Abbreviations
ABP:
acute biliary pancreatitis
ASGE:
American Society for Gastrointestinal Endoscopy
CBD:
common bile duct
CBDS:
common bile duct stones
CI:
confidence interval
CT:
computed tomography
DPOC:
direct peroral cholangioscopy
EHL:
electrohydraulic lithotripsy
EPLBD:
endoscopic papillary large-balloon dilation
ERCP:
endoscopic retrograde cholangiopancreatography
ESGE:
European Society of Gastrointestinal Endoscopy
ESWL:
extracorporeal shock wave lithotripsy
EUS:
endoscopic ultrasonography
IOC:
intraoperative cholangiography
IRR:
incidence rate ratio
LFT:
liver function test
MBC:
mother – baby cholangioscopy
MRCP:
magnetic resonance cholangiopancreatography
NR-ERCP:
non-radiation ERCP
NSAID:
nonsteroidal anti-inflammatory drug
OR:
odds ratio
OTS:
out of the scope
PTBD:
percutaneous transhepatic biliary drainage
RCT:
randomized controlled trial
RR:
relative risk
SD:
standard deviation
SEMS:
self-expanding metal stent
SOC:
single-operator cholangioscopy
TTS:
through the scope
UDCA:
ursodeoxycholic acid
This Guideline is an official statement of the European Society of Gastrointestinal
Endoscopy (ESGE). It provides practical advice on how to manage common bile duct stones.
The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system
was adopted to define the strength of recommendations and the quality of evidence.
1 Introduction
Gallstones are a very common problem in developed countries [1 ]
[2 ]
[3 ]. Most patients with gallstones remain asymptomatic throughout their lifetime [4 ]
[5 ], but 10 % – 25 % of them may develop biliary pain or complications [6 ]
[7 ]
[8 ]
[9 ], with an annual risk of about 2 % – 3 % for symptomatic disease [10 ] and 1 % – 2 % for major complications [11 ]. The development of symptomatic disease and complications is mostly related to the
migration of stones into the common bile duct (CBD). Common bile duct stones (CBDSs)
may be treated by endoscopic retrograde cholangiopancreatography (ERCP) or surgically
during cholecystectomy. The aim of this evidence- and consensus-based Guideline, commissioned
by the European Society of Gastrointestinal Endoscopy (ESGE), is to provide practical
advice on how to manage patients with CBDSs. It considers diagnostic strategies in
patients with suspected CBDSs, as well as the different therapeutic options available
for CBDSs.
2 Methods
The ESGE commissioned this Guideline (chair J.v.H.) and appointed a guideline leader
(G.M.), who invited the listed authors to participate in the project development.
The key questions were prepared by the coordinating team (G.M. and G.P.) and then
approved by the other members. The coordinating team formed task-force subgroups,
each with its own leader, and divided the key topics among these task forces (Appendix 1 s ; see online-only Supplementary Material).
Each task force performed a systematic literature search to prepare evidence-based
and well-balanced statements on their assigned key questions. The coordinating team
independently performed systematic literature searches, with PubMed/Medline, EMBASE,
the Cochrane Library, and the internet being finally searched for papers published
until April 2018. The search focused on fully published randomized controlled trials
(RCTs), meta-analyses, and prospective series. Retrospective analyses, case series,
and abstracts were also included if they addressed topics not covered in the prospective
studies. The literature search was restricted to papers published in English after
1990.
After further exploration of their content, articles that contained relevant data
were then included and summarized in the literature tables for the key topics (Tables 1s – 14 s ). All selected articles were graded by the level of evidence and strength of recommendation
according to the GRADE system [12 ]. Each task force developed a draft and proposed statements on their assigned key
questions, which were discussed and voted on during plenary meetings held in February
2017 in Düsseldorf, Germany, and in October 2017 in Barcelona, Spain. In April 2018,
a draft prepared by the coordinating team was sent to all group members.
After agreement of all group members, the manuscript was reviewed by two members of
the ESGE Governing Board, and by two external reviewers and was then sent for further
comments to the ESGE National Societies and Individual Members. The manuscript was
then submitted to Endoscopy for publication. All authors agreed on the final revised manuscript. This Guideline
was issued in 2019 and will be considered for review in 2023, or sooner if new and
relevant evidence becomes available. Any updates to the Guideline in the interim will
be noted on the ESGE website: http://www.esge.com/esgeguidelines.html .
3 General principles
3.1 Epidemiology
Gallstones are common with a prevalence as high as 10 % – 15 % in developed countries
[1 ]
[2 ]
[3 ] and an overall cumulative incidence of gallstone formation of 0.60 % per year [13 ].
According to a large Swedish registry [14 ], the prevalence of CBDSs detected during intraoperative cholangiography (IOC) is
11.6 % in patients with symptomatic gallbladder stones; other prospective studies
have described a prevalence of CBDSs detected during IOC ranging from 4.6 % to 12 %
in Europe [15 ]
[16 ], and up to 20.9 % in South America [17 ]. A prevalence of 8 % – 18 % for CBDSs in patients with symptomatic gallbladder stones
has been proposed [18 ].
No studies have focused on the prevalence of CBDSs in patients with asymptomatic gallbladder
stones, as most studies are based on IOC during cholecystectomy for symptomatic disease.
3.2 The natural history of CBDSs and recommended handling
ESGE recommends offering stone extraction to all patients with common bile duct stones,
symptomatic or not, who are fit enough to tolerate the intervention.
Strong recommendation, low quality evidence.
The natural history of CBDSs is not well described, but data from the GallRiks study
[14 ] suggest that, if CBDSs are detected, they should be removed to reduce the risk of
complications over time: of the 3969 patients with CBDSs on IOC, 594 had their CBDSs
left in place. During follow-up, ranging from 0 to 4 years, 25.3 % of patients with
CBDSs in situ developed complications (pancreatitis, cholangitis, or obstruction of
the bile duct) vs. 12.7 % of patients who had undergone CBDS removal (odds ratio [OR]
0.44, 95 %CI 0.35 – 0.55). The likelihood of an unfavorable outcome increased with
the size of the CBDS, but the incidence of complications even for CBDSs less than
4 mm was 5.9 % vs. 8.9 % for larger CBDSs (OR 0.52, 95 %CI 0.34 – 0.79).
These data support a strategy of extracting CBDSs regardless of size, although some
previous studies have suggested that small unsuspected stones can pass spontaneously
without the need for intervention [16 ]
[19 ]
[20 ]
[21 ]
[22 ] (Table 1 s ). The spontaneous passage of small CBDSs without serious complications has been documented
by Collins [15 ] in 24 of 46 patients with a filling defect observed on IOC in whom a cystic duct
catheter was left in place after laparoscopic cholecystectomy. The asymptomatic migration
of small (less than 8 mm) stones has also been noted in the interval between diagnosis
at endoscopic ultrasonography (EUS) and ERCP [23 ].
