Introduction
Since the advent to the traditional mother-daughter system [1]
[2], per oral cholangioscopy has significantly evolved to now include the digital single-operator
cholangiopancreatoscopy (D-SOCP, SpyGlass DS, Boston Scientific Corp., Natick, Massachusetts,
United States), leading to its widespread application [3]. There have been numerous studies showing its safety and efficacy in therapeutic
and diagnostic interventions, often when conventional cholangiography fails [4]
[5]
[6]
[7].
As a diagnostic tool, D-SOCP has demonstrated favorable diagnostic yield both visually
and histologically for indeterminate lesions [5]
[6]
[8]
[9]. As a therapeutic tool, D-SOCPs use in treating biliary or pancreatic complex strictures,
stones and difficult cannulation has also shown promise, achieving complete duct clearance
for biliary and pancreatic stone as high as 86 % to 97% [5]
[7]
[10]
[11]
[12].
While technological advances in D-SOCP have promoted its clinical use, data in patients
with surgically altered anatomy (SAA) are limited. There are no studies reporting
the use of D-SOCP in patients with SAA without either revising the anatomy or using
endoscopic ultrasound (EUS) for guidance. Thus, we sought to evaluate the effectiveness
and safety of D-SOCP in this patient population with the currently available endoscopic
instruments.
Patients and methods
This was an international, multicenter, retrospective study at eight tertiary centers
(6 US, 2 Europe) between February 2015 to June 2020. A total of 35 consecutive adult
patients > 18 years old with SAA who underwent digital single-operator cholangioscopy
(D-SOC) and/or digital single-operator pancreatoscopy (D-SOP) for all diagnostic or
therapeutic biliary or pancreatic indications were included. SAA includes patients
with a history of Whipple procedure, Billroth II, Roux-en-Y Gastric bypass (RYGB),
RY hepaticojejunostomy (RYHJ), and RY esophagojejunostomy (RYEJ). Patients who underwent
EUS-guided cholangiopancreatography (EUS-CP), including biliary drainage (EUS-BD),
EUS-rendezvous (EUS-RV), transmural drainage (such as hepatogastrostomy, hepatoduodenostomy,
hepatojejunostomy), antegrade stenting; percutaneous transhepatic cholangioscopy,
laparoscopy-assisted ERCP (LA- ERCP), EUS-directed trans-gastric ERCP (EDGE), per
oral direct cholangioscopy or procedures done with the legacy version of single-operator
cholangiopancreatoscopy system were excluded. Institutional review board approval
was obtained at all institutions.
Cholangiopancreatoscopy system
The D-SOC and D-SOP system (Spyglass DS, Boston Scientific Corp) was used for all
procedures. The system is composed of a catheter and integrated digital controller.
The 10.5F catheter has a tip with four-way steering facilitating visualization of
the entire biliary tree and through its working channel, multiple accessories including
the small forceps (Spybite, Boston Scientific), electrohydraulic lithotripsy (EHL)
or laser lithotripsy (LL) probes can be used. It also has a dedicated irrigation and
aspiration channels, dual LED light source and video imaging sensor. The integrated
digital controller combines the function of a processor and light emitting diode source
[2]
[5].
Cholangiopancreatoscopy procedure
All procedures were carried out under general anesthesia in either the supine or prone
position and performed by therapeutic endoscopists at each participating center who
were highly trained in diagnostic and therapeutic ERCP. The cholangioscope was advanced
through the accessory channel of an endoscope with a channel that is at least 3.7 mm
in diameter. Patients received peri-interventional antibiotic prophylaxis, according
to the institutional guidelines. The type of scopes and accessory used for biliary
or pancreatic interventions, was based on endoscopist preference.
Diagnostic D-SOCP was done for visually targeting biopsies in cases of indeterminate
biliary strictures and intraductal papillary mucinous cystic neoplasms (IPMN) [13]. Therapeutic D-SOCP was done for the therapy of large or complex biliary or pancreatic
duct stones, foreign material removal or biliary stent retrieval. The definition of
a large stone was one that was > 15 mm and a complex stone was one that was of any
size not retrieved by conventional extraction methods [7]
[14].
