Search in the Abstract Database

Search Abstracts 2016

* = Presenting author

P256 Autoimmune pancreatitis associated with inflammatory bowel disease: a multicentric study of the GETAID

D. Lorenzo*1, F. Maire2, C. Stefanescu2, J. M. Gornet3, P. Seksik4, J. C. Grimaud5, B. Bournet6, P. Marteau4, A. Amiot7, D. Laharie8, C. Trang9, B. Coffin10, G. Bellaiche11, G. Cadiot12, C. Reenaers13, A. Racine14, S. Viennot15, A. Pauwels16, G. Bouguen17, G. Savoye18, A. L. Pelletier19, G. Pineton de Chambrun20, P. Lahmek21, S. Nahon22, V. Abitbol23

1Hôpital Cochin AP-HP, Paris, France, 2CHU Beaujon, Clichy, France, 3Hôpital Saint Louis, Paris, France, 4Hôpital Saint-Antoine, Paris, France, 5Hôpital Nord, Marseille, France, 6CHU Toulouse, Toulouse, France, 7CHU Henri Mondor, Créteil, France, 8Hôpital Haut Lévêque, Bordeaux, France, 9CHU Nantes, Nantes, France, 10CHU Louis Mourier, Colombes, France, 11CH d’Aulnay, Aulnay, France, 12CHU Reims, Reims, France, 13CH Liège, Liège, Belgium, 14CH Kremlin Bicêtre, Kremlin Bicêtre, France, 15CHU Caen, Caen, France, 16CH Gonesse, Gonesse, France, 17CHU Rennes, Rennes, France, 18CHU Rouen, Rouen, France, 19Hôpital Bichat, Paris, France, 20CHU Montpellier, Montpellier, France, 21Hôpital Emile Roux, Gastroenterology Addictology, Limeil-Brevannes, France, 22Centre Hospitalier Le Raincy, Gastroenterology, Montfermeil, France, 23Hôpital Cochin AP-HP, Gastroenterology, Paris, France

Background

Autoimmune pancreatitis (AIP) associated with inflammatory bowel disease (IBD) is little known. Our aims were to describe AIP associated with IBD and to determine whether the AIP has an effect on the natural history of IBD.

Methods

From July 2012 to July 2015, patients with both AIP and IBD were included in a retrospective database of the GETAID. All files were reviewed. AIP diagnosis was definitive (histology) or probable (imaging, corticosteroids, IgG4, and other organ) according to the international diagnostic criteria. Other causes of pancreatitis were excluded. Characteristics of the IBD (type, extent, phenotype, and treatments) and the AIP (type, treatment, and evolution) were collected using a standardised questionnaire (FileMaker Pro V.12®). Results were reported as mean ± SD

Results

Out of 114 cases, 91 were included in 23 centres (47 women, 33 Crohn’s disease [CD] [36%], 58 ulcerative colitis [UC] [64%]). Age at IBD diagnosis was 32 ± 12. At inclusion, IBD and AIP durations were 8.3 ± 7 and 5.7 ± 4.9 years, respectively. There were 27% cases of extraintestinal manifestations. AIP diagnosis was definitive in 14(16%) and probable in 77(84%). AIP preceded IBD in 19(20%), was synchronous in 23(26%) and occurred after IBD in 49(54%). AIP was type 2 in 98%. Initial manifestation was acute pancreatitis (80%), abdominal pain (11%), jaundice (7%), or incidental diagnosis (2%). No pancreatitis was severe. Cholangitis was associated in 18 (20%). AIP was treated with corticosteroids in 44 cases with 100% clinical responses. Corticodependence was observed in 9. Five patients underwent pancreatic surgery for suspicion of cancer. During follow-up, AIP relapsed in 31 (34%), 11 (12%) developed diabetes, and 17(19%) exocrine pancreatic insufficiency. No patient had pancreatic or colorectal cancer. UC extent was rectal, left sided or pancolitis in 20 (34%), 18 (32%), and 20 (34%), respectively. UC was mild to moderate in 33 (57%). UC was severe in 25 (43%): immunosuppressors (n = 5), anti-TNF (n = 4), combo therapy (n = 3), colectomy (n = 13) for acute severe colitis (n = 12), and non-adenoma dysplastic lesion. CD was ileal, ileocolonic, colonic, and anoperineal in 8 (22%), 13 (40%), 12 (38%), and 6 (6%), respectively. Phenotypes were inflammatory, stricturing, and penetrating in 29 (87%), 2 (6%), and 2 (6%), respectively. Treatments were immunosuppressors (n = 9), anti-TNF (n = 13), or combo therapy (n = 2). Eight patients had at least 1 bowel surgery

Conclusion

This study is the largest reported series to date of patients with both AIP and IBD. AIP is mostly type 2, preceding IBD in 21%. Two thirds of patients have UC, often distal. One third has CD, often colonic with inflammatory phenotype. Long-term follow-up and a case-control study (in progress) would better characterise patients with both IBD and AIP.