15 p.y threshold, diagnosis of acute pancreatitis and ACP was made significantly earlier (33y and 36y versus 45 and 46y resp) (p=0.005 and 0.005 resp), irrespective of the amount of alcohol intake. Chronic pancreatic pain was also more frequent (p=0.05). At 20 p.y threshold, ACP and acute pancreatitis occurred earlier (p=0.0002 and <0.0001), and the pts hade more often, calcifications and ductal changes (p=0.05 and 0.005 resp) irrespective of alcohol intake. Similar results were observed at the 30 p.y threshold, but additionally pancreatic exocrine insufficiency occurred earlier (p=0.04). Conclusion- Tobacco intake accelerates the course of ACP in a dose-dependent fashion. ACP occurs earlier as soon as 15 p.y, and a major threshold effect is seen at 20 p.y, where the frequency of all major complications of ACP is increased. 721 Immunological Correlates of Disease Progression, Exocrine and Endocrine Failure in Patients With Chronic Pancreatitis: An Analysis of Pancreatic Juice Cytokines Bimaljit S. Sandhu, Sarah Sherman, Kshama R. Mehta, Arun J. Sanyal, Alvin M. Zfass, Doumit BouHaidar, Ravi Vachhani, Pankaj J. Pasricha BACKGROUND: Chronic pancreatitis (CP) is characterized by disabling pain and eventual progression to diabetes (DM), exocrine failure (EF), biliary stricture and in some cases pancreatic cancer. It is extremely difficult to get pancreatic tissue and currently there are no biomarkers to assess progression of CP. AIMS: 1. Perform cytokine analyses of the serum and pancreatic juice of subjects with CP and compare it to the controls. 2. Evaluate cytokines associated with progression of CP to DM and EF in serum and pancreatic juice. METHODS: A pilot study where the analyses of serum and pancreatic juice of subjects with established CP was performed and compared to controls (subjects getting esophago-gastro-duodenoscopy (EGD) for reflux symptoms and without pancreatitis). Clinical data including presence and onset of DM and/or EF, Cambridge stage, and type of pain (A or B) were recorded. Pancreatic juice was collected from the duodenum after intravenous human secretin (0.2 mcg/kg). EGD was performed and duodenal fluid collected over ice, aliquoted at 15, 30 and 45 minutes, immediately placed in liquid nitrogen and stored at -70o C. A Luminex-based assay of 12 cytokines in serum and pancreatic juice was performed(NIDDK P30 DK56339). These cytokines reflected the activity of the innate immune system (IL-1β, TNF-α, IL-6), T helper1 (TH1) cells (IFN-γ), TH2 cells (IL-4, IL-10, IL-13), and fibrogenic cytokines (TGF- β). Levels of cytokines in serum and pancreatic juice of patients with CP were compared to those of controls and with clinical features. RESULTS: 25 subjects with CP and 10 controls were enrolled. Levels of TNF-α and IL-1β were not significantly different in the serum of controls and patients with CP without pancreatic insufficiency. There was a step-wise increase in TNF-α and IL-1β in the pancreatic juice from controls to CP without EF to those with EF (p= 0.045). There were no significant differences in IFN-γ levels in serum. There was a significant increase in pancreatic juice IFN-γ in those with CP and DM (p= 0.043). IL-13 levels in the pancreatic juice were correlated with varying Cambridge stages of CP. There was a small but insignificant increase in IL-13 in mild and moderate Cambridge stage. However, severe Cambridge stage was associated with increased IL-13 in most subjects. CONCLUSIONS: Pancreatic juice cytokine levels serve as biomarkers correlating with pancre- atic exocrine and endocrine failure. Exocrine failure is associated with activation of the innate immune system. Endocrine failure is associated with TH1 cytokine profile in pancreatic juice. Severe Cambridge stage may be associated with increased TH2 activity (IL-13). 