Surgical versus endoscopic treatment of bile duct stones
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
In four separate analyses, our objectives are:
1. To compare open cholecystectomy and BDS extraction versus ERCP/ES.
2. To compare laparoscopic surgical versus endoscopic BDS clearance in patients who have previously had cholecystectomy, i.e., primary or retained BDS.
3. To compare pre‐operative ERCP/ES versus laparoscopic BDS clearance in patients who presented for laparoscopic cholecystectomy.
4. To compare laparoscopic versus post‐operative ERCP/ES BDS clearance in patients who were found to have BDS at laparoscopic cholecystectomy.
Background
Most patients with symptomatic gallstones are recommended to undergo cholecystectomy to alleviate symptoms of pain and jaundice and prevent complications such as pancreatitis, cholangitis, and cholecystitis (NIH Consensus 1993). In patients who have cholecystectomy for gall stones, approximately 10‐18 per cent also have bile duct stones (BDS) present (Lezoche 1996; Courvoisier 1890; Soltan 2001). BDS are suspected pre‐operatively by a history of jaundice, dark urine, or recent pancreatitis or cholangitis, by derangement in liver function tests, or by duct dilatation and/or duct stones demonstrated on ultrasound examination of the upper abdomen. However, up to 25 per cent of patients with BDS have unsuspected duct stones found at surgery (Millat 1995). In the pre‐laparoscopic era, most BDS found at surgery were managed surgically, and only a minority were managed endoscopically (Fletcher 1994). Studies suggested that surgical BDS extraction was the recommended option for routine cases (Neoptolemos 1989). Laparoscopic cholecystectomy has become the treatment of choice for gallbladder removal over the past two decades (NIH Consensus 1993). In the early days of laparoscopic biliary surgery, operative clearance of bile duct stones as a part of a cholecystectomy was not technically possible, and either open surgical clearance or, more commonly, endoscopic retrograde cholangiopancreatography/endoscopic sphincterectomy (ERCP/ES) became the techniques used to clear duct stones. This has resulted in a dramatic increase in the rates of ERCP/ES (Fletcher 1994). As laparoscopic clearance of BDS has become more technically feasible, it is now important that comparisons are again made between the surgical and endoscopic approaches to clearing duct stones (Memon 2000). Several studies have shown that laparoscopic treatment of BDS is possible and is as effective as ERCP/ES (Lezoche 1996; Millat 1995; Rhodes 1998). Since ERCP/ES has additional associated morbidity and costs, and moreover the patient must undergo an additional procedure, laparoscopic treatment for BDS may prove to be the preferred treatment for patients able to undergo laparoscopic cholecystectomy (Millat 1995). In addition, some centres have become so familiar with laparoscopic BDS clearance that they prefer this method to that of endoscopic clearance in the case of primary or retained BDS when the gall bladder has previously been removed. This practice, too, needs to be assessed for its suitability in the treatment of these patients.
Objectives
In four separate analyses, our objectives are:
1. To compare open cholecystectomy and BDS extraction versus ERCP/ES.
2. To compare laparoscopic surgical versus endoscopic BDS clearance in patients who have previously had cholecystectomy, i.e., primary or retained BDS.
3. To compare pre‐operative ERCP/ES versus laparoscopic BDS clearance in patients who presented for laparoscopic cholecystectomy.
4. To compare laparoscopic versus post‐operative ERCP/ES BDS clearance in patients who were found to have BDS at laparoscopic cholecystectomy.
Methods
Criteria for considering studies for this review
Types of studies
We will include randomised clinical trials (RCTs) and quasi‐randomised trials which compare ERCP/ES and surgical (open or laparoscopic) treatment for BDS. Historically controlled trials, cohort studies, and case series will be excluded. Trials will be included from journal articles, abstracts, and unpublished studies in any language irrespective of blinding.
Types of participants
Patients included:
-
Those found to have BDS having previously had a cholecystectomy (analysis 1).
-
Those suspected to have BDS prior to open or laparoscopic cholecystectomy (analysis 1 and 2).
-
Those found to have BDS at open or laparoscopic cholecystectomy (analysis 1and 3).
Patients excluded:
-
Those not fit for cholecystectomy.
-
Those not fit for ERCP/ES.
-
Those in whom ERCP was not possible (e.g., previous Billroth II operation).
Types of interventions
-
ERCP performed with stone extraction as necessary, most commonly accompanied by ES and either balloon or basket extraction.
