Identification and validation of functional gastrointestinal disorder subtypes using latent class analysis: a population-based study

Pain is a cardinal symptom cutting across Rome IV anatomical categories in disorders of gut-brain interaction: A network-based approach

INTRODUCTION

Disorders of gut-brain interaction (DGBI) are symptom-based disorders categorized by anatomic location but have high overlap and heterogeneity. Viewing DGBI symptoms on a spectrum (i.e. dimensionally) rather than categorically may better inform interventions to accommodate complex clinical presentations. We aimed to evaluate symptom networks to identify how DGBI symptoms interact.

METHODS

We used the Rome IV Diagnostic Questionnaire continuously/ordinally scored items collected from the Rome Foundation Global Epidemiology Study. We excluded participants who reported ≥1 organic/structural gastrointestinal disorder(s). We sought to (1) identify core symptoms in the DGBI symptom networks, (2) identify bridge pathways between Rome IV diagnostic categories (esophageal, bowel, gastroduodenal, anorectal), and (3) explore how symptoms group together into communities.

RESULTS

Of 54,127 adults, 20,229 met criteria for at least one DGBI (age mean = 42.2 ± 15.5; 57% female). General abdominal pain and epigastric pain were the core symptoms in the DGBI symptom network (i.e., had the strongest connections to other symptoms). Pain symptoms emerged as bridge pathways across existing DGBI diagnostic anatomic location (i.e., abdominal pain connected to chest pain, epigastric pain, rectal pain). Without a priori category definitions, exploratory network community analysis showed that symptoms grouped together into "pain," "gastroduodenal," and "constipation," rather than into groups by anatomic location.

CONCLUSION

Our findings suggest pain symptoms are central and serve as a key connection to other symptoms, crosscutting anatomic location. Future longitudinal research is needed to test symptom network relations longitudinally and investigate whether targeting pain symptoms (rather than anatomic- or disorder-specific symptoms) has clinical impact.

Unsupervised machine learning highlights the challenges of subtyping disorders of gut-brain interaction

BACKGROUND

Unsupervised machine learning describes a collection of powerful techniques that seek to identify hidden patterns in unlabeled data. These techniques can be broadly categorized into dimension reduction, which transforms and combines the original set of measurements to simplify data, and cluster analysis, which seeks to group subjects based on some measure of similarity. Unsupervised machine learning can be used to explore alternative subtyping of disorders of gut-brain interaction (DGBI) compared to the existing gastrointestinal symptom-based definitions of Rome IV.

PURPOSE

This present review aims to familiarize the reader with fundamental concepts of unsupervised machine learning using accessible definitions and provide a critical summary of their application to the evaluation of DGBI subtyping. By considering the overlap between Rome IV clinical definitions and identified clusters, along with clinical and physiological insights, this paper speculates on the possible implications for DGBI. Also considered are algorithmic developments in the unsupervised machine learning community that may help leverage increasingly available omics data to explore biologically informed definitions. Unsupervised machine learning challenges the modern subtyping of DGBI and, with the necessary clinical validation, has the potential to enhance future iterations of the Rome criteria to identify more homogeneous, diagnosable, and treatable patient populations.

Identifying Novel Subtypes of Functional Gastrointestinal Disorder by Analyzing Nonlinear Structure in Integrative Biopsychosocial Questionnaire Data