Outcomes assessment
The primary study outcome was the rate of technical success of D-SOCP in patients
with SAA. This was defined as completing the intended interventions with the use of
D-SOCP: 1) Biliary or pancreatic duct stones – successful treatment of stones with
complete duct clearance including use of electrohydraulic lithotripsy (EHL), Laser
Lithotripsy (LL); 2) Biliary or pancreatic stricture: diagnostic purposes – successful
visual impression and obtaining adequate targeted tissue samples; therapeutic purposes
– successful treatment of the stricture; and 3) Other indications: facilitating guidewire
passage and foreign body removal. The safety of the procedure was assessed by the
rate and severity of the adverse events (AEs) [15] as graded per the American Society for Gastrointestinal Endoscopy lexicon (mild,
moderate, severe, fatal) [16]. Other outcomes included number of D-SOCP sessions required to achieve technical
success; need for alternative therapies, such as interventional radiological procedures
or surgery; and procedure time which was defined as the time between insertion and
withdrawal of the endoscope.
Statistical methods
Descriptive statistics were presented as frequencies (%) for categorical variables
and mean (standard deviation), or median (interquartile range) for continuous variables.
Statistical analysis was performed using SPSS version 17.0 (SPSS, Chicago, Illinois,
United States).
Results
During the study interval, 35 patients (15 female; mean age, 67.9 ± 11.9 years) underwent
D-SOCP, of whom 34 underwent D-SOC and one patient underwent D-SOP. Billroth II was
the most common type of SAA (45.7 %), followed by Whipple reconstruction (31.4 %),
RYHJ (14.2 %) and RYEJ (8.5 %). A total of 23 patients (65.7 %) had a prior failed
ERCP, which was most commonly due to complex biliary stones (52.2 %). A therapeutic
duodenoscope was used in the majority of the cases (n = 24): eight patients had Whipple
reconstruction, 14 patients had Billroth II and two patients had RYHJ. Either a therapeutic
gastroscope (n = 8) or an adult colonoscope (n = 3) were used in the remainder of
the cases ([Table 1]). To facilitate the passage of D-SOCP, majority 28 patients (80 %) underwent sphincteroplasty
with a biliary or pancreatic balloon, three (8.5 %) underwent sphincterotomy and the
remainder either required no ampullary intervention (n = 2, 5.7 %) or had prior ampullary
intervention on index procedure (n = 2, 5.7 %) ([Table 2]).
Table 1
Baseline characteristics of patients with D-SOCP in SAA.
|
Baseline characteristics
|
N = 35
|
|
D-SOC
|
34
|
|
D-SOP
|
1
|
|
Age (mean ± SD)
|
67.9 ± 11.9
|
|
Female n (%)
|
15 (42.9)
|
|
Type of surgery, n (%)
|
|
|
11 (31.4)
|
|
|
16 (45.7)
|
|
|
5 (14.2)
|
|
|
3 (8.5)
|
|
Indication for D-SOC, n (%)
|
|
Therapeutic
|
|
|
19 (55.9)
|
|
|
10 (29.4)
|
|
|
9 (26.5)
|
|
|
|
|
|
13.5 (5.5)
|
|
|
2.6 (1.0)
|
|
|
|
|
13 (59)
|
|
|
2 (9.09)
|
|
|
7 (31.8)
|
|
|
4 (11.8)
|
|
Diagnostic
|
|
|
8 (23.6)
|
|
More than 1 indication
|
|
|
2 (5.9)
|
|
|
1 (2.9)
|
|
Indication for D- SOP n (%)
|
|
|
1 (100)
|
|
Symptoms and lab work, n (%)
|
|
|
22 (62.9)
|
|
|
21 (60.0)
|
|
|
5 (14.3)
|
|
|
4 (11.4)
|
|
|
4 (11.4)
|
|
|
3.5 (3.9)
|
|
|
222.0 (213.4)
|
|
|
210.8 (195.5)
|
|
|
200 (1888.2)
|
|
Prior attempted ERCP (total number), n (%)
|
23 (65.7)
|
|
Reason for prior failed attempt ERCP
|
|
|
12 (52.2)
|
|
|
5 (14.3)
|
|
|
1 (2.9)
|
|
|
1 (2.9)
|
|
|
1 (2.9)
|
|
|
1 (2.9)
|
|
|
2 (8.7)
|
D-SOCP, digital single-operator cholangiopancreatoscopy; SAA, surgically altered anatomy;
D-SOC, digital single-operator cholangiography; D-SOP, digital single-operator pancreatography;
SD, standard deviation; BD, bile duct; AST, aspartate aminotransferase; ALT, alanine
aminotransferase; ALP, alkaline phosphatase; ERCP, endoscopic retrograde cholangiopancreatography;
RYHJ, Roux-en-Y hepaticojejunostomy; RYEJ, Roux-en-Y esophagojejunostomy.
