-lymphocytes decreased but no differences were seen in EC cell counts or clinical endpoints of abdominal pain or diarrhoea59. An open-labelled trial of 18 patients using mesalamine 800 mg three times a day for 30 days showed improvement in scores for abdominal pain, distension, stool frequency and consistency60. This study also showed similar improvements in 43 non-PI-IBS patients treated with same dose and duration of mesalamine. Another double-blind, placebo controlled study using mesalamine (Asacol®) 1.6 g twice a day in 20 patients with PI-IBS showed improvement in global symptoms, abdominal pain, bloating, stool urgency, frequency, or consistency and quality of life61.

Conclusions

Gastrointestinal infections with pathogenic microbes constitute an important risk factor for development of IBS and other FGIDs. In spite of a significant knowledge of epidemiological risk factors, little is known about pathophysiological mechanisms of PI-IBS. Existing literature supports the role of immune dysregulation, altered barrier function, EC cell activation and host genetics. Much needs to be done to identify novel mechanisms such as the role of microbial pathogenicity factors, alterations in host commensal microbiota, and post-infectious plasticity in the neuromuscular apparatus of the gut. From the stand-point of the host, alterations in gut physiology using deep phenotyping measures in studies of GI transit, sensation and permeability and central nervous system changes using functional brain imaging will be necessary to find subsets of PI-IBS patients with unique physiological alterations. Overall, this subset of IBS provides an opportunity to prospectively recruit patients and study mechanism for PI-IBS and IBS in general. Additionally, studies involving isolated, identifiable pathogens will likely provide more relevant mechanistic information.

References