Dr N. J. Talley, Mayo Clinic College of Medicine, Charlton 8–138, 200 First Street S.W., Rochester, MN 55905, USA. E-mail: email@example.com
Functional dyspepsia and irritable bowel syndrome are currently considered to be two separate nosological entities. However, the overlap of symptoms and the evidence of a number of common pathophysiological characteristics suggest that functional dyspepsia and irritable bowel syndrome may be different presentations of the same disorder. In this review, we critically appraise points in common, as well as differences, in the epidemiology, pathophysiology and response to treatment of functional dyspepsia and irritable bowel syndrome. Population-based studies and large case series show that one- to two-thirds of subjects with irritable bowel syndrome have symptoms that overlap with functional dyspepsia. Symptom analyses have generally failed to support functional dyspepsia and irritable bowel syndrome as separate entities. An exaggerated motor response to meals, delayed gastric emptying and abnormal small bowel and colonic transit can all be found in subsets of functional dyspepsia and irritable bowel syndrome, and are not exclusive to either condition. Visceral hypersensitivity is a common feature to both entities and seems unlikely to be site or disease specific. There is good evidence for the post-infectious development of irritable bowel syndrome, and this may also apply in functional dyspepsia. Psychiatric comorbidities are similar in functional dyspepsia and irritable bowel syndrome. Several common drug classes (prokinetics, visceral analgesics, psychoactive agents) may similarly improve both functional dyspepsia and irritable bowel syndrome symptoms. The evidence available suggests that at least subsets of functional dyspepsia and irritable bowel syndrome represent different manifestations of a single entity. The identification of common pathophysiological targets for therapy should be pursued in future research.
Functional dyspepsia and irritable bowel syndrome are common, and rate amongst the most frequent conditions referred to gastroenterologists.1,2 They are generally considered to be distinct entities; irritable bowel syndrome refers to lower abdominal symptoms associated with altered bowel habits, whilst functional dyspepsia describes a syndrome of upper abdominal symptoms.3 Yet, the clinical spectrum of these conditions is wide, and the clinical features overlap or fluctuate.4,5 This raises the issue of whether the current nosology, separating irritable bowel syndrome and functional dyspepsia, is appropriate. The aim of this article is to review the features in common, as well as the critical differences, between functional dyspepsia and irritable bowel syndrome; we focus on epidemiological, pathophysiological and treatment data in examining this issue.
A consensus process involving experts in the field has led to the development of the Rome criteria for the functional gastrointestinal disorders.3,6 The prevalence of both functional dyspepsia and irritable bowel syndrome is reported to be remarkably high in Western countries. Community survey data from the US show a striking prevalence of abdominal symptoms in the general population: symptoms compatible with the definitions of irritable bowel syndrome and functional dyspepsia were present in 12% and 26% of the sample, respectively, and isolated diarrhoea, constipation and heartburn were reported by 15–20% of subjects.7 The prevalence estimates seem to vary depending on the stringency of the criteria used for disease status adjudication.1,8–11 However, a recent survey performed to compare prevalence estimates for irritable bowel syndrome using different versions of the Rome consensus criteria (Rome 1989, Rome 1990, Rome I and Rome II criteria) showed good agreement between three of the four sets of criteria.12 A systematic review of the studies of the prevalence of functional dyspepsia found significant study heterogeneity across 19 published reports with regard to the definition of dyspepsia and the prevalence rates.13 When the definition of dyspepsia was restricted to participants with upper abdominal pain, irrespective of the presence of heartburn or acid regurgitation, the prevalence rate estimates ranged from 5% to 12%.13
Many upper gastrointestinal symptoms are present in patients classified as having irritable bowel syndrome;10,14,15 in at least 40% of the patients presenting to gastroenterologists, functional dyspepsia and irritable bowel syndrome overlap considerably.14,16 In a Swedish community, this overlap reached 90% of subjects reporting abdominal symptoms.10 The symptoms of bloating, epigastric pain, early satiety, indigestion, chest pain, nausea and vomiting were the most frequently reported upper gastrointestinal complaints in irritable bowel syndrome in tertiary care (Table 1),15 and in another study early satiety and fullness were more common in constipation-predominant than in diarrhoea-predominant irritable bowel syndrome patients (Figure 1).17 A recent case series from Belgium has suggested that patients with concomitant symptoms of functional dyspepsia and irritable bowel syndrome have a significantly higher symptom severity rating compared with those with irritable bowel syndrome alone.18
Table 1. Upper gastrointestinal symptoms in patients with constipation-predominant and diarrhoea-predominant irritable bowel syndrome (IBS). Modified from Talley et al.15 with permission
Overlap is reported consistently in studies from both tertiary referral centres and primary care.11,15 Thus, the phenomenon is not restricted to those who seek care from a gastrointestinal specialist, but more likely reflects the natural pattern of these conditions.
