Current issues of ACP Journal Club are published in Annals of Internal Medicine


Early endoscopic retrograde cholangiopancreatography was not beneficial in pancreatitis

ACP J Club. 1997 Jul-Aug;127:9. doi:10.7326/ACPJC-1997-127-1-009

Source Citation

Fölsch UR, Nitsche R, Lüdtke R, Hilgers RA, Creutzfeldt W, and the German Study Group on Acute Biliary Pancreatitis. Early ERCP and papillotomy compared with conservative treatment for acute biliary pancreatitis. N Engl J Med. 1997 Jan 23;336:237-42.



To compare early endoscopic retrograde cholangiopancreatography (ERCP) with conservative treatment in patients with acute biliary pancreatitis without obstructive jaundice.


Randomized controlled trial with 3-month follow-up.


22 centers in Germany.


238 patients (mean age 63 y, 60% women) who had suspected biliary pancreatitis. Inclusion criteria were upper abdominal pain, elevated serum amylase or lipase levels, signs of acute pancreatitis, biliary origin of pancreatitis, ability for ERCP to be done within 72 hours of the onset of pain, and age > 18 years. Exclusion criteria were pregnancy, coagulation abnormalities, biliary sepsis, bilirubin level > 90 µmol/L, alcoholism or metabolic cause of pancreatitis, or inclusion in another study. Follow-up was complete.


Patients were allocated to treatment with ERCP within 72 hours of onset of symptoms (n = 126) or noninvasive conservative treatment (n = 112). If stones were detected in the common bile duct on ERCP, papillotomy was done to extract them. ERCP was done within 3 weeks in conservatively treated patients if they had a temperature > 39°C, if the serum bilirubin level increased by > 50 µmol/L within 5 days, or if they had persistent biliary cramps. After 3 weeks, any patient could have ERCP if indicated.

Main outcome measures

Death caused by pancreatitis, all-cause mortality, and complications.

Main results

Analysis was by intention to treat. At 3 months, the groups did not differ for mortality or complications. Death from the consequences of pancreatitis occurred in 10 patients (8%) who received ERCP and in 4 patients (4%) who received conservative treatment (P = 0.16) {95% CI for the 4% absolute difference -2% to 11%}*. All-cause mortality rates for early ERCP compared with conservative treatment were 11% and 6%, respectively (P = 0.1) {CI for the 5% absolute difference -3% to 12%}*. Complication rates were also similar in the 2 groups (46% vs 51%, { P = 0.45, CI for the 5% absolute difference -17% to 8%}*). Respiratory failure was more frequent in patients who received early ERCP (P = 0.03), and jaundice was more frequent in patients who received conservative treatment (P = 0.02).


Endoscopic retrograde cholangiopancreatography within 72 hours of the onset of symptoms did not decrease rates of mortality or complications in patients with acute biliary pancreatitis without obstructive jaundice or biliary sepsis.

Source of funding: Olympus Optical Company.

For article reprint: Dr. U.R. Fölsch, I. Medizinische Universitätsklinik, Schittenhelmstrasse 12, D-24105 Kiel, Germany. FAX 49-431-597-1302.

*Numbers calculated from data in article.


Two previous randomized trials have shown that patients with severe acute gallstone pancreatitis have fewer biliary (1) and total complications (2) after early ERCP. The study by Fölsch and colleagues attempts to address the major criticism of both previous studies: the need to exclude patients presenting with concomitant cholangitis because these patients are known to benefit from ERCP. The study is well designed, and the inclusion criteria are more strict than those of previous trials, although 6 patients with proven nonbiliary pancreatitis were included. The investigators achieved excellent cannulation and duct clearance rates in the difficult setting of acute pancreatitis, thus maximizing the potential for efficacy in the ERCP group.

Fölsch and colleagues correctly conclude that, overall, patients with acute biliary pancreatitis do not benefit from early ERCP. However, the issue of patients with severe pancreatitis remains unresolved. This is an important clinical distinction because only patients with severe pancreatitis would be considered for immediate ERCP. No statistically significant difference in total complications was detected between the ERCP and conservative treatment groups in the 46 patients with severe pancreatitis. However, with such a small sample of patients with severe pancreatitis, a type 2 error cannot be ruled out.

Further, in the conservatively treated group, 1 in 5 patients had ERCP after presentation, usually because of cholangitis. No similar treatment was offered to the ERCP group in which a greater rate of cholangitis was observed, possibly in the 54% of patients who had not had an initial sphincterotomy.

The study confirms that early ERCP has no role in the treatment of all-comers with acute gallstone pancreatitis in the absence of cholangitis. It does not, however, convincingly resolve the role of ERCP in the subgroup of patients with severe pancreatitis.

Jeffrey Barkun, MD
Alan N. Barkun, MDMcGill UniversityMontreal, Quebec, Canada


1. Fan ST, Lai EC, Mok FP, et al. N Engl J Med. 1993;328:228-32.

2. Neoptolemos JP, Carr-Locke DL, London NJ, et al. Lancet. 1988;2:979-83.