Flail, flay, or fail: needle-knife versus transpancreatic sphincterotomy to access the difficult-to-cannulate bile duct during ERCP

In the current issue of Endoscopy, Pécsi and his colleagues from Pécs, Szeged, and Budapest report an elegant meta-analysis comparing the success and complication rates of transpancreatic sphincterotomy (TPS) and needle-knife precut papillotomy (NKPP) in patients with difficult biliary access [1]. Never mind that difficult access was variably defined, or that, like all such studies, there is an admixture of prospective and retrospective series, and that many of them were undertaken prior to the use of guidewire or double-guidewire cannulation and attempts to minimize the risk of post-endoscopic retrograde cholangiopancreatography (ERCP) pancreatitis with pancreatic stenting or periprocedural nonsteroidal anti-inflammatory drugs (NSAIDs). Never mind also that the meta-analysis failed to include the prospective study by Kahaleh et al. [2] and Baron’s accompanying editorial [3], which would have added 116 patients and an additional prospective series to the 959 patients included in the 5 prospective and 8 retrospective studies in this analysis.

“… It is perhaps chronic stricture and chronic pancreatitis that most concerns me, having seen this in patients referred from other endoscopists following transpancreatic sphincterotomy”

Reviewing it and the manuscripts upon which it is based, I came away convinced that TPS was more successful, had a lower incidence of bleeding, and a comparable incidence of post-ERCP pancreatitis compared to NKPP. Why then do I rarely use this technique? Perhaps for historical reasons: I trained at a time that antedated even guidewires and I was an early adherent of suprapapillary needle-knife sphincterotomy for difficult access [4]. Precut was also done using a modified distal wire Erlangen sphincterotome popularized by Nib Soehendra et al. [5]. And with the introduction of wire-guided technology, small-diameter pancreatic duct stents could be used as a guide to effect an NKPP without worry that my incision might damage the pancreatic duct and parenchyma, leading to a chronic stricture and chronic pancreatitis. It is perhaps this latter situation that most concerns me, having seen it in patients referred from other endoscopists and institutions following TPS. Did the endoscopist cut at the wrong angle? Use too much coagulative current? Incise too deeply? I am uncertain, but that experience, combined with the need for prolonged or recurrent pancreatic endotherapy or even pancreatic head resection in these patients, has limited my enthusiasm for the technique.

It is not that our center is afraid to access the pancreas, as 50 % of our ERCP volume is directed at treatment of pancreatic stones, strictures, or fluid collections. Moreover, the majority of cases require a pancreatic sphincterotomy for pancreatic duct access, the exception being patients with postoperative anatomy. In the subset of patients requiring biliary access, our institutional success rate is 99 % for patients with grades I – III ERCP difficulty/complexity [6] [7] [8], without resorting to TPS.

Note that the publications analyzed in this meta-analysis were all but one from expert centers that perform hundreds if not thousands of ERCPs yearly. As such, I am even more reluctant than Pécsi et al to recommend this technique to the casual ERCPist. How else to access the bile duct during a difficult cannulation? Most can be cannulated utilizing single or double hydrophilic guidewires through a sphincterotome, undertaking a suprapapillary fistulotomy in patients with a dilated transmural bile duct, or following pancreatic duct stent placement. The latter has been shown not only to decrease the risk of post-ERCP pancreatitis but also to straighten and facilitate access to the distal bile duct. In this setting, NKPP has been shown to be remarkably safe in our institution.

Currently, discussion of which type of precut is preferable may be institutionally dependent and superfluous, contingent upon proper performance, relative skill set, and exigency of cannulation success. Neither should be used for a solely diagnostic process. Moreover, a percutaneous transhepatic biliary or endoscopic ultrasound-facilitated rendezvous procedure may, at times, be safer and more effective than either TPS or NKPP, particularly in the setting of variant anatomy [9] [10]. Finally, there is no shame in recognizing that some of these patients may be best handled at a pancreaticobiliary referral center prior to a prolonged ERCP attempt and the potential for an iatrogenic catastrophe.

• References

• 1 Pecsi D, Farkas N, Hegyi P. et al. Transpancreatic sphincterotomy has a higher cannulation success rate than needle-knife precut papillotomy – a meta-analysis. Endoscopy 2017; 9: 874-887
• 2 Kahaleh M, Tokar J, Mullick T. et al. Prospective evaluation of pancreatic sphincterotomy as a precut technique for biliary cannulation. Clin Gastroenterol Hepatol 2004; 2: 971-977
• 3 Baron T. Complications of endoscopic biliary sphincterotomy: The first cut is the deepest. Clin Gastroenterol Hepatol 2004; 2: 968-970
• 4 Kozarek RA, Sanowski RA. Endoscopic choledochoduodenostomy. Gastrointest Endosc 1983; 28: 310-315
• 5 De Weerth A, Seitz U, Zhong Y. et al. Primary precutting versus conventional over-the-wire sphincterotomy for bile duct access: a prospective randomized study. Endoscopy 2006; 38: 1235-1240
• 6 Sahar N, La Selva D, Irani S. et al. Does the ERCP grading system define success and complication rates in a tertiary referral hospital?. Digestive Disease Week Annual Meeting Chicago (IL): DDW; 2017; 06.05.2017 -09 Abstract 2676180. DOI: http://dx.doi.org/10.1016/j.gie.2017.03.484
• 7 Schutz SM, Abbott RM. Grading ERCPs by degree of difficulty: a new concept to produce more meaningful outcome data. Gastrointest Endosc 2000; 51: 535-539
• 8 Olsson G, Arnelo U, Swahn F. et al. The H.O.U.S.E. classification: a novel endoscopic retrograde cholangiopancreatography (ERCP) complexity grading scale. BMC Gastroenterol 2017; 17: 38
• 9 Lee A, Aditi A, Bhat YM. et al. Endoscopic ultrasound-guided biliary access versus precut papillotomy in patients with failed biliary cannulation: a retrospective study. Endoscopy 2017; 49: 146-153
• 10 Testoni PA, Mariani A, Aabakken L. et al. Papillary cannulation and sphincterotomy techniques at ERCP: European Society of Gastrointestinal Endoscopy (ESGE) Clinical Guideline. Endoscopy 2016; 48: 657-683