According to the Rome II criteria, a diagnosis of functional dyspepsia can only be established after excluding any upper abdominal symptoms suggestive of irritable bowel syndrome.3 Exclusion of irritable bowel syndrome may dramatically reduce the prevalence estimates for functional dyspepsia. In a classical study by Drossman et al. in a US householder volunteer population, the prevalence of functional dyspepsia was 13%, but decreased in the absence of irritable bowel syndrome and heartburn to as low as 3%.19 This estimate makes it questionable whether functional dyspepsia and irritable bowel syndrome can be separated.
Factor and cluster analyses have examined whether distinctive subgroups exist naturally in patient cohorts or the general population (Figure 2). A factor analysis of a large European tertiary referral series suggested that functional dyspepsia is a heterogeneous condition, and four subgroups were identified.20 This analysis, however, did not account for any overlap between functional dyspepsia and irritable bowel syndrome.20 However, data from the general population using, primarily, factor analysis have not consistently supported the separation of symptom groups into functional dyspepsia and irritable bowel syndrome; these analyses have tended to identify the irritable bowel syndrome subgroup more consistently than the functional dyspepsia group.21 A discrepancy between the clustering of symptoms in some tertiary referral series and community studies remains.10,20–22 This questions the generalizability of data from tertiary referral series and suggests that symptom subgroups may have limited validity. Currently available analyses have also failed to account for the relationship of symptoms to meals, a feature of both functional dyspepsia and irritable bowel syndrome.23,24 The annual onset rates for functional dyspepsia and irritable bowel syndrome reported in the US are very similar: 5.6% and 6.6%, respectively.25
Symptom patterns also appear to be unstable in both functional dyspepsia and irritable bowel syndrome. A 7-year prospective study of 99 subjects with dyspepsia showed that 15% of subjects could be classified as having irritable bowel syndrome after 1 year and 20% after 7 years, but only 30% of subjects maintained the original symptom characteristics at follow-up.4 Population-based data from Olmsted County have shown that symptom turnover occurs in at least 20% of subjects after 7 years, and may take place in as many as 50% of subjects within 1 year, but the prevalence of dyspepsia and irritable bowel syndrome does not change significantly over time.26 It is probable that the turnover of symptoms does not take place by chance; medications and environmental stressors could play a role, but this has not been studied. However, there are no reliable predictors of this symptom turnover, and no natural symptom clustering associated with such turnover has been identified thus far.10
Grouping of symptoms has been proposed as a useful tool for directing treatment in functional dyspepsia and irritable bowel syndrome.27,28 However, the subdivision of functional dyspepsia into ulcer-like, dysmotility-like and unclassifiable dyspepsia probably provides little guidance in terms of therapeutics.29 On the other hand, in irritable bowel syndrome, the distinction based on symptom patterns (constipation- and diarrhoea-predominant irritable bowel syndrome and irritable bowel syndrome with alternating bowel pattern) holds some value, and is commonly used when planning clinical trials or empirical patient management.
Both functional dyspepsia and irritable bowel syndrome can run in families.30,31 In a community survey, the reporting of a first-degree relative with abdominal pain or bowel problems was significantly associated with the reporting of irritable bowel syndrome [odds ratio (OR), 2.3; 95% confidence interval (CI), 1.3–3.9] and dyspepsia (OR, 1.8; 95% CI, 1.05–3.0) symptoms, but not constipation, diarrhoea or gastro-oesophageal reflux.31 The reporting of a spouse with abdominal pain or bowel problems was not associated with any of these disorders, suggesting that heredity or behaviour learned early in life could play a role.31 Hence, the intrafamilial occurrence of abdominal symptoms is not specific to functional dyspepsia or irritable bowel syndrome, but whether this is environmental or genetic is unclear.
