Epidemiology of Disorders of the Gut-Brain Interaction: An Appraisal of the Rome IV Criteria and Beyond

Abstract

Disorders of the gut-brain interaction (DGBIs) are presently classified into mutually exclusive anatomical area-related symptom-based categories according to the Rome IV criteria. The pathophysiology of visceral nociception, which contributes to the wide range of symptoms of DGBIs, involves complex psychobiological processes arising from the bidirectional interactions of multiple systems at the gut and brain levels, which affect symptom expression and illness behaviors. The attitude toward an illness and expression of pain and bowel habit vary across cultures with variable interpretation based on sociocultural beliefs, which may not tally with the medical definitions. Thus, psychological factors impact DGBI definitions, their severity and health care utilization. Due to the poor localization and multisegment referral of visceral pain, the anatomical site of pain may not correspond to the affected segment, and there may be a variable degree of overlap among symptoms. The somewhat restrictively defined Rome IV criteria assume one-to-one correlation of symptoms with underlying pathophysiology and ignore overlapping DGBIs, nonstandardized symptom categories, and change or shift in category over time. The microorganic nature of DGBIs resulting from systemic, metabolic or motility disorders, gut dysbiosis and inflammation are not addressed in the Rome IV criteria. Although there is a multidimensional clinical profile that does address these factors, it is not followed rigorously in practice. Threshold changes for diagnostic criteria or addition/deletion of symptoms leads to wide variation among different DGBI criteria resulting in uncertain comparability of results. Although the Rome IV criteria are excellent for research studies and therapeutic trials in homogenous populations, further improvement is needed for their wider applicability in clinical practice.

INTRODUCTION

The term “functional gastrointestinal disorder” (FGID; recently replaced by the term “disorders of the gut-brain interaction” [DGBIs]) is used to define several variable combinations of chronic or recurrent gastrointestinal (GI) symptoms in the absence of any structural or biochemical alterations and identifiable etiology. A symptom is an experiential change in the body or its parts that is reported by the patient as being different from normal and may or may not be interpreted as meaningful. However, a syndrome relates to the association of several clinically recognizable symptoms or signs that cluster together consistently to define a clinical entity like DGBI. In the absence of any objective biomarker, DGBI is classified primarily by symptom clusters that relate to the patient’s interpretation and reporting of an illness experience.¹ These are variably referred to as motility disorders but the symptoms may not be entirely related to the motility changes. The present review focuses on the shortcomings of the epidemiology of DGBIs defined by Rome IV criteria and the future perspectives in DGBI diagnosis.

PATHOPHYSIOLOGY OF VISCERAL NOCICEPTION AND ITS IMPLICATION FOR DGBIS

Altered visceral nociception is the key underlying abnormality that predisposes to the development of DGBIs. Visceral nociception involves complex pathways. Sensory neuroreceptors in abdominal organs are located in the mucosa (which respond to chemical stimuli), muscle, serosa, and mesentery (which respond to mechanical stimuli) where they mediate nociception as well as secretion, motility and blood flow via local (enteric nervous system) and central (brain/spine) reflexes. Autonomic fibers from these receptors synapse within the spinal cord with fibers coming from the brain (mid-brain, periventricular grey matter, and caudate nucleus), which inhibit or facilitate the degree of incoming impulses by gate control mechanisms by activating spinal inter-neurons. Ascending second order fibers from these synapses travel via the lateral spinothalamic tracts to relay in the thalamus and reticular formation. From here the third order fibers project to somatosensory (involved in appreciating pain intensity and localization), limbic (involved in the emotional, motivational and affective aspect of pain) and prefrontal (involved in the cognitive and evaluative aspect of pain) brain cortex, which interpret the afferent gut impulses variably according to their state. Because of the bidirectional brain-gut interactions, environmental and psychological information alter GI function but GI conditions also affect pain perception, mood and other mental functions. Visceral pain perception is thus a complex psychobiologic process, during which visceral afferent input is processed and modulated by cognitive and affective circuits in the brain. At the gut level, synchronization of the enteric, parasympathetic and sympathetic nervous system is necessary for normal control of gut function, which relies on specific GI cell types such as enteric neurons, immune cells (resident macrophages, mast cells), enteroendocrine cells (which secrete serotonin) and a normally functioning gut microbiome. Repetitive enteric and extrinsic nerve stimulation can lead to transient or long lasting neuroplastic changes (increased expression of neuronal growth factor, polymodal receptors and neurite outgrowth) in the spinal cord and brain circuits, which are responsible for gut secretory, motor, and sensory dysfunctions (like allodynia) that characterize DGBIs.² Visceral pain is poorly localized and dull (due to multi-segmental and low-density innervation), felt in the midline (due to bilateral innervation of organs from the spine), is often associated with somatic and other visceral reference (due to convergence of afferent nerves with somatic and other visceral afferents in the spinal cord) and is associated with strong negative affective reactions and secondary autonomic effects like nausea, vomiting, perspiration, pallor and restlessness.

