C-peptide at 6 m C-peptide at 12 m Insulin at 6 m Insulin at 12 m IEq; r=0.80 IEq; r=0.85 IEq; r=-0.54 IEq; r=-0.56 Islet dose; r=0.57 Islet dose; r=0.88 Islet dose; r=-0.52 Islet dose; r=-0.54 Islet quality; r=- Islet quality; r=- Islet quality; r=- Islet quality; r=0.53 0.01 0.22 0.22 % free islets; r=- % free islets; r=- % free islets; r=0.02 % free islets; r=0.15 0.17 0.14 purity; r=-0.26 purity; r=0.47 purity; r=0.35 purity; r=0.46 digestion time; r=- digestion time; r=- digestion time; r= digestion time; r= 0.35 0.64 0.32 0.47 m:months; IEq:islet equivalent Su1449 DO PATIENTS WITH CHRONIC PANCREATITIS RECEIVE OPTIMAL BONE HEALTH CARE? Padmavathi Srivoleti, Allison L. Yang, David X. Jin, Shadeah L. Suleiman, Peter A. Banks, Julia McNabb-Baltar Background: Chronic pancreatitis is associated with osteoporosis. Guidelines currently recommend that patients with chronic pancreatitis undergo bone density screening (BMD) and vitamin D serum assay (vit D). It remains unknown if providers are compliant with these recommendations. Aim: To assess BMD and vit D testing in patients with chronic pancreatitis and if the type of provider influences these metrics. Methods: Chronic pancreati- tis patients followed in the ambulatory clinic of a tertiary hospital (August 2017-18) were assessed retrospectively using the institution patient data registry. Charts were reviewed to confirm diagnosis and assess if BMD was tested and if vit D was tested within the past year. Provider was categorized as primary care provider only (PCP), gastroenterologist (GI) or pancreas specialist (P). Demographic information was also abstracted. The relationships between BMD / vit D testing and provider type were analyzed using chi-square tests. Multivariable logistic regressions were performed to assess the impact of provider type on BMD and vit D testing while controlling for confounders. Results: A total of 478 charts were reviewed, 257 met inclusion criteria. Of those, 58% were men and the mean age was 59-year-old (SE 0.86). The most common etiologies of chronic pancreatitis were alcohol (33%); idiopathic (31%); severe acute pancreatitis (8%); and obstructive (8%). Overall, BMD was performed in 133 patients (52%). BMD was abnormal in 60% of patients; 30% of those had osteoporosis. 17/58 (29%) followed by PCP, 11/38 (28%) followed by GI and 84/161 (52%) followed by P underwent BMD (p=0.002). Vit D was performed in 211 of 257 (82%). 43/58 (74%) followed by PCP, 29/38 (76%) followed by GI and 139/161 (86%) followed by P underwent vit D testing (p=0.07). When controlling for gender, age, smoking status and steatorrhea symptoms, patients followed by P were more likely to undergo BMD when compared to those followed by PCP (OR 2.73 95% CI 1.37-5.44, p=0.004). When controlling for gender and age, patients followed by P were more likely to undergo vit D testing when compared to those followed by PCP (OR 2.20 95% CI 1.04-4.64, p=0.039). In multivariable analysis, patients followed by GI did not undergo significantly more BMD and vit D testing compared to those followed by PCP. Conclusion: A significant number of patients with chronic pancreatitis do not receive optimal bone health preventive care. Pancreas specialist were more likely to perform bone density screening and vitamin D compared to PCP and GI. Physician education is needed to increase compliance to guidelines. Su1450 CHRONIC PANCREATITIS ASSOCIATED WITH CYSTIC FIBROSIS TRANSMEMBRANE CONDUCTANCE REGULATOR (CFTR) GENE MUTATIONS – A COMPARISON OF HETEROZYGOTES AND COMPOUND HETEROZYGOTES. Abubaker Y. Ahmed, John McLaughlin, Scott E. Levison, Catherine Fullwood, Alistair Makin Background Chronic pancreatitis is most commonly caused by excess alcohol consumption, it can also be linked to Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) gene mutations. Two different CFTR groups are seen within the chronic pancreatitis population, compound heterozygous with two mutations and heterozygous with a single mutation. It is unknown how these differ clinically. Aims To describe the clinical presentation and course of chronic pancreatitis in patients who are CFTR gene carriers. To assess whether CFTR genotype (heterozygous or compound heterozygous) is associated with a different clinical phenotype. Methods: Retrospective cohort analysis of all patients with chronic pancreatitis and CFTR