Recent evidence supports the relevance of a genetic milieu in functional dyspepsia. A case–control study found that the presence of a polymorphism of the GNβ3 G-protein (CC) was associated with functional dyspepsia (OR, 2.2; 95% CI, 1.1–4.3).32 A study in irritable bowel syndrome showed that functional polymorphisms of the α(2A) and α(2C) adrenergic receptor and of the soluble carrier protein member 4 (SLC6A4) promoter of the serotonin transporter were associated with functional constipation (OR, 2.48; 95% CI, 0.98–6.28) and with psychosomatic symptoms (OR, 2.2; 95% CI, 1.06–4.64).33 These associations do not imply that functional polymorphisms are the sole or even the main cause of functional bowel disorders, but the findings do represent the first identification of a putative genetic predisposition. Further studies of genetic markers in those with functional dyspepsia and irritable bowel syndrome are now awaited.
In both functional dyspepsia and irritable bowel syndrome, the pathophysiology is likely to be heterogeneous. Of the mechanisms proposed on the basis of the bio-psycho-social model for the functional bowel disorders, many aspects seem to be shared by both functional dyspepsia and irritable bowel syndrome28(Table 2). There are few studies, however, that have directly and comprehensively compared the occurrence of pathophysiological abnormalities in patients with functional dyspepsia, irritable bowel syndrome and both functional dyspepsia and irritable bowel syndrome diagnosed by accepted symptom criteria.
Table 2. Levels of evidence for pathophysiological mechanisms common to functional dyspepsia (FD) and irritable bowel syndrome (IBS)
+++, strong evidence, numerous clinical studies in substantial agreement; ++, moderate evidence, clinical studies in agreement with preliminary data to be confirmed; +, weak evidence, isolated reports, data to be confirmed; ?, no evidence.
Upper gastrointestinal dysmotility
Lower gastrointestinal dysmotility
Brain processing abnormalities
Serotonin signalling abnormalities
Disturbances in gut motor function
The proportion of patients with functional dyspepsia with underlying disturbances in upper gastrointestinal tract motility is variable across different series, but at least one in three patients will have a definable gastric motor abnormality.34 The disturbances of gastric motor function most often reported in functional dyspepsia include delayed gastric emptying,34 abnormalities in antral contractility35 and an impaired accommodation response (a vagally mediated reflex relaxation of the fundus in response to a meal).36 The relationship of these abnormalities to individual symptoms is, however, uncertain, and there is a considerable overlap between asymptomatic subjects and patients,37,38 regardless of their predominant complaint. Impaired gastric accommodation has been proposed to be a more specific finding in patients with early satiety, a symptom that occurs in functional dyspepsia and irritable bowel syndrome.36 However, a study of 309 subjects with functional dyspepsia, including a subset of patients presenting with both functional dyspepsia and irritable bowel syndrome symptoms, observed no differences in gastric emptying measured by a breath test or accommodation measured by a barostat balloon.18 Thus, the specificity of these motor alterations for functional dyspepsia is likely to be low.
In irritable bowel syndrome, classical studies with colonic manometry have shown a relationship between abdominal symptoms and excessive motor responses to stimuli, such as a high fat meal, ileal distension and infusion of cholecystokinin (CCK), normally released after a meal following the presence of fat in the duodenal lumen.39 In patients with irritable bowel syndrome, abnormalities of small bowel manometry have been reported, including the occurrence of discrete clustered contractions and prolonged propagated contractions, and the presence of shorter periodicities of migrating motor complexes.39,40 The coincidence between symptoms and motor abnormalities has, however, been established in only a few cases,40 suggesting that symptom categorization is a weak predictor of actual pathophysiology.
