SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Background
  4. Aim
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Supporting Information

Background

Irritable bowel syndrome (IBS), one of the most common gastrointestinal disorders, markedly impairing patients' quality of life. Drug development for IBS treatment has been hampered by the lack of understanding of IBS aetiology. In recent years, numerous data have emerged that suggest the involvement of immune activation in IBS, at least in a subset of patients.

Aim

To determine whether immune activation and intestinal permeabilisation are more frequently observed in IBS patients compared with healthy controls.

Methods

The scientific bibliography was searched using the following keywords: irritable bowel syndrome, inflammation, immune activation, permeabilisation, intestine, assay, histology and human. The retrieved studies, including blood, faecal and histological studies, were analysed to provide a comprehensive and structured overview of the available data including the type of assay, type of inflammatory marker investigated or intestinal segment studied.

Results

Immune activation was more frequently observed in IBS patients than in healthy controls. An increase in the number of mast cells and lymphocytes, an alteration in cytokine levels and intestinal permeabilisation were reported in IBS patients. No consistent changes in the numbers of B cells or enterochromaffin cells or in mucosal serotonin production were demonstrated.

Conclusions

The changes observed were modest and often heterogeneous among the studied population. Only appropriate interventions improving irritable bowel syndrome symptoms could highlight and confirm the role of immune activation in this pathophysiology.


Background

  1. Top of page
  2. Summary
  3. Background
  4. Aim
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Supporting Information

Irritable bowel syndrome (IBS) is a common functional gastrointestinal disorder of unknown aetiology.[1] Typical symptoms include abdominal pain and bowel habit disturbance, which lead to a tremendous impairment in patients' quality of life.[2] Deciphering the pathophysiological mechanisms underlying IBS is a major public healthcare priority because IBS affects 5–20% of the population, depending on the diagnostic criteria.[3, 4]

Thus far, numerous factors have been implicated in the development of IBS. The aetiology of IBS is thought to be genetic[5-10]; however, the genetic penetrance is incomplete,[11] suggesting that the genetic component of IBS could involve other unknown factors. Indeed, environmental factors such as diet (for a review, see[12, 13]) and stress (for a review, see[14-17]) can impact gut physiology. However, studies showing a direct link between these factors and the onset of IBS are lacking. More recent data have provided evidence that an imbalance in gut bacterial populations could contribute to the development of IBS. Dysbiosis has been notably reported through microflora taxonomy studies in the faeces of IBS patients.[18-20] The mechanisms by which dysbiosis occurs and contributes to IBS pathophysiology remain unclear. One possible explanation is that exogenous pathogenic bacterial and/or viral agents have deleterious interactions with endogenous gut bacteria. Six to 17% of patients report that their IBS symptoms began following a digestive infection. This finding is consistent with prospective studies showing a high incidence of postinfectious IBS (PI-IBS) after gastroenteritis.[21] Moreover, IBS following bacterial gastroenteritis is well documented and indicates that an acute infection can promote persistent IBS symptoms. Indeed, several studies have reported that bacterial PI-IBS can be caused by bacteria of the Campylobacter,[22] Salmonella[23, 24] and Shigella[21, 25] genera. The intriguingly high incidence of previous medical infections in IBS patients has sparked a renewed interest as to whether inflammation could play a role in the development of IBS. Indeed, for many years, the pathology of IBS has been considered sine materia because neither structural nor biochemical abnormalities could be detected in the colons of patients. In the early 1960s, Hiatt and Katz described an increased number of mast cells in the colon muscularis propria of patients displaying ‘spastic colitis’.[26] However, this striking finding did not lead to a widespread assessment of the histopathological status of IBS patients, and no other inflammatory features were further described. This effect is likely due to the lack of sensitivity of routine histology techniques and their inability to detect inflammatory soft signs. However, within the last decade, the development of endoscopic techniques and the increased availability of colonic biopsies have provided increasing evidence of the infiltration of inflammatory cells into the colon mucosa of IBS patients (for a review, see[27]). Moreover, the most recent insights into IBS pathophysiology suggest that low-grade inflammation could underlie IBS symptomatology.

Aim

  1. Top of page
  2. Summary
  3. Background
  4. Aim
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Supporting Information

The aim of this review is to clarify whether immune activation and intestinal permeabilisation are associated with IBS. For this purpose, we systematically assessed the presence of inflammatory features in studies focused on IBS. We present data from studies that quantify inflammatory markers (assays of stools or peripheral blood samples) and from studies that evaluated immune cell infiltration as well as the presence of inflammatory mediators in the gut mucosa of IBS patients. Recent studies assessing gut permeability in IBS were also included, given that permeabilisation of the intestinal mucosa and immune activation are two processes that can impact each other.[28-30] The data are summarised in six tables included in the manuscript or provided as supporting information online.

Methods

  1. Top of page
  2. Summary
  3. Background
  4. Aim
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Supporting Information

Search strategy

A search of the clinical bibliography was conducted using Medline, PubMed and Scopus to identify studies assessing immune activation and gut permeability in patients suffering from IBS. The search was restricted to clinical studies published in English through October 2011 using the filters ‘English’ language and ‘Human’ species. The following key words were used: ‘irritable bowel syndrome <AND> inflammation <AND> gut’; ‘irritable bowel syndrome <AND> histology <AND> gut’; and ‘irritable bowel syndrome <AND> permeability’.

Study selection

We determined our inclusion criteria based on the PICO formula as follows:

P/Participants. Cohort studies and case series (with 10 or more patients suffering either from irritable bowel syndrome or functional abdominal pain included in the study) using symptom-based criteria for IBS diagnosis were considered. Studies with both adults and children diagnosed with IBS were included regardless of gender.

I/Interventions. Studies involving the collection of faeces or blood samples or involving the removal of intestinal biopsies were included.

C/Comparisons. Only studies comparing IBS patients to healthy controls were included.

O/Outcome. The outcomes of studies were determined by detection of immune activation or evaluation of intestinal permeabilisation in IBS patients.

Exclusion criteria

Individual case reports (<10 IBS patients), review articles, abstracts, studies published in languages other than English and those presenting data from IBS patients displaying co-morbid symptoms affecting other visceral organs (e.g. inflammatory bowel disease, intestinal tuberculosis, colon cancer, gastritis, cystitis, etc.) were not considered in this review.

Data extraction

The articles retrieved were first evaluated based on their title and abstract. Studies were excluded when the full article was not available or when the title and abstract did not contain sufficient pertinent information. Additional relevant publications cited in systematic reviews were also considered when they matched similar selection criteria. Full papers selected through this screening were then evaluated in greater detail. The outcome measure was the detection of immune activation (mean levels of inflammatory markers) or the evaluation of gut permeabilisation (mean levels of permeability markers), which were compared between IBS patients and healthy controls.

Question addressed

Are the levels of immune activation and gut permeability markers in faecal samples, blood samples or intestinal biopsies of IBS patients higher than in healthy controls?

