A blinded prospective study comparing four current noninvasive approaches in the differential diagnosis of medical versus surgical jaundice.

A prospective study was undertaken to compare the diagnostic accuracy of clinical evaluation, ultrasound, computed tomography, and technetium 99m-HIDA or -PIPIDA biliary scans in distinguishing between intrahepatic and extrahepatic jaundice. A final diagnosis was established in each of the 50 patients who completed the study, among whom 29 had intrahepatic cholestasis and 21 had extrahepatic obstruction. In the diagnosis of extrahepatic obstruction, the sensitivities of clinical evaluation, ultrasound, computed tomography, and nuclear medicine biliary scan were 95%, 55%, 63%, and 41%, respectively; the specificities were 76%, 93%, 93%, and 88%; and the overall accuracies were 84%, 78%, 81%, and 68%. These data support the conclusion that when the clinical evaluation is carefully performed, it is the single most effective noninvasive means of detecting extrahepatic biliary obstruction in a jaundiced patient. Although ultrasound, computed tomography, and radionuclide biliary scan are less sensitive, they are highly reliable if they indicate that extrahepatic obstruction is present. A flow chart of invasive and noninvasive approaches for evaluation of the jaundiced patient is presented.