P588 Test characteristics of commonly used clinical and biochemical markers predicting mucosal healing and active inflammation in ulcerative colitis: faecal calprotectin is clearly superior
K. Carlsen*1, 2, R. Goll3, 4, H. Elsberg5, T. Thorkilgaard5, L. Maagaard5, S. S. Wergeland6, J. Florholmen4, C. Jakobsen2, V. Wewer2, L. B. Riis1, P. S. Munkholm3
1Herlev Hospital, Department of Pathology, Herlev, Denmark, 2Hvidovre Hospital, Department of Paediatrics, Hvidovre, Denmark, 3North Zealand Hospital, Department of Gastroenterology, Frederikssund, Denmark, 4University Hospital of North Norway, Department of Gastroenterology, Tromsoe, Norway, 5Herlev Hospital, Department of Gastroenterology, Herlev, Denmark, 6University Hospital of North Norway, Department of Clinical Pathology, Tromsoe, Norway
Mucosal healing has emerged as a key treatment goal in ulcerative colitis (UC); however, currently evaluation of mucosal healing requires endoscopic assessment, which limits its use in daily clinical practice. The aim of the current study was to calibrate clinical and biochemical parameters as predictors of mucosal healing and active inflammation defined by endoscopy and histology, to make a non-invasive prediction of mucosal inflammatory status possible.
In total, 108 UC patients (51 mild, 25 moderate, and 32 with severe activity) and 51 controls undergoing sigmoid-/colonoscopy were prospectively included. Two faecal calprotectin tests (FC1 and FC2 before bowel cleansing) with one to 7 days between them, blood samples (max. 5 days before procedure) and disease activity score (Simple Clinical Colitis Activity Index, SCCAI) were collected. Biopsies localised to the rectum and from any active inflammation foci underwent double blinded histopathological grading (Geboes score). The intestinal mucosa was macroscopically graded by the Mayo endoscopic score (MES). Receiver operating characteristics (ROC) analyses were used to evaluate the sensitivity and specificity of FC, SCCAI, and blood parameters (erythrocyte sedimentation rate [ESR], C-reactive protein [CRP], and orosomucoid [ORO]) in reference to mucosa healing and disease activity: mucosa healing, MES score = 0 and Geboes score = < 1, and inflammatory activity, MES = > 2 and Geboes score > 3.
Prediction of inflammatory activity: SCCAI showed moderate test characteristics, area under the curve (AUC) 0.68, p = 0.016; max Youden index 0.318 at cut-off SCCAI > 2.5 (sensitivity 0.72; specificity 0.60). For FC (mean of 2 samples), AUC was 0.77 p = 0.001; max Youden index 0.496 at cut-off > 230 mg/kg (sensitivity 0.73; specificity 0.76). Prediction of Mucosa healing: SCCAI had poor test characteristics, AUC 0.61 p = 0.076. For FC (mean of 2 samples), AUC 0.83 p < 0.0005; max Youden index 0.492 at cut-off < 54 mg/kg (sensitivity 0.64; specificity 0.86). ESR, CRP, and ORO all performed poorly in predicting both remission and active inflammation, showing non-significant AUC in ROC analyses.
Faecal calprotectin (mean of 2 samples) outperforms ESR, CRP, and ORO clearly in prediction of both mucosal healing and inflammatory activity. SCCAI can predict inflammatory activity, but it performs poorly in prediction of mucosal healing.