The clinical phenotype of DGBI emerges from the interactions of multiple systems in the periphery (noxious luminal contents, altered gut microbiome and mucosal barrier, abnormal stimulation of the gut immune system including hormones and neurotransmitters leading to inflammation, visceral hypersensitivity and altered gut motor activity) and in the brain (networks of emotional arousal, sensorimotor and autonomic function). These interact bi-directionally to produce the clinical expression of the DGBI phenotype namely, changes in digestive functions (including secretory and motor disorders) and the symptom experience, its severity and subsequent illness-related behaviors. This is shaped by the meaning of illness, the fears of continued symptoms, the perceived concerns relating to alterations in body image and social acceptability, the degree of functional impairment with its implications at work and at home, the sense of helplessness to affect symptom relief, and the difficulty of coping with disability.

IMPACT OF PATHOPHYSIOLOGY ON SYMPTOM INTERPRETATION AND EPIDEMIOLOGY

Given the complex pathophysiology of visceral nociception including deep involvement of the central nervous system in it, patients’ perception drives healthcare-seeking behavior affecting the clinical spectrum of DGBI. Patients’ sociocultural background (which formulates a person’s traditional belief, attitude to illness and its relation to diet) and expressive language (the terms used to describe a particular feeling of symptom) are important for his or her symptom perception and reporting. The expression of pain varies across cultures from dramatic expression to stoicism and denial; so is the bowel habit, e.g., diarrhea or constipation by medical definition may not be applicable from the patient’s perspective. These factors pose major challenge to the relevance of data generated by applying a questionnaire formulated in one language with its unique terms and semantics to the widely variable population across the world without corresponding vocabulary in regional languages. The direct translation may not convey the appropriate meaning in another culture and this may lead to over or under-estimation of disease prevalence. For example, the patient reported symptomatology of abdominal pain, burning, discomfort, fullness, bloating/gas trap, dyspepsia, early satiety, nausea and belching may mean the same visceral nociception as felt by different persons or the same person at different times according to the degree and level of the multi-tiered gut-brain interaction. Moreover, due to the poor localization of visceral pain, the anatomic site of pain may not correspond to the affected segment and there may be a variable degree of overlap among symptoms. These factors have implications for epidemiological definitions of DGBIs (especially overlap between clinical entities and shifting between categories)³ and treatment.

THE EXTENT OF ORGANIC DISEASE EXCLUSION NEEDED FOR DEFINING DGBI

The term “functional” will depend in reality on the exclusion of organic and specific motility disorders as well as systemic and metabolic diseases⁴ by meticulous investigations. These may include: (1) different combinations of endoscopy including biopsy (to exclude morphologically or histologically distinguishable disorders like cancer, Helicobacter pylori and other chronic infections, celiac disease, microscopic colitis, eosinophilic GI disorders, disaccharidase deficiency); (2) motility studies including manometry of different GI areas (to differentiate functional dyspepsia [FD] from gastroparesis or functional constipation [FC] and constipation predominant irritable bowel syndrome [IBS-C] from slow colon transit due to visceral neuromyopathy); (3) pH-metry with impedance study of the esophagus (to exclude gastroesophageal reflux disease [GERD]), exclusion diets (for food intolerance or allergy); or (4) investigations to diagnose small intestinal bacterial overgrowth (SIBO), bile acid diarrhea and subtle pancreatic exocrine insufficiency. If not excluded, these may lead to inflated results for precise determination of prevalence of DGBIs. However, this is an uphill task for large-scale population-based studies. Fortunately, most such disorders are sufficiently rare not to overinflate the prevalence results (except for H. pylori infection and GERD) or mandate a detailed search early in the management protocol. The other problem is that even after detailed investigations and treatment of a diagnosed problem, the symptoms may remain and not tally in time and site of occurrence with the disorder present due to the complex pathophysiology of visceral nociception in which the cause and effect cannot be disentangled between symptoms and pathophysiological mechanisms. GI cancer, a potentially fatal diagnosis, which may present sometimes with non-specific symptoms mimicking DGBIs, is extremely rare, particularly in the absence of alarm features. In a recent meta-analysis of 28,630 patients of IBS undergoing colonoscopy, the pooled prevalence of colorectal cancer, inflammatory bowel disease, and microscopic colitis were 0.78%, 4.48%, and 2.35%, respectively with no difference in the yield between IBS and non-IBS patients.⁵ Others have reported similar rates.⁶ Regarding DGBIs with upper GI symptoms, an older meta-analysis of Asian patients with FD⁷ reported an overall malignancy detection rate of 1.3% (95% confidence interval [CI], 0.80 to 2.10), which is similar to more recent studies.^8-12

The approach to DGBIs will vary in different parts of the world according to the cost, access and standards of healthcare (determining the extent of investigations possible), which might limit appropriate testing. In many Asian countries where access to endoscopic services is easy and “open access” type, this is usually the first investigation that the patients want and get done. In others, an initial course of a proton pump inhibitor along with lifestyle and dietary modifications may be prescribed. The other factor is the ready availability of over-the-counter medications or other complementary/alternative medicines. It is difficult to generalize one approach for all except for scientific and investigational purposes. In a recent Asian survey involving 196 institutes across seven countries (Republic of China, Indonesia, Japan, Korea, Philippines, Singapore, and Thailand), more than 60% of respondents used both upper GI endoscopy and H. pylori test to diagnose FD, and colonoscopy to diagnose IBS and chronic constipation. The frequency of GI motility testing was low in most countries (60% to 100% did not use it) except Singapore and Korea where it was commonly used. 50% of respondents used the gastric emptying test in Thailand and colonic transit time in China. Scintigraphy was the most common gastric emptying test used in seven countries except China, which used the ¹³C breath test. The radiopaque marker method was used the most to measure colonic transit time in seven countries except Indonesia.¹³