722 Long Term Outcome of Patients With Auto-Immune Pancreatitis Frédérique Maire, Yann Le Baleur, Vinciane Rebours, Marie-Pierre Vullierme, Anne Couvelard, Alain Sauvanet, Olivia Hentic, Gaëlle Buzaglo, Philippe Levy, Philippe B. Ruszniewski, Pascal Hammel Diagnostic criteria (HISORt) are well-defined in patients with auto-immune pancreatitis (AIP), but long term outcome of the disease are still poorly understood. Our aims were to analyse the short and long term outcome of pts with focus on pancreatic endocrine and exocrine functions, to search for predictive factors of relapse and pancreatic insufficiency and to compare HISORt + pts and pts with normal serum IgG4. Patients and methods : All consecutive pts followed in our centre between 1999 and 2008 were included. Two groups of pts were defined: HISORt+ pts, meeting HISORt criteria (histology, imaging, elevated serum IgG4, extra-pancreatic involvement and response to steroids), and HISORt - pts (typical imaging and response to steroids, with normal serum IgG4). AIP-related events and pancreatic exocrine/endocrine insufficiency were looked for during pt follow-up. Predictive factors of relapse and pancreatic insufficiency were analyzed. Results : 44 pts (22 male), median age 37.5 (19-73) years were included: 29 (66%) in HISORT+ group and 15 (34%) in HISORt - group. First-line treatment consisted of steroids or pancreatic resection (suspicion of pancreatic cancer), in 59% and 27% of the pts respectively. Median follow-up was 41 months [5-130]. Steroids were effective in all treated pts. Relapse was observed in 12 (27%) pts, after a median delay of 6 (1-70) months. Four pts received azathioprine due to steroid resistance/dependence at time of relapse. Elevated serum IgG4 and pain/jaundice at time of diagnosis were associated with relapse at univariate analysis (p=0.04 and <0.01 respectively). Complete resolution of pancreatic abnormalities on imaging was observed in 65% of the pts. Exocrine and endocrine insufficiency was observed in 34% and 39% of the pts, respectively. At univariate analysis, no factor was associated with exocrine insufficiency, while female gender (p=0.04), increasing age (p=0.006) and HISORt criteria (p=0.001) were associated with the occurrence of diabetes. Steroid/azathioprine treatment did not prevent pancreatic insuffi- ciency. HISORt - AIP was more frequently associated with inflammatory bowel disease than HISORt + AIP (33% and 3%, respectively), but relapse rates were similar in both groups. Conclusion : Relapse occurs in 27% of AIP pts and is more frequent in pts with elevated serum IgG4 levels at time of diagnosis. Pancreatic atrophy and /or pancreatic insufficiency occur in >1/3 pts within 3 years of diagnosis. The outcome of AIP in pts with normal serum IgG4 - a condition often associated with inflammatory bowel disease- is not different from that of pts with high values. S-97 AGA Abstracts 723 Is Chronic Pancreatitis Really a Patchy Disease? A Prospective Study Based on Quantitative Pancreatic Elastography Julio Iglesias-Garcia, Margarita Castineira, Jose Larino-Noia, Maria Luaces, Rocio Ferreiro, Enrique Dominguez-Munoz Chronic inflammation, pancreatic fibrosis and atrophy are the 3 main criteria for the diagnosis of chronic pancreatitis (CP). These features are considered to have a patchy distribution throughout the pancreas, thus making unreliable the pathological diagnosis based on pancre- atic FNA or biopsy. Nevertheless, studies defining the patchy distribution of CP are based on surgical specimens, which are not representative of the natural history of the disease. Aim of this study was to revisit the concept of patchy distribution of CP in a consecutive series of patients diagnosed with CP at different stages. Methods: 79 consecutive patients (mean age 49 years, range 21-76, 62 male), who underwent an endoscopic ultrasound (EUS) to confirm the diagnosis of CP were prospectively included in the study. 99 patients with healthy pancreas were included as controls. EUS was performed under conscious sedation with the radial Pentax and the Hitachi 900. EUS criteria accepted for the diagnosis of CP were evaluated, and patients were classified according to the Rosemont Classification (RC) into 3 groups: indeterminate for CP, suggestive of CP and consistent with CP. The stiffness of the pancreas at the level of the head, body and tail, as a measure of fibrosis, was quantified by EUS-elastography. Two different areas (A and B) from each region of interest were selected for the quantitative elastographic analysis: Area A represents an area of the pancreatic parenchyma and B refers to a soft peripancreatic reference area. The quotient B/A (strain ratio (SR)) is considered as the result of the elastographic evaluation. Data are shown as median and 95%CI and compared by the ANOVA test. Results: According to the RC, 20 (25.3%) patients were indeterminate for CP, 34 (43.0%) suggestive of CP, and 25 (31.6%) consistent with CP. Strain ratio was similar in the head (2.84; 95%CI: 2.63-3.05), body (3.09; 95%CI: 2.88-3.31) and tail of the pancreas (2.98; 