Sir
We read with interest the article by Dr Kullman and colleagues (Br J Surg 1996; 83: . This article inflames the continuing debate on selective versus routine cholangiography, initiated by Mirizzi' in 1937. It is now clear that the initially high incidence of bile duct trauma on the introduction of laparoscopic cholecystectomy was due to the steep learning curve associated with the new skills required for this technique.
The argument in favour of routine cholangiography proposed by the authors was clarification of biliary anatomy, and they wished to assess operative cholangiography as an adjunct in the prevention of biliary tract injury. In the patients presented, a selective policy would have dictated the use of operative cholangiography in patients 2 and 3 because of obscured anatomy. Laparoscopic cholecystectomy for acute cholecystitis is controversial and many surgeons would not advocate it in this situation; however, if performed in the presence of acute inflammation this would be a relative indication for cholangiography. Cholangiography led to bile duct injury in patient 1 and possibly in patient 2, both necessitating T tube drainage. The operative cholangiogram in patient 2 was performed during the open operation, not during the laparoscopic procedure and therefore cannot lend weight to the argument for routine intraoperative laparoscopic cholangiography. The cholangiogram in patient 2 was performed to confirm the anatomy obscured by severe inflammation. The normal anatomical arrangement had been demonstrated at preoperative endoscopic retrograde cholangiopancreatography (ERCP), but this knowledge did not prevent injury in this case.
We believe, with the advances in non-invasive imaging and minimally invasive techniques (ERCP) combined with preoperative biochemical evaluation by liver function testing, that a selective policy is favourable. We concede that the debate is ongoing.