Methods Case control study using the 2016 National Inpatient Sample (NIS) using ICD10-CM codes to identify patients who underwent EUS with and without FNA or FNB for pancreatobiliary solid lesions (PBSL). We compared cirrhotics with non-cirrhotic controls. Primary outcome was rate of procedural complications. Subgroup analysis to compare compensated and decompensated cirrhosis as per validated BAVENO VI classification.

Aims Few studies have shown an increased risk of certain psychiatric conditions as well as substance abuse disorders in CP, however; these were limited by sample size. Using a large database, we sought to describe the epidemiology and risk association of several psychiatric diseases in patients with CP.

Bajaj, J.S. MD; Brenner, D.M. MD, MSHS; Cai, Q. MD; Cash, B.D. MD; Crowell, M. PhD; DiBaise, J. MD; Gallegos-Orozco, J.F. MD; Gardner, T.B. MD; Gyawali, C.P. MD; Ha, C. MD; Holtmann, G. MD; Jamil, L.H. MD; Kaplan, G.G. MD; Karsan, H.A. MD; Kinoshita, Y. MD; Lebwohl, B. MD; Leontiadis, G.I. MD; Lichtenstein, G.R. MD; Longstreth, G.F. MD; Muthusamy, V.R. MD; Oxentenko, A.S. MD; Pimentel, M. MD; Pisegna, J.R. MD; Rubenstein, J.H. MD; Russo, M.W. MD; Saini, S.D. MD; Samadder, N.J. MD; Shaukat, A. MD, MPH; Simren, M. MD, PhD; Stevens, T. MD; Valdovinos, M. MD; Vargas, H. MD; Spiegel, B. MD, MSHS, FACG; Lacy, B.E. MD, PhD, FACGAuthor Information

Aims It is unclear if antecedent biliary interventions (BI) [ERCP, PHTC] performed for biliary obstruction pose risk of infecting evolving sterile pancreatic and or peripancreatic necrosis (PPN). We sought to determine if BIs increase the risk of infection in PPN patients.

INTRODUCTION: Acute pancreatitis (AP) is a well-recognized and generally serious complication following liver transplantation. We sought to assess AP following adult liver transplantation to describe the risk factors, natural history of the disease and outcomes. METHODS: Data from Nationwide Readmission database from 2010 to 2015 was analyzed. All adult patients who were hospitalized for a primary diagnosis of acute pancreatitis were identified and patients who had a history of liver transplantation were compared to patients with no transplantation, using the appropriate ICD 9 codes. Continuous variables were expressed as means ± standard deviation or median (IQR), and categorical variables were expressed as percentages. All statistical tests were two-sided. RESULTS: 1,575,148 AP patients were included, and out of those 1581 patients (0.1%) had a history of liver transplantation. Patients with a history of transplantation had higher rates of hypertension 59.1% vs. 53.8%, P-value < 0.001, uncomplicated diabetes mellitus, 32.6% vs. 22.9%, P-value < 0.001, diabetes mellitus with chronic complications, 5.8% vs. 3.8%, P-value < 0.001, chronic pancreatitis, 18.2% vs. 16.1%, P-value = 0.02, chronic renal failure 39.6% vs. 7.9%, P-value < 0.001. Transplant patients had higher rates of in-hospital AP complications: acute kidney injury 20.3% vs. 8.3%, P-value < 0.001, transfusion requirement 7.2% vs. 3.0%, P-value < 0.001. No significant difference in the rates of systemic inflammatory response syndrome (SIRS), 2.1% vs. 2.1%, P-value = 0.929, or ileus, 4.0% vs. 3.7%, P-value = 0.595. Transplant patients had lower rates of acute respiratory distress syndrome and/or ventilation need, 0.6% vs. 2.3%, P-value < 0.001, and sepsis 0.5% vs. 1.6%, P-value < 0.001. In-hospital mortality was higher in patients with history of liver transplantation 0.8%vs. 0.3%, P-value = 0.053. Median length of stay was not significantly different 3.00 (2–6) days vs. 3.00 (2–5), P-value = 0.17. Charges of hospitalization were significantly higher in the transplantation group 23,615 USD vs. 21,020 USD, P-value < 0.001. CONCLUSION: Post-liver transplant AP carries significant morbidity and mortality. The extensive nature of the anatomical dissection around the pancreas, the type of biliary reconstruction and anastomosis, chronic immunosuppression medications, associated comorbidities, could be contributing factors. Our study highlights possible areas for further investigation in the liver transplant population.