-
Laparoscopic cholecystectomy either with gas insufflation of the peritoneum (usually with CO2) or with an abdominal wall lifting device.
-
BDS discovered by the use of intra‐operative cholangiography, via a trans‐cystic approach (most commonly), or via the gallbladder, or intra‐operative choledochoscopy.
-
BDS clearance via either the cystic duct or common bile duct/common hepatic duct (CBD/CHD). (Discrimination between these two routes has implications for surgical morbidity and therefore requires comparison.)
-
BDS clearance achieved by either flushing (with or without the aid of medications such as glucagon or buscopan), balloon extraction, Dormia basket extraction (with or without the use of choledochoscopy), or with the aid of intra‐operative ERCP and either antegrade or retrograde sphincterotomy.
Types of outcome measures
-
Mortality at 30 days and at maximum follow‐up.
-
Duct clearance as determined by cholangiogram at the time of intervention, lack of symptoms, signs or abnormal biochemistry/liver function tests.
-
Numbers and size of stones cleared.
-
Number of procedures.
-
Duration of operation and/or procedure(s).
-
Duration of hospital stay.
-
Postoperative pain score (questionnaire or visual analogue scales), ideally administered during the first 48 hours and at day 30.
-
Use of analgesics.
-
Complications from surgery and procedures such as bile duct injuries, pancreatitis, cholangitis, pulmonary/cardiac/renal complications.
-
Quality of life.
-
Health economics.
Search methods for identification of studies
Electronic databases, such as MEDLINE, Health star, Ovid Biomedical collection, The Cochrane Hepato‐Biliary Group's Controlled Trials Register, EMBASE, and The Cochrane Collaboration Controlled Trials Register will be searched for articles on laparoscopic and ERCP/ES treatment of bile duct stones. Abstracts from Digestive Diseases Week since 1988 and the International Hepatobiliary Association will be searched. Abstracts of the United European Gastroenterological Week since 1991 will be searched. Authors of identified abstracts will be contacted if the results have not been subsequently published. Principal authors of identified RCTs will be contacted to discover if they know of additional unpublished trials.
Search terms such as laparoscopic, duct, exploration, cholecystectomy, ERCP, ERC, duct stones, and choledocholithiasis will be used to look for articles in all languages possible (see below).
Search strategy:
1 exp cholelithiasis/ or "cholelithiasis".mp.
2 (bil$ and duct$ and stone$).mp. [mp=title, abstract, registry number word, mesh subject heading]
3 exp Common bile duct calculi/
4 exp biliary tract/ or "biliary tract".mp.
5 exp biliary tract diseases/ or "biliary tract diseases".mp.
6 exp bile/ or "##'Bile$'.mp.##"/ or "bile$".mp.
7 "BILIARY$".mp.
8 choled$.mp. [mp=title, abstract, registry number word, mesh subject heading]
9 hyperbilirubin$.mp. [mp=title, abstract, registry number word, mesh subject heading]
10 choledocholithiasis.mp. [mp=title, abstract, registry number word, mesh subject heading]
11 1 or 2 or 3 or 4 or 6 or 7 or 8 or 9 or 10
12 randomized controlled trial.pt.
13 randomized controlled trials.mp. [mp=title, abstract, registry number word, mesh subject heading]
14 exp Double‐blind method/
15 exp Random allocation/
16 exp single‐blind method/
17 clinical trials.pt.
18 exp Clinical trials/
19 (clin$ adj trial$).mp. [mp=title, abstract, registry number word, mesh subject heading]
20 (clin$ adj5 trial$).ti.
21 (clin$ adj5 trial$).ab.
22 ((singl$ or double$ or treb$ or tripl$) adj 50 (blind$ or mask$)).mp. [mp=title, abstract, registry number word, mesh subject heading]
23 ((singl$ or double$ or treb$ or tripl$) adj 50 (blind$ or mask$)).ti. or ((singl$ or double$ or treb$ or tripl$) adj 50 (blind$ or mask$)).ab.
24 exp Placebos/
25 placebo$.ti. or placebo$.ab.
26 random$.ti. or random$.ab.
27 exp Research design/
28 exp controlled clinical trials/ or "controlled clinical trials".mp.
