Mo1395 Epidemiology of Hospitalizations for Acute Pancreatitis in Patients With Chronic Pancreatitis: A Population-Based Analysis From 2002-2012

Table 1 Adverse Outcomes for Mild, Moderate, and Severe Pancreatitis

LOS - length of hospital stay, OF - organ failure, d - days Table 2 Outcomes after Necrotic Collections, Peri-pancreatic Fluid Collections, and Organ Failure

Table 2: Rates of Moderate to Severe Pancreatitis in Malignancy Patients

LOS - length of hospital stay, OF - organ failure, AP - acute pancreatitis, d - days

Mo1394 Association Between Gut Hormones and Risk of Recurrence of Acute Pancreatitis : A Prospective Clinical Study Varsha Asrani Background: Acute pancreatitis (AP) is a significant burden for patients and health care systems. Recurrent AP attacks lead to chronic pancreatitis in 36% of patients with a mortality rate of up to 50%. However, the exact pathophysiological mechanisms of AP recurrence are largely unknown. We hypothesized that gut hormones may play an important role in this process. Aim: To investigate the associations between gut hormones and recurrence of AP. Design: The study design was a cross-sectional follow-up clinical study of consecutive patients with AP admitted to Auckland City Hospital (New Zealand). Fasting venous blood samples were collected from all patients and used for analysis of gastric inhibitory polypeptide (GIP), glucagon-like peptide active (GLP-1 active), peptide YY, vasoactive intestinal polypeptide (VIP), ghrelin, gastrin, cholecystokinin (CCK), and secretin. GLP-1 active, and GIP were measured using the MILLIPLEX® MAP human metabolic hormone magnetic bead panel based on the Luminex xMAP® (Luminex Corporation, Austin, Texas, USA, 1995) technology. The Luminex xPONENT® software (MILLIPLEX® Analyst 5.1) recorded the fluorescent reporter signals based on which the results were quantified (ng/mL). Peptide YY was measured using the Merck-Millipore (MA, USA) ELISA kits as per the user's manual. The Rayto Microplate Reader (V-2100C, Santa Fe, Granada, Spain) with an absorbance of 405 - 630 nm was used and results were reported in ng/mL. All assays were performed as per the instructions in user's manual. Statistical analyses were conducted using the unadjusted and adjusted linear regression. Age, sex, ethnicity, aethiology, duration since first AP attack, body mass index, and presence of diabetes mellitus were adjusted for. Results: A total of 94 patients were included in this study, of which 28 (30%) developed recurrent AP. GIP was significantly associated with recurrence of AP in both unadjusted ( B 48.38; 95%CI; p 0.008) and adjusted analyses (B 51.90; 95%CI; p 0.004). Peptide YY was significantly associated with recurrence of AP in both unadjusted ( B 51.68; 95%CI; p 0.047) and adjusted analyses (B 57.04; 95%CI; p 0.030 ). GLP-1 was significantly associated with recurrence of AP in adjusted analysis (B 11.53; 95%CI; p 0.042) only. VIP, ghrelin, gastrin, CCK, secretin were not significantly associated with recurrence of AP in either unadjusted or adjusted analyses. Conclusion: This study demonstrates, for the first time, that GIP, GLP-1, and peptide YY are significantly associated with risk of recurrence of AP. Further, this association appears to be independent of diabetes status and other patient- and AP-related characteristics. Further prospective clinical studies are now warranted to investigate the exact pathophysiological role of incretins in with the pathogenesis of recurrent AP.

