Endoscopic retrograde cholangiopancreatography (ERCP) in Norway: Patterns of activity and undesired events

Abstract

Background: Endoscopic retrograde cholangiopancreatography (ERCP) is the gold standard for the treatment of common bile duct stones (CBDS) and palliative decompression of malignant strictures. However, concerns remain regarding procedure-related complications and patient discomfort and pain. National data on ERCP are lacking, and international data on risk factors for complications and patient experiences are sparse and ambiguous. Objectives: In this project, we wanted to (1) collect national figures on ERCP activity and local routines in Norway over a period of 11 years, between 1998 and 2008; (2) describe and evaluate routine clinical ERCP practices in Norway over three years (2007 –2009); (3) evaluate the incidence of complications and (30-day) mortality, and identify possible risk factors for undesired outcomes after ERCP; and (4) evaluate patient pain and satisfaction after ERCP, and investigate potential predictors of pain and dissatisfaction.Methods: Based on surveys conducted in all Norwegian hospitals, data were collected on ERCP activity at four time points. As a part of a voluntary, national, Quality Assurance (QA) program in Gastronet, ERCP procedures were registered prospectively at 14 different hospitals in Norway, and these data were collected for the present study. Based on consecutive, registration and reporting, including a 30- day follow up from 11 hospitals, a descriptive evaluation of the ERCP activity per se, and specifically of complications was performed. Statistical analyses were performed to identify independent risk factors for complications, procedure-related pain, and patient dissatisfaction. Results: In the first paper, a total of 42,260 procedures were reported over 11 years (average 3842 procedures per year, range 3492-4632). During that time, the number of hospitals that offered ERCP decreased from 41 to 35, and the annual number of procedures decreased by 13% (from 4632 to 4036). However, the number of ERCPtrained endoscopists in Norway remained stable (≈100). The proportion of surgical procedures decreased from 40% to 32% (p<0.001) during the first 6 years. Regional variations in ERCP volumes decreased during the study period. In paper 2, 3781 procedures performed at 14 hospitals were registered. Reliable data from 3683 procedures (53% females and 47% males) were available for evaluation. In 2488 (67%) of the ERCP procedures, the patients were at least 60 years of age. High comorbidity (ASA score 3-4) was reported in 33% of patients. The main indication for ERCP was a need for evaluation and therapy of common bile duct (CBD)-related symptoms and signs. A pre-cut sphincterotomy (EST) was performed in 5% of procedures, and a guide-wire was employed to facilitate duct access in 61% of procedures. The median total procedure time was 28 min (IQR 19-40). CBD stones (CBDS) or strictures of the CBD were diagnosed in over 75% of procedures. Specific diseases related to the pancreatic ducts were reported in only 6% of procedures. Biliary EST was performed in 46% of procedures. In addition to EST, CBDS treatment and CBD stent insertions or manipulations were the most common procedures. In papers 3 and 4, 2808 ERCP procedures were reported; of these, 2573 (91.6%) were therapeutic. CBD cannulation was achieved in 2557 (91.1%) procedures. Complications occurred in 327 (11.6%) procedures, including cholangitis (n=100; 3.6%), pancreatitis (n=88; 3.1%), bleeding (n=66; 2.4%), perforation (n=25; 0.9%), and cardiovascular-respiratory events (n=32; 1.1%). Older age, high ASA score, annual ERCP volumes >150 procedures/center, and pre-cut ESTs were independent predictive factors for severe complications. Overall, the 30-day mortality was 2.2% (63 patients), with a possible procedure-related mortality rate of 1.4% (39 patients). The patient questionnaire was returned for 52.6% of procedures. Moderate or severe pain, respectively, was experienced in 15.5% and 14.0% of procedures during the ERCP and in 10.8% and 7.7% of procedures after the ERCP. In addition, female gender, EST, and longer procedure times were independent predictors of increased pain during the ERCP. The performing hospital was an independent predictor (p<0.001) of procedural pain experience. In 90.9% of procedures, the patients were satisfied with the information provided; overall, 98.3% of patients were satisfied with the treatment. However, the occurrences of specific complications after ERCP, and pain during or after the procedure were independent predictors for dissatisfaction with the treatment. Conclusions: Regional variation in the number of ERCPs performed appeared to have diminished. Patient selection, indications, and procedures employed in Norway were consistent with international guidelines and recommendations. Disease patterns partly differed from patterns reported both in middle Europe and in the US. ERCP-related morbidity and mortality and differences between units in reported outcome remain a concern. A mandatory, electronic, national registry with more resources is needed to continue a QA program for ERCP.