29 exp prospective studies/ or "prospective studies".mp
30 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25 or 26 or 27 or 28 or 29
31 limit 30 to animal
32 limit 30 to (human and animal)
33 31 not 32
34 30 not 33
35 exp cholecystectomy, laparoscopic/ or "##'Laparoscopic cholecystec$'.mp##"/ or "laparoscopic cholecystec$".mp.
36 exp laparoscopy/ or "laparoscopy".mp.
37 exp Cholangiopancreatograpy, endoscopic retrograde/
38 endoscopic retrograde cholangiopancreatography.mp. [mp=title, abstract, registry number word, mesh subject heading]
39 erc$.mp. [mp=title, abstract, registry number word, mesh subject heading]
40 exp cholangiography/ or "cholangiography".mp.
41 exp biliary tract surgical procedures/ or "biliary tract surgical procedures".mp.
42 "CHOLEC$".mp.
43 exp endoscopy/ or "endoscopy".mp
44 exp sphincterotomy, endoscopic/ or "endoscopic sphincterotomy".mp
45 35 or 36 or 37 or 38 or 39 or 40 or 41 or 42 or 43 or 44
46 11 and 34 and 45
Data collection and analysis
The review will consist of four parts, as described above in 'Objectives'.
Studies considered for inclusion will be considered by two reviewers (JJ & DV) for eligibility.
As previously described, only RCTs and quasi‐randomised trials are eligible. Historically controlled studies, cohort studies, and case series will be excluded.
In each case, details of the methods of duct clearance will be extracted from the included trials. Success rates for duct clearance, including size and number of stones retrieved/cleared will be extracted. Duration of procedures, number of procedures, equipment used, and cost of procedures will be extracted. All surgical manoeuvres to obtain BDS clearance will be incorporated together, similarly for all endoscopic techniques. In the event that there are a significant number of different modalities used within each study, sub‐group analyses will be performed.
Laparoscopic operations which are converted to open operations will remain in the laparoscopic group on an intention‐to‐treat basis. Retained stones from any treatment will be seen as a failure of that treatment. Patients with clear bile ducts at one month or more post‐operatively as determined by cholangiography (preferred), normal biochemistry, or lack of symptoms will be regarded as success for that treatment. Successful combinations of treatments will also be assessed.
In order to assess trial quality, the following components will be assessed:
-
Patient numbers. As part of a satisfactory prospective study design, evidence of power analysis calculations should be displayed by the study authors. In order to achieve this, a realistic estimate of baseline variability in the primary outcome measure must be stated.
-
Allocation concealment. The preferred method to conceal allocation is to involve a third party not participating in administration of therapies who is contacted at enrolment of the patient and who allocates the treatment arm. In practice, however, envelopes or card systems are more commonly used. The trials will be subjected to sensitivity analysis, forming a subgroup with adequate allocation concealment (i.e., centralised randomisation, or opaque, serially numbered, sealed envelopes, or similar) versus trials with inadequate methods (i.e., in which allocation concealment is not described or is performed with methods open to the breaking of the concealment) versus quasi‐randomised trials.)
-
Blinding. Due to the nature of the interventions in this study, patient blinding is not really practicable. Surgeon, endoscopist, and carer blinding may be more feasible; assessor blinding would certainly be feasible and could reasonably be expected. Sensitivity analyses will be prepared subdividing the trials into groups with or without adequate blinding (assessor or otherwise).
-
Description of withdrawals and dropouts. This includes treating operations converted to open procedures on an intention‐to‐treat basis, rather than as exclusions.
A Jadad score (Jadad 1996) will be calculated, bearing in mind the problems with blinding. Only trials with a score of 3 or more will be considered high quality trials.
Trials included in the meta‐analysis will have heterogeneity between trials investigated by visual inspection of the forest plot and by use of the Chi square test as calculated by RevMan using a P value of 0.1 (i.e., if the Chi square is found to be less than the degrees of freedom, the trials are likely to be homogeneous and can be combined). If significant heterogeneity is found between the trials, careful examination of reasons for these differences will be looked for in these trials. Any resulting post‐hoc subgroup sensitivity analysis will be clearly stated and explained.
Fixed and random effects models will be used to analyse and present data from the trials included in the meta‐analysis. The Peto odds ratio will be used as a fixed effect model and the Der Simonian method will be used to examine random effects.
A funnel plot of trials undergoing meta‐analysis will be examined to determine if any bias is present (Egger 1997).
The trials not included in the meta‐analysis will have their quality assessed and the results from each trial will be subjected to graphical analysis for discussion of efficacy of the treatments.
