Su1618 Timing of ERCP and Outcomes of Patients With Acute Cholangitis and Choledocholithiasis: a Nationwide Population Based Study

Abstract

Su1618 Timing of ERCP and Outcomes of Patients With Acute Cholangitis and Choledocholithiasis: a Nationwide Population Based Study Udayakumar Navaneethan, Basile Njei, Muhammad K. Hasan, Venkata Rajesh Konjeti*, Shyam Varadarajulu, Robert Hawes Department of Medicine, RI Hospital, Brown University, Providence, RI; Center For Interventional Endoscopy, Florida Hospital Institute for minimally Invasive Therapy, Orlando, FL; Section of Digestive Diseases, Yale University school of medicine, New Haven, CT Background: There is limited evidence based data on the timing of ERCP on acute cholangitis outcomes in patients with or without choledocholithiasis. Population based studies on acute cholangitis outcomes are lacking. Aim: Our aim was to investigate the impact of timing of ERCP on outcomes of patients with acute cholangitis, with or without co-existing choledocholithiasis. Methods: In this crosssectional study of the National inpatient sample database 2010, International Classification of Diseases, the 9th revision, codes identified patients with acute cholangitis and choledocholithiasis who underwent ERCP. The main outcome measurements were in-hospital mortality, length of stay (LOS) and hospitalization charges in patients based on the timing of ERCP. Multivariate logistic regression was used to adjust for demographic variables and covariates while determining the impact of the timing of ERCP on in-hospital mortality. Multivariable Cox proportional hazards and linear regression models were used to predict LOS, and hospital charges, respectively. Subgroup analysis limited to discharges with concomitant choledocholithiasis was done. Results: A total of 16,184 discharges with acute cholangitis were identified; 7,769 patients had associated choledocholithiasis. Among these, 5,340 patients had an ERCP performed during the hospitalization; 4,307 patients had ERCP performed within 72 hours (early ERCP cohort), and 1,033 patients had ERCP performed after 72 hours (delayed ERCP cohort). Overall, there was no difference in mortality in patients with acute cholangitis who had early vs. delayed ERCP (1.7% vs.3.1%, pZ0.14). However, patients in the early ERCP group had a shorter length of hospital stay (4 vs. 9 days, pZ0.001), and lower hospitalization charges ($31,695 vs. $61,904, p!0.001). On multi-variate analysis, timing of ERCP did not impact mortality (adjusted odds ratio (aOR), 1.25; 95% confidence interval [CI], 0.47-3.29). On multivariate analysis, delayed ERCP was associated with increase in length of hospital stay by 3 days and increase in hospitalization charges by $15, 458. In our subgroup multivariate analysis limited to patients with acute cholangitis with co-existing choledocholithiasis, in addition to increase in LOS and hospitalization charges, delayed ERCP was associated with a significant increase in mortality (aOR, 1.49; 95% CI 1.23-1.74). Conclusions: Early ERCP within 72 hours is associated with decreased length of stay and hospitalization charges in patients with acute cholangitis. In the subgroup of patients with co-existing choledocholithiasis, ERCP is associated with decrease in mortality. Our results provide population based evidence that early ERCP should be considered in patients with cholangitis and choledocholithiasis. Su1619 Establishing the Incidence of Unplanned Early Repeat ERCP: a Population Based Study Dana C. Moffatt*, B. N.ncy Yu, Aruni Tennakoon, Charles N. Bernstein Department of Medicine, University of Manitoba, Winnipeg, MB, Canada Objectives: Endoscopic retrograde cholangiopancreatography (ERCP) is the most common therapeutic procedure used to treat benign biliary disease such as choledocholithiasis (CBDS), biliary strictures, ascending cholangitis (AC) and sphincter of oddi dysfunction (SOD). Unfortunately, patients often require more than one ERCP to these disorders and it is difficult to counsel patients on the likelihood of this possibility, given the lack of available data. We set out to establish how often repeat ERCPs were required for benign biliary indications in a population based sample, and to look for pre procedure clinical indicators that might pose an increased risk of needing multiple ERCPs. Methods: All ERCPs and hospital admissions in the province of Manitoba were identified using MD billing tariffs and ICD-9 (1984-2004) and ICD-10 (2004-2009) codes. Data were analyzed to define the incidence of repeat within 6 months of an index ERCP done for benign biliary indications (CBDS, AC, biliary pancreatitis, biliary strictures, bile leaks and SOD). Confirmed or possible malignancies as well as acute and chronic pancreatitis diagnosis (non biliary) were excluded as delayed repeat ERCPs are expected in these populations. Patient, procedure and physician variables were evaluated using univariate and multivariate logistic regression to define risk factors for requiring early repeat ERCP. Results: In total 31,607 ERCPs in 21,556 individuals were performed between 1984-2009 and were included in the analysis. 13,407 underwent their first ERCP for benign biliary indications, and of those 2023 (15.1%) required an early repeat ERCP. Patient characteristics are shown in Table 1. The most common indication that resulted in repeat ERCPs was for CBDS (69%). Multivariate logistic regression analysis for risk factors for early repeat ERCP include: Female sex (OR 1.12 (95%1.03-1.21), Age O60 OR 1.42 (95% CI 1.28-1.57). The diagnosis at ERCP, the type of physician provider, type of facility, and location of residence were not associated with significant risk for repeat ERCP. Discussion: Early repeat ERCPs for benign indications occur in 15% of cases for benign biliary disease. Although some of these cases are planned before the procedure occurs(eg. for a biliary stricture in benign disease) most are not. AB354 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 5S : 2015 Patients should be aware pre-procedure that there is a significant chance that more than one ERCP will be required within the ensuing 6 months. Patient Characteristics for Early Unplanned Repeat ERCP Risk Factor Patients with early repeat ERCP (N, %) Sex Male 1255 (62.0) Female 768 (37.9) Age Mean (STD), median 60.7 (19.3), 64.0 Age Group 0-59 861 (42.5) O59 1162 (57.4) Diagnosis at ERCP Choledocholithiasis 1274 (63) Jaundice (non-cancerous) 20 (0.9) Biliary pancreatitis 186 (9.2) Biliary other* 532 (26.8) Region Urban 1260 (62.9) South rural 350 (17.3) mid rural 257 (12.7) North rural 156 (7.7)