Table 2
Procedure characteristics.
|
Procedure characteristics
|
D-SOCP
|
|
Scope used, n (%) overall
|
|
|
24 (68.6)
|
|
|
8 (22.7)
|
|
|
3 (8.5)
|
|
D-SOC
|
|
|
24
|
|
|
8
|
|
|
2
|
|
D-SOP
|
|
|
1
|
|
Interventions performed, n (%)
|
|
Ampullary interventions performed to facilitate passage of D-SOCP
|
|
|
3 (8.5)
|
|
|
28 (80)
|
|
Other interventions
|
|
|
2 (5.7)
|
|
|
2 (5.7)
|
|
|
10 (28.6)
|
|
|
5 (14.3)
|
|
|
3 (8.6)
|
|
|
1 (2.9)
|
|
More than 1 intervention (mechanical lithotripsy, stent removal, stone removal, stricture
dilation, stent placement)
|
12 (34.3)
|
D-SOCP, digital single-operator cholangiopancreatoscopy; D-SOP, digital single-operator
pancreatoscopy; EHL, electrohydraulic lithotripsy; LL, Laser Lithotripsy.
D-SOC procedure
Majority of the patients (n = 26, 76.4 %) underwent D-SOC for therapeutic indications.
The main therapeutic indication was choledocholithiasis (n = 22, 64.7 %), most of
which were complex choledocholithiasis (n = 19, 55.9 %); and removal of prior migrated
stent (n = 4, 11.8 %). Diagnostic indications were for evaluation of indeterminate
bile duct (BD) stricture (n = 8, 23.6 %). Three patients (8.8 %) had multiple reasons
to undergo the procedure ([Table 2]). Among those with biliary stones, the mean number of stones was 2.6 ± 1.1, mean
stone size was 13.5 ± 5.5 mm, and were located in the common BD (n = 13, 59 %), common
hepatic and intrahepatic ducts (n = 9, 40.9 %). The majority of patients required
multiple interventions during the procedure (n = 12, 34.3 %). The most common interventions
performed were EHL (n = 10, 28.6 %), followed by LL (n = 5, 14.3 %) and stricture
dilation (n = 2, 5.7 %) ([Table 2]). [Fig. 1a–e] is that of a 52-year-old female with a history of pylorus-preserving Whipple in
which D-SOC aided in confirming absence of recurrent BD stone.
Fig. 1 52-year-old female with history of pylorus-preserving Whipple for pancreatic neuroendocrine
tumor underwent ERCP 2 months prior for left hepatic duct stone successfully removed,
but presented for concern of recurrent stone vs stricture. She underwent digital cholangioscopy
using a therapeutic gastroscope to further evaluate. a Bisectoral hepaticojejunal anastomoses noted are widely patent. b Cholangiogram showing a short narrowing of left hepatic branch to liver segment III
(arrow). c Digital single-operator cholangioscopy advanced into the left hepaticojejunal anastomosis.
d, e Cholangioscopic images of left intrahepatic duct with benign-appearing stricture
(yellow arrow), guidewire seen in background (black arrow). No stones noted. Spybite
biopsies were obtained and returned benign.
D-SOP procedure
One patient with Whipple anatomy underwent a D-SOP for imaging that showed dilation
of the pancreatic duct (PD) to 9 mm. Initially a duodenoscope and a therapeutic gastroscope
were used, but the pancreatico-jejunal (PJ) anastomosis could not be identified. An
adult colonoscope with a cap was successful in identifying the PJ anastomosis and
PD cannulation was achieved. D-SOP confirmed proximal PD dilation, a benign-appearing
distal PD stricture with an obstructing 7-mm stone; thus, excluding the presence of
a main duct mucinous neoplasm. Two D-SOP sessions (with EHL, stricture dilation and
stent placement) were performed ([Fig. 2a], [Fig. 2b]).