INTRODUCTION: Cirrhosis affects the outcome of non–liver-related illnesses requiring hospitalization, but its impact on acute pancreatitis (AP) patients requiring hospitalization has not yet been studied. We aim to investigate the prevalence of cirrhosis among AP patients and outcomes of mortality, morbidity and cost. METHODS: Data from Nationwide Readmission database from 2010 to 2015 was analyzed. All adult patients who were hospitalized for a primary diagnosis of acute pancreatitis were identified and patients who had a history of cirrhosis were compared to patients with no cirrhosis, using the appropriate ICD 9 codes. Continuous variables were expressed as means ± standard deviation or median (IQR), and categorical variables were expressed as percentages. All statistical tests were two-sided. RESULTS: A total of 1,575,148 patients were admitted for AP were included in this cohort, and out of those 46,904 patients (3.0%) had a history of cirrhosis. Patients in the cirrhosis group were significantly older, mean age 53.42 ± 11.885 vs. 51.55 ± 17.627 years, P-value < 0.001, mostly males 62.0% vs. 52.3%, P-value < 0.001, and admitted emergently, 97.0% vs. 95.5%, P-value < 0.001. Alcohol abuse was more common among cirrhosis patients 69.1% vs. 28.7%, P-value < 0.001 as well as drug abuse, 12.1% vs. 6.7%, P-value < 0.001. Cirrhosis patients had higher rates of in-hospital AP complications: acute kidney injury 11.7% vs. 8.2%, P-value < 0.001, pneumonia 3.6% vs. 2.8%, P-value < 0.001, acute respiratory distress syndrome and/or ventilation need, 3.0% vs. 2.2%, P-value < 0.001, transfusion requirement 8.9% vs. 2.8%, P-value < 0.001, sepsis, 2.1% vs. 1.6%, P-value < 0.001. SIRS: no difference. In-hospital mortality was higher in patients with cirrhosis 1.8% vs. 0.8%, P-value < 0.001. Median length of stay (LOS) was not significantly different 4.00 (2–6) days vs. 4.00 (2–6), P-value = 0.751. Charges of hospitalization were significantly higher in the cirrhosis group 24,819 USD vs. 20,918 USD, P-value < 0.001. CONCLUSION: Acute pancreatitis patients with liver cirrhosis have higher mortality, possibly related to cirrhosis-related comorbidities (sepsis, AKI, GI bleed/transfusion requirement, and ARDS) and not due to AP. Interestingly, they had lower prevalence of SIRS and LOS was not different.