1. Impact of psychological factors on defining DGBI

Genetics, early life events, sociocultural influences and environmental factors may affect the psychosocial development of a person in terms of personality traits, susceptibility to life stresses, psychological state, and cognitive and coping skills. Psychosocial factors influence the physiological functioning of the GI tract via the gut-brain axis (motility, sensitivity, barrier function, and secretion), are modulators of the patient’s pain experience and symptom behavior, and ultimately affect treatment selection and the clinical outcome. These also influence the susceptibility to gut dysfunction like abnormal motility or sensitivity, mucosal immune dysfunction or inflammation, the microbial environment as well as the effect of food and nutritional substances. Furthermore, these gut-brain variables reciprocally influence central nervous system expression.¹⁴

The presence of psychological comorbidity influences health-seeking behavior quite significantly. In modern times, stress presents in varied forms with increasing severity day by day. The rapid pace of modern lifestyle with its associated psychological stress puts the interaction of the nerves at enteric, spinal, and brain levels with the gut hormones, neurotransmitters, immune system, and gut microbiome on an all-time high. They are all in a state of constant flux due to the variety of environmental stressors like unhealthy dietary patterns, smoking, alcohol, pollution and intoxicants. This alters nociception and may change the type of DGBI or cause overlapping DGBIs due to which the epidemiology is likely to become very dynamic and changeable as compared to other chronic diseases. The modern-day rapid population flux from rural to urban areas, from less to more industrialized countries with its attendant multi-faceted changes in living and work environment, lifestyle, work and dietary pattern, stress level, exposure to processed foods and antibiotics, environmental toxins and pathogens add to this dynamicity. It can be anticipated that the development of DGBIs will vary in different environments. Multiple studies have shown DGBIs change over time in the same individual.^15-17

Notwithstanding the complexity of the pathophysiology, which makes scientific study a difficult matter, the first effort toward a systematic study of DGBIs was made by Manning and Thompson in 1978. Subsequently, Rome criteria have been formulated for betterment and addition, alterations were made according to accumulating evidence over time (Rome I in 1994, Rome II in 1999–2000, and Rome III in 2006). The recently updated Rome IV criteria in 2016 (classifying 33 diseases in six categories and 22 subgroups by anatomic location: esophageal, gastroduodenal, bowel, biliary, anorectal, and centrally mediated disorders of gastrointestinal pain sub-categories) were designed not only to provide guidance for research studies in general and therapeutic trials in particular but also to facilitate making a positive diagnosis of individual DGBI, based on both the presence of characteristic symptoms and absence of alarm features and on objective findings from a limited number of standard diagnostic tests and investigations.¹ However, such symptom-based sub-classifications are narrowly defined assuming a one-to-one correlation of symptoms with underlying pathophysiology. The most common DGBIs in the community belong to the first three groups: GERD spectrum disorders, FD, and IBS.

EPIDEMIOLOGY OF GERD BY ROME IV

The term GERD includes a spectrum of disorders that present with similar symptoms like heartburn, belching, and regurgitation. They include erosive esophagitis (EE, mucosal abnormality with abnormal acid exposure) and those with normal mucosa, e.g., nonerosive reflux disease (NERD; abnormal acid exposure with positive or negative symptom reflux association), functional heartburn (normal acid exposure with negative symptom reflux association) and reflux hypersensitivity (normal acid exposure with positive symptom reflux association) by Rome IV criteria. In addition, all the other symptoms like globus, chest pain, and dysphagia may also be present in GERD patients, and to label them as “functional,” Rome IV criteria endorsed the exclusion of organic disease by endoscopy, ambulatory pH-metry, and pH-impedance testing.¹⁸ However, in the Rome Foundation global study¹⁹ in which even when such investigations were not performed, the prevalence of functional esophageal disorders were surprisingly low compared to the reported worldwide prevalence of GERD (Fig. 1).²⁰

Fig. 1.

Prevalence of gastroesophageal reflux disease in different countries. Adapted from Nirwan JS, et al. Sci Rep 2020;10:5814.²⁰

