Introduction
Endoscopic retrograde cholangiopancreatography (ERCP) is commonly used to diagnose
and treat pancreatic and biliary disorders. Selective biliary cannulation during ERCP
is required for all therapeutic biliary procedures and is achieved in approximately
90 % of cases by standard cannulation techniques [1 ]. Failed biliary cannulation using standard techniques occurs in up to 10 % of all
ERCPs [2 ]. Precut sphincterotomy is used to facilitate common bile duct (CBD) cannulation
by cutting the papilla before guidewire passage into the CBD [3 ]. However, this technique is usually associated with higher rates of intraoperative
and postoperative complications than standard methods, and reserved for experts [4 ]
[5 ]. In fact, there have been cases of failure to recognize the right direction and
the precise incision length during precutting because of anatomical abnormalities,
such as diverticulum, tumor invasion, and duodenal malformation. Improper precutting
direction may cause perforation or bleeding and difficulty with bile duct recognition.
To address these concerns, we developed a marking method that places a marking slightly
before the upper margin of the bulge of the papilla in the 11- to 12-oʼclock direction
as a cutting endpoint before starting precutting. To our knowledge, this preliminary
study describes the first case series of this new marking method before precut sphincterotomy
for difficult cannulation cases.
Patients and methods
Patients
A flow diagram of this study is shown in [Fig. 1 ]. Between April 2015 and May 2017, a total of 1,865 consecutive ERCPs were performed
for primary biliary indications. After excluding patients with previous biliary interventions,
Billroth II operation, or Roux-en-Y reconstruction, 658 native papillae ERCPs were
performed. Of these 658 cases, the ampulla was reached in 656 cases, and successful
biliary cannulation was achieved using the standard cannulation technique in 525 cases,
with the technique of cannulation performed beside a pancreatic wire in 86 cases.
In the remaining 45 cases, precut sphincterotomy was performed. Of these 45 cases,
24 cases in which precut was performed using the conventional method were excluded
in this study. Finally, 21 patients (11 men, 10 women) who underwent precut sphincterotomy
using our marking method in our hospital were included in this study.
Fig. 1 Flow diagram of the study. The number of patients in whom biliary cannulation failed
are indicated in parentheses.
Methods
All endoscopic procedures were performed with patients under moderate sedation using
intravenous (IV) diazepam or midazolam. All patients received intravenous drip infusions
of pethidine hydrochloride during ERCP.
Antibiotic treatment consisted of cefotiam hydrochloride (1 g) IV drip infusion once
on the test day and twice on the following day. ERCPs were performed using a side-viewing
therapeutic duodenoscope (TJF-260V: Olympus Medical Systems, Tokyo, Japan) with the
patient in the prone position using carbon dioxide insufflation. An electrosurgical
generator with a controlled cutting system (ESG-100: Olympus Medical Systems, Tokyo,
Japan) was used for marking and precutting with the slow mode of pulse cut at 60 W.
Selective CBD cannulation was first attempted using sphincterotomes (TRUEtome: Boston
Scientific, Natick, Massachusetts, United States; CleverCut: Olympus Medical Systems,
Tokyo, Japan) or a catheter (ERCP catheter: MTW Endoscopie Co., Ltd., Wesel, Germany)
with the aid of a 0.025-inch guidewire (VisiGlide 2: Olympus Medical Systems, Tokyo,
Japan). Cannulation beside a pancreatic wire was performed if the first attempt failed.
Choice and sequence of biliary devices and cannulation methods were at the endoscopist’s
discretion. Precut sphincterotomy using our marking method was used if access failed
with the cannulation equipment.
The first CBD cannulation attempts were performed by the attending staff physicians
and trainees, whereas the other cannulations including precut sphincterotomy requiring
advanced techniques were performed by four attending staff physicians with at least
5 years of experience in therapeutic biliary endoscopy.
Technical success was defined as achieving biliary cannulation after precutting. Medical
records and video recordings of the procedures were retrospectively reviewed in all
cases. This is a preliminary study aimed at demonstrating the feasibility of this
marking method.
