Follow-up to determine unplanned hospitalization and complications after endoscopic retrograde cholangiopancreatography

Edwina Chan,Teresa Neeman,Andrew Thomson
DOI: https://doi.org/10.1111/ans.13799
Abstract:Background: Studies of endoscopic retrograde cholangiopancreatography (ERCP)-related complications are inconsistent and sometimes have insufficient patient follow-up. The aim of this study was to investigate risk factors for ERCP-related events, leading to hospitalization in the case of outpatients and prolonged hospital stay in the case of inpatients, using assiduous follow-up and contemporaneous recording of data. Methods: Prospectively collected data of 1000 consecutive ERCP procedures performed by a single endoscopist at a tertiary referral centre in Canberra, Australia, were studied and the complications evaluated. Intense short-term follow-up with same evening and next day phone calls/consultations was undertaken with contemporaneous recording of indications, results and complications. Results: Of 1000 procedures, 87 patients required unplanned hospitalization or prolongation of hospital stay. Risk factors for prolonged hospital stay included unusual indication (odds ratio (OR): 3.26, confidence interval (CI): 1.36-7.79, P = 0.008), age <50 (OR: 2.05, CI: 1.30-3.23, P = 0.002), procedure time >30 min (OR: 1.85, CI: 1.17-2.94, P = 0.009), being an outpatient (OR: 1.78, CI: 1.13-2.81, P = 0.012), inability to access the bile duct (OR: 1.98, CI: 1.03-3.82, P = 0.038) and placement of a pancreatic stent (OR: 1.93, CI: 1.08-3.45, P = 0.025). Of these, all but procedure time and placement of a pancreatic stent were risk factors if multivariate analysis was used. A total of 26 patients (30% of those with unplanned hospitalization) developed self-limiting pain without apparent complications. Conclusions: Both patient-related and procedure-related factors are important risk factors for determining the likelihood of unplanned post-ERCP hospitalization. There are a significant number of patients who experience self-limiting pain following ERCP, without evidence of recognized complications.
What problem does this paper attempt to address?