In spite of the absence of controlled studies, some factors favor a policy of stone
extraction in asymptomatic CBDSs: the occurrence of unfavorable outcomes is not different
in patients classified as asymptomatic or symptomatic [14 ]; the lifetime risk of untreated CBDSs is unknown and may be higher than that reported;
severe complications such as cholangitis, pancreatitis, or obstructive jaundice can
occur without preceding warning symptoms [24 ]. A conservative approach can only be considered in patients where the risks of surgical
or endoscopic CBDS extraction are higher than the risks of leaving stones in situ.
When offering stone extraction to asymptomatic patients with CBDSs, patients should
be made aware of the limited evidence regarding this recommendation and of the risk
of ERCP, which may be elevated in asymptomatic patients [25 ].
4 Defining the risk of having CBDSs
4 Defining the risk of having CBDSs
4.1 Initial evaluation
ESGE recommends liver function tests and abdominal ultrasonography as the initial
diagnostic steps for suspected common bile duct stones. Combining these tests defines
the probability of having common bile duct stones.
Strong recommendation, moderate quality evidence.
Patients at risk of having CBDSs, such as patients with gallstones who present with
symptoms, undergo non-invasive tests such as liver function tests (LFTs) and abdominal
ultrasound as triage to determine the need for further evaluations to confirm the
presence of CBDSs.
A recent systematic review including five studies assessed the diagnostic accuracy
of LFTs (1 study) and ultrasonography (5 studies) for CBDSs [26 ]. All studies were of poor methodological quality. The sensitivities of bilirubin
(cutoff > 22.23 μmol/L or > 1.3 mg/dL) and alkaline phosphatase (cutoff > 125 U/L)
for CBDSs were 84 % (95 % confidence interval [CI] 64 % – 95 %) and 91 % (95 %CI 74 % – 99 %),
respectively; the specificities were 91 % (95 %CI 86 % – 94 %) and 79 % (95 %CI 74 % – 84 %),
respectively. Regarding ultrasonography, sensitivity was 73 % (95 %CI 44 % – 95 %)
and specificity was 91 % (95 %CI 84 % – 95 %). Ultrasonography findings were considered
positive if there was visualization of CBDSs and/or CBD dilatation.
Multidetector multiphase computed tomography (CT), when used to investigate patients
with CBDSs, had a sensitivity of 78 % and a specificity of 96 % in a retrospective
study [27 ]. The size and composition of the stones significantly affects CT accuracy, which
is significantly lower when stones are less than 5 mm (56.5 % vs. 81.2 %) or have
a similar density to bile [28 ]. Coronal reconstruction does not increase the diagnostic efficiency of CT scanning
[29 ].
The pretest probability of CBDSs in suspected patients is essential to select which
patients will benefit most from a more accurate assessment. Several predictive models
have been developed combining clinical, biochemical, and ultrasonography findings
in order to identify high risk patients [30 ]
[31 ]
[32 ]
[33 ]
[34 ] (Table 2s ).
The risk of having CBDSs in spite of normal LFTs and ultrasonography has been adequately
evaluated in two studies [35 ]
[36 ]. In a large study including 765 patients with ERCP-proven CBDSs, 541 had previously
documented LFTs and 29 (5.4 %) had normal LFTs. Age more than 55 years and the presence
of pain were independently associated with normal LFTs in patients with CBDSs [35 ]. A more recent retrospective study including 413 patients with gallstones who underwent
ultrasonography and magnetic resonance cholangiopancreatography (MRCP) for suspected
CBDSs showed that 109 /413 (26.3 %) had CBDSs revealed on the MRCP, but in 7 /109
(6.4 %) ultrasonography and LFTs (one or more of total bilirubin, ALP, AST, ALT, or
GGT) were normal [36 ] (Table 2 s ).
4.2 Role of EUS and MRCP
ESGE recommends endoscopic ultrasonography or magnetic resonance cholangiopancreatography
to diagnose common bile duct stones in patients with persistent clinical suspicion
but insufficient evidence of stones on abdominal ultrasonography.
Strong recommendation, moderate quality evidence.
The diagnostic accuracy of EUS and MRCP for the detection of CBDSs has been widely
investigated. Meeralam and co-workers in a recent meta-analysis of five head-to-head
studies [37 ] demonstrated that diagnostic accuracy was high for both methods (sensitivity 97 %
vs. 90 % and specificity 87 % vs. 92 % for EUS and MRCP, respectively), but the overall
diagnostic OR of EUS was significantly higher (P = 0.008). They showed that this was mainly because of the significantly higher sensitivity
of EUS, as compared with that of MRCP, especially in the detection of small stones,
while the specificity was not significantly different. High accuracy for both methods
was demonstrated by another meta-analysis including 18 studies (2 comparative, 5 evaluating
MRCP alone and 11 EUS alone) [38 ]. Sensitivity and specificity were respectively 95 % (95 %CI 91 % – 97 %) and 97 %
(95 %CI 94 % – 99 %) for EUS, and 93 % (95 %CI 87 % – 96 %) and 96 % (95 %CI 90 % – 98 %)
for MRCP.
Various considerations may help to select the most adequate procedure if both are
available and the patient does not present any factors that would impede MRCP, such
as claustrophobia, obesity, cardiac pacemaker, or metal clips. Sonnemberg and colleagues
[39 ], performing a threshold analysis on costs, concluded that, for a pretest probability
of CBDSs < 40 %, MRCP would represent the procedure of choice. For a pretest probability
in the range 40 % – 91 %, EUS should be the preferred imaging modality, because it
allows an ERCP to be performed in the same session if the EUS results are positive
for CBDSs. However, the applicability of their results is limited because they are
strictly influenced by the costs of each procedure and local rules of reimbursement.
Furthermore, logistic issues regarding the scheduling of an EUS and an ERCP during
the same examination slot should be taken into consideration. The minimally invasive
nature of MRCP, its suitability if there is altered gastroduodenal anatomy, and its
ability to visualize the whole biliary tree should also be considered when deciding
between the two methods.
4.3 An algorithm for investigating suspected CBDSs
[Fig. 1 ] depicts an algorithm for investigating suspected CBDSs. ERCP can be performed in
patients without cholangitis only when CBDSs are visible on imaging modalities that
have a high specificity. Normal LFTs and ultrasonography indicate a low risk of CBDSs
and no further evaluations are recommended, unless the patient continues to have symptoms
that suggest CBDSs. All other pictures depict an intermediate risk of CBDSs, which
should prompt further investigation by EUS or MRCP. In the absence of a morphological
diagnosis of CBDSs, ERCP should be performed immediately only in patients with a clinical
picture of cholangitis (see section 8.1).