Results

  1. Top of page
  2. Summary
  3. Background
  4. Aim
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Supporting Information

Using Medline and PubMed, we identified a total of 352 entries. Of the 136 studies that appeared to be relevant to the systematic review, 74 fulfilled the eligibility criteria as depicted in the flow diagram describing the study selection process (Figure 1). Evidence of immune activation collected from blood and stool samples might be more reflective of IBS-related neural alterations than more superficial biopsies, for which correlations with afferent or efferent neural dysfunction have not been clearly demonstrated yet. Therefore, a breakdown of the included studies was performed to classify the data into blood/stool or tissue evidence of immune activation in IBS patients.

image

Figure 1. Flow diagram illustrating the identification of the studies to be assessed in the systematic review using PubMed. The search was limited to clinical studies published in English-language journals up until October 2011 using the filters ‘English’ language and ‘Human’ species. CFS, chronic fatigue syndrome; GERD, gastro-oesophageal reflux disease; IBD, inflammatory bowel disease; IBS, irritable bowel syndrome.

Download figure to PowerPoint

Findings suggesting immune activation in the blood and stool of IBS patients

Evidence for immune activation in IBS was obtained from biochemical analyses of faecal or peripheral blood samples (Table S1). Assays of inflammatory markers have been performed on samples from IBS and healthy control patients based on the hypothesis that their presence and/or increase would indicate the presence of gut inflammation.

Levels of inflammatory mediators in the blood of IBS patients

Cytokine assays have revealed no major changes in the basal levels of cytokines in IBS patients[31-34] with the exception of IL6 and IL8, which have been found to be moderately increased in some studies.[31, 34, 35] A decrease in the levels of the anti-inflammatory cytokine IL10 has also recently been reported in children with IBS,[36] although this decrease was not observed in adults.[31, 32] However, several studies that assessed the effect of immune challenges on cytokine production by peripheral blood mononuclear cells isolated from IBS patients have reported the abnormal secretion of both pro- and anti-inflammatory cytokines.[32, 34, 35, 37, 38] For example, IBS mononuclear cells released more tumour necrosis factor-α (TNF-α), IL1β and IL6 following exposure to lipopolysaccharides produced by Escherichia coli when compared with mononuclear cells isolated from healthy control patients.[35] Alterations in the production of these pro-inflammatory cytokines (primarily produced by monocytes and macrophages) are suggestive of an increase in innate immune activity in IBS patients. Accordingly, data collected from D-IBS patients indicate an increase in protease activity in the stool[39] and jejunal fluid[40] and an increase in NO release in the rectal mucosa,[41] indicating the potential involvement of mast cells in the increased release of inflammatory mediators. This finding is consistent with data indicating an increase in blood levels of lipid inflammatory mediators such as prostaglandin E2 and leukotriene B4, which are known to be released from newly formed mast cells.[42]

The potential activation of the innate gut immune system in IBS could be a sign of previous infection and/or ongoing low-grade inflammation. In this regard, increased faecal levels of the human beta-defensin HBD2,[43] an antimicrobial peptide produced by intestinal epithelial cells, and elevated blood concentrations of the A4-Fla2 and Fla-X bacteria-specific antibodies[44] have been found in IBS patients.

Key points
  1. No clear changes in the basal levels of cytokines have been described in the blood of IBS patients.
  2. Mononuclear cells from IBS patients secrete cytokines abnormally in response to immune challenge.
  3. Biomarkers of mast cell activity are increased in both the blood and stool of D-IBS patients.
Blood lymphocytic status in IBS patients

The number of activated regulatory CD4+, CD25+[32, 45] and CD8+[32, 37] T cells is unchanged in the blood of IBS patients.[32, 45] However, some authors have suggested that T cells could be specifically targeted to the gastrointestinal tract because IBS patients have an increased number of T cells expressing the gut homing integrin-β7+.[38, 46] Similarly, the expression levels of type G immunoglobulin (IgG) and the co-stimulatory molecules CD80 and CD86 are increased specifically in B cells expressing integrin-β7+.[47]

Several studies have also assessed the reactivity of circulating T cells to immune challenge in IBS patients. However, the findings were somewhat inconsistent. One group reported that the stimulation of T cells isolated from patients with anti-CD3/CD28 antibodies resulted in both an increased production of IL1β and a decrease in T-cell proliferation.[38] Another approach involved the exposure of lymphocytes from IBS patients to E. coli lipopolysaccharide, which resulted in either an increased production of IL5 and IL13 and a decreased production of the anti-inflammatory cytokine IL10[32] or in a decreased production of TNF-α.[37] Finally, another study assessing the expression level of the soluble IL2 receptor (sIL2R) found that sIL2R levels were decreased in the blood of IBS patients,[48] suggesting a possible lack of T-cell activation.

Key points
  1. The number of circulating activated T cells is similar in IBS patients and healthy control patients.
  2. The blood of IBS patients contains an increased number of T cells expressing the gut homing integrin-β7+, and B cells exhibit increased activation levels.
Faecal levels of calprotectin and lactoferrin in IBS patients

Calprotectin and lactoferrin are inflammatory mediators released by activated mononuclear immune cells that play an important antimicrobial role in innate immunity. Several studies have shown that faecal calprotectin and lactoferrin levels are reliable indicators of the gut inflammatory state.[49-52] However, most studies found no difference in the faecal levels of calprotectin and lactoferrin between IBS patients and healthy controls.[43, 44, 49, 50, 52-59] Only one publication reported that increased levels of faecal calprotectin were correlated with pain severity; however, this result was observed in children displaying both functional abdominal pain and IBS.[60]

Key point
  1. Typically, no changes in the levels of the faecal inflammatory markers calprotectin and lactoferrin are reported in IBS patients.

Findings suggesting immune activation in the intestinal tissue of IBS patients

Several biochemical studies using stool and blood samples have indirectly suggested that immune activation occurs in IBS. However, the data are scarce and are not sufficiently robust to draw any conclusions due to a lack of replicated studies. Moreover, the vast majority of these findings was not specific to IBS and thus requires further corroboration. We now present an extensive overview of the results obtained from direct analysis of the intestinal mucosa of IBS patients by means of molecular, cellular and histological methods.

Mast cell numbers in the gut mucosa of IBS patients

Mast cells are innate immune cells involved in food allergy, wound healing and protection against pathogens. They undergo a degranulation process, which leads to the release of compounds such as histamine, tryptase and chymase. Based on their biochemical content, morphology, location and behaviour, two types of phenotypically distinct mast cell subpopulations have been described in humans: connective tissue-type mast cells and mucosal-type mast cells.[61, 62] However, none of the papers reviewed here differentiated between these two subtypes. The data presented herein illustrate the quantification of gut mucosa-associated mast cell density in IBS patients (Table 1).