EPIDEMIOLOGY OF FD BY ROME IV CRITERIA

The FD subgroups are postprandial distress syndrome (PDS), epigastric pain syndrome (EPS) and their overlap. The reported prevalence of uninvestigated dyspepsia (UD) varies considerably in different populations due to different interpretations and expressions of symptoms, diagnostic criteria adopted, environmental factors and local prevalence of organic diseases like peptic ulcer and gastric cancer.²¹ According to the Rome IV criteria, such structural disorders need exclusion by gastroduodenoscopy. A recent meta-analysis showed pooled global prevalence rate of UD to reduce from 17.6% (95% CI, 9.8% to 27.1%) by Rome I criteria to 13% (95% CI, 8.4% to 18.5%) by Rome II criteria to 11.5% (95% CI, 9.8% to 13.3%) by Rome III criteria and 6.9% (95% CI, 5.7% to 8.2%) by Rome IV criteria. PDS was the commonest subtype (46.2% by Rome III criteria, and 62.8% with Rome IV). Even when uniform criteria were used, prevalence rates varied between countries suggestive of environmental, cultural, ethnic, dietary or genetic influences determining symptoms.²² In addition, symptoms of H. pylori gastritis and gastroparesis²³ often overlap with FD. Therefore, ideally, these should be excluded before the diagnosis of FD is made. The prevalence of FD parallels the prevalence of H. pylori infection in most countries except Australia, United States, and some European and Asian countries due to above-mentioned factors (Fig. 2).^19,24 In an endoscopic study of UD patients from rural Bangladesh, 67% had FD and 28.6% with organic dyspepsia had a peptic ulcer (9% overall). But the prevalence of H. pylori (78%) was not different between FD (75.2%) and organic dyspepsia (82.1%).²⁵ A similar study from Vietnam reported 3.2% prevalence of gastroduodenal ulcer, being lower in FD with versus without reflux symptoms (0.6 and 4.7%, respectively).²⁶ In yet another middle-aged and elderly population-based study from a cancer prevalent zone of China, the prevalence of organic dyspepsia was 4% (peptic ulcer 3%) overall and 10.2% in UD. The H. pylori infection rate in FD patients was not higher than that in the control group (34.23% vs 42.26%).⁸ Though the overall prevalence of UD in the Rome Foundation global study¹⁹ tallies with the meta-analysis, bias may arise from areas of high H. pylori and peptic ulcer prevalence. Also, a very low prevalence of dyspepsia in the house-to-house survey in India is quite interesting.

Fig. 2.

(A) Prevalence of functional dyspepsia in different countries. Adapted from Sperber AD, et al. Gastroenterology 2021;160:99-114.¹⁹ (B) Prevalence of Helicobacter pylori infection in different countries. Adapted from Hooi JKY, et al. Gastroenterology 2017;153:420-429.²⁴

Another problem with the symptom-based definition is exemplified by the variable prevalence of PDS, EPS and their overlap by different criteria.²⁷ Using Rome III criteria, 25% of participants had PDS, 8% had EPS, and 67% had an overlap whereas by Rome IV criteria, the overlap group was reduced to 35%, 57% of patients had PDS and 8% EPS. The symptoms of pain, fullness, bloating and discomfort, which define these criteria, may have the same pathophysiological basis yet are variably interpreted by patients possibly due to psychological factors, which are intricately related to the central pain processing mechanism.²⁸ Multidimensional clinical profile of FD needs to be considered while diagnosis and treatment of these patients.²⁹

EPIDEMIOLOGY OF IBS BY ROME IV

In a recent meta-analysis of 92 studies comprising 4,23,362 participants, the pooled prevalence of IBS in 53 studies that used the Rome III criteria from 38 countries comprising 3,95,385 participants was 9.2% (95% CI, 7.6% to 10.8%) versus 3.8% (95% CI, 3.1% to 4.5%) among six studies that used the Rome IV criteria from 34 countries comprising 82,476 individuals. IBS with diarrhea (IBS-D) was the most common subtype with the Rome IV criteria (reported by 31.5%; 95% CI, 23.2% to 40.5%).³⁰

Table 1 depicts a comparative analysis of ROME III versus ROME IV criteria used for diagnosing IBS by a number of recent studies^30-39 among different populations in different settings. Universally, ROME IV criteria are more restrictive and so are likely to diagnose more severe disease while excluding patients with milder disease or those with less specific symptoms (unclassified functional bowel disease). Thus, Rome IV criteria may be good for application in secondary/tertiary care whereas Rome III is likely to perform better in primary care. Among subgroups, there is clear surge of IBS-D, fall in IBS-M (mixed) and U (unclassified), with IBS-C showing variable change by Rome IV criteria, which also increase the prevalence of FC/FD due to the elimination of the word “discomfort.” It has correctly been hypothesized that IBS-C and FC are spectrum disorders rather than distinct clinical or pathophysiological entities.³ It also appears that more Europeans with IBS by Rome III also fulfil Rome IV criteria (50% to 85%)^32,33 compared to Chinese Asians (20% to 66%)^34,35,40 where there is a tendency to report bloating/distension as discomfort and not pain.⁴⁰ Newer methods of subgrouping IBS patients which includes psychological criteria are being explored.⁴¹

Table 1.