INTRODUCTION: The triad of hypertriglyceridemia (HTG) induced acute pancreatitis (AP) coexisting with diabetes ketoacidosis (DKA) has been reported. Patients with the triad are a special subgroup of AP patients that may need different management due to potential life-threatening conditions presenting all at once. Outcomes of pancreatic fluid collections (PFCs) in this population has not yet been studied. We sought to investigate AP mortality, morbidity and PFC outcomes in patients with the triad. METHODS: We perform a cohort study of a prospectively maintained database of patients admitted with AP at a tertiary center in the last 15 years. We strictly included patients who met AP diagnosis by Revised Atlanta Classification. Study groups were: AP-only vs. HTG-AP + DKA, AP + DKA, and HTG-AP. Primary outcome was inpatient mortality, morbidity (SIRS, ICU, AKI, ARDS), severity of AP and complication of AP. Multivariable logistic regression models were constructed using STATA software version 9.4. RESULTS: Over two thousand patients were reviewed, of whom 124 patients had the triad (HTG-AP, DKA), 100 had HTG-AP only (triglyceride levels >1,000 mg/dL), 67 had AP + DKA and we included 101 with AP-only (control) for the analysis. Patients with HTG-AP + DKA had higher rates of PFCs 59% compared to AP-only 22%, HTG-AP 29% and AP + DKA 20% ( P < 0.001). HTG-AP + DKA were more likely to have acute peripancreatic fluid collections (APFC) 46% compared to AP-only 17%, HTG-AP 19% and AP + DKA 13% ( P < 0.001). HTG-AP were more likely to have pseudocysts 11% compared to AP-only 4%, HTG-AP + DKA 4%, and AP + DKA 2% ( P = 0.06). No difference in PFC drainage was found ( P = 0.97). Patients with AP+DKA were more likely to develop AKI 63% compared to AP-only 8%, HTG-AP + DKA 34% and HTG-AP 33% ( P < 0.001). HTG-AP patients were more likely to have ARDS 25% compared to AP-only 2%, AP + DKA 10% and HTG-AP + DKA 17% ( P < 0.001). HTG-AP were more likely to require enteral tube feeding and total parenteral nutrition (Table 1). No difference inpatient mortality was found. CONCLUSION: Patients with the triad of HTG-AP + DKA have significantly higher rates of pancreatic fluid collections. But PFCs seems to resolve in its own as there was not difference in interventions for the PFCs. Less aggressive strategy for PFCs management in this patient population can be followed.

AIM:To evaluate GI safety of celecoxib compared with 2 nonselective (ns) NSAIDs, as a secondary objective of a large trial examining multiorgan safety.METHODS:This randomised, double-blind controlled trial analysed 24 081 patients. Osteoarthritis or rheumatoid arthritis patients, needing ongoing NSAID treatment, were randomised to receive celecoxib 100-200 mg b.d., ibuprofen 600-800 mg t.d.s. or naproxen 375-500 mg b.d. plus esomeprazole, and low-dose aspirin or corticosteroids if already prescribed. Clinically significant GI events (CSGIE-bleeding, obstruction, perforation events from stomach downwards or symptomatic ulcers) and iron deficiency anaemia (IDA) were adjudicated blindly.RESULTS:Mean treatment and follow-up durations were 20.3 and 34.1 months. While on treatment or 30 days after, CSGIE occurred in 0.34%, 0.74% and 0.66% taking celecoxib, ibuprofen and naproxen. Hazard ratios (HR) were 0.43 (95% CI 0.27-0.68, P = 0.0003) celecoxib vs ibuprofen and 0.51 (0.32-0.81, P = 0.004) vs naproxen. There was also less IDA on celecoxib: HR 0.43 (0.27-0.68, P = 0.0003) vs ibuprofen; 0.40 (0.25-0.62, P < 0.0001) vs naproxen. Even taken with low-dose aspirin, fewer CSGIE occurred on celecoxib than ibuprofen (HR 0.52 [0.29-0.94], P = 0.03), and less IDA vs naproxen (0.42 [0.23-0.77, P = 0.005]). Corticosteroid use increased total GI events and CSGIE. H. pylori serological status had no influence.CONCLUSIONS:Arthritis patients taking NSAIDs plus esomeprazole have infrequent clinically significant gastrointestinal events. Co-prescribed with esomeprazole, celecoxib has better overall GI safety than ibuprofen or naproxen at these doses, despite treatment with low-dose aspirin or corticosteroids.