Table 1. Mast cell infiltration in the gut mucosa of IBS patients
Site of biopsyFindingsMagnitude of changeIntestine layerMethodsClinical subtypeN (IBS:HC)IBS sex ratio (M:F)Proportion of abnormal values (IBS)Proportion of abnormal values (HC)Diagnostic criteriaReference
  1. ns, not specified; IBS, irritable bowel syndrome (IBS subtypes were either not determined during diagnostic analysis or not differentiated for data analysis); A-IBS, alternating IBS; C-IBS, constipation-predominant IBS; D-IBS, diarrhea-predominant IBS; + defines a 25–50% change; ++ defines a 50–100% change; +++ defines a change greater than 100%; EM, electron microscopy; F, female; H, histology; HC, healthy control; IHC, immunohistochemistry; M, male; PI-IBS, post-infectious IBS.

DuodenumIncrease+MucosaH + IHCD-IBS20:297:1325% (5/20)10% (3/29)Rome II(Foley et al.,2011 )
 No differenceMucosaIHCC-IBS + D-IBS38:2017:21nsnsRome III(Wang et al., 2007)
IleumIncrease+MucosaIHCC-IBS + D-IBS38:2017:21nsnsRome III(Wang et al., 2007)
  +MucosaIHCD-IBS18:1510:8nsnsRome II(Park et al., 2006)
  ++MucosaIHCPI-IBS56:1225:31nsnsRome II(Wang et al., 2004)
  +++MucosaH + IHCD-IBS20:15-nsnsManning(Weston et al., 1993)
JejunumIncrease+++MucosaH + IHCD-IBS20:146:14nsnsRome II(Guilarte et al., 2007)
 No differenceMucosaIHCC-IBS + D-IBS38:2017:21nsnsRome III(Wang et al., 2007)
CaecumIncrease+++MucosaIHCC-IBS + D-IBS + A-IBS34:156:28nsnsRome III(Vivinus-Nebot et al., 2011)
  ++MucosaIHCD-IBS15:50-nsnsRome III(Goral et al., 2010)
  +++Lamina propriaH + IHCC-IBS + D-IBS50:219:41nsnsRome II(Piche et al., 2008)
  +++MucosaH + IHCC-IBS + D-IBS + A-IBS11:58:3100% (11/11)0% (0/5)Rome II(Tunc et al., 2005)
  +Lamina propriaIHC + EMD-IBS14:146:8nsnsRome II(Park et al., 2003)
  ++MucosaH + IHCIBS14:72:12nsnsRome I(O'Sullivan et al., 2000)
ColonIncrease+MucosaIHCD-IBS but not C-IBS25:189:16nsnsRome II(Coeffier et al., 2010)
  ++MucosaIHCPI-IBS11:107:4nsnsRome II(Kim et al., 2010)
  +++MucosaIHCC-IBS + D-IBS11:44:7nsnsRome II(Buhner et al., 2009)
  +++MucosaIHCC-IBS + D-IBS48:2413:35nsnsRome II(Cremon et al., 2009)
  +MucosaIHCD-IBS18:1510:8nsnsRome II(Park et al., 2006)
  +++MucosaIHC + EMIBS44:2213:3177% (34/44)nsRome II(Barbara et al., 2004)
  ++MucosaH + IHCC-IBS + D-IBS + A-IBS77:2817:6051% (39/77)nsRome I(Chadwick et al., 2002)
  Muscularis propriaHspastic colitisns:4ns(Hiatt and Katz, 1962)
 No differenceMucosaIHCD-IBS11:107:4nsnsRome II(Kim et al., 2010)
  MucosaH + IHCIBS14:72:12nsnsRome I(O'Sullivan et al., 2000)
Sigmoid colonNo differenceMucosaIHCD-IBS11:107:4nsnsRome II(Kim et al., 2010)
Rectosigmoid colonIncrease+MucosaIHCIBS23:223:20nsnsRome II(Akbar et al., 2008)
  ++MucosaH + IHCC-IBS + D-IBS + A-IBS77:2817:6051% (39/77)nsRome I(Chadwick et al., 2002)
 No differenceMucosaIHCPI-IBS + D-IBS11:107:4nsnsRome II(Kim et al., 2010)
  MucosaIHCPI-IBS56:1225:31nsnsRome II(Wang et al., 2004)
RectumIncrease++MucosaIHCD-IBS57:50-nsnsRome III(Goral et al., 2010)
  +Lamina propriaIHCD-IBS but not C-IBS and A-IBS37:1216:2138% (14/37)nsRome III(Lee et al., 2008)
  +Lamina propriaIHCPI-IBS5:121:4100% (5/5)nsRome III(Lee et al., 2008)
  +MucosaIHCD-IBS18:1510:8nsnsRome II(Park et al., 2006)
  +MucosaIHCIBS but not PI-IBS28:3411:17nsnsRome I(Dunlop et al., 2003)
  +Lamina propriaEMD-IBS14:146:8nsnsRome II(Park et al., 2003)
  ++MucosaH + IHCC-IBS + D-IBS + A-IBS77:2817:6051% (39/77)nsRome I(Chadwick et al., 2002)
 No differenceMucosaIHCPI-IBS + D-IBS11:107:4nsnsRome II(Kim et al., 2010)
  MucosaIHCIBS9:101:80% (0/9)0% (0/10)Rome II(Cenac et al., 2007)
  MucosaH + IHCIBS14:72:12nsnsRome I(O'Sullivan et al., 2000)
 Decrease+Lamina propriaH + IHCIBS29:15nsnsnsRome II(Klooker et al., 2010)

In patients with all clinical subtypes of IBS (D-IBS, C-IBS, A-IBS and PI-IBS), an increase in mast cell density (Figure 2a) is frequently observed in the mucosa and throughout the intestinal tract, including the duodenum,[63] ileum,[25, 64-66] jejunum,[40] caecum,[48, 65, 67-71] colon[58, 64, 72-77] and rectum.[48, 64, 68, 78-80] These results are particularly robust for the caecum (100% of studies, 6/6) and the colon (80% of studies, 8/10), where mast cell densities can be anywhere from 50% to more than 100% higher than in healthy control patients. When available, individual data from caecum and colon samples indicate that the proportion of abnormal values (values that would be considered as outliers in healthy control datasets) in IBS patients is high, which suggests that overlap with normal tissue is limited. In contrast, all studies focused on the terminal ileum (4/4) found that IBS patients had an increased mast cell density; however, these increases were more modest. Finally, only half of the studies performed in the rectum reported a higher mast cell density in the mucosa of IBS patients. Although an increase in mast cell density is usually consistently found in the colon and rectum of patients with D-IBS and PI-IBS subtypes, a few groups could not replicate this finding.[25, 58, 67, 81] While discrepancies exist, the involvement of mast cells in IBS is further supported by an increased secretion of tryptase[71, 74, 81-84] and histamine[74, 82, 85] in the colon and rectum of IBS patients (Table S3). Only one group reported a decrease in rectal levels of mast cell tryptase in a set of IBS patients,[85] which is in accordance with their data showing a reduced mast cell density in the rectal lamina propria of these patients. These controversial data may be attributable to differences in immunostaining techniques, sample size, IBS-related recruitment criteria and/or the definition of IBS.

image

Figure 2. Immune activation and increased intestinal permeability in IBS patients. (a) Increased area of the colonic lamina propria occupied by mast cells in IBS patients. (b) Increased numbers of intra-epithelial lymphocytes in the rectum of IBS patients (left: per 100 epithelial cells). Increased area of the colonic lamina propria occupied by lymphocytes in IBS patients (middle: CD4+; right: CD8+). (c) Increased paracellular permeability to FITC-sulphonic acid in colonic biopsies of IBS patients. Data were extracted from Barbara et al., [73] (mast cells); Kim et al., [58] (IELs); Cremon et al., [77] (CD4+ and CD8+ lymphocytes); and Piche et al., [110] (permeability).

