Pancreatitis: In search for a cu RE) cohort. Methods: Demographic and clinical information of children with acute recurrent pancreatitis (ARP) and chronic pancreatitis (CP) ≤19 years of age at the time of enrollment were entered into the REDCap database at sixteen pediatric centers. Presence of PD was confirmed with magnetic retrograde cholangiopancreatography (MRCP), endoscopic retrograde cholangiopancreatography (ERCP), or both. Differences between children with or without PD were analyzed using two-sample t-test or Wilcoxon- rank sum test for the continuous variables, and Pearson Chi-square or Fisher's exact test for categorical variables. Results: Of 288 registry subjects, PD was found in 39 (13.5%). Females were more likely to have PD (72% with PD vs 53% without PD, p=0.025). Children with PD were not different in ethnicity, age at first presentation of acute pancreatitis (AP) or CP, or time from the diagnosis of first AP to CP. Medical, endoscopic, and surgical interventions were similar between children with or without PD. The frequency of PD was not different in those with SPINK1, CFTR or CTRC mutations or with other obstructive, and/or toxic-metabolic risk factors. INSPPIRE children with PD were less likely to have a family history of pancreatitis (p<0.05) and had a lower rate of PRSS1 mutations (10% with PD vs 36% without PD; p=0.004). Although PD did not have any impact on the frequency or pattern of the abdominal pain or school attendance, children with PD reported a higher number of lifelong emergency room visits and hospitalizations (p=0.042 and 0.021, respec- tively) compared with non-PD. Finally, children with PD were less likely to be exocrine pancreatic insufficient (p=0.032). Conclusion: The frequency of PD was only slightly higher in the INSPPIRE cohort compared with reported rates in the general population. Presence of PD did not seem to associate with genetic mutations to influence the risk of pediatric pancreatic disease. Even though children in our cohort with ARP and CP with PD had more ER visits and hospitalizations over their lifetime than children without PD, PD did not appear to have a major impact on the risk for pancreatitis or disease progression in childhood. Tu1465 Comparison of Outcomes From Patients With Prior Surgery in Total Pancreatectomy With Islet Autotransplantation For Refractory Chronic Pancreatitis: A Single-Institution Experience With 120 Consecutive Cases Gumpei Yoshimatsu, Morihito Takita, Rauf Shahbazov, Charles Chang, Prathab B. Saravanan, Mazhar A. Kanak, Michael C. Lawrence, Peter Kim, Nicholas Onaca, Bashoo Naziruddin, Marlon F. Levy Introduction: Chronic pancreatitis (CP) causes narcotic dependence and poor quality of life due to refractory severe pain. Surgical procedures such as Whipple procedure, distal pancreatectomy and pancreatico-jejunostomy are performed when CP patients showed poor response to medication, endoscopic intervention or nerve block. Total pancreatectomy (TP) or completion of TP is considered to effectively eliminate the severe pain if prior procedures failed. Severe hypoglycemia due to surgical diabetes, however, is a common adverse event of TP. Islet autotransplantation following TP (TPIAT) is a treatment option to prevent such serve hypoglycemic event and to remove intractable abdominal pain. The benefits of completion of TPIAT in CP patients with history of prior pancreatic surgery are unclear. Patients and Method; The patients who underwent total or completion pancreatectomy for refractory chronic pancreatitis from 2006 to 2015 at Baylor University Medical Center were retrospectively analyzed for patient characteristics, islet isolation, and endocrine function after islet autotransplantation. Patients were divided into the following two groups: prior surgery group with history of Whipple procedure, Peustow procedure or distal pancreatec- tomy before TPIAT (PS, n = 15) and the group without any pancreatic surgery (NPS, n= 105). Results; No significant differences were found in age, gender, body mass index and etiology of CP between the two groups while pre-operative HbA1c was significantly higher in PS than NPS group (5.9 ± 1.1% and 5.5 ± 0.6%, respectively. p = 0.03). The pancreas weight and islet count after pancreas digestion showed significant difference between two groups (50.2 ± 22.8 and 72.9 ± 26.3 g, p = 0.005 and 295,737 and 412,640 IEQ p = 0.049 between PS and NPS groups respectively). Tissue volume in final preparation was significantly lower in PS group when compared to NPS (7.2 ± 5.3 and 11.7 ± 8.0 mL in PS and NPS groups; p = 0.016). No statistically significant differences between the two groups were observed in post-TPIAT C-peptide and HbA1c. In sub-group analysis, the total islet yield of patients with Whipple procedure (PD) is significantly higher than those with distal pancreatectomy (DP) (423,625 and 160,594 IEQ in PD and DP; p = 0.02); however, there is no significant difference between Puestow procedure and PD or DP groups. Islet dose DP group were significantly lower in islet dose than both PD and Puestow groups (1,980, 6,266 and 3,949 IE/kg in DP, PD and Puestow groups. p < 0.05 for between DP and either PD or Puestow groups). Conclusion: No differences were found in islet isolation outcomes as well as post-operative graft function between the PS and NPS groups. Completion of TP followed by IAT should be a feasible procedure even after the CP patients have prior history of pancreatic surgery. Figure1. Comparison of post digestion islet count (A), total islet yield (B), Pellet size (C) and transplanted outcome (D) between TPIAT patients with or without prior sugery. S-909 AGA Abstracts Tu1466 A Novel Set of microRNA Biomarkers Can Predict Outcomes of Pancreatic Islet Isolation and Graft Function Following Autologous Islet Transplantation for the