Mo1393 Assessing the Definition of Moderately Severe Acute Pancreatitis - Time to Redefine? Wilson Kwong, Alena Ondrejkova, Santhi Swaroop Vege Background and Aims: A new category of moderately severe acute pancreatitis (MSAP) was incorporated in the revised Atlanta classification of AP in 2012 and defined as acute pancreatitis (AP) with transient organ failure (TOF), or local (peripancreatic fluid collections, and necrotic collections). We sought to determine whether patients in the MSAP category with these complications have disparate outcomes compared to those with mild AP (MAP) and severe AP (SAP). We also compared the outcomes of patients with peri-pancreatic fluid collections, necrotic collections, and TOF to see whether they are similar enough to merit placement into the same category which has not been previously studied. Methods: We retrospectively identified 514 consecutive cases of AP directly admitted to Mayo Clinic Rochester from November 2010 to December 2013. Results: The mean length of hospital stay (LOS) was 5.2 days (d). 26 patients (5.1%) developed necrotic collections, 33 patients (6.4%) developed an acute peri-pancreatic fluid collections. In-hospital mortality was 1% (5 deaths).There were significant differences in the morbidity and mortality among the three categories of mild, moderate, and severe AP (Table 1). MSAP had longer length of hospital stay (p<0.0001), higher rates of ICU admission (p<0.0001), and higher mortality rate (p= 0.04) compared to MAP. SAP experienced longer LOS (p<0.0004), more ICU admissions (p<0.0001), more interventions for necrosis (p<0.0001), and had higher mortality rate (p= 0.002) compared to MSAP Patients with TOF had mean LOS 1 day longer than in MAP (4.7d vs 3.3d; p=0.04) and demonstrated similar rates of ICU admission (p=0.16) (Table 2). The mean LOS in patients with acute fluid collections was longer than in patients with mild AP (5.9d vs 3.3d; p=0.0003) but the two groups had similar rates of ICU admission (p=0.16). There was no mortality in the MAP, TOF, and acute fluid collection subsets. Patients with TOF demonstrated similar LOS and need for ICU admission as patients with acute fluid collections. Patients with necrotic collections had longer LOS (p=0.0001) and higher rates of ICU admission (p=0.02) compared with patients with TOF. Compared to those with acute fluid collections, patients with necrosis had longer LOS (p<0.0001), higher rates of ICU admission (p=0.0005), required more interventions (p=0.001), and demonstrated higher mortality (p=0.003). Conclusions: 1. Necrotic collections required longer LOS, higher ICU admission, more interventions, and even some mortality not seen in TOF or acute fluid collections. 2. The LOS for TOF (4.7d) and fluid collections (5.9d) is closer to that of MAP (3.3d) than for necrotic collections (21.1d). 3. The heterogeneous components of MSAP have disparate clinical outcomes and MSAP may need to be refined in future classifications as necrotic collections without persistent OF

Mo1395 Epidemiology of Hospitalizations for Acute Pancreatitis in Patients With Chronic Pancreatitis: A Population-Based Analysis From 2002-2012 Amrit Kamboj, Alice Hinton, Phil Hart, Darwin Conwell, Somashekar G. Krishna Background: Acute pancreatitis (AP) is the most common gastrointestinal cause of hospitalization. Chronic pancreatitis (CP) is associated with a high likelihood of re-admissions due to AP. The objective of this study was to investigate the epidemiology of AP-related hospitalizations in patients with CP. Methods: Utilizing the Nationwide Inpatient Sample, all patients (‡18 years) with a primary diagnosis of AP from 2002-2005 were compared to those from 2009-2012. Univariate and multivariate comparisons were performed to identify predictors of CP amongst all AP-admissions adjusting for demographics, time period, etiology, and associated factors on AP-related hospitalizations. Results: A total of 2,016,045 APrelated admissions were evaluated during the study periods. Compared to 2002-2005, there

S-699

X : 10052$CH01 03-28-16 00:57:27 PDFd : 10052B : o

Page 699

AGA Abstracts

AGA Abstracts

pancreatic duct cannulation and performance of biliary sphincterotomy, rectal indomethacin had an OR of 0.36 (95% CI 0.10 - 0.79) (p<0.001) for PEP and an adjusted OR of 0.18 (95% CI .05 - 0.64) (p<0.001) for moderate to severe PEP. Interestingly, the majority of benefit for both outcomes was seen in patients with pancreatic adenocarcinoma (1.88 % vs. 7.57% for PEP, p<0.001, 0% vs. 4.32% for moderate to severe PEP). Discussion1. Our retrospective cohort study demonstrated that patients with malignant obstruction, particularly those with pancreatic adenocarcinoma, had high rates of PEP at baseline and received significant benefit with rectal indomethacin in real-world usage. 2. Our findings suggest that rectal indomethacin should be adopted for use in these patients and that they warrant routine inclusion in any subsequent studies involving interventions for the reduction of post-ERCP pancreatitis. Table 1: Rates of Post-ERCP Pancreatitis in Malignancy Patients

