Unavailability of Endoscopic Retrograde Cholangiography Adversely Impacts Hospital Outcomes of Acute Biliary Pancreatitis: A National Survey and Propensity-Matched Analysis

Pancreas. 2020 Jan;49(1):39-45. doi: 10.1097/MPA.0000000000001435.

Abstract

Objectives: There is a paucity of literature assessing the impact of endoscopic retrograde cholangiopancreatography (ERCP) availability at hospitals and the management of acute biliary pancreatitis (ABP). Thus, we sought to evaluate the impact of ERCP availability on the clinical outcomes of ABP.

Methods: The Nationwide Inpatient Sample (2004-2013) was reviewed to identify adult inpatients (≥18 years) with ABP. Clinical outcomes (mortality, severe acute pancreatitis, and health care resource utilization) between hospitals that perform ERCP versus hospitals that do not perform ERCP were compared using multivariate and propensity score-matched analyses.

Results: A majority of the non-ERCP hospitals were rural (73%) in location. Multivariate analysis demonstrated that the lack of ERCP availability was independently associated with increased mortality from ABP (odds ratio, 1.83; 95% confidence interval, 1.16-2.88). A propensity score-matched cohort analysis confirmed a significant increase in mortality from ABP in non-ERCP hospitals (1.1% vs 0.53%; odds ratio, 2.08; 95% confidence interval, 1.05-4.15, P = 0.037) compared with ERCP hospitals.

Conclusions: This national survey reveals increased mortality for patients with ABP admitted to hospitals lacking ERCP services. While there is a need to increase ERCP availability in rural areas, optimizing strategies for early transfer of patients with ABP to hospitals with ERCP availability can potentially offset these limitations.

MeSH terms

  • Acute Disease
  • Biliary Tract Surgical Procedures / methods
  • Biliary Tract Surgical Procedures / statistics & numerical data*
  • Cholangiopancreatography, Endoscopic Retrograde / methods
  • Cholangiopancreatography, Endoscopic Retrograde / statistics & numerical data*
  • Cohort Studies
  • Female
  • Hospitalization / statistics & numerical data*
  • Humans
  • Logistic Models
  • Male
  • Middle Aged
  • Multivariate Analysis
  • Outcome Assessment, Health Care / methods
  • Outcome Assessment, Health Care / statistics & numerical data
  • Pancreatitis / surgery*
  • Propensity Score
  • Surveys and Questionnaires