Studies Presenting Comparative Analyses of Rome III and Rome IV Criteria

Reference	Country	Cohort	Subject number	ROME III	ROME IV	Remarks
Sperber et al.19	Worldwide	Questionnaire via internet	54,127	IBS 10.1%	IBS 4.1% IBS-D 1.2% IBS-C 1.3% IBS-M 1.3% IBS-U 0.3% FC 11.7% FDr 4.7% FB 3.5%	Prevalence was much lower when face to face interview was used than when self-completed via internet
		Questionnaire via interview	18,949	IBS 3.5%	IBS 1.5.% IBS-D 0.4% IBS-C 0.6% IBS-M 0.3% IBS-U 0.2% FC 6.6% FDr 1.2% FB 1.2%
Oka et al.30	Worldwide	Metanalysis of 92 separate populations, Questionnaire via internet and interview	423,362	IBS 9.2% IBS-D 27.8% IBS-C 20.0% IBS-M 33.8% IBS-U14.1%	IBS 3.8% IBS-D 31.5% IBS-C 29.3% IBS-M 26.4% IBS-U 11.9%	Prevalence of IBS much lower when questionnaire was administered in interview than when self-completed online
Chuah et al.31	Malaysia	Primary care, health screening, face to face questionnaire	1,002	IBS 4% IBS-D 40% IBS-C 27.5% IBS-M 7.5% IBS-U 25% FDr 1.2% FC 10.5%	IBS 0.8% IBS-D 50% IBS-C 25% IBS-M 0% IBS-U 25% FDr 3.3% FC 11.7%	Cases reduction in ROME IV–24 due to elimination of the word “discomfort”; 8 due to change pain frequency to once a week
Palsson et al.32	USA, UK, Canada	Population-based study by internet questionnaire	6,300	IBS 9% IBS-D 19.8% IBS-C 17.95 IBS-M 59.7% IBS-U 2.6% FDr 0.9% FC 5.6%	IBS 4.6% IBS-D 35% IBS-C 28.5% IBS-M 31% IBS-U 5.5% FDr 4.7% FC 7.8%	Cases reduction in ROME IV–62 due to elimination of the word “discomfort”; 296 due to change in pain frequency to once a week; 9 due to changing the association between pain and abnormal stools
Aziz et al.33	Sweden	Tertiary care FGID OPD, IBS by Rome III	542	IBS 100% IBS-D 35% IBS-C 25% IBS-M 10% IBS-U 21%	IBS 85% IBS-D 34% IBS-C 25% IBS-M 11% IBS-U 21%	May not be applicable to primary care with milder symptoms
Wang et al.34	China	Database of FGIDs by ROME III, tertiary care OPD	307 IBS (of total 946 FGIDs)	IBS 30.9% IBS-D 5.8% IBS-C 7.9% IBS-M 13.4% IBS-U 72.9%	IBS 6.2% IBS-D 25.9% IBS-C 1% IBS-M 19% IBS-U 44.8%	With ROME IV there was more overlap with FDr and surge of U-FBD
Bai et al.35	China	Tertiary care OPD Face to face questionnaire	175 IBS	IBS 12.4% IBS-D 58.8% IBS-C 30% IBS-M 1.2% IBS-U 10%	IBS 6.1% IBS-D 61.9% IBS-C 26.2% IBS-M 2.4% IBS-U 9.5%	Cases reduction in ROME IV–61 due to elimination of the word “discomfort”; 27 due to change pain frequency to once a week; Bloat was very common among both ROME III and IV patients
Goyal et al.36	India	College students	1,309	IBS 9.5% IBS-D 38.7% IBS-C 14.5% IBS-M 39.5% IBS-U 7.3% FDr 0.92% FC 1.3%	IBS 6.2% IBS-D 63% IBS-C 27.2% IBS-M 8.6% IBS-U 1.2% FDr 2.9% FC 2.1%	Cases reduction in ROME IV–25 due to elimination of the word “discomfort”; 23 due to change pain frequency to once a week; 4 due to changing the association between pain and abnormal stools
Black et al.37	UK	Self-identified IBS	1,375	IBS 78.9% IBS-D 31.1% IBS-C 11.5 % IBS-M 55.6 % IBS-U 1.7 %	IBS 59.1% IBS-D 38.8% IBS-C 17.5% IBS-M 40.8% IBS-U 3.2%	Of those Rome III IBS who did not meet the Rome IV criteria, only 11.5% were reclassified into another FBD (FC, FDr, FB) where licensed and evidence-based therapies are available. 23.4% U-FBD
Vork et al.38	Netherlands	Secondary/tertiary care OPD (72% cases), face to face ROME III questionnaire, abdominal pain severity used as surrogate for ROME IV	404	IBS 100% IBS-D 34.7% IBS-C 20% IBS-M 39.4% IBS-U 5.9%	IBS 87.4% IBS-D 33.4% IBS-C 20.7% IBS-M 40.2% IBS-U 5.7%	More ROME IV positives from secondary/tertiary care with higher symptom score. ROME IV negatives had FC, FDr, FB
Black et al.39	UK	Referred from primary to secondary care	577	IBS 75.3% IBS-D 38.2% IBS-C 21.5% IBS-M 38% IBS-U 2.3%	IBS 69% IBS-D 37.3% IBS-C 22.8% IBS-M 37.6% IBS-U 2.3%	ROME IV more specific in secondary care. Higher symptom score

IBS, irritable bowel syndrome; IBS-D, diarrhea predominant IBS; IBS-C, constipation predominant IBS; IBS-M, mixed IBS; IBS-U, unclassified IBS; FC, functional constipation; FDr, functional diarrhea; FB, functional bloating; FGID, functional gastrointestinal disorders; OPD, outpatient department; U-FBD, unclassified functional bowel disease.

