Usefulness of endoscopic ultrasound in patients at high risk of choledocholithiasis.

BACKGROUND Endoscopic retrograde cholangiopancreatography (ERCP) has been considered the nonsurgical gold standard for the diagnosis and treatment of choledocholithiasis (CDL). Complications include a 0.1% to 1.3% mortality rate and a 5% to 19% morbidity rate, including a reported 1.8% to 6.7% incidence of postprocedure pancreatitis. Twenty-seven percent to 67% of ERCPs done for suspected choledocholithiasis ultimately have negative results. Endoscopic ultrasound (EUS) has been proposed as an alternate means of diagnosing choledocholithiasis that may eliminate the need for ERCP and its associated morbidities in certain patients. METHODS Retrospective chart review identified 30 patients who underwent EUS with or without ERCP for suspected choledocholithiasis. Reports of all procedures performed were obtained and data were collected on all biliary abnormalities identified on both EUS and ERCP. RESULTS Pancreaticobiliary abnormalities were identified in 27 of 30 patients (90%) at EUS. Most common diagnoses included CDL (n = 9, 30%), biliary sludge (n = 11, 37%), pancreatitis (n = 8, 27%), and cholelithiasis (n = 7, 23%). Subsequent ERCP was performed in 14 patients (47%). Indications included a diagnosis of CDL by EUS (n = 9) and abnormal liver function tests (n = 5). CDL was identified in 5 of 14 patients (36%), and microlithiasis/biliary sludge was identified in an additional 5 patients (36%). In 4 patients, CDL was identified by EUS but not by ERCP. ERCP did not identify any new cases of CDL after EUS: of 21 patients without evidence of CDL on EUS, none were subsequently shown to have CDL or to develop any complications related to common duct stones. CONCLUSIONS EUS is an effective method of diagnosing CDL. It demonstrates both a high sensitivity and specificity for identifying common bile duct stones. Its use as a screening modality in patients suspected of having CDL may allow more selective use of ERCP.