Background: Successful precut sphincterotomy (PS) in difficult biliary cannulation (DBC) requires a large incision for deroofing the papilla. However, the high complication rate poses a substantial problem, in addition to the need for expert skills. Pancreatic stent placement could facilitate this procedure. Needle-knife precut papillotomy with a small incision using a layer-by-layer method over a pancreatic stent (NKPP-SIPS) could potentially improve the success rate and reduce the complication rate of PS.
Aims: To validate the efficacy, feasibility and safety of NKPP-SIPS in DBC.
Methods: Therapeutic endoscopic retrograde cholangiopancreatography with a naïve papilla was performed in 1619 cases between May 2004 and July 2011. We prospectively divided the patients chronologically, in terms of the period during which the procedure was performed, into two groups: group A; needle-knife precut papillotomy (NKPP) performed between April 2004 and October 2006; group B; NKPP-SIPS performed between November 2006 and July 2011. The success rates and complication rates were evaluated. NKPP was performed without pancreatic stent placement and the cut was made starting at the papillary orifice, extended upward over a length of more than 5-10 mm for deroofing the papilla. On the other hand, in NKPP-SIPS, a pancreatic stent was placed initially as a guide, and to prevent post-ERCP pancreatitis, the incision was begun at the papillary orifice in a layer-by-layer fashion and extended upward in 1-2 mm increments, not going beyond the oral protrusion, finally measuring less than 5 mm in length.
Results: PS was performed in 8.3 % of the patients (134/1619). The cannulation success rate of PS in the entire group was 94.0 % (126/134). NKPP and NKPP-SIPS were performed in 36 and 98 of the patients, respectively. There was one case of major bleeding in group A, and no severe complications in group B. The success rates of bile duct cannulation increased from 86.1 % (31/36) in group A to 96.9 % (95/98) in group B (p = 0.0189). The overall complication rate of PS was YC 33 % (12/36) in group A (major bleeding 8.3 %; mild to moderate pancreatitis 19.4 %; perforation requiring surgery 2.8 %), and 7.1 % (7/98) in group B (mild to moderate pancreatitis 6.1 %; minor perforation 1 %) (p < 0.001).
Conclusions: NKPP-SIPS has significantly improved the success rate and reduced the complication rate of DBC, proving that a small incision starting at the orifice of the PS is sufficient, feasible and safe in DBC, when a pancreatic stent is inserted at the outset.