Abnormal transit has also been used as a proxy for bowel dysfunction in irritable bowel syndrome, and is currently applied as an end-point in many pharmacodynamic studies in irritable bowel syndrome.41 Accelerated transit is associated with diarrhoea symptoms and with increased stool output in irritable bowel syndrome, and patients with constipation-predominant irritable bowel syndrome can have delayed emptying of the ascending colon. However, patients with irritable bowel syndrome often have normal transit.42
In a recent series of 146 irritable bowel syndrome patients, in whom 66% had functional dyspepsia symptoms, only the group with overlapping symptoms showed delayed gastric emptying; this abnormality was independently associated with postprandial fullness and nausea.14 On the other hand, patients with functional dyspepsia have been reported to have abnormal orocaecal,43,44 small bowel45,46 and colonic transit.45,47 In a proof-of-concept study in healthy volunteers, Tjeerdsma et al. showed that voluntary suppression of defecation over 4 days was associated with a decrease in the gastric emptying rate.48 These observations support the hypothesis of a ‘colo-gastric’ brake that might explain the occurrence of upper abdominal symptoms in patients with constipation and, possibly, constipation-predominant irritable bowel syndrome. These data show that physiological abnormalities traditionally considered to be exclusive for irritable bowel syndrome are found in patients with functional dyspepsia. However, it is still conceivable that regional motor abnormalities partially dictate the phenotype in those presenting with an irritable gut.
Postprandial symptoms and CCK
In both irritable bowel syndrome and functional dyspepsia, patients may present with an exacerbation of symptoms immediately following a meal. An exaggerated response to the release of CCK after a meal has been invoked in the pathogenesis of postprandial symptoms.49 CCK is known to delay gastric emptying in health and dyspepsia.50 Feinle et al. studied the effect of a CCK-1 antagonist on dyspeptic symptoms during lipid infusion and gastric distension in functional dyspepsia.51 They reported that CCK-1 antagonism was able to suppress symptoms and decrease gastric accommodation and compliance.51 These effects of CCK-1 antagonism on symptoms and gastric accommodation are counterintuitive, as a decrease in postprandial symptoms would be expected to occur with an amelioration of the accommodation response, and suggest that CCK probably does not induce postprandial symptoms by impairing gastric accommodation. In irritable bowel syndrome, Kellow et al. showed symptom exacerbation and increased colonic motility after CCK infusion.39 However, the effects of CCK on colonic motility and symptoms are inconsistent across studies52–56 and, in functional dyspepsia, this has not been evaluated. The postprandial gastrointestinal motor response is most probably the result of an interaction of several different neurohormones, including serotonin and somatostatin, in response to meal ingestion;57 thus, CCK is very unlikely to be the only player in determining postprandial symptoms in functional dyspepsia and irritable bowel syndrome.
Disturbances in visceral sensitivity
Increased sensitivity to visceral stimuli, i.e. a lower threshold for sensation, has been reported consistently in functional bowel disorders and, even in the case of irritable bowel syndrome, has been touted as a ‘biological marker’ of the condition.58,59 The exact localization of the abnormality in the brain–gut axis is a matter of debate; whether this condition results from hyperexcitability of visceral afferents or abnormalities in the central perception of noxious stimuli is unclear.60
In early studies, Whitehead et al. showed that, in irritable bowel syndrome compared with health, a higher proportion of patients reported pain at predefined increases in the volume of a balloon in the rectosigmoid (Figure 3A).61 These data have been replicated using different techniques and sensation ratings; moreover, visceral hypersensitivity correlates with symptom ratings reported over time.62 Visceral hypersensitivity in irritable bowel syndrome does not seem to be limited to the distal bowel. Compared with controls, irritable bowel syndrome patients may also have a lower threshold for pain in the upper gastrointestinal tract.63 A recent study applied rectal distension using the barostat in irritable bowel syndrome, chronic constipation, functional dyspepsia and healthy controls; rectal hypersensitivity had a high sensitivity (95%) and intermediate specificity (72%) for distinguishing between irritable bowel syndrome patients and controls.38 On the other hand, the rectal sensory pressure thresholds in patients with functional dyspepsia were intermediate between those of irritable bowel syndrome patients and controls.38
Mearin et al. compared the sensory ratings in health and functional dyspepsia in response to isobaric gastric distension.64 They found that dyspeptic patients had higher perception ratings than controls to the same distending pressures64 (Figure 3B). In a study measuring visceral sensitivity to distension in the stomach and the jejunum in functional dyspepsia, gastric hypersensitivity was found to be an exclusive feature of functional dyspepsia.65 Other studies, however, have failed to confirm these findings; in functional dyspepsia, hypersensitivity to distension has also been observed in the duodenum and other gut regions.16,63