gene carrier status who attended the Manchester Foundation Trust Pancreatology Unit between 1999 to 2018. These were divided into two groups; compound heterozygous and heterozygous. Data were reported using frequencies and percentages, Fisher's exact was used to compare the two groups. Results: In total 53 were identified. Two had died (3.8%), one had missing genotype information and one had a negative genotype but had previously been diagnosed on a positive sweat test. Eight (16.3%) of the remainder were compound heterozygous, and 41 (83.7%) heterozygous. The mean age at presentation with pancreatitis was 34.6 years in the heterozygous group and 34.7 years in the compound heterozygous. Their clinical profiles are presented in the table. Conclusion Chronic pancreatitis caused by CFTR gene mutations may manifest as different diseases in the compound heterozygous and the heterozygous genotypes. In our cohort none of the compound heterozygous developed diabetes, necrotizing pancreatitis, common bile duct strictures, vascular complications, or required either endoscopic or surgical interventions. This suggests that the compound heterozygous have a milder form of pancreatitis when compared with the heterozygous group. A larger prospective cohort is needed to further explore these preliminary findings. Results S-555 AGA Abstracts Heterozygous n= Compound Heter- Fisher's exact p- 41 ozygous n=8 value Recurrent pancre- 30 (73.2%) (57.1- atitis Acute attack 85.8%) 10 5 (62.5%) (24.5- Presentation of pancreatitis (24.3%) (12.4- 91.5%) 3 (37.5%) 0.665 Unexplained 40.3%) 1 (2.4%) (8.5-75.5%) weight loss (0.1-12.9%) 22 (53.6%) (37.4- 6 (75%) (34.9- Simple analgesia Comparison 69.3%) 3 (7.3%) 96.8%) 1 (12.5%) Opioid when between constant Analgesia score (1.5-19.9%) 16 (0.3-52.7%) 1 needed Constant opioid to other p= (39%) (24.2- (12.5%) (0.3- opioid 0.233 55.5%) 52.7%) Diabetes devel- 10(24.4%) (12.4- 0/8 0 (0.0-36.9%) p=0.180 oped 40.3%) Pancreatic exo- 19 (46.3%) (30.7- 4 (50.0%) (15.7- difference 0.702 crine Insufficiency 62.6%) 84.3%) Average Time to develop exocrine 66.5 29.5 failure (months) 3 (7.3%) (1.5- 2 (25.0%) (3.2- Pancreatic stones 0.182 19.9%) 65.1%) Pancreatic Stric- 12 (29.2%) (16.1- 1 (12.5 %) (0.3- 0.663 tures 45.5%) 52.7%) Pancreatic Pseu- 13 (31.7%) (18.1- 1 (12.5%) (0.3- 0.411 docyst 48.1%) 53.7%) 10 (24.3%) (12.4- 0 (0.0%) (0.0- Pancreas Necrosis 0.180 40.3%) 36.9%) Endoscopic Ther- 10 (24.3%) (12.4- 0 (0.0%) (0.0- 0.180 apy 40.3%) 36.9%) Common Bile 11 (26.8%) (14.2- 0 (0.0%) (0.0- 0.172 Duct Strictures 42.9%) 36.9%) Hepatobiliary Sur- 3 (7.3%) (1.5- 0 (0.0%) (0.0- 1.000 gery 19.9%) 36.9%) Vascular complica- 7 (17%) (7.2- 0 (0.0%) (0.0- tions, venous 0.581 32.1%) 36.9%) thrombosis Su1451 AMMANN PAIN TYPE AND THE NATURAL HISTORY OF PAINFUL CHRONIC PANCREATITIS Christina Tsai, Stuart R. Gordon, Jeffrey M. Adler, Kerrington Smith, Timothy B. Gardner Objectives: Chronic pancreatitis (CP) is an inflammatory disease that usually causes severe abdominal pain. The pain is categorized as Ammann Type A, characterized by intermittent pain episodes, or Ammann Type B, characterized by daily chronic pain. This study examines pain progression and its associated variables throughout the course of the natural history of chronic pancreatitis. Methods: This retrospective cohort chart review study evaluated patients who were diagnosed with chronic pancreatitis and followed longitudinally at Dart- mouth-Hitchcock Medical Center from 2000 to 2018. Contingency table and t-test analysis were performed for categorical and continuous variable associations, respectively, between Ammann pain types and patient and procedural characteristics. Results: 70 patients met inclusion criteria, with 33 having Ammann Type A and 37 Ammann Type B pain. The primary clinical outcome was change of pain over time, reported among 31 (94%) Ammann Type A and 37 (100%) of Ammann Type B patients (P = 0.13). 23 (70%) of Ammann Type A and 22 (59%) of Ammann Type B patients experienced less pain (P = 0.37) compared to the 8 (24%) Ammann Type A and 15 (41%) Ammann Type B who experienced more pain (P = 0.15). Conclusions: More than half of patients with pain from chronic pancreatitis, irrespective of Ammann pain type, have improvement in pain over time. However, patients with Ammann type A pain are more likely to improve. AGA Abstracts