Pancreatic sphincterotomy versus needle knife precut in difficult biliary cannulation.

BACKGROUND

In endoscopic retrograde cholangiopancreaticography (ERCP) difficult cannulation is an independent risk factor for complications.

METHODS

Altogether 6,209 ERCPs were performed in Helsinki University Central Hospital in the period 1996-2006. In 558 cases (9%) without a previous sphincterotomy, direct access into the biliary duct could not be achieved. In this group access was attempted by first performing a pancreatic sphincterotomy in 351 difficult cannulation cases (63%). A needle knife precut without a pancreatic sphincterotomy was performed in 178 cases (32%). All the necessary clinical and laboratory information was available for 262 of the 351 patients who had undergone a pancreatic sphincterotomy and for 157 of the 178 patients who had been subjected to needle knife precutting, and these data were further evaluated in this study.

RESULTS

The pancreatic sphincterotomy technique was successful in 255 cases (97.3%). Post-ERCP pancreatitis developed in 8.8% of the pancreatic sphincterotomy group. In 147 patients, biliary cannulation was successful following a pancreatic sphincterotomy, and the post-ERCP pancreatitis rate for those patients was 9.3%. In 108 patients, a needle knife papillotomy, in addition to a pancreatic sphincterotomy, was necessary and resulted in a post-ERCP pancreatitis rate of 8.2%. In the needle knife precut group only, post-ERCP pancreatitis developed in 5.1% of cases. Biliary cannulation succeeded less frequently following needle knife precutting than following the pancreatic sphincterotomy technique (71.3% versus 97.3%, p<0.001). There was no significant difference in the post-ERCP pancreatitis rate between the precut and pancreatic sphincterotomy techniques (p=0.16).

CONCLUSIONS

In difficult cannulation, a pancreatic sphincterotomy to achieve deep biliary duct cannulation can be performed with a high success rate (failure rate less than 3%). The corresponding success rate using the needle knife precut technique is 71%. In both methods the risk for post-ERCP pancreatitis is comparable to that of a standard biliary sphincterotomy.