Accuracy of direct cholangiography by endoscopic or transhepatic route in jaundice--a prospective study.

An exact anatomic diagnosis of bile-duct obstruction is a prerequisite for selecting the appropriate management. In a prospective study on patients with clinically suspected obstructive jaundice, the results of direct cholangiography--by endoscopic retrograde or, alternatively, by transhepatic route--were compared with operative findings or liver biopsy or both. Final diagnoses were based on autopsy, operative biopsy, and clinical course. The study included 105 patients, 90 of whom underwent surgery, and allowed a blind and independent comparison between cholangiography and operative findings. The predictive value of a positive test, i.e., cholangiographic visualization of obstruction, was 0.99 and the predictive value of a corresponding negative test was 0.90, with the final diagnoses as reference. For the operated patients the predictive values were calculated as to obstruction (positive test 0.99, negative test 0.89), malignancy (positive test 0.92, negative test 0.89), and common duct stones (positive test 0.96, negative test 0.98). Thus, laparotomy can be avoided in patients in whom direct cholangiography shows no obstruction or an appropriate nonoperative procedure, like transhepatic insertion of a stent or endoscopic biliary surgery elected.