[1435] Pathology of Pancreas in Inflammatory Bowel Disease: An Etiology of Idiopathic Ductocentric Chronic Pancreatitis?

M Mino-Kenudson, TC Smyrk, K Notohara. Massachusetts General Hospital, Boston, MA; Mayo Clinic, Rochester, MN; Kurashiki Central Hospital, Kurashiki, Japan

Background: Autoimmune pancreatitis (AIP) is probably a heterogenous category, including (at least) lymphoplasmacytic sclerosing pancreatitis (LPSP) and idiopathic ductocentric chronic pancreatitis (IDCP). LPSP appears to represent the pancreatic manifestation of systemic IgG4-related fibrosclerosing disease; however, the nature of IDCP has not been well studied. IDCP has been reported in patients with inflammatory bowel disease (IBD). Thus, we reviewed resected pancreata of IBD patients to characterize the morphology, and to assess the possible link between IBD and IDCP.
Design: The study group consisted of 9 patients with IBD (7 ulcerative colitis [UC] and 2 Crohn's disease [CD]). Their resected pancreata were classified into IDCP, LPSP, and others. Study cases classified as IDCP were compared with IDCPs of non-IBD patients (n=21) for the following characteristic features of AIP: the extent of inflammation in interlobular and intralobular ducts, lobules and fibrosis, reactive epithelium of ducts, inflammatory pseudotumor-like changes, obliterative phlebitis, lymphoid aggregates, and an IgG4/IgG1 ratio. The demographic data were also compared between the 2 groups.
Results: In 6 of 7 UC cases, the pancreatic pathology was classified as IDCP. LPSP was seen in the remaining UC patient who had undergone ileocolectomy 7 years prior to Whipple resection. One CD patient showed patchy neutrophilic ductal inflammation with diffuse lobular atrophy and perivenular granulomatous inflammation, recapitulating her colonic morphology, and the other, periductal lymphocytic inflammation with mild, diffuse lobular atrophy and no granulomas. When compared with IDCPs of non-IBD patients (M:F = 11:10, age: 15 - 73 years [mean: 45 years]), IDCPs of the 6 UC patients (M:F = 3:3, age: 43 - 80 years [mean: 57 years]) exhibited more prominent eosinophilic infiltrate (p = 0.023) and a markedly smaller IgG4/IgG1 ratio (mean 0.09 opposed to 0.31 of the non-IBD group, p = 0.001). There was no difference in other clinicopathological features between the 2 groups.
Conclusions: In this study, although limited by a small sample size, pancreatic manifestations in UC are often classified as IDCP. When compared to non-IBD cases, UC-related IDCPs are associated with more prominent eosinophilic infiltrate and a markedly smaller IgG4/IgG1 ratio. The results may support the hypothesis that IDCP represents the morphology of a heterogeneous group that includes UC.
Category: Liver & Pancreas

Tuesday, March 10, 2009 1:00 PM

Platform Session: Section E, Tuesday Afternoon