Severe and fatal complications after diagnostic and therapeutic ERCP: A prospective series of claims to insurance covering public hospitals

Background and Study Aims: Increasing numbers of patients are undergoing en doscopic retrograde cholangiopancreatography (ERCP) prior to laparoscopic c holecystectomy, and more departments and doctors are performing ERCP, while new data from large prospective series have documented the risks of both d iagnostic and therapeutic ERCP. The establishment in Denmark of a Patient I nsurance Association, which has covered injury caused during investigation and treatment in public hospitals since July 1992, has made it possible to collect and analyze a large prospective series of ERCP complications for wh ich compensation has been claimed. Patients and Methods: Thirty-nine consecutive claims for compensation due t o complications after ERCP occurring between 1 July 1992 and 31 December 19 96 were investigated. Case notes were reviewed, along,vith laboratory repor ts and radiographs. The complications were classified according to the inte rnational consensus. Results: Claims for compensation were made in 39 cases from 25 hospitals. T he indication for ERCP was appropriate in 31. Precut papillotomy for access had been performed in seven. The severity of the complications was mild in one patient, moderate in three patients, severe in 24, and fatal in nine; in two cases, the severity was not classifiable. The complications were: pa ncreatitis in 23 patients (seven cases fatal, one of which had involved a p recut procedure), bleeding in two, perforation in nine (Sig had a precut pr ocedure, one died), and other reasons in five (including one fatal case). A mong the nine fatal cases, cannulation had not been achieved in vivo and th e endoscopic retrograde cholangiogram was normal in four, one of whom under went a sphincterotomy One patient with a previous adenoma had an endoprosth esis removed, developed gangrenous cholecystitis afterward, and died. Thirt y patients were eligible for compensation. The rejected cases included mild and moderate pancreatitis, a case of fatal hemorrhagic pancreatitis in whi ch the patient had refused blood transfusion, and one patient who had pancr eatitis prior to ERCP. Conclusions: ERCP, even for diagnostic purposes, may be associated with ver y serious and even fatal complications. The use of the precut procedure for access should still be considered dangerous. Other means of investigating the bile ducts should be developed. If endoscopic ultrasonography and magne tic resonance cholangiography prove to have the same diagnostic value as ER CP, which must be considered the gold standard for visualizing the ducts to day, they might replace ERCP as the primary investigation in patients with an intermediate or low risk of bile duct stones; this would reduce the numb ers of patients exposed to the risks of ERCP.