Download figure to PowerPoint

Key points
  1. Increased mast cell numbers are consistently reported in the ileum, caecum and colon mucosa of IBS patients.
  2. Increased levels of mast cell tryptase have been observed in the caecum and colon mucosa of IBS patients.
  3. Discrepancies exist regarding increases in mast cell numbers in the rectal mucosa of IBS patients.

Lymphocyte density in the gut mucosa of IBS patients

Mucosal T- and B-type lymphocytes are key components of the gut adaptive immune response to pathogens. In addition to conventional T-type lymphocytes, intestinal intraepithelial T-type lymphocytes (IELs) are of particular relevance because they naturally reside in the gut mucosa where they constitute the first host immune barrier against pathogens (GALT: Gut-Associated Lymphoid Tissue). IELs also stimulate mucosal healing following acute injury. Therefore, numerous studies have assessed lymphocyte populations (T, IEL and B) in IBS patients (Table 2).

Table 2. Lymphocyte infiltration in the gut mucosa of IBS patients
Site of biopsyFindingsMagnitude of changeTypeIntestine layerMethodsClinical subtypeN (IBS:HC)IBS sex ratio (M:F)Proportion of abnormal values (IBS)Proportion of abnormal values (HC)Diagnostic criteriaReference
  1. ns, not specified; A-IBS, alternating IBS; CD, cluster of differentiation; C-IBS, constipation-predominant IBS; D-IBS, diarrhoea-predominant IBS; ELISA, enzyme-linked immunosorbent assay; F, female; H, histology; HC, healthy control; IBS, irritable bowel syndrome; Ig, immunoglobulin; IHC, immunohistochemistry; M, male; PI-IBS, postinfectious IBS.

DuodenumIncrease+++Intra-epithelialMucosaH + IHCD-IBS20:297:1375% (15/20)7% (2/29)Rome II(Foley et al., 2011)
IleumNo differenceLymphocyte + B cellMucosaH + IHCD-IBS25:13nsnsnsRome II(Park et al., 2006a)
  Lymphocyte + B cellMucosaIHCD-IBS18:1510:8nsnsRome II(Park et al., 2006b)
JejunumIncrease+Intra-epithelialMucosaH + IHCD-IBS20:146:1410% (2/20)0% (0/14)Rome II(Guilarte et al., 2007)
  +T cellMyenteric plexiH + IHCC-IBS + D-IBS + A-IBS10:15 (autopsy)2:890% (9/10)nsRome I(Tornblom et al., 2002)
  +Intra-epithelialMucosaH + IHCC-IBS + D-IBS + A-IBS10:15 (autopsy)2:840% (4/10)nsRome I(Tornblom et al., 2002)
 No differenceLymphocyte + B cellMucosaH + IHCD-IBS25:13nsnsnsRome II(Park et al., 2006a)
CaecumIncrease++LymphocyteLamina propriaH + IHCC-IBS + D-IBS50:219:41nsnsRome II(Piche et al., 2008)
 No differenceLymphocyte + B cellLamina propriaH + IHCC-IBS + D-IBS + A-IBS14:72:12nsnsRome I(O'Sullivan et al., 2000)
  Intra-epithelial + B cellLamina propriaH + IHCC-IBS + D-IBS50:219:41nsnsRome II(Piche et al., 2008)
ColonIncrease++T cellMucosaFlow cytometry + ELISAC-IBS + D-IBS + A-IBS11:102:9nsnsRome II(Ohman et al., 2009)
  ++T cellLamina propriaIHCC-IBS + D-IBS48:2413:35nsnsRome II(Cremon et al., 2009)
  ++Intra-epithelialMucosaH + IHCIBS77:2817:6049% (38/77)nsRome I(Chadwick et al., 2002)
  +++T cellMucosaH + IHCIBS77:2817:6049% (38/77)nsRome I(Chadwick et al., 2002)
  ++T cellLamina propriaIHCPI-IBS + D-IBS11:107:4nsnsRome II(Kim et al., 2010)
  ++CD8+ T cellMucosaFlow cytometry + IHC + ELISAIBS33:1514:19nsnsRome II(Ohman et al., 2005)
  ++CD8+ T cellLamina propriaIHCPI-IBS but not D-IBS11:107:4nsnsRome II(Kim et al., 2010)
  +++CD25+ T cellLamina propriaH + IHCIBS77:2817:6049% (38/77)nsRome I(Chadwick et al., 2002)
 No difference-Intra-epithelialMucosaIHCC-IBS + D-IBS + A-IBS34:156:28nsnsRome III(Vivinus-Nebot et al., 2011)
  Intra-epithelialMucosaIHCPI-IBS + D-IBS11:107:4nsnsRome II(Kim et al., 2010)
  Lymphocyte + B cellMucosaIHCD-IBS18:1510:8nsnsRome II(Park et al., 2006b)
  Lymphocyte + B cellLamina propriaH + IHCC-IBS + D-IBS + A-IBS14:72:12nsnsRome I(O'Sullivan et al., 2000)
  CD8+ T cellLamina propriaH + IHCIBS77:2817:60nsnsRome I(Chadwick et al., 2002)
  CD25+ T cellLamina propriaFlow cytometryC-IBS + D-IBS + A-IBS34:2610:24nsnsRome II(Holmen et al., 2007)
  CD79+ B cellLamina propriaIHCC-IBS + D-IBS48:2413:35nsnsRome II(Cremon et al., 2009)
  IgG and IgM(+) B cellsMucosaIHCC-IBS + D-IBS + A-IBS12:105:6nsnsRome II(Forshammar et al., 2008)r
 Decrease+IgA(+) B cellsMucosaIHCC-IBS + D-IBS + A-IBS12:115:7nsnsRome II(Forshammar et al., 2008)
Sigmoid colonIncrease++Intra-epithelialMucosaIHCPI-IBS but not D-IBS11:107:4nsnsRome II(Kim et al., 2010)
  ++T cellLamina propriaIHCPI-IBS + D-IBS11:107:4nsnsRome II(Kim et al., 2010)
  ++CD8+ T cellLamina propriaIHCPI-IBS but not D-IBS11:107:4nsnsRome II(Kim et al., 2010)
Rectosigmoid colonIncrease+CD8+ T cellLamina propriaIHCIBS23:223:20nsnsRome II(Akbar et al., 2008)
 No difference+CD4+ T cellLamina propriaIHCIBS23:223:20nsnsRome II(Akbar et al., 2008)
RectumIncrease++Intra-epithelialMucosaIHCPI-IBS +D-IBS11:107:4nsnsRome II(Kim et al., 2010)
  +T cellLamina propriaIHCPI-IBS5:121:4100% (5/5)nsRome III(Lee et al., 2008)
  ++LymphocytesMucosaH + IHCD-IBS25:13nsnsnsRome II(Park et al., 2006a)
  ++Intra-epithelialMucosaH + IHCIBS77:2817:6049% (38/77)nsRome I(Chadwick et al., 2002)
  ++T cellMucosaHacute PI-IBS but not non PI- and post PI-IBS29:188:1430% (3/10)0% (0/18)Rome I(Gwee et al., 1999)
  ++T cellLamina propriaIHCPI-IBS + D-IBS11:107:4nsnsRome II(Kim et al., 2010)
  ++T cellLamina propriaH + IHCIBS77:2817:6049% (38/77)nsRome I(Chadwick et al., 2002)
  +++Intra-epithelial + lymphocyteMucosaH + IHCPI-IBS21:128:13nsnsRome I(Spiller et al., 2000)
  +++CD4+ T cellMucosaH + IHCPI-IBS21:128:13nsnsRome I(Spiller et al., 2000)
  +++CD8+ T cellMucosaH + IHCPI-IBS21:128:13nsnsRome I(Spiller et al., 2000)
  +++CD8+ T cellMucosaH + IHCpostdysenteric IBS10:124:6nsnsRome I(Spiller et al., 2000)
 No difference-T cellLamina propriaIHCD-IBS + C-IBS + A-IBS37:1216:2114% (5/37)nsRome III(Lee et al., 2008)
  -Lymphocyte + B cellMucosaIHCD-IBS18:1510:8nsnsRome II(Park et al., 2006b)
  -Intra-epithelialMucosaIHCPI-IBS + IBS28:3411:17nsnsRome I(Dunlop et al., 2003)
  -Lymphocyte + B cellLamina propriaH + IHCC-IBS + D-IBS + A-IBS14:72:12nsnsRome I(O'Sullivan et al., 2000)
  -CD8+ T cellLamina propriaIHCPI-IBS +D-IBS11:107:4nsnsRome II(Kim et al., 2010)
  -CD8+ T cellLamina propriaH + IHCIBS77:2817:60nsnsRome I(Chadwick et al., 2002)
  Intra-epithelial + lymphocyteMucosaIHCIBS (children)12:123:9nsnsRome III(Faure et al., 2010)