AGA Abstracts

was a 13.2% increase in AP admissions in recent years. The greatest absolute increase in associated etiology was chronic pancreatitis (CP)-related AP. Multivariate analysis of a secondary diagnosis of CP among AP-related admissions revealed that patients with CP were more likely to be male, African American race, lower socioeconomic status, and covered by Medicaid or Medicare (table 1). Patients with CP were also more likely to be admitted to large volume, urban teaching hospitals. Etiologically, these patients had associated smoking (OR 1.43, 95% CI 1.39, 1.47) and alcohol-related (OR 1.33, 95% CI 1.29, 1.38) disorders. Although pancreatic anomalies constituted a small portion of the study population, they had the highest odds of AP-on-CP (OR 4.82, 95% CI 4.11, 5.65). The mortality rate for patients with AP-on-CP and AP-without-CP decreased during the study period (0.65% to 0.26%, p<0.001 and 1.7% to 0.89%, p<0.001, respectively), with the former having a significantly lower mortality rate (p<0.001). Conclusion: Patients with CP who are hospitalized for AP likely represent a select subset of AP patients with multiple re-admissions. Although these patients have lower mortality rates, the increasing rates of hospitalizations in this subset require additional attention. Further research focusing on preventing the progression of AP to CP is needed. This study adds to growing data suggesting that cessation of tobacco and alcohol use is a potential strategy to address this challenge.

Table 1: Multivariable regression model showing significant predictors for Physical Quality of Life after 12 months of AP

* NS: No significant

Mo1397 Idiopathic Pancreatitis: Endoscopic Outcomes With Long-Term Follow-Up C. Mel Wilcox, Toni Seay, Hwasoon Kim, Shyam Varadarajulu Background: Although idiopathic pancreatitis is common, the natural history is poorly studied and the best diagnostic approach to both single and multiple attacks is unclear. Aim: To study the impact of endoscopic interventions in idiopathic pancreatitis with longterm follow-up. Methods: This is a prospective study of consecutive patients with idiopathic pancreatitis evaluated over a 10-year (2004-2014) period. EUS was performed in all patients and clinical information for each episode of pancreatitis was reviewed. Patients with microlithiasis or bile duct stones were referred for cholecystectomy and ERCP, respectively. For patients with a single attack, if EUS was normal or if the diagnosis was chronic pancreatitis or pancreas divisum, the patient was followed for recurrence. For those with multiple attacks and a negative EUS, ERCP and Sphincter of Oddi manometry with endoscopic therapy was undertaken. All patients were followed long-term to evaluate for recurrent pancreatitis. Results: Over the study period, 201 patients were identified (80 single attack, 121 multiple attacks; mean age 53 years; 53 % female). Following EUS, among patients with a single attack, 53.8% were categorized as idiopathic, 16.3% pancreas divisum, 15% microlithiasis/ cholelithiasis, 11.3% chronic pancreatitis and 4.6% other diagnosis. For patients with multiple attacks, following EUS and appropriate work-up, 40.5% were diagnosed with sphincter of Oddi dysfunction (SOD), 24.8% pancreas divisum, 14.1% idiopathic, 10.7% chronic pancreatitis, 6.6% microlithiasis/cholelithiasis and 3.3% other diagnosis. Long-term follow-up (median 37 months; range = 1-129 months) documented recurrence of pancreatitis in 24% (95% CI, 15% to 38%) of patients with a single attack and 49% (95% CI, 38% to 62%) of patients with multiple attacks. Despite endoscopic therapy, patients with pancreas divisum and SOD had a relapse of 50% (95% CI, 35% to 68%) and 55% (95% CI, 31% to 82%), respectively. Conclusions: EUS may be an ideal, minimally invasive, tool for the diagnostic evaluation of idiopathic pancreatitis. Following a single idiopathic attack of pancreatitis and a negative EUS examination, relapse was infrequent. Despite endoscopic therapy, SOD and pancreas divisum were associated with a high relapse rate in patients with multiple attacks of pancreatitis.

