Feb. 11, 2023

Irritable bowel syndrome (IBS) is currently the most commonly diagnosed of the recognized functional gastrointestinal disorders affecting adults. Last updated in 2016, the Rome diagnostic criteria define IBS as a disorder of gut-brain interaction characterized by chronic and frequent abdominal pain associated with defecation, and changes in the consistency or frequency of bowel movements.

Research has demonstrated that factors such as diet and lifestyle and psychiatric, musculoskeletal and neurological comorbidities can influence the clinical course and outcomes associated with IBS. However, the diagnostic classification approach outlined in the Rome criteria is focused on predominant bowel habits. And there is a limited understanding about the impact of these nongastrointestinal factors.

To address this knowledge gap, Mayo Clinic researchers and colleagues conducted a survey-based case-control study. Their goal was to determine if patients with IBS cluster according to various parameters, including clinical, dietary, lifestyle and psychosocial factors. The results of this study were published in Clinical Gastroenterology and Hepatology in 2022.

According to Mayo Clinic gastroenterologist Madhusudan (Madhu) Grover, M.B.B.S., the corresponding author on the study publication, the research team hypothesized that patients with IBS can be segregated into subsets that reflect more than just the predominant bowel habit categories. Dr. Grover is a clinician and researcher at Mayo Clinic in Minnesota who studies the mechanisms involved in the pathophysiology of IBS.

"We suspected that by examining the detailed clinical metadata from a large cohort of patients with IBS, we would be able to discern unique subsets of IBS, in which patients have variable levels of associations with pain, bowel dysfunction, comorbidities, and diet and lifestyle changes," explains Dr. Grover.

Study methods

To test this hypothesis, Dr. Grover and colleagues accessed the Mayo Clinic Biobank and examined 40,291 patient responses to a questionnaire incorporating Rome III criteria that were received between 2013 and 2020. Using this survey data, they determined several factors associated with IBS. And they performed latent class analysis, a model-based clustering, on data from respondents diagnosed with IBS.

The IBS cohort consisted of 4,021 individuals with IBS (mean age of 64 years) and a control group of 12,063 individuals without IBS. Within the IBS cohort, 1,280 had diarrhea-predominant IBS (32%), 1,021 had constipation-predominant IBS (25%), 1,597 had mixed IBS (40%) and 81 had unsubtyped IBS (2%).

Study results

Using 26 variables to separate the IBS cohort from the control group, the optimal clustering model helped the researchers identify seven latent patient clusters:

• Cluster 1 and Cluster 2 were characterized by the level of perceived health impairment: Individuals in Cluster 1 (17% of the IBS cohort) had moderate impairment, and individuals in Cluster 2 (16% of the IBS cohort) had severe impairment. Individuals in both Cluster 1 and Cluster 2 demonstrated more pain. Individuals in Cluster 2 also had more psychiatric comorbidities.
• Cluster 3 (10% of the IBS cohort) was characterized by above-average scores for both psychiatric comorbidities and neurological comorbidities, including anxiety, depression, migraine, and any other mental health or neurological disorders.
• Cluster 4 (20% of the IBS cohort) was characterized by above-average scores for psychiatric comorbidities only.
• Cluster 5 (11% of the IBS cohort) was characterized by above-average scores for neurological comorbidities only.
• Cluster 6 (12% of the IBS cohort) and Cluster 7 (12% of the IBS cohort) were characterized by bowel dysfunction. Individuals in Cluster 6 had diarrhea predominance. Individuals in Cluster 7 had constipation predominance. Individuals in both Clusters 6 and 7 had lower scores for pain, perceived health impairment and other comorbidities. Individuals in Cluster 6 were most likely to report pain improvement with defecation. Individuals in Cluster 7 had the highest scores related to exercise and the consumption of fruits, vegetables and alcohol.

The distribution of clusters remained similar when Rome IV criteria were applied. Physiological tests were available on a limited subset (6% of the IBS cohort), and there were no significant differences between clusters for these variables.

Overall, Dr. Grover and colleagues note that these findings suggest that the IBS phenotype has multiple dimensions, and that these disorders have a unique pathophysiology, with many variables affecting the associated clinical course and outcomes.

"This new knowledge may be useful in helping us design clinical trials and personalized treatment approaches that target specific clusters of patients with IBS," explains Dr. Grover. "The insight provided by these study results will allow us to go beyond the current treatment approaches that just focus on predominant bowel patterns, and that will help us further individualize patient management."

For more information

Byale A, et al. High-dimensional clustering of 4000 irritable bowel syndrome patients reveals seven distinct disease subsets. Clinical Gastroenterology and Hepatology. In press.

Refer a patient to Mayo Clinic.