T cells. Several groups have reported an increased density of T lymphocytes in the mucosa of IBS patients compared with healthy control patients (Figure 2B). T-cell density was higher in the duodenum,[63] jejunum,[86] caecum,[69] colon,[38, 58, 77] rectosigmoid colon[58] and rectum.[22, 58, 64, 79, 80] However, a similar number of studies has found that IBS patients have a normal lymphocyte density in different intestinal tissue segments.[58, 64, 67, 72, 75, 80, 87, 88] Such discrepancies might result from the use of different lymphocyte markers and different quantification methods. It is interesting to note that Park and colleagues ([89]) did not find any change in lymphocyte density in the ileum, jejunum, colon or rectum,[64, 89] but they used a semi-quantitative assessment of cell number (a score of 1–4), which likely explains their contrasting results. In addition, these authors used very restrictive inclusion criteria and only considered D-IBS patients without a history of PI-IBS, atopy, food allergy or asthma, which might also reduce the possibility that lymphocyte density would be elevated in their sample set. Indeed, it appears that changes in lymphocyte density are more frequently observed in the PI-IBS subtype.[22, 58, 90] Thus, the possibility that other pathologies classically associated with abnormal immune system reactivity might confound certain experimental designs cannot be ruled out.

IELs. Studies assessing IEL density in IBS patients also report heterogeneous findings. An increase in IEL density has been observed in the duodenum,[63] jejunum,[40, 86] colon[75] and rectum[22, 58, 75]; however, these results were not replicated by other groups.[58, 69, 71, 79, 87]

B cells. Studies exploring the involvement of B cells in the intestinal tract of IBS patients have failed to report any alterations in this lymphocyte population. No changes in B-cell density were found in the ileum,[64, 88] jejunum,[88] caecum,[67, 69] colon[64, 67, 77, 91] or rectum.[64, 67] Only one study described a decrease in secretory B cells [Ig(A)+] in the colon.[91]

Key points

  1. Discrepant studies show either no change or an increase in T-type and intra-epithelial lymphocyte density in the gut mucosa of IBS patients.
  2. No changes in B-cell density have been observed in the gut mucosa of IBS patients.

Enterochromaffin cell numbers and serotonin levels in IBS patients

Enterochromaffin (EC) cells are enteroendocrine cells found in the mucosa of the gastrointestinal tract. EC cells account for 90% of serotonin (5-hydroxytryptamine or 5HT) production in the body.[92] 5HT is a key mediator of both secretory and peristaltic gastrointestinal reflexes and can also activate intestinal afferent nerves. Several modulators of 5HT receptors have some beneficial effects on the intestinal symptoms of IBS.[3, 93] These 5HT-targeted drugs act by modulating intestinal motility[94, 95] and blunting visceral sensitivity.[96] Based on these findings, several investigators have examined EC cell density and 5HT levels in IBS patients (Tables S2 and S3). No changes in EC cell counts were observed in the duodenum,[63, 65, 97] ileum[65, 88] or jejunum.[65, 88] Unfortunately, studies focused on the lower gastrointestinal tract have provided more ambiguous data (Table S2). Although increases in EC cell number were found in the colon[58, 83, 84] and the rectum,[22, 58, 78, 80] other authors reported no changes in similar gut regions.[58, 80, 87, 88, 98, 99] Similarly, the results of studies exploring potential serotonin level alterations in IBS patients reflect the difficulty in interpreting these EC cell findings (Table S2). Alterations in serotonin expression are often reported throughout the gastrointestinal tract of IBS patients. However, these changes are rather inconsistent. The data indicate either an increase[25, 84, 87, 100] or a decrease[25, 65, 100] in serotonin levels in the small intestine and colon of IBS patients, and the expression of 5HT was found to be unaltered in the ileum[65] and the colon.[74, 83] Moreover, no region- or symptom-specific patterns could be correlated with these changes, nor have these changes been homogeneously observed across different studies. Some authors have addressed the possibility that alterations in the intestinal serotonergic system of IBS patients could result from a disruption in the bioavailability of 5HT and/or mucosal 5HT release. Indeed, an increased mucosal 5HT level could be the consequence of a decrease in 5HT turnover as suggested by decreased levels of 5HT mRNA and/or the 5HT metabolite 5-hydroxyindoleacetic acid.[63, 98, 99] Accordingly, several reports have indicated a decrease in the expression of the 5HT transporter (SERT).[63, 87, 98]

Key points

  1. No changes in EC cell counts have been observed in the small intestine of IBS patients.
  2. Discrepant studies show either no change or an increase in EC cell density in the colon and rectum of IBS patients.
  3. The aetiology of the alterations in 5HT levels in the gut mucosa of IBS patients remains equivocal.

