Faecal calprotectin increases the diagnostic yield in patients with suspected small bowel disease – a multicenter retrospective cohort study

Abstract

Diagnosing small bowel pathology is challenging, and the diagnostic yield of small bowel capsule endoscopy is highly variable. Faecal calprotectin is a non-invasive intestinal inflammation marker that could be used as a selection tool to identify patients who might benefit from small bowel capsule endoscopy and increase its diagnostic yield. This study aimed to investigate the value of faecal calprotectin in detecting small bowel lesions in an unselected patient population. We performed a retrospective analysis of consecutive patients who underwent small bowel capsule endoscopy at the University Hospital Basel and the University Medical Clinic Cantonal Hospital Baselland, Switzerland, between 2010 and 2018. Patients without faecal calprotectin testing were excluded from the analysis. The primary endpoint was the presence of a clinically significant small bowel finding. Patients with positive faecal calprotectin results were more likely to have small bowel findings (66.7% vs 39.4%;P= 0.007). The optimal faecal calprotectin cut-off to identify clinically significant small bowel lesions was 63 μg/g with 78.3% (95% confidence interval: 66.7-87.9) sensitivity, 47.9% (33.3-62.8) specificity, 1.50 (1.1-2.0) positive and 0.45 (0.3-0.8) negative likelihood ratios, 68.4% (61.6-74.4) positive and 60.5% (47.3-72.4) negative predictive values, and 65.0% overall accuracy. Faecal calprotectin performed best in patients with suspected inflammatory bowel disease, with 66.7% (48.2-82.0) sensitivity, 70.6% (44.0-89.7) specificity, 2.27 (1.0-4.9) positive and 0.47 (0.3-0.8) negative likelihood ratios, 81.5% (67.0-90.5) positive and 54.5% (39.7-68.6) negative predictive values, and 71.4% overall accuracy. Faecal calprotectin testing increases the diagnostic yield of small bowel capsule endoscopy and may help identify patients at risk of small bowel disease.