Patients with combined functional dyspepsia and irritable bowel syndrome have been shown to have a lower threshold for first sensation and discomfort during gastric barostat balloon distension compared with subjects with symptoms of functional dyspepsia alone.18 Overall, in patients with concomitant functional dyspepsia and irritable bowel syndrome, the prevalence of visceral hypersensitivity (using healthy volunteer data as the reference standard) was 44%, compared with 28% in patients with functional dyspepsia alone.18 Hence, the boundaries between irritable bowel syndrome and dyspepsia seem to be poorly defined when using hypersensitivity as a marker. The finding of rectal and oesophageal hypersensitivity in functional dyspepsia is a further clue that the current symptom classification could be artificially separating two different presentations of a single disease process that has altered visceral afferent signals throughout a part of the gastrointestinal tract.
Derangements in the central processing of visceral afferent stimuli have been hypothesized to explain this hyperalgesia.60 Studies with functional brain imaging techniques (functional magnetic resonance and positron emission tomography) have been performed in irritable bowel syndrome patients vs. healthy controls, and consistently greater activation of areas physiologically involved in the processing of visceral pain (such as the anterior cingulate cortex and the thalamus) has been reported in those with irritable bowel syndrome.66 No studies have been published in functional dyspepsia patients. Ladabaum et al. performed positron emission tomography scanning during distal gastric distension in healthy volunteers, and found increases in the activation of somatic and visceral pain processing areas that were proportional to the distension-evoking stimuli.67 This study did not provide specific insight into the pathophysiology of dyspeptic symptoms, but identified the cortical and subcortical structures activated during the generation of painful stimuli from the stomach. Future studies should establish whether the functional involvement of brain areas for the processing of pain stimuli in functional dyspepsia follows a pattern similar to that shown in irritable bowel syndrome. It should be borne in mind, however, that these functional imaging studies still have limits. Indeed, the currently available techniques permit a spatial resolution that can be insufficient for area discrimination, and they cannot provide information on whether any increased blood flow actually corresponds to functional activation of the brain areas studied. Moreover, the results are often presented as summary probabilistic models of activation, and the degree and localization of brain activation may have large intra- and intersubject variability.
Acute past or ongoing infection may be involved in the pathogenesis of the functional bowel disorders.60Helicobacter pylori infection seems to be a cause only relevant to dyspepsia,68 but ongoing H. pylori infection has not been linked to disturbed gastric motor or sensory physiology.69,70 The concept that an acute infectious gastroenteritis can lead to prolonged persistence of symptoms has been evaluated in functional dyspepsia and irritable bowel syndrome.71,72 Studies in animal models of persistent bowel dysfunction post-inoculation with Trichinella spiralis support the concept that an infection could be responsible for widespread changes in the visceral motor and sensory apparatus, despite subsequent resolution of the infection.73
Longitudinal studies are available only in irritable bowel syndrome, and suggest the persistence of irritable bowel syndrome-like symptoms in up to 26% of patients 6 months after a documented acute infectious gastroenteritis.72 A 6-year follow-up of a similar series of patients showed that symptoms persisted in 4%.74 Comparison with cases of new-onset irritable bowel syndrome in the same cohort (presumably not infectious) showed that subjects with persistent post-infectious irritable bowel syndrome were more likely to have diarrhoea and features of anxiety and depression.74
Post-infectious dyspepsia has also been recognized as a possible clinical entity. In a large, retrospective, tertiary referral centre study, Tack et al. grouped functional dyspepsia patients into ‘presumed post-infectious’ (based on a positive response to a questionnaire) and ‘unspecified’.71 They found that impairment in gastric accommodation, detected by the gastric barostat balloon, and hypersensitivity to balloon distension were both more common in ‘post-infectious’ patients.71 In this study, potential ascertainment bias and the lack of prospective design limit the validity of the results. However, the concept that an infection may persistently disrupt gastric function in certain functional dyspepsia patients, leading, as it does in irritable bowel syndrome, to symptom persistence, is plausible and deserves further study.