Background/Objectives: Given the limited success in treating functional gastrointestinal disorders (FGIDs) through conventional methods, there is a pressing need for tailored treatments that account for the heterogeneity and biopsychosocial factors associated with FGIDs. Here, we considered the potential of novel subtypes of FGIDs based on biopsychosocial information. Methods: We collected data from 198 FGID patients utilizing an integrative approach that included the traditional Korean medicine diagnosis questionnaire for digestive symptoms (KM), as well as the 36-item Short Form Health Survey (SF-36), alongside the conventional Rome-criteria-based Korean Bowel Disease Questionnaire (K-BDQ). Multivariate analyses were conducted to assess whether KM or SF-36 provided additional information beyond the K-BDQ and its statistical relevance to symptom severity. Questions related to symptom severity were selected using an extremely randomized trees (ERT) regressor to develop an integrative questionnaire. For the identification of novel subtypes, Uniform Manifold Approximation and Projection and spectral clustering were used for nonlinear dimensionality reduction and clustering, respectively. The validity of the clusters was assessed using certain metrics, such as trustworthiness, silhouette coefficient, and accordance rate. An ERT classifier was employed to further validate the clustered result. Results: The multivariate analyses revealed that SF-36 and KM supplemented the psychosocial aspects lacking in K-BDQ. Through the application of nonlinear clustering using the integrative questionnaire data, four subtypes of FGID were identified: mild, severe, mind-symptom predominance, and body-symptom predominance. Conclusions: The identification of these subtypes offers a framework for personalized treatment strategies, thus potentially enhancing therapeutic outcomes by tailoring interventions to the unique biopsychosocial profiles of FGID patients.

Latent profile analysis reveals the central role of psychological symptoms in driving disease severity in chronic constipation

BACKGROUND

Chronic constipation (CC) is defined by symptom criteria reflecting heterogenous physiology. However, many patients with CC have significant psychological comorbidities-an alternative definition using a biopsychosocial classification model could be warranted to inform future treatments. We sought to: (1) empirically derive psychological symptom profiles of patients with CC using latent profile analysis and (2) validate these profiles by comparing them on symptom severity, GI-specific anxiety, body mass index (BMI), and anorectal manometry findings.

METHODS

Participants included adults presenting for anorectal manometry for CC (N = 468, 82% female, Mage = 47). Depression/anxiety symptoms and eating disorder (ED) symptoms (EAT-26) were used as indicators (i.e., variables used to derive profiles) representing unique psychological constructs. Constipation symptoms, GI-specific anxiety, BMI, and anorectal manometry results were used as validators (i.e., variables used to examine the clinical utility of the resulting profiles).

KEY RESULTS

A 5-profile solution provided the best statistical fit, comprising the following latent profiles (LPs): LP1 termed "high dieting, low bulimia;" LP2 termed "high ED symptoms;" LP3 termed "moderate ED symptoms;" LP4 termed "high anxiety and depression, low ED symptoms;" and LP5 termed "low psychological symptoms." The low psychological symptom profile (61% of the sample) had lower abdominal and overall constipation severity and lower GI-specific anxiety compared to the four profiles characterized by higher psychological symptoms (of any type). Profiles did not significantly differ on BMI or anorectal manometry results.

CONCLUSIONS AND INFERENCES

Profiles with high psychological symptoms had increased constipation symptom severity and GI-specific anxiety in adults with CC. Future research should test whether these profiles predict differential treatment outcomes.

Re-investigation of functional gastrointestinal disorders utilizing a machine learning approach

BACKGROUND

Functional gastrointestinal disorders (FGIDs), as a group of syndromes with no identified structural or pathophysiological biomarkers, are currently classified by Rome criteria based on gastrointestinal symptoms (GI). However, the high overlap among FGIDs in patients makes treatment and identifying underlying mechanisms challenging. Furthermore, disregarding psychological factors in the current classification, despite their approved relationship with GI symptoms, underlines the necessity of more investigation into grouping FGID patients. We aimed to provide more homogenous and well-separated clusters based on both GI and psychological characteristics for patients with FGIDs using an unsupervised machine learning algorithm.

METHODS

Based on a cross-sectional study, 3765 (79%) patients with at least one FGID were included in the current study. In the first step, the clustering utilizing a machine learning algorithm was merely executed based on GI symptoms. In the second step, considering the previous step's results and focusing on the clusters with a diverse combination of GI symptoms, the clustering was re-conducted based on both GI symptoms and psychological factors.