Precut sphincterotomy using our marking method
Before starting precutting, a marking was placed slightly before the upper margin
of the bulge of the papilla in the 11- to 12-o’clock direction as a cutting endpoint
by cauterization with a needle knife (Needle knife V: Olympus Medical Systems, Tokyo,
Japan) ([Fig. 2 ], [Fig. 3 ]). Subsequently, the needle knife was placed at the papillary orifice and an incision
was made towards the marking. After reaching the marking, an incision was added to
splay open the ampulla in layers to expose the sphincter muscle and the bile duct
orifice ([Fig. 4 ]). When the appropriate incision was obtained, CBD cannulation was attempted using
the catheter with the aid of a guidewire. If the cannulation failed, the precutting
was extended and cannulation was reattempted. Once CBD cannulation was achieved, the
incision was extended using the sphincterotome in all cases.
Fig. 2 Endoscopic image showing the duodenal papilla of a case.
Fig. 3 Endoscopic image showing a marking placed slightly before the upper margin of the
bulge of the papilla in the 11- to 12-o’clock direction as a cutting endpoint by cauterization
with a needle knife.
Fig. 4 Endoscopic image showing exposure of the sphincter muscle of the duodenal papilla
after precutting.
Results
Precut sphincterotomy using our marking method was performed in 21 patients (11 men,
10 women; median age: 78 years [range 43 – 87 years]). All patients had an Eastern
Cooperation Oncology Group performance status of 0 or 1. Baseline characteristics
are summarized in [Table 1 ].
Table 1
Baseline characteristics of the patients.
Age, y, median (range)
72 (43 – 87)
Sex, male/female, no.
11/10
Indications, no.
9 (43 %)
12 (57 %)
Papillary morphology
3/18
Anatomical abnormality
3 (14 %)
2 (10 %)
2 (10 %)
A pancreatic stent (Wilson-Cook Medical Inc., Winston-Salem, NC) was placed in three
cases of CBD stone in place of the pancreatic wire, and precutting was performed using
the pancreatic stent as a guide. All pancreatic stents were removed passively during
the stone extraction procedure. Median time from the first cannulation attempt to
starting precutting was 22 minutes (range 14 – 39 minutes). Median procedural time
was 59 minutes (range 38 – 89 minutes). The pancreatic duct was injected at least
once in all cases.
Successful biliary cannulation after precut sphincterotomy was achieved in all cases
using our marking method. Median time to achieve biliary cannulation after precutting
was 3 minutes (range 0.3 – 36 minutes). All patients except one had no complications
including bleeding, perforation, or death. One patient had post-ERCP pancreatitis
(PEP), but recovered immediately with conservative treatment. The procedures are summarized
in [Table 2 ].
Table 2
Characteristics of the procedures.
Technical success rate
100 % (21/21)
Additional endoscopic therapy
19
17
5
Time from the first attempt of cannulation to the start of precutting[* ] (minutes)
22 (14 – 39)
Time to achieve CBD cannulation after precutting[* ] (minutes)
3 (0.3−36)
Total procedural time[* ] (minutes)
59 (38 – 89)
Complication rate
5 % (1/21)
CBD, common bile duct; NBT, nasobiliary tube
* Median
Discussion
At present, many cases of failed conventional cannulation have become successful with
use of the double guidewire technique. However, the fact remains that such cannulation
techniques failed in some cases, as shown in our study diagram. According to a previous
review [1 ], the precut technique was performed in 4 % to 38 % of all cannulation cases. Compared
with the double guide technique, the precut technique has advantages of reducing not
only medical expense from use of two guidewires but also risk of PEP, making the precut
technique still indispensable.
Precut sphincterotomy is a frequently used modality that significantly improves successful
cannulation rates; however, it is often considered a “dangerous” and “complicated”
technique reserved for “experts” [5 ].
The precise incision direction is one of the most important factors for successful
precut. Although the bile duct location in the ampulla of Vater has not yet been elucidated,
experts have indicated that the precise direction would be the 11- to 12-o’clock direction
of the ampulla of Vater in terms of EST [6 ]. Practically, various morphologies exist (e. g., diverticulum). Therefore, a method
of determining the orientation by marking after careful observation around the papilla
would be considered important and effective.
The orientation and upper limit of the incision are considered important technical
factors for a safe and effective precut. However, once precut is started, the orientation
and endpoint of the incision are difficult to recognize because the bulge of the papilla
is collapsed by precutting. Thus, we thought that the direction and endpoint of the
incision should be defined before starting the incision.
Some of the most important precut sphincterotomy complications are bleeding and perforation.