Fig. 1 Diagnostic algorithm for suspected common bile duct stones (CBDSs). LFTs, liver function
tests; US, ultrasound; CBD, common bile duct; EUS, endoscopic ultrasonography; MRCP,
magnetic resonance cholangiopancreatography; ERCP, endoscopic retrograde cholangiopancreatography.
5 Performing ERCP
5.1 Antibiotic prophylaxis
ESGE suggests against the use of routine antibiotic prophylaxis before ERCP for bile
duct stones.
Weak recommendation, moderate quality evidence.
The ERCP procedure is often associated with the occurrence of bacteremia [40 ], which is mostly transient. The occurrence of cholangitis is an infrequent event,
which occurs mainly in a subgroup of patients at higher risk, such as those with biliary
obstruction and incomplete biliary drainage [41 ].
The role of antibiotic prophylaxis in reducing the rate of cholangitis has been evaluated
by several RCTs, which differed significantly in terms of type of antibiotic, duration
of administration, and indications for ERCP [42 ]
[43 ]
[44 ]
[45 ]
[46 ]
[47 ] and three meta-analyses (Table 3 s ) [48 ]
[49 ]
[50 ].
The most recent meta-analysis of nine RCTs [50 ] (1573 patients)
indicated that antibiotic prophylaxis could reduce bacteremia and may prevent cholangitis
and septicemia in patients undergoing elective ERCP. However, in random-effects meta-analyses,
only the effect on bacteremia remained significant; if ERCP resolved the biliary obstruction
at the first procedure, there was no significant benefit in using antibiotic prophylaxis
to prevent cholangitis (relative risk [RR] 0.98, 95 %CI 0.35 – 2.69, only three trials)
[50 ].
Cotton et al. [51 ] reported in a retrospective series of 11 484 ERCPs performed over 11 years that,
in spite of a progressive reduction in the use of antibiotic prophylaxis over the
years (from 95 % to 25 % of ERCP patients), the incidence of infections decreased
from 0.48 % to 0.25 %. In the multivariate model, endoscopic treatment of CBDSs was
not associated with an increased risk of developing cholangitis after ERCP. All these
data suggest that not all patients benefit from antibiotic prophylaxis and that patients
with CBDSs should not routinely receive antibiotic prophylaxis before ERCP (Table 3 s ).
Patients with ongoing acute cholangitis should already be receiving antibiotics at
the time of intervention and additional antibiotics are not recommended.
Antibiotic prophylaxis should be considered for patients with refractory CBDSs undergoing
extracorporeal shock wave lithotripsy (ESWL) for CBD clearance [52 ]
[53 ]. No data are available for patients undergoing cholangioscopy-assisted lithotripsy;
nevertheless, antibiotic prophylaxis is likely to be advisable as two recent prospective
studies have demonstrated that cholangioscopy per se may carry a risk of bacteremia
that ranges from 8.8 % to 13.9 % and that up to 9.7 % of patients may develop infective
complications despite the use of post-procedure antibiotics [54 ]. Biopsy sampling, older age, previous stent placement, and laser lithotripsy or
electrohydraulic lithotripsy (EHL) were likely to increase the risk of developing
either infection or persistent bacteremia.
Antibiotic prophylaxis in some special conditions, such as in liver transplant patients,
was considered to be out of the scope of this guideline.
5.2 Gaining access to the biliary tree
ESGE recommends that an adequate exit for the stones that are to be removed should
be provided according to the papilla and common bile duct anatomy and the stone size.
Strong recommendation, low quality evidence.
The various technical aspects, either of deep biliary cannulation or endoscopic sphincterotomy,
have been reviewed in other guidelines [55 ]
[56 ]. A critical step to obtain successful stone extraction is to provide an adequate
exit for the stones that are to be removed by endoscopic sphincterotomy alone, endoscopic
papillary balloon dilation alone, or a combination of both [55 ]
[57 ]. Papillary balloon dilation alone however remains unpopular and is not advocated
for routine use as it is associated with a lower technical success for stone clearance,
the need for mechanical lithotripsy more frequently than with endoscopic sphincterotomy,
and a presumed increased risk of pancreatitis [55 ]
[58 ]
[59 ]. At present, the use of primary papillary balloon dilation without endoscopic sphincterotomy
is considered mainly in patients with coagulopathy or with altered anatomy who have
stones smaller than 8 mm [55 ]. The appropriate length of endoscopic sphincterotomy should be adjusted according
to the papillary anatomy and stone size. Data on the effect of endoscopic sphincterotomy
length on the rate of stone recurrence are presently contradictory [60 ]
[61 ].
5.3 Stone extraction
ESGE recommends that balloon and basket catheters are equally effective and safe for
common bile duct stone removal.
Strong recommendation, moderate quality evidence.
Two multicenter RCTs have compared the efficacy of balloon vs. basket catheters for
the extraction of CBDSs sized ≤ 10 mm or < 11 mm after endoscopic sphincterotomy [62 ]
[63 ]. In one RCT (158 patients), the balloon catheter achieved a higher clearance rate
than the basket catheter (92.3 % vs. 80.0 %) [62 ]. The other RCT (184 patients) reported similar efficacies for basket and balloon
catheters for stone extraction, but a stone diameter of < 6 mm was independently associated
with failed stone removal within 10 minutes using a basket catheter, because of the
inability to grasp the stone with the basket [63 ]. No differences in safety were reported in the two studies.
Stone extraction baskets and balloons are commercially available in various configurations.
As yet, no comparative studies between various models of basket catheters exist [64 ]. In general, choosing which device to use depends mainly on the anatomy of the bile
duct, the stone characteristics, financial considerations, and personal preferences.
5.4 Biliary stenting for incomplete removal of CBDSs
ESGE recommends endoscopic placement of a temporary biliary plastic stent in patients
with irretrievable biliary stones that warrant biliary drainage.
Strong recommendation, moderate quality of evidence.
Endoscopic sphincterotomy with stone extraction has success rates of 80 % – 90 % in
the treatment of CBDSs [65 ]. When CBDSs cannot be completely removed, a plastic stent is often placed to relieve
the obstruction, before a second attempt at stone extraction is made or a subsequent
surgical intervention is undertaken. An indwelling endoprosthesis may reduce the volume
and number of stones, as reported by nine studies (three prospective [66 ]
[67 ]
[68 ] and six retrospective [69 ]
[70 ]
[71 ]
[72 ]
[73 ]
[74 ]) involving a total of 364 patients (Table 4 s ). The success rate for stone removal after previous ERCP with biliary stenting has
been reported to range from 44 % to 96 % (Table 5 s ) [66 ]
[67 ]
[68 ]
[69 ]
[70 ]
[71 ]
[72 ]
[73 ]
[75 ]
[76 ].
