P288 Gut barrier failure biomarkers in IBD: Is there anything new beyond „The Wall’?

N. Sipeki MD1, P. Kovats1, B. Balogh1, Z. Shums2, G.L. Norman2, P. Antal-Szalmas3, M. Papp1

1Division of Gastroenterology, Department of Internal Medicine, University of Debrecen, Debrecen, Hungary, 2Inova Diagnostics, Inc., San Diego, USA, 3Department of Laboratory Medicine, University of Debrecen, Debrecen, Hungary


Several defects in the many specialised components of mucosal barrier have been reported in inflammatory bowel disease (IBD). These alterations may represent a primary dysfunction in Crohn’s disease(CD), but they may also perpetuate chronic mucosal inflammation in ulcerative colitis(UC). Changes in intestinal permeability can predict IBD course.


We aim to determine the predictive potential of a panel of serological markers that reflect either mechanical or immunological gut barrier dysfunction regarding determination of disease phenotype, therapeutic strategy and long-term disease course in a prospective referral adult IBD patient cohort.

Sera of 266 CD (m/f:112/154, median age:25 years, B1:80.1%, P1:18.0%) and 187 UC (m/f:86/101, median age:33 years, extensive colitis:28.3%) patients were assayed for intestinal fatty acid-binding protein(I-FABP) and various immunoglobulin A (IgA) molecules, anti-F-actin[AAA IgA/IgG] and anti-gliadin[AGA IgA/IgG]) by enzyme-linked immunosorbent assay (ELISA) along with 155 healthy controls (HCONT). Clinical data were available on unfavourable disease outcome as well as disease activity and medical treatment during the prospective follow-up (median: 143 and 135 mths for CD and UC respectively).


In UC, median I-FABP level was significantly lower than in HCONT(177.2 vs. 244.3 pg/ml; p < 0.05). sIgA level was higher in both CD and UC compared with HCONT(51.2 and 46.0 vs. 29.2 μg/ml; p < 0.0001). AAA positivity with IgA type predominance was significantly higher in CD(40.2vs UC:15.7; HCONT:6.2%). AGA was also more prevalent in CD(16.5vs UC:6.7; HCONT:7.2%). There was an association between the presence of IgA type AAA or AGA and antimicrobial antibodies. ACA IgA was also more prevalent in case of AAA IgA positivity(52.9 vs. 32.7%, p = 0.009). PS/PT IgA positivity was higher in AGA IgA positive patients (33.3 vs. 7.6%, p = 0.001). Complicated disease behaviour at sample procurement was associated with the presence of AAA and AGA IgA positivity. In Kaplan–Meier analysis concomitant presence of IgA and IgG type AAA was associated with a shorter time to resective surgery (HR[95% CI]: 2.03[1.06–3.90]) along with a higher risk of a second surgery needed (HR[95% CI]: 2.82[1.14–6.98]). Sensitivity analysis showed that the later remained significant in B1 patients (p = 0.002) and also remained independent predictor in multivariate Cox-regression analysis comprising relevant clinical factors(HR[95% CI]: 2.88[1.07–7.78], p = 0.037).


The presence of AAA and AGA reflects the ongoing mucosal damage in IBD rather than has a value in predicting the disease course.