Introduction: Acute pancreatitis (AP) is a common gastrointestinal discharge diagnosis in the US, but outcomes in cirrhotic patients has not yet been studied. We aim to investigate these outcomes. Methods: Propensity score matching was used to match patients with and without cirrhosis on a 1:2 basis. The Cleveland Clinic acute pancreatitis registry database was used. Outcomes included inpatient mortality, organs failure, systemic inflammatory response syndrome (SIRS), length of hospital stay (LOS). Subgroup analysis of cirrhotic patients according to Child-Pugh and Model of End-stage Liver Disease (MELD) scores was performed. Multivariable logistic regression models were constructed using SAS software version 9.4 Results: 819 patients with acute pancreatitis were analyzed. Cirrhosis prevalence was 6.1%. Cirrhosis patients were mostly males and younger (table 1). Mean Child-Pugh score was 8.2, mean MELD score was 16 (table 1). Etiology of AP was similar in both groups. Cirrhosis subjects had significantly higher inpatient mortality (7.5% vs. 1.3%, P=0.1), severe AP (17.5% vs. 7.5%), Shock (7.9% vs. 3%), ARDS (10% vs. 3.8%), ICU requirement (15% vs. 6.3%) although not statistically significant, the trend was present. Surprisingly, cirrhotics had lower prevalence of admission SIRS (22.5% vs. 32.5%), SIRS day 2 (25% vs. 15%). Cirrhotics had similar rates of pancreatic necrosis, ileus, BISAP score, Marshall score, admission hematocrit, BUN and LOS (table 2). Finally, cirrhotic patients who had severe AP and/or required ICU and/or die in-hospital were mainly Child-C and had high MELD score (17-38) but they had low BISAP and Marshall scores (table 5). Conclusion: Limitations: Retrospective design. Small sample size due to low prevalence of cirrhosis. Strength: First study to analyze outcomes of cirrhosis in AP and compare according to Child-Pugh and MELD scores. Adjusted for age, gender, race, CCI. Conclusion: Acute pancreatitis patients with liver cirrhosis have higher inpatient mortality, possibly related to decompensated cirrhosis - as they appear to have lower prevalence of SIRS, low BISAP and Marshall scores as well as similar rates of pancreatic necrosis compared to non-cirrhotics. Importantly, cirrhotic patients who died were mostly Child-Pugh C and had very high MELD score.32_A Figure 1. Demographic and Baseline Characteristics32_B Figure 2. Outcomes: Characteristics of cirrhotic patients with severe pancreatitis, required ICU and/or died in the hospital32_C Figure 3. Outcomes of Cirrhotic patients: Child Pugh Class A vs. B/C

Introduction: The triad of HTG-induced AP and DKA is a special subgroup of AP that may need different management due to potential life-threatening conditions presenting all at once. Our aim was to investigate if plasmapheresis was beneficial in this subgroup of patients. Methods: Retrospective review of patients admitted with AP at the Cleveland Clinic in the last 15 years. We strictly included patients who met AP diagnosis by Revised Atlanta Classification attributed primarily to HTG and had concomitant DKA on admission. Primary outcome was inpatient mortality. We compare three different groups: Plasmapheresis + insulin drip vs. insulin drip only, plasmapheresis + insulin drip vs. none, insulin drip only vs. none. Multivariable logistic regression models were constructed using STATA software version 9.4 Results: Over two thousand patients were reviewed and 124 patients with the triad were analyzed. 11 patients received plasmapheresis + insulin drip but no patient received plasmapheresis alone. 60 patients received insulin drip only and 53 patients did not receive plasmapheresis or insulin drip. No benefit in mortality reduction was found in patients who received plasmapheresis (p=0.44). Plasmapheresis patients had higher odds of having pancreatic necrosis, pancreatic fluid collections, admission SIRS, persistent SIRS (48h), ARDS, sepsis, Venous thromboembolic event, pseudoaneurysm, pleural effusion, being NPO for longer time, end-up receiving total parenteral nutrition (TPN) more often as well as enteral feeding, and had longer requirement of insulin drip (p<0.05, CI 95%) (table 1, 2). Conclusion: Limitations: Retrospective data. All patients received some amount of insulin drip. Strength: Biggest patient cohort to date. Conclusion: No benefit of plasmapheresis was found on inpatient mortality. Patients who received plasmapheresis had higher rates of pancreatic necrosis and fluid collections as well as SIRS and multi-organ failures. Randomized Clinical Trials are needed to assess the benefit of plasmapheresis.39_A Figure 1. Plasmapheresis + insulin drip vs insulin drip only.39_B Figure 2. Plasmapheresis + insulin drip vs None.39_C Figure 3. Insulin drip only vs None.