The mechanism by which stones change in number and size is unclear. It is likely that
continuous friction between the plastic stent and the stones produces stress forces
that facilitate the disintegration of stones and reduce their size [71 ].
There are no studies comparing the different types of biliary plastic stents or plastic
vs. metal stents. Similarly, there are no specific prospective comparative data with
regard to whether one or more than one biliary stent is preferable in patients with
incomplete stone removal. In the only retrospective published study, 64 elderly patients
(≥ 65 years) with large (≥ 20 mm) or multiple (≥ 3) CBDSs underwent placement of single
or double plastic stents at the time of initial ERCP. Approximately 3 months later,
stone removal was attempted at a second ERCP using standard techniques. Double plastic
biliary stenting (7 or 8.5 Fr) was superior to single stenting (8.5 Fr) in maintaining
higher 3-month stent patency rates (P = 0.008), but was similar in terms of reducing the size and number of stones [77 ]. No differences in complications were found.
In recent years, some studies with small patient series have evaluated the management
of incomplete stone removal using fully covered self-expanding metal stents (SEMSs)
(Table 6 s ) [78 ]
[79 ]
[80 ]. In the largest retrospective case series [80 ], 44 patients received covered SEMSs (diameter 10 mm, length 60 mm). After a median
in-stent duration of 8 weeks, 36/42 stents (82 %) were removed with successful duct
clearance. The median post-procedure follow-up was 15 months. Four patients (9 %)
developed post-ERCP pancreatitis (mild in 3, moderate in 1), two patients (4 %) developed
post-procedure cholangitis, and one (2 %) hematemesis. During follow-up, 10 patients
(22.7 %) had incidental stent migration (distally in 6, proximally in 4), but in none
of them was it clinically significant, with all being discovered at the time of subsequent
ERCP.
At present, covered SEMSs can be considered as an alternative to plastic stents to
drain the bile ducts after unsuccessful stone removal, but there are uncertainties
over how long the stents should be left in place and the cost – benefit ratio of the
treatment.
5.5 Timing of stent removal/exchange
ESGE recommends that a plastic stent placed because of incomplete common bile duct
stone clearance should be removed or exchanged within 3 – 6 months to avoid infectious
complications.
Strong recommendation, moderate quality of evidence.
ESGE recommends against the use of definitive biliary stenting in patients with incomplete
common bile duct stone clearance because of the high complication and mortality rates
on medium-term follow-up.
Strong recommendation, moderate quality of evidence.
Intervals of 3 – 6 months for routine ERCP and stent change are commonly recommended
to reduce the rate of complications, mainly cholangitis [70 ]
[76 ]. One randomized prospective study including 78 patients with primary failure for
biliary stone removal who had undergone insertion of a 10-Fr plastic stent compared
two different managements: either systematic stent exchange every 3 months or stent
exchange on demand if symptoms occurred. Cholangitis was significantly more frequent
in the group with on-demand stent exchange (35.9 % vs. 7.7 %; P < 0.03) [81 ].
Definitive stenting has been suggested for difficult CBDSs in the elderly with co-morbidities
and a limited life expectancy, given that ERCP in patients aged > 90 years may carry
risks of bleeding, cardiopulmonary events, and mortality that are increased two to
three fold (incidence rate ratio [IRR] 2.4, 95 %CI 1.1 – 5.2; IRR 3.7, 95 %CI 1.0 – 13.9;
and IRR 3.8, 95 %CI 1.0 – 14.4, respectively), and that patients aged > 80 years had
a two-fold risk of procedure-related death (IRR 2.4; 95 %CI 1.3 – 4.5) [82 ]. However, definitive stenting for CBDSs should be approached with caution. Six series,
including 230 patients [83 ]
[84 ]
[85 ]
[86 ]
[87 ]
[88 ], have reported a complication rate for definitive biliary stenting, mainly cholangitis,
of 34 % – 63 %, with a 2.3 % – 23.5 % mortality rate during 16 – 39 months of follow-up
(Table 7 s ).
5.6 Role of dissolution therapy
EGSE suggests against the use of ursodeoxycholic acid or other choleretic agents,
either for the treatment of CBDSs or to prevent the recurrence of CBDSs after endoscopic
clearance.
Weak recommendation, moderate quality of evidence.
Ursodeoxycholic acid (UDCA) with or without terpene preparation (Rowachol) has been
suggested as a complementary treatment to induce stone reduction when used together
with biliary endoprostheses, but in two RCTs the addition of UDCA therapy to endoprosthetic
treatment showed no effect on stone size reduction or successful duct clearance [66 ]
[68 ].
UDCA has been administered with the aim of reducing the rate of stone recurrence after
successful removal of CBDSs in patients with risk factors such as CBD dilatation,
delayed biliary emptying (biliary stricture, papillary stenosis), or the presence
of gallstones, a periampullary diverticulum, or systemic diseases that cause stone
formation [89 ]
[90 ]
[91 ]. Two RCTs have investigated this issue and both revealed no significant difference
regarding stone recurrence [92 ]
[93 ].
6 Difficult stones
“Difficult” biliary stones are defined according to their diameter (> 1.5 cm), number,
unusual shape (barrel-shaped), or location (intrahepatic, cystic duct), or because
of anatomical factors (narrowing of the bile duct, distal to the stone, sigmoid-shaped
CBD, stone impaction, shorter length of the distal CBD, or acute distal CBD angulation
< 135°) [94 ]
[95 ]. Clearance of a difficult stone cannot usually be obtained using standard techniques,
so multiple procedures and additional interventional techniques (large-balloon dilation,
mechanical lithotripsy, cholangioscopy-assisted electrohydraulic/laser lithotripsy,
or ESWL) may be required [96 ].
6.1 Gaining access to the biliary tree and basic treatment for the management of difficult
stones
ESGE recommends limited sphincterotomy combined with endoscopic papillary large-balloon
dilation as the first-line approach to remove difficult common bile duct stones.
Strong recommendation, high quality evidence.
Since the original description in 2003 by Ersoz et al., the use of endoscopic papillary
large-balloon dilation (EPLBD) after endoscopic sphincterotomy has become widespread
for the management of difficult CDBSs [97 ]. Overall, seven RCTs [98 ]
[99 ]
[100 ]
[101 ]
[102 ]
[103 ]
[104 ] and five meta-analyses [105 ]
[106 ]
[107 ]
[108 ]
[109 ] have compared the efficacy and safety of EPLBD with endoscopic sphincterotomy vs.
endoscopic sphincterotomy alone (Table 8 s ).
