An investigation of the correlations between subjective and objective measures of bowel inflammation in Spondyloarthritis

Abstract

Introduction: Spondyloarthritis (SpA) and Inflammatory Bowel Disease (IBD) are related conditions of unknown aetiology demonstrating both common clinical features and common genetic and immunological pathomechanisms. Patients with SpA frequently exhibit intestinal inflammation and many develop significant gastrointestinal symptoms. Conversely, many patients with IBD develop an inflammatory arthritis. It has been proposed that an increase in intestinal permeability is an important mechanism in the aetiology of both conditions. However, to date the association between symptoms, intestinal pathology and altered gut permeability has been poorly elucidated.

Methods: Patients who fulfilled the Assessments in Spondyloarthritis Internation Society (ASAS) criteria for axial spondyloarthritis were recruited to the study. Gastrointestinal symptoms were assessed using the Dudley Inflammatory bowel symptom Disease Questionnaire (DISQ). Intestinal permeability was measured using the three sugars test, which measures the differential urine recovery of ingested sucralose, L-rhamnose, and lactulose. Small-intestinal lesions were assessed with wireless capsule endoscopy (WCE). An indirect measure of intestinal inflammation (faecal calprotectin) was also used. Drug therapy – including the use and dose of NSAIDs was recorded. Patients with SpA were assessed clinically including a measure of disease activity - the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and CRP.

Results: In total, 35 patients and 15 healthy controls completed the DISQ and underwent a sucralose, lactulose and L-rhamnose absorption test to assess intestinal permeability. Ten patients underwent WCE. The majority of patients were taking NSAIDs (25/35). DISQ scores were significantly increased in patients compared to controls (p<0.0001) and were significantly correlated with BASDAI scores (p<0.001). Intestinal permeability was not significantly different between patients and controls, and was not associated with DISQ or BASDAI scores. Faecal calprotectin results were high in some patients, but the correlation with DISQ scores was not significant (p=0.169) although a trend was apparent. WCE showed mild to severe ulcerations/erosions to be remarkably common (8 out of 9 complete studies). Macroscopic lesions of the duodenum appeared to be associated with bowel symptoms, while some severe lesions in the jejunum and ileum 3 were often asymptomatic.

Conclusions: We conclude that both gastrointestinal symptoms and intestinal lesions are common in SpA patients. The use of non-steroidal anti-inflammatory drugs appears to be a associated with intestinal symptoms in SpA patients, but ileocolonic ulceration is commonly asymptomatic. The DISQ appears to be a good screening tool for identifying patients with bowel symptoms which are more common with upper GI involvement and especially with lesions in the duodenum. Faecal calprotectin and WCE identified a high proportion of SpA patients as having asymptomatic lesions of the jejunum and ileum, common sites of inflammation in IBD.