RESULTS

The first phase clustering of all participants based on GI symptoms resulted in the formation of pure and non-pure clusters. Pure clusters exactly illustrated the properties of most pure Rome syndromes. Re-clustering the members of the non-pure clusters based on GI and psychological factors (i.e., the second clustering step) resulted in eight new clusters, indicating the dominance of multiple factors but well-discriminated from other clusters. The results of the second step especially highlight the impact of psychological factors in grouping FGIDs.

CONCLUSIONS

In the current study, the existence of Rome disorders, which were previously defined by expert opinion-based consensus, was approved, and, eight new clusters with multiple dominant symptoms based on GI and psychological factors were also introduced. The more homogeneous clusters of patients could lead to the design of more precise clinical experiments and further targeted patient care.

Latent tree analysis for the identification and differentiation of evidence-based Traditional Chinese Medicine diagnostic patterns: A primer for clinicians

BACKGROUND

A supplementary chapter on the diagnostic patterns of Traditional Medicine, including Traditional Chinese Medicine (TCM), was introduced into the latest edition of the International Classification of Diseases (ICD-11). However, evidence-based rules are yet to be developed for pattern differentiation in patients with specific conventional medicine diagnoses. Without such standardised rules, the level of diagnostic agreement amongst practitioners is unsatisfactory. This may reduce the reliability of practice and the generalisability of clinical research.

PURPOSE

Using cross-sectional study data from patients with functional dyspepsia, we reviewed and illustrated a quantitative approach that combines TCM expertise and computer algorithmic capacity, namely latent tree analysis (LTA), to establish score-based pattern differentiation rules.

REVIEW OF METHODS

LTA consists of six major steps: (i) the development of a TCM clinical feature questionnaire; (ii) statistical pattern discovery; (iii) statistical pattern interpretation; (iv) TCM diagnostic pattern identification; (v) TCM diagnostic pattern quantification; and (vi) TCM diagnostic pattern differentiation. Step (i) involves the development of a comprehensive questionnaire covering all essential TCM clinical features of the disease of interest via a systematic review. Step (ii) to (iv) required input from TCM experts, with the algorithmic capacity provided by Lantern, a dedicated software for TCM LTA.

MOTIVATIONAL EXAMPLE TO ILLUSTRATE THE METHODS

LTA is used to quantify the diagnostic importance of various clinical features in each TCM diagnostic pattern in terms of mutual information and cumulative information coverage. LTA is also capable of deriving score-based differentiation rules for each TCM diagnostic pattern, with each clinical feature being provided with a numerical score for its presence. Subsequently, a summative threshold is generated to allow pattern differentiation. If the total score of a patient exceeded the threshold, the patient was diagnosed with that particular TCM diagnostic pattern.

CONCLUSIONS

LTA is a quantitative approach to improving the inter-rater reliability of TCM diagnosis and addressing the current lack of objectivity in the ICD-11. Future research should focus on how diagnostic information should be coupled with effectiveness evidence derived from network meta-analysis. This will enable the development of an implementable diagnostics-to-treatment scheme for further evaluation. If successful, this scheme will transform TCM practice in an evidence-based manner, while preserving the validity of the model.

Characteristics of patients with overlap functional gastrointestinal disorders

BACKGROUND AND AIM

Functional gastrointestinal disorders (FGIDs) are frequently overlapped. The present study was designed to (i) search the clinical differences between patients with single FGID and overlap FGIDs and (ii) define the most common FGIDs associations to identify homogenous subgroups of patients.

METHODS

A total of 3555 outpatients with FGID filled out the Rome III adult diagnostic questionnaire, Bristol stool form, and four 10-point Likert scales to report the severity of constipation, diarrhea, bloating, and abdominal pain. An unsupervised algorithm was used to estimate the number of groups directly from the data. A classification tree separated patients into different subgroups, according to FGIDs. Multinomial logistic regression was used to characterize the groups of patients with overlap disorders.

