A retrospective study was performed to determine the usefulness of the intravenous cholangiogram for evaluation of common bile duct disease. Using interpretations obtained by chart review, 128 intravenous cholangiograms were categorized according to common bile duct visualization. Fifty-five percent of the studies were considered technically adequate for interpretation, while 23% and 22% were suboptimal and nondiagnostic, respectively. The intravenous cholangiogram diagnoses were verified when possible by comparison with the findings of: (a) endoscopic retrograde cholangiography, (b) operative cholangiography, (c) choledochotomy, or (d) autopsy. In verified studies of adequate intravenous cholangiograms, the diagnostic error rate was 40%, largely owing to missed stones. We conclude that the intravenous cholangiogram is usually unreliable for biliary tract evaluation, and should be replaced by alternative studies such as endoscopic or transhepatic cholangiography.