In summary, endoscopic sphincterotomy + EPLBD reduces the need for mechanical lithotripsy
by about 30 % – 50 % in comparison with endoscopic sphincterotomy alone [100 ]
[102 ]
[103 ], while the overall rate of successful stone removal remains identical [105 ]
[106 ]
[107 ]
[108 ]. The rate of major adverse events, mainly pancreatitis, bleeding, and perforation,
between the two groups was similar in 6 of 7 RCTs [99 ]
[100 ]
[101 ]
[102 ]
[103 ]
[104 ], whereas it was significantly lower for EPLBD plus endoscopic sphincterotomy compared
with endoscopic sphincterotomy alone in the study by Stefanidis et al. [98 ]. In a systematic review (30 studies considered), the rate of overall adverse events
(pancreatitis, bleeding, perforation) was lower for endoscopic sphincterotomy with
EPLBD than for endoscopic sphincterotomy alone (8.3 % vs. 12.7 %, OR 1.60; P < 0.001) [110 ].
Based on these data, if large bile duct stones are seen on ERCP or cross-sectional
imaging, endoscopic sphincterotomy combined with EPLBD can be used as a first-line
approach to facilitate difficult biliary stone removal [111 ]. Another possible indication for performing EPLBD is the treatment of recurrent
CBDSs in individuals with a previous endoscopic sphincterotomy because extension of
an endoscopic sphincterotomy may be associated with a high risk of bleeding and perforation
[112 ]
[113 ]
[114 ]
[115 ] ([Fig. 2 ]).
Fig. 2 Therapeutic algorithm for management of common bile duct stones when ERCP is selected
as the primary treatment. ERCP, endoscopic retrograde cholangiopancreatography; EPLBD,
endoscopic papillary large-balloon dilation (12 – 20 mm); ESWL, extracorporeal shock
wave lithotripsy.
EPLBD can be performed after either a large [97 ]
[98 ]
[114 ]
[116 ]
[117 ]
[118 ]
[119 ]
[120 ]
[121 ] or limited endoscopic sphincterotomy [99 ]
[120 ]
[122 ]
[123 ]
[124 ]
[125 ]
[126 ]
[127 ]. A multicenter retrospective analysis from Asia including 946 patients [120 ] found large endoscopic sphincterotomy before EPLBD to be independently associated
with an increase in overall adverse events (OR 3.4, 95 %CI 1.8 – 6.6; P < 0.001). The risk of bleeding was higher in the large vs. limited endoscopic sphincterotomy
group (OR 6.2, 95 %CI 2.4 – 16.3; P < 0.001). Perforation was found in only nine patients but it was fatal in three of
them. Although only distal CBD stricture and not size of endoscopic sphincterotomy
was an independent predictor of perforation, two of the three fatal cases were associated
with a large endoscopic sphincterotomy. A recent literature review suggested performing
a small or midsized endoscopic sphincterotomy (1/3 to 1/2 of the distance to the papillary
roof) rather than a large one before EPLBD [128 ]. Nevertheless, in real life most endoscopists decide to perform EPLBD when their
attempts to remove the stones have failed after having already performed a complete
endoscopic sphincterotomy.
EPLBD is performed with a dilation balloon diameter that ranges from 12 to 20 mm.
Criteria for deciding the balloon size for EPLBD have not been specifically evaluated
in prospective studies. In most published studies, the diameter of the distal part
of the CBD has been used as the criterion to select the size of the balloon [98 ]
[99 ]
[100 ]
[120 ]
[121 ]. The risk of perforation increases when the diameter of the balloon is larger than
the diameter of the distal part of the CBD and in the presence of a stricture [111 ].
The vast majority of studies have reported a dilation duration of 10 – 180 seconds
from the disappearance of the waist, with only three studies reporting a duration
in excess of 60 seconds [110 ]. One RCT has demonstrated that the rate of complications is similar whether EPLBD
duration is either 30 or 60 seconds [121 ]. Moreover, a meta-analysis has demonstrated that a short duration (< 1 minute) vs.
a long duration (≥ 1 minute) for EPLBD does not significantly affect the rate of CBD
clearance [105 ]. According to these data, the duration of balloon dilation should be between 30
and 60 seconds from the disappearance of the waist [111 ].
6.2 Mechanical lithotripsy
ESGE recommends mechanical lithotripsy for difficult stones when sphincterotomy plus
endoscopic papillary large-balloon dilation has failed or is inappropriate.
Strong recommendation, moderate quality evidence.
Mechanical lithotripsy is the simplest available method of fragmenting CBDSs. It consists
of entrapping the stone within a reinforced basket and then crushing it by closing
the basket against a metal spiral sheath. Two techniques of mechanical lithotripsy
are used: out of the scope (OTS) and through the scope (TTS). The OTS technique represents
a “salvage” procedure to be performed when a standard basket engages a large stone
and becomes impacted in the papilla, while the TTS technique is preferred in elective
cases.
Mechanical lithotripsy has been reported to be an effective and safe technique, but
multiple sessions may be required. The reported success rates range between 76 % and
91 % and overall complications from 3 % to 34 % with minimal mortality [129 ]
[130 ]
[131 ]
[132 ]
[133 ]
[134 ] (Table 9 s ). Three studies have evaluated the predictors of mechanical lithotripsy failure using
multivariate analysis. In a retrospective study [130 ], stone size was the only variable that affected the success rate. A subsequent prospective
study [129 ] reported that stone size should be considered together with the diameter of the
bile duct, suggesting that only the presence of stone impaction significantly predicted
the failure of mechanical lithotripsy. In another more recent retrospective study
[132 ], stone impaction, stone size > 30 mm, and stone to CBD diameter ratio > 1 were significant
predictors of mechanical lithotripsy failure.
The most common and feared complications of mechanical lithotripsy are entrapment
of the basket, a broken basket, a traction wire fracture, or a broken handle. In a
multicenter study by Thomas et al. [135 ], including 643 patients and using the TTS technique, the incidence of mechanical
lithotripsy-related technical complications was 3.5 %. These complications are usually
treated by other types of lithotripsy (OTS, ESWL, or cholangioscopy-assisted lithotripsy),
sphincterotomy extension, or stenting.
6.3 Cholangioscopy-assisted lithotripsy
ESGE recommends the use of cholangioscopy-assisted intraluminal lithotripsy (electrohydraulic
or laser) as an effective and safe treatment of difficult bile duct stones.
Strong recommendation, moderate quality evidence.
ESGE suggests that the type of cholangioscopy and lithotripsy should depend on local
availability and experience.
Weak recommendation, low quality evidence.
Intraductal shock wave lithotripsy represents an alternative method to fragment bile
stones and allow their removal. There are two methods of generating shock waves in
a fluid, using either a bipolar probe capable of generating a spark in the case of
EHL or a pulsed dye laser system in the case of laser lithotripsy. Both EHL and laser
lithotripsy are preferably performed under direct visualization with cholangioscopic
guidance.