ENTITIES UNADDRESSED IN THE ORIGINAL ROME IV CRITERIA

1. Post-infection DGBI

Acute gastroenteritis due to the pathogen (bacteria, virus, or protozoa) is the most common acute GI disorder worldwide. Though most are self-limiting with complete recovery, some can go on to develop DGBI over time revealing its microorganic nature with prolonged alteration in gut microbiome being the most important cause. The two most common are post-infection (PI)-IBS and PI-FD.

The incidence of PI-IBS is widely variable (3.7% to 85.5%) depending at least partly on the type of pathogen and associated frequency of risk factors.⁴² A recent meta-analysis showed the pooled prevalence to be 11.5% (95% CI, 8.2% to 15.8%) with comparable results after a follow-up duration of 3, 6, 12, 13–59, or ≥60 months after gastroenteritis.⁴³ The incidence of PI-FD is less studied; two recent estimates are 7.4% (overlap with IBS 4.6%)⁴⁴ and 20%.⁴⁵ An older meta-analysis of 19 studies showed the mean prevalence to be 9.55% (909/9,517) with a summary odds ratio of 2.54 (95% CI, 1.76 to 3.65) at more than 6 months after gastroenteritis compared to the prevalence in controls within the same population.⁴⁶

SIBO has been linked to FD, IBS, and FC.⁴⁷ Two recent meta-analyses found a higher frequency of SIBO among FD and IBS patients compared to healthy controls.^48,49 Small intestinal fungal overgrowth has also been associated with symptoms of DGBIs. The improvement of FD associated with H. pylori eradication has been surmised to be due to the settling of dysbiosis rather than due to H. pylori eradication itself. The current standards of diagnosing small intestinal dysbiosis (breath tests, aspirate culture) are suboptimal and molecular assays may better delineate the nature of dysbiosis and help in the disentanglement of the cause-effect relationship.⁵⁰

Viruses associated with gastroparesis include rotavirus, parvovirus, norovirus, herpes family viruses, Norwalk virus⁵¹ and very recently coronavirus disease-19.^52,53 However, virus-induced DGBI usually normalizes due to the less invasiveness of the virus.³⁸ At 6‐month follow‐up, 15 (5.3%), six (2.1%), and five (1.8%) of 280 coronavirus disease-19 patients developed IBS by Rome III criteria (of these 14 fulfilled Rome IV criteria also), UD, and IBS‐UD overlap, respectively.⁵²

Though this entity was unaddressed in the original ROME IV criteria, a subsequent Working Team Report formulated diagnostic guidelines endorsing its importance in any future large-scale study on the prevalence of these DGBIs.⁵⁴

2. Overlap among DGBIs

Overlap between different DGBIs like FD-gastroparesis, FD-IBS, FC-IBS-C, FD-GERD-IBS, and among subgroups like PDS-EPS are common. In a recent review,⁵⁵ summarizing the results of studies on the overlap of GERD-FD-IBS conducted before Rome IV (2016) utilizing various criteria in a different population, the prevalence of GERD-IBS overlap was 3% to 79% in questionnaire-based community studies with the positive association for all geographic regions and degree of overlap, 10% to 74% in studies using endoscopy (higher with NERD) and 11% to 79% for pH-metry with impedance and manometry-based studies (associated in descending order with functional heartburn, NERD and EE). FD-GERD overlap was 21% to 63% (higher with more frequent symptoms) in questionnaire-based studies, 28% in endoscopy-based studies, and 23% in pH-metry-based studies (associated in descending order with functional heartburn, NERD, and reflux hypersensitivity). In a German study analyzing nationwide data on IBS from a database, 55% had a preexisting GI diagnosis within 1 year before the first IBS diagnosis like intestinal infection (26%), gastritis/duodenitis (21%), esophageal disease (15%), non-infectious enteritis or colitis (7.4%), FD (6%), and ulcers (1%). One of these diagnoses was also rendered to 41% within 1 year after the first IBS diagnosis including infectious diseases (18%), gastritis/duodenitis (16%), diseases of the esophagus (13%), non-infectious enteritis or colitis (3%), FD (2%), and ulcers (1%).⁵⁶ With a better understanding of pathophysiology, it is becoming clear from more recent studies,^{25,26,36,45,57-73} that overlap is more common than any individual DGBI and there can be overlap among more than two categories (Table 2). Though this entity was unaddressed in the original Rome IV criteria, the Rome committee has appreciated its importance and is likely to formulate guidelines. The symptom severity increases with the number of DGBIs with concomitant worsening of mental, physical, and global health-related quality of life. The correlates of multiple DGBIs are higher severity of GI symptoms, somatization, higher depression and anxiety rates, lower quality of life and increased physician consultations. With increasing illness severity, psychosocial variables gain more relevance. More severe courses are associated with multiple and persistent somatic symptoms, numerous psychosocial stressors, high emotional distress, disproportionate illness anxiety, high functional impairment, frustrating physician-patient relationships and dysfunctional healthcare utilization.⁷⁴ Thus, from a management aspect, better pathophysiological delineation with targeted treatment is the need of the hour and all these issues have been addressed recently in an Asia-Pacific Consensus on overlap disorders.⁷⁵ The microorganic nature of DGBI has already been shown in studies on gut motility and dysbiosis. Studies also implicate sleep disturbance in its pathophysiology.^34,76 With the availability of functional MRI scans of the brain, alterations in multiple brain areas are being detected in different DGBIs which can be future targets for better treatment.^28,77-79

Table 2.