95%CI: 2.77-3.19) (p=0.9). No differences in strain ratio of the different pancreatic areas were also found when patients were classified according to the number of EUS criteria and the RC. Mean strain ratio of the 3 pancreatic regions increases with the number of EUS criteria and Rosemont groups (1.80 in healthy controls, 2.40 for indeterminate group, 2.85 for suggestive group, and 3.62 for consistent group, p<0.001) suggesting that these groups represent different stages of the CP. Conclusion: The analysis of tissue stiffness by quantitative EUS-elastography in patients with different stages of CP does not support the concept of CP as a patchy disease. The 3 groups of the RC most probably represent different stages of the disease. 724 The Functional Role of Cellular Prion Protein in the Enteric Nervous System Gary R. Martin, Mohammad Bashashati, Catherine M. Keenan, Wallace K. MacNaughton, Frank Jirik, Keith A. Sharkey Although misfolding of normal cellular prion protein (PrP C ) into its pathological isoform (PrP Sc ) is thought to be responsible for the transmissible spongiform encephalopathies, the functional role of PrP C is unclear. In the gut, there is evidence that PrP C is expressed in the enteric nervous system (ENS), gut-associated lymphoid tissue and the epithelium. However, a physiological role for PrP C in the gastrointestinal tract has not been identified. Loss of PrP C has been linked to atypical synaptic activity, Ca 2+ homeostasis, and alterations in copper (Cu) metabolism. As the N-terminal region of PrP C is able to bind up to 5 Cu atoms, we hypothesized that the lack of PrP C , by increasing local free [Cu]-induced s-nitrosothiols, would raise nitric oxide (NO) concentrations, thereby decreasing intestinal contractility and tone. Distal ileal tissues obtained from C57BL/6 PrP C -deficient (Prnp -/- ) or wild type littermate control (WT) male mice were mounted in organ baths, and electrical field stimulation (EFS) was applied in the presence or absence of either a nitric oxide synthase (NOS) inhibitor (L- NAME, 100 μM) or an NO-donor (sodium nitroprusside, 100 μM). Whole gut transit and colonic bead expulsion latency was also examined In Vivo. Ileal tissues were mounted in Ussing chambers to examine short-circuit current (Isc) responses to EFS and to epithelial secretagogues. PrP C expression was assessed by immunoblot and by immunohistochemical analysis of whole mount preparations of the ENS. PrP C was expressed in lamina propria cells and in neurons and glia of the ENS. PrP C was mostly expressed in enteric glia and was co-localized with glial fibrillary acidic protein and S-100β in the myenteric and submuco- sal plexus of WT animals. In contrast, PrP C expression was lacking in Prnp -/- mice. Compared to WT, Prnp -/- mice exhibited a marked reduction in contractility in response to EFS that was augmented in the presence of the NOS-inhibitor. Moreover, relaxation responses to EFS in the presence of the NO-donor were significantly increased. There was no significant difference in whole gut transit or colonic bead expulsion latency. The Isc response to forskolin was significantly reduced in PrP C deficient mice, but the EFS response was unaffected. Reductions in forskolin-induced ileal Isc in Prnp -/- mice suggest that PrP C may be involved in cAMP/PKA signaling at the epithelium, but not in neurally evoked secretory responses. Although under resting conditions no alterations in motility were observed in Prnp -/- mice In Vivo, our In Vitro findings raise the possibility that PrP C might be involved in the regulation of gut function during pathophysiological states. 725 Cholinergic Agonists Attenuate Inflammation in Macrophages via Interaction With Neuropeptides Esmerij P. van der Zanden, Klaus Michel, Michael Schemann, Guy E. Boeckxstaens, Wouter de Jonge BACKGROUND AND AIMS Vagus nerve activity attenuates production of pro-inflammatory cytokines by macrophages and inhibits inflammatory processes via the release of acetylcholine (ACh). However, vagus nerve stimulation may also lead to release of neuropeptides with immune-modulatory potential, such as substance P (SP), vasoactive intestinal polypeptide (VIP) en neuropeptide Y. Here, we studied if cholinergic agonists can affect the immuno- modulatory effects of different neuropeptides on peritoneal macrophages. METHODS RAW264.7 peritoneal mouse macrophages were pretreated with SP, VIP and neuropeptide AGA Abstracts