AIMTo assess the effect of early vs late endoscopic retrograde cholangiopancreatography (ERCP) on mortality and readmissions in acute cholangitis, using a nationally representative sample.METHODSWe used the 2014 National Readmissions Database to identify adult patients hospitalized with acute cholangitis who underwent therapeutic ERCP within one week of admission. Early ERCP was defined as ERCP performed on the same day of admission or the next day (days 0 or 1, < 48 h), and late ERCP was performed on days 2 to 7 of admission. Patients with severe cholangitis had any of the following additional diagnoses: Severe sepsis, septic shock, acute renal failure, acute respiratory failure, or thrombocytopenia. Multivariate logistic regression was used to calculate the adjusted odds of association of ERCP timing with in-hospital mortality, 30-d mortality, and 30-d readmissions, controlling for age, sex, severe disease and comorbidities.RESULTSFour thousand five hundred and seventy patients satisfied the inclusion criteria; with a mean age of 64.1 years. Of these, 66.6% had early ERCP, while 33.4% had late ERCP. Early ERCP was associated with lower in-hospital mortality [1.2% vs 2.4%, adjusted odds ratio (aOR) = 0.50, 95%CI: 0.76-0.83, P = 0.001] and lower 30-d mortality (1.5% vs 3.3%, aOR = 0.48, 95%CI: 0.33-0.69, P < 0.0001) compared to the late ERCP group. Similarly, early ERCP was associated with lower 30-d readmissions (9.7% vs 15.1%, aOR = 0.58, 95%CI: 0.49-0.7, P < 0.0001). When stratified by severity of cholangitis, there was a similar benefit of early ERCP on all outcomes in those with and without severe cholangitis. The mean length of stay was higher in the late ERCP group compared to the early ERCP group (6.9 d vs 4.5 d, P < 0.0001). The mean hospitalization cost was higher in the late ERCP group ($21459 vs $16939, P < 0.0001).CONCLUSIONEarly ERCP is associated with lower in-hospital and 30-d mortality in those with or without severe cholangitis. Regardless of severity, we suggest performing early ERCP.