The infectious hypothesis could account for localized abnormalities in sensorimotor function producing different predominant symptom patterns. In such a case, the distinction between functional dyspepsia and irritable bowel syndrome would be a mere academic exercise.
Psychiatric and psychological comorbidities, including anxiety, neuroticism and somatization, have been reported in both functional dyspepsia and irritable bowel syndrome.75 However, in neither disorder have these abnormalities been established to be causative.75 The association with different personality traits has also been studied in patients with functional dyspepsia and irritable bowel syndrome;76 these traits were similar across the conditions, and the Minnesota Multiphasic Personality Inventory scores were very poorly predictive of symptoms of functional dyspepsia or irritable bowel syndrome.76
The commonalities in pathophysiology suggest that certain drug categories should prove to be efficacious in both functional dyspepsia and irritable bowel syndrome. The armamentarium of drugs with potential use in both functional dyspepsia and irritable bowel syndrome is vast. Prokinetic drugs with pharmacological properties in both the upper and lower gastrointestinal tract may be of use in a subset of functional dyspepsia patients, as well as in patients with constipation-predominant irritable bowel syndrome with slow colonic transit.77,78 Selective serotonergic agents may, in theory, resolve the impairment of gastric accommodation in functional dyspepsia and abnormalities in colonic transit (and, seemingly, visceral perception) in irritable bowel syndrome.79,80 Even gastric acid suppression has been reported to be beneficial in irritable bowel syndrome.81
Hyperalgesia potentially unifies the pathophysiology of functional dyspepsia and irritable bowel syndrome, and is an obvious common therapeutic target. For instance, opioid receptor agonists may be efficacious for increasing the threshold to painful sensation in both functional dyspepsia and irritable bowel syndrome.82 Several experimental visceral analgesics with potential usefulness in both syndromes are under development, including N-methyl-d-aspartate receptor antagonists, neurokinin receptor antagonists and natural and synthetic modulators of the vanilloid receptors.83
Antidepressants are used in practice for patients with epigastric or lower abdominal pain.84 The premise for their usefulness in both functional dyspepsia and irritable bowel syndrome is based on the similar prevalence of comorbid psychiatric conditions in these disorders, and on the hypothesis that they benefit gut sensorimotor function and central processing of visceral stimuli. Current evidence for the use of antidepressants is much stronger in irritable bowel syndrome than in functional dyspepsia.84 The evidence of a response to psychoactive drugs is not sufficient to confirm that these drugs affect peripheral and central mechanisms equally in functional dyspepsia and irritable bowel syndrome.
In conclusion, it can be argued that the similar response of functional dyspepsia and irritable bowel syndrome to many drugs is not unexpected, considering that they may represent different manifestations of the same disorder.
Issues for future research
The current evidence is insufficient to determine whether functional dyspepsia and irritable bowel syndrome are, indeed, two distinct processes or simply different manifestations of a single, broad-spectrum condition (Figure 2). We hypothesize that the distinction between functional dyspepsia and irritable bowel syndrome is artificial based on the pathophysiological data at hand. The relationship of the purported disease mechanisms to specific gut symptoms is poor; it can be argued that we are not dealing with two conditions with different pathophysiologies, but with a single disease leading to various disturbances and symptoms. To test this hypothesis, the multiple pathophysiological mechanisms hypothesized to be relevant in functional dyspepsia and irritable bowel syndrome need to be studied in a large group of patients with overlap, from both tertiary care and the community, in order to factor in possible referral bias. The regulation of the symptom response to a meal and the central processing of increased afferent signals may be identically disturbed in functional dyspepsia and irritable bowel syndrome, and are amongst the issues worthy of further investigation.