Levels of inflammatory mediators in the gut mucosa of IBS patients

Five of the reviewed studies evaluated the expression levels of pro-inflammatory (IL-1, IL-2, IL-6, IL-8, IL-12, TNF-α or IFN-γ) and anti-inflammatory (IL-10 or TGF-β) cytokines in the gut mucosa of IBS patients (Table S3). The heterogeneity of reported findings might be due to the diverse array of quantitative methods used (e.g. qRT-PCR, microarray, fluorometric assay, cytometric bead array assay, ELISA) and the different intestinal segments analysed. Overall, most studies reported altered cytokine expression levels. Pro-inflammatory cytokines known to be associated with activated mononuclear cells (IL-1β, IL-6, IL-8, IL-12) and activated lymphocytes (IL-2, IFN-γ) were found to be either down-regulated[76, 101-103] or up-regulated.[25, 76] Increased TNFα[76] and NO[41] levels were observed in rectal and colonic mucosa. Finally, expression levels of IL-10 and TGF-β anti-inflammatory cytokines were found to be decreased in the colon and rectum of IBS patients.[76, 101, 102]

Key points

  1. Changes in the expression levels of pro-inflammatory cytokines in the gut mucosa of IBS patients are equivocal.
  2. The expression levels of the anti-inflammatory cytokines IL-10 and TGF-β are decreased in the colonic and rectal mucosa of IBS patients.

Trypsin and thrombin levels in the gut mucosa of IBS patients

The luminal proteases trypsin and thrombin, the derived peptides of which exhibit antimicrobial activity, may contribute to the induction of gut inflammation via the activation of protease-activated receptors (PAR) and subsequent gut sensitisation.[104] Accordingly, recent clinical evidence has shown that trypsin and thrombin levels are increased in colonic[81, 83] and rectal biopsies[81] of IBS patients (Table S3).

Key point

  1. Two groups have assessed trypsin and thrombin levels in IBS patients and found that their expression is increased in colonic and rectal mucosa.

Intestinal permeability in IBS patients

Recently, repeated observations of increased mast cell numbers in the gut of IBS patients have prompted an interest in uncovering a potential correlation with gut permeability. Indeed, mast cells are involved in the regulation of intestinal permeability and the integrity of the epithelial barrier.[105-107] Studies focused on permeability assessment using urinary recovery of orally administered sugars or radioisotopes (Table 3) reported increased permeability in the small intestine,[22, 108, 109] with the exception of one study performed in children.[60] Several studies also highlighted an increased colonic permeability[60, 71, 109-111]; however, this result could not be consistently replicated by other groups.[108, 112] Two groups have examined whether the altered intestinal permeability observed in IBS patients could arise from an improper response of the intestinal barrier to noxious agents rather than intrinsic defects. These groups found that rectal permeability was increased in D-IBS patients following exposure to mast cell tryptase,[113] and immune system depression induced by nonsteroid anti-inflammatory drugs resulted in an increased intestinal permeability,[112] suggesting that the ability of the intestinal barrier to react to luminal aggression is more restricted in patients with IBS. Finally, recent studies assessing the integrity of the epithelial barrier have observed a decrease in the expression of the tight junction proteins occludin, claudin-1 and zonula occludens-1 in the colon of IBS patients.[76, 110, 114]

Key points

  1. Increased permeability was observed in the small intestine and colon of IBS patients.
  2. A decrease in the expression of several tight junction proteins was observed in the colonic mucosa of IBS patients.
Table 3. Evidence for intestinal permeabilisation in IBS
FindingsMagnitude of changeSiteMethodsClinical subtypeN (IBS:HC)IBS sex ratio (M:F)Proportion of abnormal values (IBS)Proportion of abnormal values (HC)Diagnostic criteriaReference
  1. ns, not specified; A-IBS, alternating IBS; C-IBS, constipation-predominant IBS; D-IBS, diarrhoea-predominant IBS; F, female; FAP, functional abdominal pain; HC, healthy control; HRP, horseradish peroxidase; IBS, irritable bowel syndrome; M, male; PI-IBS, postinfectious IBS; RT-PCR, reverse transcription polymerase chain reaction; qRT-PCR, quantitative RT-PCR; WB, western blot.

Intestinal permeability
IncreasensSmall intestine and colon5-h and 24-h urinary recoveries of orally administered sugars (lactulose/mannitol urinary ratio)D-IBS19:100:1942% (8/19)0% (0/10)Rome III(Zhou et al., 2010)
 nsns24-h urinary recoveries of orally administered sugars (lactulose/mannitol urinary ratio)D-IBS54:2213:4139% (21/54)0% (0/22)Rome III(Zhou et al., 2009)
 ++Proximal gastrointestinal tract3-h urinary recoveries of orally administered sugars (sucrose/lactulose urinary ratio)FAP/IBS (children)93:5224:69nsnsRome II(Shulman et al., 2008)
 +Colon3-h urinary recoveries of orally administered sugars (sucralose/lactulose urinary ratio)FAP/IBS (children)93:5224:69nsnsRome II(Shulman et al., 2008)
 +++Small intestine5-h urinary recoveries of orally administered 51Cr-EDTAPI-D-IBS but not C-IBS15:15nsnsnsRome II(Dunlop et al., 2006)
 ++Small intestine5-h urinary recoveries of orally administered 51Cr-EDTAD-IBS + PI-D-IBS15:16nsnsnsRome II(Dunlop et al., 2006)
 +++Small intestine6-h urinary recoveries of orally administered sugars (lactulose/mannitol urinary ratio)PI-IBS31:1212:1923% (7/31)nsRome I(Spiller et al., 2000)
No changeSmall intestine3-h urinary recoveries of orally administered sugars (lactulose/mannitol urinary ratio)FAP/IBS (children)93:5224:69nsnsRome II(Shulman et al., 2008)
 -Colon24-h urinary recoveries of orally administered polyethylene glycols or sugars (lactulose/mannitol urinary ratio)IBS14:154:10nsnsRome II(Kerckhoffs et al., 2010)
 -Colon24-h urinary recoveries of orally administered 51Cr-EDTAPI-D-IBS + C-IBS15:12nsnsnsRome II(Dunlop et al., 2006)
Increased permeability in response to tryptase+RectumHRP diffusion through the rectal biopsies in Ussing chambersD-IBS20:30nsnsnsRome II(Lee et al., 2010)
Increased paracellular permeability++colonFITC diffusion through the colon biopsies in Ussing chambersIBS13:5ns77% (10/13)0% (0/5)Rome II(Piche et al., 2009)
 +colonFITC diffusion through the colon biopsies in Ussing chambersC-IBS + D-IBS + A-IBS34:156:28nsnsRome III(Vivinus-Nebot et al., 2011)
Increased permeability in response to NSAIDs administrationnsColon24-h urinary recoveries of orally administered sugars (lactulose/mannitol urinary ratio)IBS14:154:10nsnsRome II(Kerckhoffs et al., 2010)
Decreased transepithelial resistance++ColonFITC diffusion through the colon biopsies in Ussing chambersIBS13:5nsnsnsRome II(Piche et al., 2009)
Epithelial barrier integrity
Decreased tight junction protein occludin++Colon mucosaWBD-IBS but neither C-IBS nor A-IBS50:3312:3832% (6/19); 18% (9/50 total IBS)9% (3/33)Rome III(Bertiaux-Vandaele et al., 2011)
 ++Colon mucosaWB (no change in qRT-PCR)IBS25:189:16nsnsRome II(Coeffier et al., 2010)
Decreased tight junction protein claudin-1++Colon mucosaWB (no change in qRT-PCR)D-IBS but neither C-IBS nor A-IBS50:3312:3853% (10/19); 36% (18/50 total IBS)6% (2/33)Rome III(Bertiaux-Vandaele et al., 2011
Decreased tight junction protein zonula occludens-1+++Colon mucosaqRT-PCRIBS21:12nsnsnsRome II(Piche et al., 2009)
 ++Colon mucosaWBIBS50:3312:38nsnsRome III(Bertiaux-Vandaele et al., 2011)

