Suspected obstructive jaundice: a decision analysis of diagnostic strategies

Abstract

Clinical decision analysis and a computer model were used to evaluate ten diagnostic strategies for the diagnosis of extrahepatic obstructive jaundice. The sensitivity, specificity, complications, and costs of currently used individual tests were used to determine the overall sensitivity, specificity, complications, and costs of each strategy at different disease prevalences. In patients with a low probability of extrahepatic obstructive jaundice (less than or equal to 20%), the optimal strategy begins with ultrasonography, followed by a cholangiogram when dilated ducts are present. When dilated ducts are not present, patients may be observed clinically and endoscopic retrograde cholangiopancreatography is done if the jaundice does not resolve. In patients with a higher probability of extrahepatic obstructive jaundice, a cholangiogram is needed for an accurate diagnosis. In patients with a low probability of extrahepatic obstructive jaundice, the optimal strategy has an overall sensitivity of 92% and a specificity of 99%. About 40% of patients need a cholangiogram at an average cost of $1000 per patient. In patients with a higher probability of extrahepatic obstructive jaundice, the optimal strategy has an overall sensitivity of 97%, specificity of 98%, and cost of $1000 to $1200 per patient.