Article Text

Download PDFPDF

Mind over matter in the postinfective irritable bowel
  1. UNC Functional GI Disorders Centre
  2. Division of Digestive Diseases,
  3. 726 Burnett-Womack Bldg. CB 7080,
  4. University of North Carolina,
  5. Chapel Hill, N.C. 27599–7080, USA

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

See article on page 400

In recent years, our understanding of the irritable bowel syndrome (IBS) has evolved from being considered solely a disorder of motility, to one that is characterised by dysregulation of brain–gut function, manifest by gut motor hyper-reactivity and enhanced visceral sensitivity to various stresses.1 In this regard, several psychosocial factors, namely life stress, abuse, psychological co-morbidity, and illness behaviour, have been implicated in the clinical expression of IBS.2 However, although most published studies have shown an associationof psychosocial factors with IBS, only one study until now, has reported that psychosocial distress in patients with acute infectious diarrhoea predicted the later development of IBS symptoms.3 In a follow up study in this issue, Gweeet al (see page 400) elaborate upon this observation by describing the psychosocial, physiological and histopathological features of individuals who develop postinfectious IBS symptoms.

In a sample of 94 patients with acute gastroenteritis studied in an infectious disease unit, 22 (23%) were found three months later to fulfil Rome criteria for IBS.4 When comparing this group of symptomatic patients (IBS+) to the remaining sample (77%) who resolved their gut symptoms (IBS−), three major findings were noted. (1) Psychosocial difficulties were again shown to influence who with acute gastroenteritis would remain symptomatic. Specifically, in a regression analysis, pre-existent life stress and hypochondriasis were the strongest predictors of the IBS+ group, and the development of IBS-like symptoms was not explained by illness behaviours. (2) Acute gastroenteritis was associated with increased numbers (relative to biopsy samples taken from a healthy control group) of rectal inflammatory cells. However, three months later, only the IBS+ group continued to have increased mucosal inflammation. (3) When compared with healthy controls, those with the acute gastroenteritis went on at three months to develop physiological evidence of gut dysfunction—reduced whole gut transit time, decreased sensation threshold to rectal distension, decreased rectal compliance and reduced number of rectal contractions. However, there were no significant differences physiologically between the IBS+ and IBS− groups.

One conclusion noted by the authors harmonises with a biopsychosocial understanding of IBS5: “. . .physiological changes predispose to IBS, but psychological factors are required to bring out the perception of these physiological changes as symptoms.” So, the presence of dysmotility or visceral hypersensitivity is not sufficient to explain fully the symptoms of IBS; the individual’s psychosocial state must also be considered. Another conclusion, that psychosocial distress may perpetuate gut inflammation, offers relatively new information that needs elaboration with regard to IBS. Presumably this may occur via psychoimmunological mechanisms that converge on neuropeptide function in the gut lamina propria.6 In a study on rats, previous inflammation upregulated or rekindled the host’s inflammatory response to later stress.6Wistar-Kyoto and Sprague-Dawley rats that recovered from experimental colitis induced by trinitrobenzene sulphonic acid when subsequently stressed by physical restraint, showed a fourfold increase in tissue myeloperoxidase activity, a measure of tissue inflammatory response. Once inflammation exists in the gut, the activation of local cytokines and possibly other mechanisms may induce hypermotility and enhanced visceral sensitivity even after the inflammation subsides.7

Figure 1 conceptualises possible associations relating inflammation, psychosocial distress, physiological dysfunction, and symptoms in individuals with postinfectious gastrointestinal symptoms. Firstly, an acute enteric infection will lead to mucosal inflammation and dysmotility (either directly or via inflammatory effects). Secondly, mucosal inflammation or dysmotility, or both, may eventually lead to visceral hypersensitivity—that is, low thresholds to rectal distension.8 Thirdly, IBS-like symptoms may result from visceral hypersensitivity in the gut and dysmotility. However, in Gweeet al’s study, these gut related physiological factors were not sufficient to explain the IBS symptoms as they were also present in the postinfectious group without IBS symptoms. Thus, the study suggests that psychosocial factors may have an overarching effect on symptom development in postinfectious IBS. This may occur through putative effects on mucosal inflammation (and inflammation may have reciprocal effects on psychological state via activation of peripheral or central, or both, cytokines9), through central influences on peripheral visceral sensitisation or dysmotility (via descending corticofugal systems), or by direct effects on IBS symptoms via central nervous system modulation of incoming afferent input.

Figure 1

Possible associations relating inflammation, psychosocial distress, physiological dysfunction, and symptoms in individuals with postinfectious gastrointestinal symptoms.

There are several limitations in this study which may affect the interpretation of the results and the conclusions drawn from the data. Firstly, postinfectious gut dysfunction may not actually be IBS, or it may represent only a small subset of patients with IBS. Three months is at the lower time limit for diagnosing IBS based on Rome criteria,4 and it is uncertain what proportion of patients with symptoms three months after enteric infection would still be symptomatic six or 12 months later. Furthermore, the proportion of patients with constipation-like IBS is not reported; it is likely that patients with postinfectious bowel dysfunction would primarily have symptoms of diarrhoea. A second limitation relates to a possible acquisition bias. In addition to the 94 patients studied at intake and three months, 15 other patients initially evaluated did not return for the three month follow up tests. Furthermore, although all patients were invited for histopathological and physiological investigation, only 10 (45%) of 22 IBS+ and 19 (26%) of 72 IBS− patients had rectal biopsy, and 16 (73%) of 22 IBS+ and 16 (22%) 72 IBS− patients underwent physiological testing. So it is possible that the relatively small proportion of patients tested are not truly representative of the total population under question. Thirdly, almost half of the patients did not have a positive stool culture for enteric pathogens. Although many enteric infections are culture negative, for the purpose of a clinical research study, the possibility exists that a substantial proportion of the patients may in fact have had IBS in an early clinical presentation. A final consideration relates to whether increased rectal inflammation is typical for IBS. In one study, comparing 61 patients fulfilling Rome criteria for IBS with 59 controls,10 there was no difference between groups with regard to histopathology on rectal biopsy, and none of the patients had evidence of microscopic colitis. Although quantitative cell counts were not done, there is little qualitative evidence for increased levels of mucosal inflammation in IBS. We also need to question whether the inflammation resulted from patient related behaviours. For example, would a psychologically distressed individual having symptoms be more likely to take aspirin or non-steroidal medication, thereby increasing the likelihood of having mucosal inflammation?

These limitations are likely to be tackled in future studies by the authors and other investigators. What is important is that the findings are new and provocative. They lead us to consider additional questions for future study. (1) Are there pre-morbid psychosocial difficulties that predict who will develop an enteric infection, or does an infection unmask patients having high susceptibility to respond with psychological distress? This question could only be answered by studying patients psychosocially before the inoculation of the enteric infection. (2) What is the role for gut motor dysfunction in the symptoms of IBS? The evidence in this study suggests that dysmotility does not relate to symptoms directly, but rather develops in response to the infection. Future studies may determine whether gut motor dysfunction has indirect effects on symptoms, possibly by influencing the development of visceral hypersensitivity. Finally, (3) in the psychologically susceptible individual, would early psychopharmacological (or psychological) treatment prevent the full blown development of IBS (e.g. by reducing mucosal inflammation or visceral sensitisation)?


Linked Articles