There are three major techniques for cholangioscopy: (i) a dual-operator dedicated
mother – baby cholangioscopic (MBC) system; (ii) a single-operator catheter-based
cholangioscopic system (SOC); and (iii) direct use of an ultraslim endoscope or slim
gastroscope (direct peroral cholangioscopy [DPOC]). The procedures vary with respect
to the number of operators, maneuverability, image quality, and method of access,
resulting in variable success rates. A detailed ESGE technology review on cholangioscopy
techniques was published recently [136 ]. All three techniques allow laser lithotripsy and EHL.
Korrapati et al. have reviewed the efficacy of peroral cholangioscopy for difficult
bile duct stones [137 ]. They estimated an overall rate of stone clearance of 88 % (95 %CI 85 % – 91 %),
with SOC showing a high technical success rate. No attempt was made to compare EHL
and laser lithotripsy.
Both EHL and laser lithotripsy are effective methods for the removal of difficult
bile duct stones, with a 69 % – 81 % clearance rate in one session and a 97 % – 100 %
clearance rate after multiple sessions [138 ]
[139 ]
[140 ]
[141 ]. However, no direct comparisons between the different methods have been published.
In one recent RCT, patients with bile duct stones > 1 cm were treated with either
laser lithotripsy or conventional therapy (included EPLBD and mechanical lithotripsy)
and achieved one-session endoscopic clearance rates of 93 % and 67 %, respectively
[142 ].
When looking at the rough data of Korrapati et al. [137 ], the complication rate ranged between 0 % and 25 % (mean 7 %, 95 %CI 6 % – 9 %).
Cholangitis is the most frequently reported complication [139 ]
[140 ]
[141 ]. Pancreatitis is a rare complication, probably owing to the high percentage of pre-existent
sphincterotomies [139 ].
Overall, the available data suggest that intraluminal lithotripsy is an effective
and safe method to treat difficult biliary stones (Table 10 s ; [Fig. 2 ]), but there are no data supporting the superiority of one method over another.
6.4 Extracorporeal shock wave lithotripsy
ESGE suggests considering extracorporeal shock wave lithotripsy when conventional
techniques have failed to achieve bile duct clearance and the intraluminal lithotripsy
techniques are not available.
Weak recommendation, low quality evidence.
ESWL uses electrohydraulic or electromagnetic energy to generate shock waves that
then travel through the soft tissues of the body to fragment CBDSs [143 ].
ESWL is a complex and technically demanding procedure. A nasobiliary drain is inserted
to allow fluoroscopic identification and targeting of CBDSs and to perform continuous
irrigation of the bile duct with saline during ESWL. In addition, multiple ESWL sessions
and subsequent ERCP procedures to extract stone fragments are required.
Ductal clearance rates of 70 % – 90 % have been reported with ESWL [52 ]
[144 ]
[145 ]
[146 ]
[147 ]
[148 ]
[149 ]
[150 ].
Several controlled trials have compared ESWL with EHL or laser lithotripsy for stone
disruption. These studies suggest that the efficacy of final duct clearance with laser
lithotripsy is superior to that of ESWL (83 % – 97 % vs. 53 % – 73 %) [146 ]
[151 ], while it is similar for EHL and ESWL (74 % vs. 78.5 %) [145 ].
ESWL-related adverse events range from 9 % to 35.7 %, including mostly cholangitis
and pancreatitis [143 ]
[145 ]
[146 ]
[152 ]
[153 ]. Minor side effects such as pain, local hematoma formation, and microhematuria are
common.
7 Endoscopic CBDS management and surgery
7 Endoscopic CBDS management and surgery
ERCP with stone clearance represents the primary and definitive treatment in patients
with CBDSs and previous cholecystectomy. In patients with CBDSs and in situ gallbladder,
both the management of CBDSs and gallbladder removal should be considered.
When ERCP is the selected technique to treat CBDSs, different options are available
with regards to the sequencing of endoscopy and surgery. Basically, ERCP can be performed
prior to (preoperative ERCP), during ongoing (intraoperative ERCP), or after (post-operative
ERCP) cholecystectomy. Preoperative ERCP is most commonly practiced, as it is highly
effective and both the endoscopist and the surgeon treat the patient in an environment
that is tailored to their own needs and routines.
7.1 The sequential strategy
ESGE recommends performing a laparoscopic cholecystectomy within 2 weeks from ERCP
in patients treated for choledocholithiasis to reduce the conversion rate and the
risk of recurrent biliary events.
Strong recommendation, moderate quality evidence.
Laparoscopic cholecystectomy represents the standard treatment for patients with CBDSs
and gallbladder stones following endoscopic CDBS clearance. A Cochrane review in 2007
[154 ], which considered five RCTs involving 662 patients treated for choledocholithiasis
with cholecystolithiasis, revealed an advantage of cholecystectomy. Over a follow-up
time varying from 17 months to more than 5 years, mortality was higher in the wait-and-see
group compared with the cholecystectomy group (14.1 % vs. 7.9 %; RR 1.78, 95 %CI 1.15 – 2.75)
and the difference persisted when only patients at high surgical risk were considered.
Similarly, endoscopic sphincterotomy followed by “wait and see” also resulted in a
higher risk of biliary events, such as cholangitis, pancreatitis, jaundice, and biliary
colic, as well as a higher risk for repeated biliary intervention (i. e. ERCP or percutaneous
procedure): 35 % of the patients managed with endoscopic sphincterotomy followed by
“wait and see” eventually underwent rescue cholecystectomy. The outcome of rescue
cholecystectomy in patients with an ASA > 3 was not significantly different compared
to elective cholecystectomy; however, patients unfit for surgery (i. e. ASA 4 and
5) were excluded in three of the five selected RCTs [155 ]
[156 ]
[157 ]. In the study by Suc et al. [158 ], 20 % of the included patients were classified as ASA 3 – 4, and mortality was not
significantly different between the two groups in the intention-to-treat analysis
(3.1 vs. 0.9 %). Also, in the RCT by Targarona et al. [159 ], mortality was not significantly different between the groups but, in the multivariate
analysis, age, and not surgical risk, was an independent predictor of mortality.