Fig. 2 A 61-year-old man with a history of pylorus-preserving Whipple for side branch Intraductal
pancreatic mucinous cystic neoplasm( IPMN), presents with imaging evidence of dilated
main pancreatic duct distal to the pancreaticojejunal (PJ) anastomosis, concerning
for main duct IPMN. The patient underwent digital single-operator pancreatoscopy using
a therapeutic gastroscope. PJ anastomosis cannulated, pancreatogram showed a dilated
main pancreatic duct measuring approximately 1 cm in diameter with an upstream stricture
in the pancreatic tail. a Fluoroscopic image showing single-operator pancreatoscope passed into the pancreatic
duct re-vealing a dilated duct with distal stricture (arrow). b Pancreato-scopic image showing a normal pancreatic duct mucosa with non-mucin-like
fluid with small solid debris inside. An obstructive stone was seen at the level of
the distal tail PD. IMPN was less likely, given these findings.
Outcomes
Technical success was achieved in in all 35 patients (100 %) in our study cohort.
In 22 patients with a BD stone and one patient with a PD stone, complete duct clearance
was achieved in all patients using D-SOCP. Eight patients (23.6 %) underwent D-SOC
for indeterminate stricture evaluation, of whom three patients had malignant stricture.
Among these three patients, the visual impression was that of abnormal dilated tumor
vessels (n = 1) and presence of friable mass with abnormal vessels (n = 2). D-SOC
based on visual impression and targeted tissue samples with small forceps (SpyBite,
Boston Scientific) biopsies were obtained in all three patients whose final diagnosis
was cholangiocarcinoma. The remainder five of eight patients had benign strictures:
BD stricture from stone disease (n = 2), postoperative BD stricture from extended
right hepatic hepatectomy (n = 1), anastomotic stricture following liver transplant
(n = 1) and chronic pancreatitis-induced benign biliary stricture (n = 1). D-SOC-based
visual impression assisted in all five cases while biopsies were performed in two
of five cases, confirming benign diagnosis. In the one patient who underwent D-SOP,
the PD stricture was found to be benign, secondary to an obstructing PD stone.
The majority of patients (n = 32, 91.3 %) required a single D-SOCP session, while
the remainder (n = 3, 8.5 %) required two D-SOCP sessions. No patients had technical
failure and none required alternative interventions such as interventional radiology
or surgery.
The mean procedure time was 61.0 ± 22.6 minutes. Overall, two AEs (6.4 %) occurred,
including one case of pancreatitis and one case of transient bacteremia. Both were
graded as mild in severity per the American Society of Gastrointestinal Endoscopy
lexicon. Both patients were successfully treated with conservative management ([Table 3]).
Table 3
Overall procedure outcomes.
|
Procedure characteristics
|
D-SOCP
|
|
Scope used, n (%) overall
|
|
|
24 (68.6)
|
|
|
8 (22.7)
|
|
|
3 (8.5)
|
|
D-SOC
|
|
|
24
|
|
|
8
|
|
|
2
|
|
D-SOP
|
|
|
1
|
|
Interventions performed, n (%)
|
|
Ampullary interventions performed to facilitate passage of DOSCP
|
|
|
3 (8.5)
|
|
|
28 (80)
|
|
Other interventions
|
|
|
2 (5.7)
|
|
|
2 (5.7)
|
|
|
10 (28.6)
|
|
|
5 (14.3)
|
|
|
3 (8.6)
|
|
|
1 (2.9)
|
-
More than 1 intervention (mechanical lithotripsy, stent removal, stone removal, stricture
dilation, stent placement)
|
12 (34.3)
|
D-SOCP, digital single-operator cholangiopancreatoscopy; D-SOC, digital single-operator
cholangiography; EHL, electrohydraulic lithotripsy; LL, Laser Lithotripsy.
Discussion
This is the first study to demonstrate the safety and feasibility of D-SOCP in SAA
without endoscopically or surgically altering the anatomy. The technical success rate
was 100 % despite the presence of SAA. We believe this is due to multiple factors,
an important one being appropriate scope choice allowing the passage of a 10.5F D-SOCP.