Introduction: HTG-induced AP usually happens when triglycerides are >1,000 mg/dL, but it could happen when levels are <1,000 mg/dL. In addition, HTG-induced AP can happen with concomitant DKA (so-called “enigmatic triangle”), which is a special subgroup of AP that may need a different management. We sought to investigate if there were different outcomes between severe and very-severe HTG in the settings of the triad of HTG-AP-DKA. Methods: Retrospective review of patients admitted with AP at the Cleveland Clinic in the last 15 years. We strictly included patients who met AP diagnosis by Revised Atlanta Classification attributed primarily to HTG and had concomitant DKA on admission. Primary outcome was inpatient mortality. We compare severe (TG 500-999 mg/dL) and very-severe HTG (TG >1,000 mg/dL). Multivariable logistic regression models were constructed using STATA software version 9.4. Results: Over two thousand patients were reviewed and 124 patients with the triad were analyzed. 90 patients with very-severe HTG and 33 patients with severe HTG were found and compared. Very-severe HTG patients were younger (47 vs. 52 years old), had less comorbidities, lower rates of acute kidney injury and altered mental status (p<0.05) (table 1). No difference in inpatient-mortality was found between groups. Moreover, Kaplan-Meier 5-years survival curves did not show significant difference (fig. 1). Predictors of mortality were: pancreatic necrosis (HR: 5.1, p<0.05), readmission within 30-days (HR: 5.8, p<0.05), presence of ascites (HR: 5.4, P=0.016) or pleural effusion (HR: 4.6, P=0.02) (table 2). Conclusion: Limitations: Retrospective data. Strength: Biggest patient cohort to date. Conclusion: No difference in mortality and morbidity was found when comparing severe vs. very-severe HTG in our “enigmatic triangle” cohort. Interestingly, patients re-admitted to the hospital within 30-days had almost 6 times higher changes of dying. Our study findings are clinically significant and warrants further investigations.38_A Figure 1. Outcomes according to triglyceride levels38_B Figure 2. Predictors of mortality38_C Figure 3. Severe vs. Very severe triglyceridemia: 5 years survival

Introduction: Acute pancreatitis (AP) is a frequent cause of hospitalization. Cannabis is the most frequently used illicit drug in the world. There is increasing inconclusive evidence about the impact of cannabis in AP. The aim of this study was to identify the prevalence of cannabis use among all patients with AP in the US and investigate the impact of cannabis use on mortality, morbidity, and cost. Methods: The National Inpatient Sample (NIS) database from 2003 to 2013 was queried for all patients with a discharge diagnosis of AP (ICD-9 577.0). Active exposure to cannabis was ascertained based on ICD-9 305.2. Outcomes included in-hospital mortality, length of stay (LOS), inflation adjusted charges, acute kidney failure, ARDS, and shock. Results were adjusted for age, gender, race, Charlson comorbidity index (CCI), median income quartile, and hospital characteristics. Multivariable logistic regression models were constructed using SAS software (version 9.4, The SAS Institute, Cary, NC). Results: Over 2.8 million patients with AP patients were analyzed [Table 1]. Cannabis use prevalence was 0.3%. Patients exposed to cannabis were younger and mostly males. After adjusting for these factors, the cannabis group had significantly lower inpatient mortality compared to the non-cannabis group (OR 0.17, CI: 0.06-0.53). Cannabis-users also had decreased LOS (4.1 vs. 5.2 days), inflation-adjusted charges ($21,398 vs. $27,488), acute kidney failure (OR: 0.82), ileus (OR: 0.71), shock (OR: 0.39), ARDS (OR: 0.40), and need for parenteral nutrition (OR: 0.66) (95% CI, table 2). Conclusion: Limitations: Inpatient data only (missing post discharge outcomes lacking), potential for misclassification (many cannabis users may not have the ICD9 code). Strength: Largest patient cohort to date. Adjusted for age, gender, race, and Charlson comorbidity index (CCI), median income quartile, and hospital characteristics. Conclusion: In AP, cannabis users had lower age-adjusted mortality, morbidity and hospitalization-cost. The interplay of these factors remains unknown and further clinical and molecular studies on how cannabis affects the pancreas during AP are needed.10_A Figure 1. Patient Characteristics10_B Figure 2. Association Between Cannabis and Binary Outcomes of AP10_C Figure 3. Trends in charges

