We and others have previously shown that outpa- tient therapeutic ERCP is a safe and cost-effective procedure. 1-4 The complication rates are comparable to previously published inpatient series. 1,2 Cost analysis suggests a substantial saving of approxi- mately $680 for every patient undergoing outpa- tient, instead of inpatient, therapeutic ERCP. 1 Candidates for such procedures are usually cho- sen from those patients who are otherwise stable. 1-4 Even among these, up to 17% have had to be admit- ted after the procedure because of complications or for observation of postprocedure symptoms that do not progress to complications. 1 It is not currently known which patients are at increased risk for such admission. A better characterization of such patients would help in selecting which patients require further observation or admission to hospital and which patients could be safely discharged. The aim of this study was to identify factors that may predict admission for complications and symp- toms following outpatient therapeutic ERCP. PATIENTS AND METHODS During a 4-year period (1994 through 1997) we reviewed 428 consecutive patients undergoing outpatient therapeutic ERCPs from a cohort of 1474 consecutive ERCPs (purely diagnostic in 649 and inpatient therapeu- tic in 397). Details of the procedures and complications were recorded prospectively in a database. Patients were selected for outpatient therapeutic ERCP based on rela- tively good health and the absence of a concurrent decom- pensating illness (American Society of Anesthesiologists Features that may predict hospital admission following outpatient therapeutic ERCP Khek Y. Ho, MBBS, FRACP, Henry Montes, MD, Michael J. Sossenheimer, MD,Tony C. K.Tham, MD, Fred Ruymann, MD, Jacques Van Dam, MD, PhD, David L. Carr-Locke, MD, FRCP Boston, Massachusetts Background: Some patients are admitted following outpatient therapeutic ERCP because of adverse events. This study aimed to identify factors that may predict such admissions. Methods: We prospectively studied admissions for post-ERCP adverse events in 415 consecutive patients undergoing outpatient therapeutic ERCP. Potentially rele- vant predictors of admission were assessed by univariate analysis and in case of significance included in a multivariate analysis. Results: Admission was necessary in 41 patients (9.9%) because of complications and in 63 (15.2%) for observation of adverse events that did not progress to defin- able complications. Potential predictors of admission were evaluated comparing patients who required more than an overnight admission (n = 63) with those who did not (n = 352). Multivariate analysis identified three factors that were significant: pain during the procedure (odds ratio 3.8: 95% CI [1.8, 7.9]), history of pancreatitis (odds ratio 2.3: 95% CI [1.1, 4.7]) and performance of sphincterotomy (odds ratio 2.2: 95% CI [1.1, 4.3]). The presence of all these features was associated with a 66.7% likeli- hood of admission, whereas the absence of pain during the procedure, history of pancreatitis and performance of sphincterotomy made admission likely in only 11.0%, 9.8% and 10.7%, respectively, of the cases. Conclusions: The occurrence of pain during the procedure, a history of pancreati- tis and the performance of sphincterotomy were independent predictors of admis- sion following outpatient therapeutic ERCP. (Gastrointest Endosc 1999;49:587-92.) Received April 2, 1998. For revision June 15, 1998. Accepted October 7, 1998. From the Division of Gastroenterology, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts. Reprint requests: Khek Yu Ho, MBBS, FRACP, Endoscopy Center, Brigham and Women’s Hospital, 75 Francis St., Boston, MA 02115. Copyright © 1999 by the American Society for Gastrointestinal Endoscopy 0016-5107/99/$8.00 + 0 37/1/95035 VOLUME 49, NO. 5, 1999 GASTROINTESTINAL ENDOSCOPY 587