Laparoscopic cholecystectomy after ERCP with endoscopic sphincterotomy is more difficult
and when compared to standard laparoscopic cholecystectomy is mostly associated with
a higher conversion rate and a higher rate of recurrent biliary events [157 ]
[160 ]
[161 ]. In this way, the timing of cholecystectomy performance after ERCP is a critical
issue [155 ]
[157 ]
[162 ]
[163 ]
[164 ]
[165 ]
[166 ]
[167 ] (Table 11 s ). The timing of cholecystectomy may be defined as early, delayed, or on demand, but
definitions of “early” or “delayed” differ among the studies. In general, with the
exception of the study by Donkervoort et al. [168 ], where the timing of cholecystectomy did not affect the outcomes, conversion rate
results are lower in the “early group” in all studies (4 % – 23 % vs. 8 % – 55 %);
recurrent biliary events are lower when the laparoscopic cholecystectomy is performed
“early” vs. “delayed or on demand” (2 % – 10 % vs. 24 % – 47 %) [155 ]
[157 ]
[162 ]
[163 ]
[164 ]
[165 ]
[166 ]
[167 ]. Overall, data are in favor of “early” laparoscopic cholecystectomy, but the exact
timing remains controversial; despite this, waiting no longer than 2 weeks to perform
laparoscopic cholecystectomy after ERCP seems to be advisable.
In patients with acute biliary pancreatitis (ABP) and in situ gallbladder, cholecystectomy
is recommended to avoid a recurrence of pancreatitis. Some of these patients may have
previously undergone ERCP and endoscopic sphincterotomy. The timing of cholecystectomy
in mild ABP has been examined in two RCTs that randomized patients either to cholecystectomy
within 48 hours of admission vs. after resolution of abdominal pain and normalizing
trend of laboratory enzymes (n = 50) [169 ], or to cholecystectomy during the same admission vs. 4 weeks later (n = 266) [170 ]. Both studies concluded in favor of early cholecystectomy because it prevents recurrent
gallstone-related complications (one study), shortens hospitalization (one study),
and is equally safe (both studies). Similar conclusions were reached in a meta-analysis
(eight cohort studies and one RCT, 998 patients) [171 ]. For severe ABP, data are limited and, based on observational studies [172 ]
[173 ], it is recommended that cholecystectomy is performed once peripancreatic collections
and local complications have resolved, generally beyond 6 weeks, to minimize the risk
of infection in the peripancreatic collection.
In patients who do not undergo cholecystectomy following ABP, endoscopic biliary sphincterotomy
reduces biliary events, in particular pancreatitis, during follow-up [171 ]
[174 ]
[175 ]. The most recent retrospective study (1119 patients) found that recurrent pancreatitis
developed in 8.2 % vs. 17.1 % of patients with their gallbladder left in situ after
ABP who had ERCP vs. no ERCP, respectively [174 ]. However, the gallbladder should be left in situ only in patients who are unfit
for surgery as a meta-analysis (five RCTs, 662 patients) has shown that endoscopic
CBD clearance alone is inferior to prophylactic cholecystectomy associated with CBD
clearance in terms of mortality and recurrent biliary events [154 ].
7.2 Intraoperative ERCP
ESGE suggests considering intraoperative rendezvous ERCP in patients with common bile
duct stones undergoing cholecystectomy.
Weak recommendation, moderate quality evidence.
Intraoperative ERCP can be performed during laparoscopic cholecystectomy when an IOC
demonstrates the presence of CBDSs; alternatively, it can be planned either as a one-stage
approach in the treatment of combined cholecysto-choledocholithiasis or after the
failure of a preoperative endoscopic attempt at CBDS clearance.
Conventional ERCP can be performed intraoperatively, but it exposes the patient to
similar risks to a conventional ERCP performed preoperatively, albeit it is performed
during the same anesthesia [176 ]
[177 ]. Conversely, intraoperative ERCP with rendezvous cannulation offers the advantages
of being a single-stage procedure and decreasing the risk of post-ERCP pancreatitis.
Although each individual clinical trial is underpowered to validate this, there are
six RCTs [176 ]
[178 ]
[179 ]
[180 ]
[181 ]
[182 ] and approximately 15 observational studies pointing in the same direction [177 ]
[183 ]
[184 ]
[185 ]
[186 ]
[187 ]
[188 ]
[189 ]
[190 ]
[191 ]
[192 ]
[193 ]
[194 ]
[195 ]
[196 ]
[197 ] (Tables 12 s and 13 s ). These results have been confirmed by six recent meta-analyses [198 ]
[199 ]
[200 ]
[201 ]
[202 ]. The most recent of these, comparing intraoperative rendezvous ERCP with sequential
management, mainly laparoscopic cholecystectomy and preoperative ERCP, reported equal
efficacy in terms of stone clearance rate (93 % vs. 95 %), but a significantly lower
rate of morbidity (6 % vs. 11 %; OR 0.54, 95 %CI 0.31 – 0.96; P = 0.03), including post-ERCP pancreatitis (0.6 % vs. 4.4 %; OR 0.19, 95 %CI 0.06 – 0.67;
P = 0.01) and length of hospital stay in the intraoperative ERCP group [202 ]. In addition, the Swedish GallRiks registry, comprising 12 718 ERCP procedures,
demonstrated a substantial 50 % risk reduction in post-ERCP pancreatitis (3.6 % vs.
2.2 %; OR 0.5, 95 %CI 0.2 – 0.9; P = 0.002) when rendezvous cannulation was practiced [203 ].
Intraoperative rendezvous ERCP does however carry logistical problems related to the
prolonged surgical time and the need to perform ERCP in an environment that is not
adapted for endoscopy [180 ]
[182 ]
[189 ]
[191 ]
[204 ]. Failure to pass the guidewire along a narrow cystic duct or papilla is reported
in about 8 % of cases (Table 12 s ); if this happens, the endoscopist must rely on conventional cannulation techniques
and their associated risks.
7.3 Surgical treatment of CBDSs
ESGE suggests that, in patients undergoing laparoscopic cholecystectomy, transcystic
or transductal exploration of the common bile duct is a safe and effective technique
for common bile duct stone clearance. The recommendation takes into account that management
is dependent on local expertise and resources.
Weak recommendation, moderate quality evidence.
The surgical treatment of CBDSs can be performed during both laparoscopic and open
cholecystectomy. It offers the valuable opportunity to definitively treat patients
with combined cholecystolithiasis and choledocholithiasis in a one-stage procedure.
Several studies have compared laparoscopic bile duct exploration during laparoscopic
cholecystectomy with pre- or postoperative ERCP and have demonstrated no significant
differences in clinical outcomes [205 ]
[206 ]
[207 ]. However, one-stage procedures, such as laparoscopic CBD exploration or combined
endo-laparoscopic approaches, usually result in a shorter hospital stay [208 ]
[209 ]
[210 ]
[211 ]
[212 ]
[213 ]
[214 ]
[215 ]
[216 ]
[217 ]. Moreover, a recent meta-analysis has demonstrated that the one-stage laparoscopic
procedure has a higher success rate than the sequential endo-laparoscopic approach
[218 ].
