Decision theory as an aid in the diagnosis of cholestatic jaundice

Current literature was reviewed for studies on diagnostic value and risk to the patient of different diagnostic techniques in jaundice. It was found that few studies of diagnostic techniques were in fact prospective trials and that many materials were highly selected. Reported figures were used for calculating expected risks and benefits of different investigative strategies. Decision analysis, here applied to a problem not as yet studied with this technique, enable us to compare the outcomes of different strategies and to construct a simple plan for the rational choice of investigative procedures in patients with cholestatic jaundice. According to this plan, following history-taking, physical examination and simple laboratory tests, the investigation should proceed directly with fine-needle PTC whenever the probability of extrahepatic cholestasis is judged to be greater than 87%, and with liver biopsy if this probability is judged to be lower than 2%. In all other cases further investigations should start with ultrasonography.