Refractory Chronic Pancreatitis Gumpei Yoshimatsu, Morihito Takita, Prathab B. Saravanan, Charles Chang, Carly M. Darden, Mazhar A. Kanak, Giovanna Saracino, Michael C. Lawrence, Peter Kim, Marlon F. Levy, Bashoo Naziruddin Introduction: Total pancreatectomy with islet autotransplantation (TPIAT) is a curative treatment option to eliminate refractory abdominal pain as well as to preserve pancreatic endocrine function at single operation for patients with intractable chronic pancreatitis (CP). Due to invasive nature of TPIAT patient selection has to be carefully considered. Prediction of outcomes for the islet isolation and post-transplant graft function is important to ensure benefits of TPIAT. Circulating microRNAs (miR) can potentially be a surrogate biomarker to predict the graft function or damage for TPIAT. We determined if pre-operative circulating microRNAs can predict outcomes of islet isolation and the endocrine function following TPIAT in this study. Patients and Method: Thirty one patients who underwent TPIAT at Baylor University Medical Center between 2006 and 2014 and with a follow-up for more than one year were included in this study. Circulating microRNAs miR-7, miR-9, miR- 184, miR-200a, miR-200c, miR-216, miR-320 and miR-375 in plasma before TPIAT were measured by RT-PCR. Preoperative patient characteristics and miR levels were analyzed to predict islet yield and patient exogenous insulin requirement one-year after TPIAT using uni- and multi-variate regression model. Result: The levels of circulating miR-200a, miR- 200c, miR-320 and miR-375 were significantly higher in CP patients than healthy donors. There was linear regression between number of islets and levels of either miR-200c or miR- 375 extracted in in vitro study with human islets, indicating that these two miRNAs should be a candidate for predictors of TPIAT outcomes. Circulating miR-375, BMI and ΔC-peptide were correlated with post-digestion islet yield in univariate analysis ( p = 0.039, < 0.001 and = 0.007, respectively) while the other miRNAs showed no correlation. MiR-200c, age, hemoglobin A1c, and ΔC-peptide were significantly correlated with patient exogenous insulin requirement at one-year post-TPIAT in the univariate model ( p < 0.001, = 0.049, < 0.001 and < 0.001, respectively). Circulating miR-200c and hemoglobinA1c were selected as independent predictors for insulin requirements one-year post-TPIAT after multivariate regression analysis (miR200c; t=-3.76, p=0.0011, HbA1c; t=3.53, p=0.0019, respectively. Adjusted R 2 =0.66). Conclusion: Preoperative circulating miR-375 can be correlated with islet isolation outcome and miR-200c can be a novel biomarker to predict endocrine function post-TPIAT. Further investigation with large cohort is warranted to verify the utility of circulating miRNAs as biomarkers for TPIAT. Figure 1: Inverse correlations with statistical significance were observed between miR-375 and post-digestion islet count (Fig A. r = -0.373 and p = 0.039) or miR-200c and exogenous insulin requirements one-year post-TPIAT (Fig B. r=-0.639 and p < 0.001). Tu1467 Narcotic Use and Quality of Life in Chronic Pancreatitis Patients Boskey Patel, Julien Fahed, Samuel Han, Joan Kheder, Lisa Bocelli, Wahid Wassef INTRODUCTION Chronic pancreatitis (CP) is a disease characterized by progressive destruction of the pancreatic gland. While there are many associated symptoms, severe and often debilitating pain is by far the most common and the most detrimental. As such, CP may lead to a considerable reduction in quality of life. While several studies have looked at different approaches to pain reduction and control, very few have assessed the quality of life as related to narcotic use in CP patients. The aim of this study was to examine this relationship using the Pancreatitis Quality Of Life Instrument (PANQOLI), a validated questionnaire that evaluates the overall quality of life as subjectively perceived by the patient. METHODS This study included patients with chronic pancreatitis (diagnosed by endoscopic ultrasound) who regularly receive their outpatient care in the Pancreatitis Clinic at UMass Memorial Healthcare. Quality of life measures were assessed using the PANQOLI. Demo- graphics were collected directly from the patient, and information on each patient's daily narcotic use was obtained from the patient's pain contract via electronic medical health record. A multivariable regression was performed to assess the relationship between the amount of narcotics and total PANQOLI scores. RESULTS Sixty-eight patients with CP were recruited for this study. The average age was 49.3 years, and 44 (64.7%) were female. Fifty-nine (86.8%) patients identified themselves as Caucasian. There was a high rate of unemployment (58.8%), substance abuse (51%), and single, unmarried, or divorced status (53%) (Table 1). A multivariable regression was performed looking for a relationship between the PANQOLI scores and daily narcotic use controlling for age, race, and gender. There was found to be a significant correlation (p=0.025) with a coefficient of -0.027. However, R squared was only 0.205. CONCLUSION While there is a mild to moderate association between the quality of life of chronic pancreatitis patients and the amount of narcotics that they take on a daily basis, a causal relationship cannot be confirmed. It is believed that patients who require more narcotics have a lower quality of life because they have more pain (physical or psychological). Multivariable regression showed that for every 100mg increase in daily morphine equivalents, the PANQOLI scores decreased (improved) by 2.7 points. However, only 20% of the variance in the PANQOLI scores can be predicted by difference in age, gender, or race. The next step is to conduct follow up surveys at regularly AGA Abstracts