The Association of Clinical Symptom Clusters With Underlying Mechanisms in Functional Gastrointestinal Disorders

Functional gastrointestinal disorders (FGID) are amongst the most common causes of abdominal pain and dysfunction seen in clinical practice, affecting between 10 to 15% of most populations (1). FGID are defined by symptoms without demonstrable underlying organic pathology (2). Within the currently used Rome definitions of FGID, there is a broad range of gastrointestinal and multi-organ symptoms, indicating heterogeneous underlying pathophysiological mechanisms (3). There is evidence of central nervous system and motility dysfunction, dysbiosis, as well as immune activation in various subgroups of patients with FGID (2). Most mechanistic studies have been performed in small and heavily selected groups of patients. Consequently, the link between different symptomatic subgroups of patients and underlying mechanisms is unclear and unconfirmed in larger and representative patient cohorts. FGID patients with different underlying pathologies are likely to benefit from divergent specific treatments, even if they fall within the same Rome classification of FGID.

Discrete clusters of clinical characteristics in a large cohort of patients with FGID will be sought using hypothesis-free cluster analysis and latent-class analysis models. Associations to underlying mechanisms will be examined using data from fermentable sugar breath, blood and stool tests. This will allow recommendations regarding improved mechanistic-based classifications of patients with FGID, with potential for more effective mechanistic-based treatments.

The investigators will use coded clinical and medical history characteristics obtained by standardized questionnaires and laboratory and breath test results from all successive patients above the age of 18 years referred to the Gastroenterology Group Practice in the last 10 years for diagnosis and treatment of FGID for statistical analysis The data is stored in a database, without any personal identifiers. Explorative statistical analysis will be performed in approximately 5000 patients.