RESULTS

Patients reported 3.3 ± 1.9 FGIDs (range 1-10, median = 3); 736 reported only one FGID, while 2819 reported more than one FGID (3.8 ± 1.7). Patients with single FGID had higher body mass index (P < 0.001), never report irritable bowel syndrome (IBS), and rarely report fecal incontinence and anorectal pain (< 1% for each disorder). The non-supervised clustering of the 2819 patients with overlap FGIDs divided this population into 23 groups, including five groups associated with only one disorder (IBS-diarrhea, dysphagia, functional constipation, levator ani syndrome, and IBS-unspecified). Ten groups were related to two overlap disorders and eight groups to three or more disorders. Three disorders were not explicitly associated with a given group: IBS-mixed, proctalgia fugax, and nonspecific anorectal pain.

CONCLUSION

Patients with FGID mostly report overlap disorders in a limited number of associations, each significantly associated with a few disorders.

Exhaled Hydrogen as a Marker of Intestinal Fermentation Is Associated with Diarrhea in Kidney Transplant Recipients

BACKGROUND

Diarrhea is common among kidney transplant recipients (KTR). Exhaled hydrogen (H2) is a surrogate marker of small bowel dysbiosis, which may drive diarrhea. We studied the relationship between exhaled H2 and diarrhea in KTR, and explored potential clinical and dietary determinants.

METHODS

Clinical, laboratory, and dietary data were analyzed from 424 KTR participating in the TransplantLines Biobank and Cohort Study (NCT03272841). Fasting exhaled H2 concentration was measured using a model DP Quintron Gas Chromatograph. Diarrhea was defined as fast transit time (types 6 and 7 according to the Bristol Stool Form Scale, BSFS) of 3 or more episodes per day. We studied the association between exhaled H2 and diarrhea with multivariable logistic regression analysis, and explored potential determinants using linear regression.

RESULTS

KTR (55.4 ± 13.2 years, 60.8% male, mean eGFR 49.8 ± 19.1 mL/min/1.73 m2) had a median exhaled H2 of 11 (5.0-25.0) ppm. Signs of small intestinal bacterial overgrowth (exhaled H2 ≥ 20 ppm) were present in 31.6% of the KTR, and 33.0% had diarrhea. Exhaled H2 was associated with an increased risk of diarrhea (odds ratio 1.51, 95% confidence interval 1.07-2.14 per log2 ppm, p = 0.02). Polysaccharide intake was independently associated with higher H2 (std. β 0.24, p = 0.01), and a trend for an association with proton-pump inhibitor use was observed (std. β 0.16 p = 0.05).

CONCLUSION

Higher exhaled H2 is associated with an increased risk of diarrhea in KTR. Our findings set the stage for further studies investigating the relationship between dietary factors, small bowel dysbiosis, and diarrhea after kidney transplantation.

Symptom-based latent classes of persons with irritable bowel syndrome

A large amount of interindividual variability exists in symptom experiences of persons with irritable bowel syndrome (IBS). Thus, consideration of multiple symptoms to identify distinct symptom subgroups may be useful in directing personalized health strategies for symptom management. We aimed to identify latent classes (i.e., subgroups) of persons with IBS who share similar patterns of symptoms using symptom-related variables (six groups of daily diary symptoms, cognitive beliefs about IBS, and IBS quality of life [QOL]); and to examine how subgroups differed in patient characteristics. Data were derived from a baseline assessment of men and women enrolled in two cognitively-focused intervention trials (N = 332). Using latent class analysis, four latent classes were identified: Class 1 (low symptoms and good QOL, n = 153), Class 2 (low symptoms and moderate QOL, n = 106), Class 3 (high symptoms with diarrhea and poor QOL, n = 38), and Class 4 (high symptoms with low diarrhea and moderate QOL, n = 35). Diarrhea, being female, less formal education, unemployment, and previous history of major depressive disorder were associated with membership in Class 3. Using these distinct symptom profiles, the next step is to explore underlying mechanisms accounting for symptom burden with the goal of designing tailored interventions to reduce that burden.