Search in the Abstract Database

Search Abstracts 2015

* = Presenting author

P090 Evaluation of NT-proBNP in Inflammatory Bowel Disease

A. Wacrenier*1, A.-M. Bourgeois2, J. Ausseil2, C. Szymanski3, H. Thiebault1, F. Brazier1, A. Galmiche2, E. Nguyen-Khac1, J.-L. Dupas1, J.-f. Colombel4, M. Fumery1

1Amiens University Hospital, Hepatogastroenterology, Amiens, France, 2Amiens University Hospital, Biochimie, Amiens, France, 3Amiens University Hospital, Cardiology, Amiens, France, 4Icahn School of Medicine at Mount Sinai Hospital, Gastroenterology, New York, NY, United States


N-Terminal pro-Brain Natriuretic peptide (NT-proBNP) is currently used as a diagnostic marker in heart failure. Elevated NT-proBNP and significant correlation between NT-proBNP and CRP was recently reported in rheumatoid arthritis (1) and in Crohn's Disease (CD) (2). The purpose of this study was to evaluate NT-proBNP as a marker of inflammatory bowel disease (IBD) activity.


In this single-center study, NTproBNP assay was performed in patients with CD or ulcerative colitis (UC), irritable bowel syndrome (IBS) and in healthy controls. Clinical characteristics (Harvey-Bradshaw score (HB), Mayo Clinic score (Mayo)), biological (CRP) and endoscopic (CDEIS, UCEIS) data were collected at the time of NTproBNP dosage. Echocardiography was performed in case of NT-proBNP serum levels> 124pg/ml.


To date, 118 patients with CD, 26 with UC, 20 with IBS and 18 healthy controls were included. The median serum NT-proBNP was 50.5 pg/ml [Q1=28-Q3=94] in CD group, 72 pg/ml [46-138] in the UC group, 43.5 pg/ml [27-111] in IBS Group and 34,5pg/ml [26-49] in the control group. A significant increase of NT-proBNP levels was observed in CD patients with HB>4 (p=0.02) or UC with Mayo>2 (p=0.03) as compared to inactive patients. There was a significant correlation between NTproBNP and, the HB score (r=0.31, p=0.0005), and the Mayo score (r=0.45, p=0.01) respectively. A significant increase in NT-proBNP was observed in CD (p<0.001) and UC (p< 0.0016) patients having CRP >10 mg/l as compared to patients with CRP<10. There was a significant correlation between NT-proBNP and CRP in both CD (r=0.64, p<0.001) and UC (r=0.8, p<0.001). Among the 34 patients who had a colonoscopy a significant correlation between NT-proBNP and, CDEIS (r=0.85, p< 0.001), and UCEIS (r=0.78, p=0.007) was observed respectively. A NTproBNP level of 57pg/ml had a sensitivity (Se) of 70% and a specificity (Sp) of 63% (AUC 0.73, 95%CI [0.63-0.83]) to predict a CRP >10 mg/l in CD. In UC, a rate of NTproBNP of 75 pg/ml had a Se of 88% and a Sp of 75% (AUC 0.87, 95%CI [0.71-1]) to predict a CRP>10 mg/l. All echocardiograms performed were normal.


NT-proBNP is correlated to the clinical, biological and endoscopic activity of CD and UC and may be a new biomarker in IBD.

(1) Provan SA et al. Arthritis Res Ther 2008

(2) Colombel JF et al. ECCO 2012.