Discussion

  1. Top of page
  2. Summary
  3. Background
  4. Aim
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Supporting Information

The aim of this systematic review was to determine whether relevant markers reflecting immune activation were more frequently observed in the gastrointestinal tract of IBS patients than in healthy controls. To this end, we examined studies published in English-language journals and assembled available evidence from faecal samples, blood samples or intestinal biopsies. The results of the individual studies that we reviewed were often heterogeneous, partly due to the heterogeneity of the populations recruited (e.g. different IBS subtypes, no age- or gender-matched controls, etc.) and the variety of experimental methods employed. Moreover, numerous biases such as allergy or psychological co-morbidities were not controlled for in these study populations. For these methodological reasons, the data for each individual marker were considered as a whole.

Our analysis indicates that IBS patients presented high numbers of mucosa-associated mast cells and T lymphocytes more frequently than healthy controls. Altered mucosal cytokine production and increased intestinal permeability were also more frequent in IBS patients. However, no significant changes were observed in the numbers of B cells or EC cells or in the mucosal production of serotonin. Taken together, these results suggest that alterations in immune cell populations and in the integrity of the epithelial barrier are associated with IBS. None of the studies reported changes in mucosal neutrophil or eosinophil counts or in faecal levels of lactoferrin and calprotectin, which rules out the occurrence of inflammation in IBS. Rather, the data support a state of heightened immune activation, which could subsequently underlie a low-grade inflammatory response. Nevertheless, functional readouts of the changes in these biomarkers and dose-response data are still lacking, which makes data interpretation difficult in terms of pathophysiology.

Immune activation in IBS patients

The most consistent outcome across all the reviewed studies is an increased number of mucosal-type mast cells in the gut of IBS patients. Although there was variability between the studies (especially depending on the region of the gut that was examined), the data were robust, and the vast majority of the papers reviewed (23/24) reported higher densities of mast cells in the gut of IBS patients. A single-discrepant finding that reported decreased mast cell numbers in the rectal mucosa of IBS patients could be imputable to an abnormally high mast cell density in the control samples.[85] Most studies have assessed the contribution of mast cells to IBS pathophysiology quantitatively (i.e. mast cell quantification) but not qualitatively (i.e. mast cell activation). However, several factors suggest that these quantitative changes could have functional relevance. First, the mucosal levels of mast cell tryptase and histamine, two proteases that are released upon activation, were consistently observed to be increased in IBS patients presenting higher mast cell numbers.[63, 71, 73, 74, 81, 83, 84] Moreover, such mast cell products can induce the permeabilisation of the intestinal epithelium,[115] and increased gut permeability in IBS patients is a robust finding (see Table 3). Two studies have reported a correlation between mast cell density and increased intestinal permeability in IBS patients.[39, 71] Finally, several studies have correlated mast cell number with the severity of abdominal pain symptoms,[64, 68, 73, 116] especially when mast cells were found close to enteric nerves.[68, 73] Further investigation is required to determine the activation state of mast cells (e.g. by coupling classical immunohistochemical staining with more specific markers, such as CD117 or the IgE Fc receptor β chain FcεRIβ), and to address the question of whether patients with gut hypersensitivity or increased permeability also exhibit increased mast cell numbers.

T lymphocytes represent another main immune cell population that is altered in the intestinal mucosa of IBS patients. More than 50% of IBS patients exhibit an increased density of lymphocytes and IELs.[22, 38, 58, 69, 75, 78-80, 88, 90] However, these changes are quantitatively modest and are far below the accepted criteria for diagnosis of intra-epithelial lymphocytosis (>30 IELs per 100 enterocytes); therefore, it is difficult to determine their functional significance. Using the CD25 marker of lymphocyte activation, two studies have attempted to evaluate whether distinct activated lymphocyte subpopulations are specifically involved in IBS pathophysiology and found either no change[45] or an increased density of CD25-positive lymphocytes.[75] Other studies assessing lymphocyte reactivity in IBS patients reported alterations in cytokine production in response to immune challenge.[32, 34, 37, 38] It is also important to note that B cells isolated from the blood of IBS patients exhibit an enhanced activation level with increased expression of type G IgG and the co-stimulatory molecules CD80 and CD86,[47] suggesting that defective communication between T and B cells could result in local overactivation of the adaptive immune system. Additional data related to the mucosal activation status of lymphocytes, production of T cell-associated cytokines (e.g. IL2 or IL4) and correlation between lymphocyte density and symptom severity would clarify the potential involvement of lymphocytes in IBS pathophysiology.

Potential neuro-immune pathophysiological mechanisms in the mucosa of IBS patients

Although further research is required to confirm that the quantitative changes observed in immune activation markers are related to alterations in gut function, these changes provide some insight into the potential mechanisms underlying IBS. Indeed, the data reported herein suggest that gut pathogens (especially infectious agents) or other as-yet unidentified agents could induce mast cell and lymphocyte recruitment within the mucosa (Figure 3). The increased density of immune cells in close proximity to enteric neurons could be a prerequisite for the activation of visceral afferents by inflammatory mediators, thus leading to visceral sensitisation and subsequent abdominal pain.[117] In fact, the distance between the enteric nervous system fibres and inflammatory cells (predominantly mast cells and lymphocytes) was reduced in IBS patients compared with healthy controls. It has been shown that mast cells gather around enteric nerves in IBS patients[68, 73] and mast cell mediators (histamine, tryptase) released into the mucosa contribute to the hyperexcitability of nociceptive visceral sensory nerves.[82] Furthermore, these neuro-immune interactions could be facilitated by an increased density of nerve fibres stemming from nerve growth in the mucosa as suggested by a study that reported a higher density of substance P- and serotonin-positive enteric afferents in the rectosigmoid biopsies of IBS patients.[67] These fibres are located around mast cell aggregates, suggesting that reciprocal interactions between immune cells and nerves could have a synergistic effect.