Mo1396 Quality of Life Is Significantly Impaired Following an Attack of Acute Pancreatitis: A Prospective Study Jorge D. Machicado, Amir Gougol, Kimberly Stello, Gregory Owens, Dhiraj Yadav, Adam Slivka, David C. Whitcomb, Georgios I. Papachristou Background: Acute pancreatitis (AP) is an inflammatory event associated with considerable morbidity and mortality. Few small prospective studies have evaluated the effect of AP on quality of life (QOL) and have been inconsistent. The objective of this study was to prospectively assess the QOL at 3 and 12 months following AP. Methods: We used a prospective cohort of AP patients admitted to the University of Pittsburgh between March 2011 and June 2015, who were prospectively followed at 3 and 12-month intervals as part of the PROOF study. Participants responded a telephone survey that included the Short Form-12 questionnaire (SF12 v2). The SF-12 measures 8 domains summarized into the physical component (PCS) and mental component summary (MCS). The results of this survey were compared with published data on US general population. Multivariable linear regression models were used to determine independent predictors of QOL. Results: Out of 188 consecutive AP patients that were contacted after 3 months of AP attack, 100 subjects (53%) responded to the survey; while out of 138 contacted at 12 months, 76 subjects (55%) responded. Mean age of patients was 53.1±19 years, 65% were female, 91% were white, 52% had biliary etiology, 63% enrolled with first AP attack, and 10% were classified as severe AP based on Revised Atlanta Classification. PCS was significantly lower following AP when compared to controls both at 3 (43.4 vs. 49.6, P < 0.001) and 12 months (46.7, P = 0.01). The physical functioning, role physical, bodily pain, general health, vitality, and social functioning domains were significantly lower at 3 months following AP compared with controls. Only the general health domain remained lower at 12 months after AP (Figure 1). On multivariate analysis, persistent abdominal pain was an independent predictor for lower physical QOL at 3 months following AP. In addition to persistent abdominal pain, preexisting diabetes, and obesity predicted lower PCS at 12 months independently of age, gender, severity of disease, and other comorbidities (Table 1). Variables in the final multivariate models explained 20% and 35% of the variance in physical QOL scores at 3 and 12 months respectively. Conclusion: The impact of an attack of AP on QOL appears substantial. Physical QOL remains impaired even at 12 months following an AP attack. Pre-existing diabetes, obesity and persistent abdominal pain significantly affect the QOL of patients with AP. Further research should focus on management strategies to improve QOL following an AP attack.

AGA Abstracts

X : 10052$CH01 03-28-16 00:57:27 PDFd : 10052B : e

Mo1398 Aggressive Fluid Resuscitation during Acute Pancreatitis: Is there Increased Risk of Acute Kidney Injury (AKI)? Carlos Roberto Simons Linares, Bashar M. Attar, Michael J. Bartel, William Trick, Palashkumar Jaiswal, Diana Plata, Helen Zhang Background & Aim: Among acute pancreatitis patients, aggressive fluid resuscitation may improve mortality during hospital day 1, especially during the first 6-12 hrs and this approach is recommended by guidelines in the US and elsewhere. However, studies have raised concerns about possible increased morbidity and mortality with extended aggressive fluid resuscitation. Among acute pancreatitis patients, we evaluated the association between persistent systemic inflammatory response syndrome (SIRS at 48 hours) and acute kidney injury (AKI), by volume of IV fluid resuscitation used during the first 12 hours of hospitalization. Methods: We performed a retrospective cohort study of acute pancreatitis patients hospitalized at a large public hospital in Chicago during 1/13-12/14. We identified acute pancreatitis by ICD9 code or lipase ‡ 3 times the normal upper limit. Two physicians reviewed each case to include only first episodes of acute pancreatitis, and exclude patients with other causes of hyperlipasemia or no documentation of clinical acute pancreatitis. We collected fluid volume administered during the first 12hrs, 24 hrs and 48 hrs of admission; we created 3 categories for average administration rate: IVF1 (<150 ml/hr), IVF2 (150-250 ml/hr) and

S-700

Page 700