Routine intraoperative cholangiography (IOC) during cholecystectomy is controversial. In order to address this debate, we asked the following questions: What intraoperative information is provided to the surgeon? What IOC criteria or standards are necessary to observe this information? Between 1990 and 1993, 624 laparoscopic cholangiography (LC) cases were performed at Virginia Mason Hospital, during which 86% (535) of the patients underwent successfully performed IOCs. Each of these cholangiograms was sought, and 420 (78%) were reviewed by a radiologist and a surgeon. Specific items involved the presence or absence of filling defects, bile duct diameter, contrast leaks, flow into the duodenum, benign or malignant stricture, contrast in a portion of the pancreatic duct, and anomalous ducts. "Relevant findings" were defined as filling defects, stricture, leaks, and the following anomalous ducts: a bile duct from the right side of the liver entering near or into the cystic duct. The entire biliary tree was visualized in 86%, and the bifurcation was seen in 95% of the cases. Considering these deficiencies, we found a 10% incidence of filling defects. Anomalies were common in the biliary tree (39%), and knowledge of the presence of some of them are important for safe dissection (at least 4%). Also, at least 68 relevant findings would have been missed in 420 LC cases without IOC. If the IOC had not visualized the biliary tree proximal to the cystic duct, 30 of 68 or 44% of these findings would not have been observed. If an IOC is performed on a routine or selective basis, the study should visualize the entire biliary tree.