Clinical Value of Fecal Calprotectin in Predicting Mucosal Healing in Patients With Ulcerative Colitis

Aim: This study aimed to evaluate the clinical significance of fecal calprotectin (FC) in assessment of ulcerative colitis (UC) patients' endoscopic patterns and clinical manifestation. Methods: A total of 143 UC patients who received colonoscopy and 108 controls were included. After providing stool samples, patients underwent total colonoscopy. FC was measured by an enzyme-linked immunosorbent assay (ELISA). Clinical activity was based on the Mayo score. Endoscopic findings was scored by the Ulcerative Colitis Endoscopic Index of Severity (UCEIS). Correlation analysis and receiver-operator characteristic (ROC) analysis were carried out to determine the significance of measurements. Results: The median (interquartile range, IQR) of FC levels was 211 (43-990) mu g/g in UC and 87.5 (40.50 similar to 181) mu g/g in the control group. Fecal calprotectin correlated significantly with both Mayo and UCEIS scores (Spearman's r 0.670 and 0.592, P < 0.01). With a cut-off value of 164 mu g/g for fecal calprotectin concentration, the area under the curve (AUC) in receiver operator characteristic analysis was 0.830, sensitivity was 85.42%, specificity was 73.68%, positive predictive value (PPV) was 62.12%, and negative predictive value (NPV) was 9.10% in predicting clinical active disease. Similarly, the power of FC to predict mucosal healing (MH) was modest. With a cut-off value of 154.5 mu g/g, the AUC was 0.839, sensitivity was 72.34%, and specificity was 85.71%. Conclusion: For evaluating the disease activity of UC, FC is a clinically relevant biomarker for both clinically active disease and MH in patients with UC. But the cut-off value still needs large and multicenter studies for confirmation.

Automated summary

This study demonstrated the clinical significance of fecal calprotectin in evaluating endoscopic patterns and clinical manifestation in ulcerative colitis patients. The cut-off value of 164 μg/g for FC concentration was found to predict clinical active disease with high sensitivity and specificity. However, further large-scale studies are needed to confirm the reliability of the cut-off value.