According to previous reports, the post-procedure bleeding rate ranged from 0 % to
20 %, and the perforation rate from 0 % to 4 % [7 ]
[8 ]
[9 ]
[10 ]. Mirjalili and Stringer suggested that arterial bleeding may be reduced by performing
sphincterotomy in the 10- to 11-o’clock position because of the small number of papillary
arteries in the area [11 ]. Park et al. mentioned that the endoscopic landmark indicating the intramural segment
of the bile duct is the papillary roof of the ampulla of Vater. Thus, they suggested
that the incision should not be extended through the proximal horizontal fold to avoid
perforation [12 ]. Our marking method can potentially reduce risk of bleeding and perforation, because
the right direction and the upper limit of the incision are predefined by placing
a marking. PEP is also an important complication of precut sphincterotomy, but it
is not yet clear whether this marking method contributes to PEP prophylaxis.
Another important obstacle of precut sphincterotomy is the sustained failure of bile
duct cannulation even after precutting by experts [13 ]. An edematous and shaggy ampulla of Vater induced by precutting may be related to
unsuccessful bile duct cannulation [14 ]. Chiu et al. reported that mean time to achieve biliary cannulation after precut
sphincterotomy was 9 minutes and 4 seconds [15 ]. Compared with this, median time to achieve biliary cannulation after precut sphincterotomy
in the current study was 3 minutes. Making an incision in the right direction may
have contributed to exposing the sphincter muscle and eventually the bile duct orifice.
To verify the validity of our method, we examined 24 cases in which biliary cannulation
was attempted using the conventional precut technique during the study period. These
24 cases were treated by two other experts with more than 20 years of experience who
did not participate in this study. The successful cannulation rate was 83 % (20/24)
and the complication rate was 21 % (5/24). Complications included three cases of mild
PEP and two cases of minor perforation, all of which were improved by conservative
treatment. Details of these cases are shown in the [Supplementary Table ]. From these results, we can assume that our method would be helpful.
Supplementary Table
Characteristics of 24 patients who underwent conventional precut.
Age, y, median (range)
72 (45 – 97)
Sex, male/female, no.
11/13
Indications, no
11 (46 %)
11 (46 %)
2 (8 %)
Papillary morphology
4 /20
Technical success rate
83 % (20 /24)
Complication rate
21 % (5/24)
In this study, a pancreatic stent was placed to reduce risk of PEP in three CBD stone
cases because the pancreatic duct was cannulated several times unintentionally. Although
placement of the pancreatic stent occasionally made it more difficult to perform precut,
precutting was achieved in these three cases. Complications including PEP were not
observed, although the pancreatic stent was removed passively during the procedure.
There are several concerns regarding our marking method. First, there is a risk that
the marking itself will cause bleeding or perforation, although this may be avoided
by carefully marking the right spot. This was evidenced by absence of bleeding or
perforation in the current study. Another concern with precut sphincterotomy is perforation
caused by deep incision. This can occur even if the incision is made in the right
direction. Future improvements to avoid this are needed.
Second, there were only a few cases that showed papillary morphology owing to tumor
invasion or diverticulum. We could not clarify in this study whether our technique
would be effective for papillary morphology cases. To date, randomized controlled
trials or prospective cohort studies comparing the success and safety of precut techniques
in elation to papillary morphology have not been conducted [6 ]. Further examinations, therefore, are necessary.
Third, the depth of cutting cannot be strictly guided with use of our marking method.
However, this marking method makes it easy to precisely incise towards the correct
direction, which may avoid unnecessary deeper incision. This advantage may possibly
reduce complications such as bleeding and perforation.
Our work has limitations in terms of being a retrospective study, the small number
of patients, and the lack of comparisons. Also, the involved physicians were expert
endoscopists from a large-volume center, which may be related to the high procedural
success rate without complications.
Conclusion
In conclusion, the current case series showed that our novel marking method before
precutting enables precise incision and quick bile duct cannulation without causing
severe complications. Further prospective studies, in which trainees are enrolled,
should be conducted to compare clinical outcomes and effectiveness of our new marking
method with those of conventional precut sphincterotomy.
Video A marking was placed slightly before the upper margin of the bulge of the papilla
in the 11 – to 12-o’clock direction as a cutting endpoint by cauterization with a
needle knife before precutting. Subsequently, the needle knife was placed at the papillary
orifice and an incision was made towards the marking. After reaching the marking,
an incision was added to splay open the ampulla in layers to expose the sphincter
muscle. When the appropriate incision was obtained, CBD cannulation was attempted
using the catheter with the aid of a guidewire.