image

Figure 3. Key players involved in immune activation and potential pathophysiological mechanisms in IBS patients. Common peripheral mechanisms could explain the development of abdominal pain in irritable bowel syndrome (IBS). (a) At the epithelial barrier level, environmental factors (infectious agents, nutrients, microflora, etc.) could permeabilise the intestinal epithelial barrier, notably via PAR-mediated mechanisms. This effect could result in increased contact between the intestinal microflora and the intestinal immune system and consequently initiate immune cell activation (mainly mast cells and lymphocytes) in the mucosa. (b) In the mucosa, inflammatory mediators (e.g. tryptase, histamine, cytokines, etc.) released by mast cells and lymphocytes in the vicinity of colonic nerves induce enteric nervous sensitisation. (c) The sensitisation of intestinal sensory afferents and possibly myenteric neurons leads to subsequent neurogenic inflammation that maintains a local inflammation, thereby contributing to ongoing alterations in sensitivity and motility. CD, cluster of differentiation; DRG, dorsal root ganglia; EC, enterochromaffin cell; GC, goblet cell; IEC, intestinal epithelial cell; IFNg, interferon gamma; IL, interleukin; PAR, protease-activated receptor; TGFb, transforming growth factor beta; TNFa, tumour necrosis factor alpha.

Download figure to PowerPoint

This scenario could be especially relevant in PI-IBS, in which the development of symptoms is likely to result from inflammatory phenomena during the acute stage of infection. Indeed, previous digestive infections and long-lasting inflammatory features in PI-IBS patients have been consistently correlated,[22, 58, 80, 90] further supporting the hypothesis that immune activation is a major contributor to the pathophysiology of IBS or at least confers vulnerability to specific IBS subtypes.[22, 58, 80, 90] The development of chronic symptoms would thus correspond to a switch from an acute level of inflammation to an ongoing low-grade inflammatory response.

However, this model, which is based on the existence of active immune cells in the colonic mucosa, cannot explain the development of symptoms in IBS patients who are not presenting inflammation-related features. The lack of inflammation observed in these patients could be due to the detection threshold of the experimental methodology in terms of its sensitivity, specificity and the populations included. It could also result from other relevant factors that have yet to be investigated (e.g. gut dendritic cells, which are immune cells in the mucosa that play a key role in the communication between innate and adaptive immunity[118]). Immune cell reactivity assays are also necessary for determining whether IBS-related alterations in the gut immune system are qualitative or quantitative. To this end, an assessment of the involvement of ‘sentinel’ molecules, such as Toll-like receptors, in immune activation has recently begun.[34, 119] Another issue is that non-inflammatory mechanisms are also involved in IBS aetiology. For example, stress or depression can induce sensory-motor alterations in the gut and may be associated with a subsequent ‘top-down’ induced inflammatory response.[120-125] Many other risk factors, such as surgery, exposure to anaesthesia, diet or food allergies, could also have as-yet unknown impacts on the immune system in IBS. This idea highlights one of the major limitations of the current data: with the exception of the IBS-PI subtype, the question of whether inflammation is a cause or a consequence of IBS remains to be addressed.

Limitations and further directions

The success of future research aimed at clarifying whether inflammation is a cause or a consequence of IBS will be linked to the researchers' ability to define the relative involvement of the different causative factors of IBS and to overcome the confounding factors. The following questions remain to be addressed: (i) What is the peripheral cause of immune activation in IBS? The specific impact on gut integrity and the physiology of elements such as previous infections, nutrition factors and dysbiosis remains unclear in IBS pathophysiology. (ii) Does the origin of intestinal inflammation in IBS truly reside within the gut? The majority of patients report a lifetime psychiatric history of affective disorders, of which stress and depression are the most common. As reciprocal and synergistic effects between stress and inflammation exist,[126-128] the potential contribution of stress and cognitive factors to inflammation (brain-gut axis) should be further investigated. In particular, stress is an important confounding factor because it induces an increase in pro-inflammatory cytokine levels in the blood.[129, 130] This result has also been demonstrated for depression, which has been associated with disturbances in pro-inflammatory cytokine levels and circulating immune cells.[121, 122] In addition, disturbances in tryptophan metabolism have been described in affective disorders[131-133] and are thus very likely to impact 5HT levels. These results question the specificity and validity of the cytokine- and 5HT-related findings in IBS and stress the crucial importance of taking co-morbidities into account as potential confounders. (iii) What is the functional significance of the changes in immune cell counts observed in IBS? These changes are often modest, and IBS values frequently overlap with those obtained in normal tissue. Consequently, functional readouts must be defined and used to address the relevance of the quantitative results reported. Accordingly, dose-response data and assessments of immune cell activation status would be highly informative. (iv) Finally, some IBS histopathological data remain difficult to interpret due to discrepancies between methodologies, including quantification and detection methods, nature and site of biopsies, variability in population samples, ethnicity and proper recruitment of healthy controls. Moreover, various IBS subtype diagnosis criteria are currently in use worldwide. These methodological issues should be considered cautiously and improved when necessary to define the appropriate controls (e.g. patients should be matched in terms of age, gender and clinical history).

Methodological considerations

In addition to the biases inherent to experimental designs in clinical studies (case and case-control selection biases, information and misclassification biases),[134] the bibliography review process itself is subjected to several other potential biases.[135] First, a publication bias could occur. Positive findings are more likely to be published and to reach a larger scientific community as opposed to papers with negative result. This can lead to both a one-sided reference bias and to an inflated size effect. Consequently, if publication bias is the case, larger studies should report smaller effects.[136] It has to be noted that this was not the case during our review process. Secondly, as described in the Methods section, we restricted our search to articles published in English and to the Medline, PubMed and Scopus databases. Therefore, some relevant studies might have been neglected. Finally, one must keep in mind that the studies reviewed here mainly recruited IBS patients likely to display the most severe symptoms and pathophysiological features. Studies conducted among the general population of IBS patients (i.e. including individuals showing milder symptoms) might not completely reproduce the findings reported herein.

Acknowledgements

  1. Top of page
  2. Summary
  3. Background
  4. Aim
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Supporting Information

JM wrote the first draft of the manuscript. All authors contributed to and have approved the final manuscript. JM constructed the tables and illustrations. EM, MM, AG, MD and TP contributed to the data discussion. EM and DA supervised and provided their final approval of the manuscript. Declaration of personal and funding interests: None.

References

  1. Top of page
  2. Summary
  3. Background
  4. Aim
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Background
  4. Aim
  5. Methods
  6. Results
  7. Discussion
  8. Acknowledgements
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
apt12080-sup-0001-TableS1.docxWord document80KTable S1. Findings suggesting immune activation in the blood and stool of IBS patients.
apt12080-sup-0002-TableS2.docWord document468KTable S2. Enterochromaffin cell counts in IBS patients.
apt12080-sup-0003-TableS3.docWord document565KTable S3. Inflammatory mediators assessment in the gut of IBS patients.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.