Introduction: Acute pancreatitis (AP) is a frequent gastrointestinal discharge diagnosis in the United States with an estimated cost of $2.6 billion per year. Cirrhosis is the eighth leading cause of death in the United States. Cirrhosis impact on acute pancreatitis patients requiring hospitalization has not yet been studied. We aim to investigate the prevalence of cirrhosis among acute pancreatitis patients and outcomes of mortality, morbidity and cost. Methods: The National Inpatient Sample (NIS) database from 2003 to 2013 was queried for patients with a discharge diagnosis of AP (ICD-9 577.0) and liver cirrhosis (ICD9 571.5). Cirrhosis was further classified as compensated cirrhosis (CC) and decompensated cirrhosis (DC) as per the well validated Baveno IV criteria. Outcomes included inpatient mortality, organs failure, systemic inflammatory response syndrome (SIRS), parenteral nutrition, length of hospital stay (LOS), US$ inflation-adjusted Charges. Results were adjusted for age, gender, race, Charlson comorbidity index (CCI), median income quartile, and hospital characteristics. Multivariable logistic regression models were constructed using SAS software version 9.4 Results: Over 2.8 million patients with acute pancreatitis were analyzed. Cirrhosis prevalence was 2.8% (80,093). Cirrhosis patients were mostly likely to be males, Hispanics and older age. Both Compensated and Decompensated cirrhosis subjects had significantly higher mortality and worse outcomes (table 2). Highest odds ratios (OR) were:: inpatient mortality (OR 3.4, P<0.001), Shock (OR 1.5, P=0.02), Ileus (OR: 1.3, p=0.02, ARDS (OR 1.2, p=0.03), upper endoscopy performed (OR 2.0, p<0.001), blood transfusions (OR 3.1, p<0.001), gastrointestinal bleed (OR 5.5, p<0.001), sepsis (OR 1.3, p=0.005), portal vein thrombosis (PVT) (OR 7.2, p<0.001), acute cholecystitis (AC) (OR 1.3, p<0.001). Interestingly, cirrhosis patients had lower length of stay, (OR 0.16, p<0.001), AKI (OR 0.93, p=0.06), myocardial infarction (OR 0.31, p<0.001), SIRS (OR 0.62, p<0.001), parenteral nutrition (OR 0.84, p=0.002). Decompensated cirrhosis had higher inflation-adjusted hospital charges ($39,585 vs. $27,313; p<0.001) compared to noncirrhosis patients. Conclusion: Acute pancreatitis patients with liver cirrhosis have higher inpatient mortality, but unlikely to be due to AP severity as in our cohort they had lower prevalence of SIRS, AKI. It is possibly related to complications of cirrhosis.31_A Figure 1. Association Between Cirrhosis and Binary Outcomes of AP.31_B Figure 2. Association Between Cirrhosis and Continuous Outcomes of AP31_C Figure 3. Trends in charges

Total pancreatectomy (TP) with auto-islet transplant (AIT) is an extreme treatment for chronic pancreatitis, and we reviewed our experience to assess the impact on quality of life (QOL).