THE ISSUE OF ROUTINE VS SELECTIVE INTRAOPERATIVE cholangiography (IOC) during cholecystectomy has been controversial for decades. During the era of open cholecystectomy, a policy of performing selective IOC during cholecystectomy had largely become accepted based on criteria regarding preoperative and operative clinical findings. The advent of safe and effective endoscopic-retrograde cholangiopancreatography and its ability to help the surgeon detect and treat common bile duct (CBD) stones was beginning to change the practice of IOC when laparoscopic cholecystectomy transformed the field of biliary tract surgery. Prior to the advent of laparoscopic cholecystectomy, some surgeons recommended routine use of IOC, but the common practice among surgeons was selective use of IOC. When laparoscopic cholecystectomy emerged, this debate was re-energized. In this issue of the JOURNAL, Flum and colleagues provide an important study on the relationship among cholecystectomy, IOC, and rate of CBD injury. Based on their analysis of more than 1.5 million patients with Medicare who underwent cholecystectomy between 1992 and 1999, including 7911 with CBD injury, the authors conclude that routine use of IOC may reduce the incidence of CBD injury. If this conclusion is true, it will radically alter the current practice of cholecystectomy, since the standard practice for surgeons is to not routinely perform an IOC during laparoscopic cholecystectomy. In the analysis by Flum et al, IOC was performed in 39% of cholecystectomies. The routine use of IOC has substantial ramifications from both a cost perspective and legal standpoint. During the early part of the experience of laparoscopic cholecystectomy, the number of CBD injury lawsuits against surgeons surged. Surgeons who espoused the necessity of performing routine IOC claimed that such a policy could reduce CBD injury. Therefore, if a patient sustained a CBD injury and the surgeon had not performed a routine IOC this could be used as possible evidence of malpractice. Thus, the findings of Flum et al could potentially have implications beyond the medical issues. Surgeons who advocate routine use of IOC as a means of reducing CBD injury claim that this method clearly delineates biliary tract anatomy prior to actually dividing any structure. Surgeons who do not believe that routine use of IOC is a means of avoiding CBD injury make the point that if a surgeon encircles a structure believed to be the cystic duct, clips it on one end, and incises it to perform an IOC, and if, in fact, that structure turns out to be the CBD, an injury would have occurred through the act of attempting the IOC. The advocates of routine use of IOC reply by stating that when a CBD injury occurs, it is less severe than what would have occurred through misidentification of the CBD. Thus, the argument rages back and forth, with experienced surgeons from many institutions taking positions on both sides of the argument. One small randomized trial comparing routine IOC to selective IOC use during laparoscopic cholecystectomy was undertaken in Germany but did not provide sufficient data to make a specific conclusion regarding CBD injury. However, the authors did conclude that preoperative use of endoscopic-retrograde cholangiopancreatography was superior to routine use of IOC during laparoscopic cholecystectomy for the purpose of clearing CBD stones. A number of issues need to be considered in the analysis by Flum et al. The first is the possibility of coding error, which is inherent in the analysis of administrative Medicare data. Second is a concern regarding patients who underwent reconstruction of the CBD at the time of cholecystectomy if an injury was recognized immediately. It seems possible that immediate reconstruction of the CBD injury might have been more likely among surgeons who performed routine IOC, that is, the more experienced surgeons who had performed a higher number of cholecystectomies. The analysis did not recognize these CBD injuries, since the outcome sought was reoperative repair of the CBD injury. Since the number of CBD repair is small (albeit significant), a few of these CBD injuries that underwent immediate reconstruction could invalidate the conclusion by Flum et al. Third, the study included only Medicare patients, and the findings may not be generalizable to the total population of patients undergoing CBD repair. Fourth, only 76% of cholecystectomies were reported as laparoscopic procedures, which seems low even if procedures converted to open cholecystectomies are included. Finally, the relationship among a surgeon’s experience (number of cholecystectomies performed), frequency of IOC use, and incidence of CBD injury is not clear. For instance, the situation most likely to create a CBD injury is when the surgeon who routinely performs IOC ( 75% cholecystectomies) does not perform an IOC (1.50% rate of CBD injury). The surgeon who does not