It is of note that the results of surgical treatment of CBDSs, which are generally
excellent in published reports, usually originate from laparoscopic centers of excellence,
and there are hardly any data on outcomes by less experienced surgeons. Moreover,
there is a trend over the last decades that the use of endoscopic management is increasing
and surgical trainees are not gaining adequate experience in CBD exploration [219 ].
8 Special situations
Acute cholangitis and ABP may complicate CBDSs, resulting in a more difficult therapeutic
approach. Moreover, CBDSs may occur in special clinical settings, such as in pregnant
women. The endoscopic management of ABP was the object of the ESGE Guideline on endoscopic
treatment of necrotizing pancreatitis [220 ].
8.1 Acute cholangitis
The majority of patients with gallstone cholangitis have mild-to-moderate disease
that usually responds to antibiotic therapy. However, 15 % – 30 % of patients have
severe disease that needs to be handled with urgent biliary decompression [221 ].
Identification and stratification of cholangitis severity is fundamental to selecting
the appropriate treatment.
ESGE recommends using the 2018 revision of the Tokyo Guidelines to classify the severity
of acute cholangitis.
Strong recommendation, low quality evidence.
The 2013 revision of the Tokyo Guidelines [221 ], recently confirmed by the 2018 revision [222 ], classifies acute cholangitis as:
severe, dysfunction of at least one of the following systems: cardiovascular, neurological,
respiratory, renal, hepatic, or hematological system (specific criteria are stated
for each item)
moderate, any of the following: white blood cell count > 12 000 or < 4000 /mm3 , fever ≥ 39 °C, age ≥ 75 years, total bilirubin ≥ 5 mg/dL, or hypoalbuminemia
mild, no criteria of moderate/severe cholangitis.
Companion mobile applications of the 2018 Tokyo Guidelines allow easy assessment of
the severity of acute cholangitis (http://www.jshbps.jp/modules/en/index.php?content_id=47 Accessed 30 January 2019).
8.2 Timing of ERCP in acute cholangitis
ESGE recommends the following timing for biliary drainage, preferably endoscopic,
in patients with acute cholangitis, classified according to the 2018 Tokyo Guidelines:
severe, as soon as possible and within 12 hours for patients with septic shock
moderate, within 48 – 72 hours
mild, elective.
Strong recommendation, low quality evidence.
Twelve studies (18 206 patients), all retrospective, have analyzed the relationship
between the timing of biliary drainage and different outcomes (Table 14 s ). An international study from 28 intensive care units published in 2016 included
260 patients with septic shock (defined as hypotension requiring vasopressors plus
several other criteria); it found that waiting longer than 12 hours from the onset
of shock to successful biliary drainage was associated with higher in-hospital mortality
(OR 3.4, 95 %CI 1.12 – 10.31) [223 ]. Overall, in-hospital mortality was 37 % and median time to biliary drainage was
12 hours, with 10 % of patients having drainage after 48 hours [223 ].
The other 11 studies were not restricted to patients with disease that was so severe
[224 ]
[225 ]
[226 ]
[227 ]
[228 ]
[229 ]
[230 ]
[231 ]
[232 ]
[233 ]
[234 ]; they revealed, among the studies that analyzed the specific matter, that: mortality
was associated with delayed ERCP in two of four studies [223 ]
[233 ]; organ failure (alone or as part of a composite index) was associated with delayed
ERCP in three of five studies [226 ]
[227 ]
[230 ]; length of hospital stay was associated with the timing of ERCP in seven of eight
studies [225 ]
[227 ]
[229 ]
[230 ]
[232 ]
[233 ]
[234 ]; hospitalization costs were higher when ERCP was delayed in both studies that analyzed
that association [230 ]
[233 ].
ESGE recommends other biliary drainage modalities (percutaneous, surgical) in patients
with acute cholangitis due to common bile duct stones when ERCP is not feasible/successful
within the recommended timeframes.
Strong recommendation, low quality evidence.
Failure of biliary drainage is a strong determinant of mortality, particularly in
patients with severe cholangitis. For example, in the abovementioned study of patients
with septic shock [223 ], 40 of 42 patients with failed biliary drainage (95.3 %) died as compared with 55
of 213 patients with successful biliary drainage (25.8 %). In that study, biliary
drainage was achieved by ERCP, percutaneous transhepatic biliary drainage (PTBD),
and surgery in 91, 90, and 34 patients, respectively. Similarly, in a study not restricted
to patients with severe disease [225 ], three of six patients with failed biliary drainage (50 %) died as compared with
two of 321 patients with successful biliary drainage (0.6 %).
8.3 Management of CBDSs in pregnant woman
ESGE recommends that therapeutic ERCP is a safe and effective procedure in pregnant
women, provided that it is performed by experienced endoscopists and the radiation
exposure to the fetus is kept as low as possible.
Strong recommendation, moderate quality of evidence.
According to six retrospective studies (144 patients), ERCP in pregnant women seems
to be a relatively safe examination throughout the whole gestation [235 ]
[236 ]
[237 ]
[238 ]
[239 ]
[240 ]. ERCP should only be performed for therapeutic purposes as EUS and MRCP are highly
accurate for the diagnosis of biliary obstruction. Furthermore, it should be performed
by experienced endoscopists as radiation dose, as well as the overall complication
rate, decreases with the experience of the endoscopist [241 ]
[242 ]
[243 ]
[244 ].
With respect to the potential harm related to X-rays, ERCP is best carried out during
the second trimester of pregnancy; during the first trimester, the phase of organogenesis,
the fetus is especially sensitive to radiation and, during the third trimester, there
is a close topographic proximity of the growing fetus to the path of the X-rays.
Guidelines have usually recommended using as little radiation as reasonably achievable
[243 ]
[245 ]. A threshold radiation dose is assumed for deterministic effects only (10 mGy),
not for stochastic effects (cancer induction) [246 ]. Therefore, as many steps as possible should be taken to keep radiation exposure
as low as possible. These are described in the ESGE Guideline on radiation protection
in digestive endoscopy [243 ]. Non-radiation ERCP (NR-ERCP) has also been proposed; it uses various techniques
such as aspiration of bile through the cannulation catheter to confirm biliary cannulation,
ultrasound guidance, peroral cholangioscopy, or a two-stage approach consisting of
biliary stenting followed by stone extraction after parturition. A systematic review
summarized 22 case reports and retrospective studies that used NR-ERCP (180 patients
in total) [247 ]. They concluded that pregnancy outcomes were not significantly affected by NR-ERCP,
although whether the avoidance of radiation is beneficial for the baby remains unknown,
but noted that NR-ERCP is technically demanding.
Disclaimer
The legal disclaimer for ESGE Guidelines [12 ] applies to the current Guideline.