Worldwide Prevalence of Overlapping DGBIs in Different Countries

Country	No. of patients	Overlap among DGBIs (FGIDs)	Study type (place) and tool
Australia57	1,127	UD-IBS 64%	Tertiary care OPD, Local questionnaire
South Korea58	632	UD-IBS 17.4%	University hospital OPD, Rome III questionnaire
India59	2,774	UD-IBS 4.1%. In UD group, PDS-EPS 64%	Rural community based, Translated Rome III questionnaire
India60	2,774	GERD-UD 41%. In UD, PDS-EPS 83.6%	Rural community based, Translated validated Asian Rome III questionnaire
India36	1,309	Overall 9.3% (of 26.9% DGBIs), most common UD-IBS 4.4% (among IBS 50.6% UD, among UD 20.6% IBS)	College students. Rome IV questionnaire
China61	8,906	FH-IBS 16.3%, FD-IBS 36.7%	Multicenter tertiary care hospital OPD, Rome III questionnaire
China62	319	29.8% had two, 18.5% had three, 4.7% had four overlapping FGIDs	Multicenter tertiary care hospital OPD, Rome III questionnaire
Vietnam26	439	GERD-FD 51.3% (PDS more in overlap than FD only)	Tertiary care OPD. Reflux symptoms ≥2 times a week for GERD or fulfilling the Rome IV criteria for FD. Endoscopy done
Bangladesh25	3,351	16.3% UD, FD 67.05% after endoscopy (PDS-EPS 7.7%)	Community based. Translated and validated enhanced Asian Rome III questionnaire. Endoscopy done
Taiwan63	222	45.8% NERD-FD, 43.8% NERD-IBS, 41.3% EE-FD, 44.4% EE-IBS	Tertiary care hospital. Chinese GERD Q and Rome III questionnaire, Endoscopy done
South Korea64	759 patients of constipation FC/IBS-C (40.6%/59.4%)	Overlap (35.2%) with GERD 17.9% (16%/20.8%), with FD 10.5% (9.5%/12%) and with both 6.7% (4.9%/9.4%)	Multicenter study using GERD Q and Rome III questionnaires
Saudi Arabia65	2,811	IBS-FD 44%, FD-FH 19%, PDS-EPS 3.4%	Internet-based national survey, Rome IV questionnaire
Denmark66	1,00,000	Overlap between two or three DGBIs 30.7%	Population-based, Rome III questionnaire
Worldwide67	54,127 adults in 26 countries	32.7% have >1 DGBI (22.3% two, 7.1% three, 2.3% four)	Internet-based, Rome IV questionnaire
France68	3,555	>2 DGBIs in 79.3% (10 groups had 2 disorders, 8 groups had 3 disorders)	Hospital OPD based, Rome III questionnaire
Romania69	2,049	Multiple DGBIs in 11.5% overall (two [19.9%], three [6.8%] or four [2.1%])	Internet-based, Rome IV questionnaire
UK, USA, Canada70	5,931	Average DGBI/person=1.5, 36% have >1 anatomic region involved. Bowel-anorectal (9.1%), bowel-gastroduodenal (7.5%) and bowel-esophageal (4.2%)	Internet-based, Rome IV questionnaire
UK, USA, Canada71	5,931	PDS-EPS 21% (overlap more associated with IBS, FH)	Internet-based
Belgium45	459 FD	PDS-EPS 55%	Rome IV questionnaire
Bulgaria72	1,896	IBS-FD 11.7%, PDS-EPS 28%	Internet-based, Rome IV questionnaire
Russia73	9,254	10.1% FD-IBS	Retrospective study from OPD records from 67 cities, Rome IV questionnaire
USA23	338 Patients of constipation	71.5% delayed gastric emptying, 88.2% gastroparesis-UD, and 58% constipation. Delayed colonic transit in 32%, IBS-C overlap?	Hospital based. Rome III questionnaire. Gastric emptying scintigraphy, wireless motility capsule

DGBI, disorders of the gut-brain interaction; FGID, functional gastrointestinal disorders; UD, uninvestigated dyspepsia; IBS, irritable bowel syndrome; OPD, outpatient department; PDS, postprandial distress syndrome; EPS, epigastric pain syndrome; FH, functional heartburn; FD, functional dyspepsia; GERD, gastroesophageal reflux disease; NERD, nonerosive reflux disease; EE, erosive esophagitis; FC, functional constipation; IBS-C, constipation predominant-IBS.