Prior studies report the success rate of gastroscopy, duodenoscopy, and colonoscopy
in SAA as 84.6 %, 62.5 %, and 93.5 %, respectively [17]
[18]
[19]. In our study, the majority of patients had a Billroth II or Whipple reconstruction,
in which a duodenoscope, therapeutic gastroscope or a colonoscope with cap were utilized.
It must be noted that while using a forward-viewing endoscope with a cap has its advantages
in SAA, one must bear in mind that the lack of elevator capability could further make
the procedure challenging.
The majority of patients in our cohort had prior failed ERCP (n = 23,65.7 %) due to
presence of complex choledocholithiasis (n = 12, 52.2 %), which was successfully managed
by D-SOCP resulting in complete BD clearance in all cases, the majority of which were
achieved in one session (91.3 %). Our findings were similar to multiple published
studies on the use of D-SOC with EHL/LL; however, these were not in SAA [20]. In the largest multicenter, retrospective cohort study of D-SOC to date, Gutierrez
et al. achieved a technical success rate for duct clearance of 97.3 %, of which 77.4 %
were in a single session [7]. In a meta-analysis and systematic review published by Korrapati et al, the efficacy
and safety of cholangioscopy in difficult BD stones was 88 % stone clearance and 7 %
AE. Limitations included study heterogeneity and variability in the type of per oral
cholangioscope used [21]. Navaneethan et al, in an observational study, performed D-SOC with LL in 31 patients
with 87.1 % of BD stone clearance in 1 session. Impacted stones were present in 13
of 31 patients (36.1 %). Twenty-three of the 31 patients (74.2 %) referred for stone
removal had prior ERCPs and had failed stone removal by using conventional methods
[5]. Further a 2:1 randomized trial compared cholangioscopy-guided LL to conventional
therapy, found endoscopic stone clearance was achieved in 39 of 42 patients (93 %)
in the cholangioscopy-guided group compared to 12 of 18 patients (67 %) in the conventional
therapy group, supporting the reduced need for mechanical lithotripsy in the D-SOC
arm. Regarding D-SOP, we only report one patient with a PD stone who also achieved
successful duct clearance, thus our data was insufficient for conclusion.
D-SOCP carries a high sensitivity and specificity for both visual impression and tissue
sampling in the evaluation of indeterminate strictures [5]
[22]. A systemic review by Kulpatcharapong et al comparing different cholangioscopy systems
for biliary strictures, showed that the digital SOC provided a higher negative predictive
value than the fiberoptic SOC (89 % vs. 69% – 84 %) [23]. Navaneethan et al performed an observational study using the newer D-SOC, among
105 patients, of whomw 44 had indeterminate stricture. They reported a tissue adequacy
of 97.7 %, sensitivity and specificity for D-SOC visual impression for diagnosis of
malignancy as 90 % and 95.8 %, respectively and sensitivity and specificity of D-SOC-guided
biopsies for diagnosis of malignancy as 85 % and 100 %, respectively [5]
[22]. These findings are supported in our analysis in patients with SAA, further emphasizing
its pivotal role in indeterminate BD strictures. We found that D-SOC was successful
in diagnosing both malignant (3/8) and benign biliary strictures (5/8), with an overall
technical success rate of 100 %.
This study has several limitations. It was retrospective with a relatively small sample
size. Neither the procedure strategy nor the devices used for D-SOCP in SAA were not
standardized. Also, the details of SAA including length of afferent limb were not
specified. The factors associated with successful or unsuccessful procedures could
not be assessed. Finally, we acknowledge that technical success can be an overestimate
of the true success and we cannot rule out bias related to the retrospective nature
of the study. Nonetheless, this is the first multicenter study reporting the utility
of D-SOCP in diagnosis and treatment of pancreatobiliary disorders in patients with
SAA.
Conclusions
In conclusion, D-SOCP results in high rates of technical success and low AE rates
in patients with altered upper gastrointestinal anatomy. In instances in which standard
ERCP attempts failed, D-SOCP proved to be a safe and reliable management strategy
for complex disease.