SHORTCOMINGS OF ROME IV CRITERIA WHICH CAN AFFECT EPIDEMIOLOGIC STUDY RESULTS

During the period of development of the Rome criteria, due to a parallel increase in sophisticated investigations, specific diseases previously thought to be DGBIs due to similar symptoms have been identified (non-celiac wheat sensitivity, bile acid diarrhea, microscopic colitis, pancreatic exocrine deficiency, SIBO, GI motility disorders), which has put a question mark on the extent of investigations (see above) required to exclude organic diseases for defining DGBIs. GI motility disorders and DGBIs are intertwined with common pathophysiologic mechanisms, which may make their absolute separation well-nigh impossible, particularly in the initial stages.⁸⁰ Though these disorders are rare enough for any large overestimation, they can undermine precise measurement of prevalence in some populations.

Classification into mutually exclusive categories based on anatomic areas is useful for performing epidemiological studies in homogenous populations but disregard subjects with overlapping DGBIs, those with shifting categories or those with a not sufficiently standardized symptom presentation e.g. a bowel symptom cluster with meal trigger and a gas cluster was found to be common in Asian population.⁸¹ A categorical diagnosis may exclude patients who do not fully meet these criteria but who could be treated similarly. Emerging data on the overlap between different DGBIs (see above) are being addressed, which will be especially important for studying mechanisms and deciding on the treatment strategies.

Psychological factors are not required for diagnosis by Rome IV criteria (though this was partially addressed by introduction of a multidimensional clinical profile), but these are integral part of symptom development, its severity and consultation behavior in patients with DGBIs. In a similar vein introduction of centrally mediated disorders of gastrointestinal pain as a separate category when most of the DGBIs have a strong central component may be somewhat repetitive. Many studies have clearly shown that the burden of psychiatric disorders increases with the number of DGBIs present or in their overlap. Whether the inclusion of psychologic criteria in symptom clusters can better define a DGBI is being explored.^29,41

Threshold changes for diagnostic criteria or addition, or deletion of symptoms thereof which have similar pathophysiological basis leads to wide variation in symptom-based definitions of DGBI in Manning, Rome I, II, III, IV criteria. Studies using these criteria report a widely variable prevalence of DGBIs and are not comparable, so definite conclusions from their similarities and differences become uncertain. This is shown in studies of IBS-C/FC or PDS/EPS overlap as mentioned above.

Differences in the study population regarding sociocultural and psychological issues, symptom perception and reporting, religious and illness beliefs, dietary factors, socio-demographic factors (prevalence of GI infection, infestation [including H. pylori, PI-IBS] and cancer, differences in bowel habit, stool frequency and form leading to different interpretation of diarrhea, constipation) may not be fully addressed by Rome IV criteria, which affect estimates of global epidemiology. Sixty percent of the world's population lives in Asia. Compared to Europe and America, they are much more diverse in socioculture, religion, race, and ethnicity with unique ethnocultural and religious beliefs and practices, culinary patterns, illness beliefs and attitudes, and gut microbiome. A large part of such population lives in rural areas (of which a good percentage are tribals) and are handicapped in socioeconomic status, clean living, access to healthcare and means of livelihood (most are agro-based with exposure to infection and infestation) compared to their more industrialized and urban European and American counterparts. These factors affect the epidemiology of DGBIs and may necessitate a different approach to diagnosis and treatment in different areas of the world.⁸²

The Rome Foundation Global Epidemiological study was conducted via internet survey in the majority of the countries and also a door-to-door survey in some countries where it tried to address these issues partly by questionnaires formulated in some local languages. While this highly commendable herculean task has brought out important epidemiological landmarks in DGBIs worldwide, there are chances of bias arising from the survey method and study instrument. A population-based house-to-house interview may give different point prevalence and epidemiology than internet-delivered questionnaire as the latter may be prone to recruitment bias. This is well reflected in the above study where door-to-door surveys gave much lower disease prevalence than internet surveys of all DGBIs. To address some of the above issues Rome IV stresses the Multidimensional Clinical Profile of DGBIs (especially the patient perspective and Multicultural Aspects of DGBIs) with individualized treatment based on identifying and integrating the multiple components (psychosocial, clinical, physiological, quality of life, and impact aspects) of the symptom experience. However, for a proper study of epidemiology, the above factors may need more attention.

Several microorganic factors are associated with DGBI. For example, gut microbiota is an important factor in pathophysiology and epidemiology of DGBIs. Several studies showed an association between gut microbiota dysbiosis including qualitative and quantitative changes in gut microbiota (referred to as SIBO) and DGBIs, particularly IBS.^47,49,83,84 Should these patients then be called as IBS or should their diagnostic label be changed to SIBO presenting as IBS? For example, according to the Rome IV algorithm for diagnosis of FD, H. pylori infection needs to be ruled out and if present, must be eradicated. If the symptoms persist after eradication of H. pylori, then only they will get a diagnostic label of FD. Why should a similar approach not be followed for SIBO?

CONCLUSION

Though the Rome IV criteria are presently used to classify and positively diagnose DGBIs, they undermine the complex psychobiologic pathophysiology of visceral nociception and the variation of pain expression and bowel habit across cultures. They fail to address overlapping DGBIs, nonstandardized symptom categories, variation in symptom reporting and expression across cultures, or a change or shift in DGBI category over time, all of which may affect the epidemiology. The microorganic natures of DGBIs are unaddressed. Though these are excellent for research studies and therapeutic trials, their further improvement is needed for wider clinical applicability.