One-step laparoscopic cholecystectomy with common bile duct exploration and stone extraction versus two-step endoscopic retrograde cholangiography with stone extraction plus laparoscopic cholecystectomy for patients with common bile duct stones: a statistical analysis and central data monitoring plan for a pilot and feasibility randomised clinical trial

doi:10.21203/rs.3.rs-3951475/v1

Background

Endoscopic retrograde cholangiography (ERC) with stone extraction and papillotomy with subsequent laparoscopic cholecystectomy — the two-step approach — is the standard treatment of common bile duct stones in many countries. However, ERC is associated with a high risk of complications, and more than half of patients require multiple ERCs. Meta-analyses of randomised clinical trials find no major differences of the two-step approach in comparison with laparoscopic cholecystectomy with intraoperative laparoscopic stone clearance — the one-step approach. Currently, there are insufficient data to ascertain superiority.

Methods

The preGallstep trial is an investigator-initiated, multicentre randomised pilot and feasibility clinical trial with blinded outcome assessment. Eligible participants are patients with common bile duct stones (identified by magnetic resonance cholangio-pancreatography), age 18 years or above, and with the possibility to perform both interventions within a reasonable time. We intend to randomise 150 participants allocated 1:1. The experimental intervention is the one-step approach. This consists of laparoscopic common bile duct exploration plus laparoscopic cholecystectomy. The control intervention is the two-step approach which consists of ERC plus sphincterotomy (first step) and subsequent laparoscopic cholecystectomy (second step). The pilot outcome is the proportion of participants with at least one postoperative complication according to the Clavien-Dindo score grade II and above until 90 days after randomisation. This outcome will be used for a future sample size calculation of a larger pragmatic trial. A range of feasibility outcomes will be assessed to ascertain feasibility of the trial. Furthermore, a range of clinical outcomes will be assessed exploratorily. After completion of the feasibility trial, blinded data will be analysed by two independent statisticians blinded to the intervention, where ‘A’ and ‘B’ refers to the two groups. A third party will compare these reports and discrepancies will be discussed. The statistical report with the analyses chosen for the manuscript is being tracked using a version control system, and both statistical reports will be published as supplementary material. Based on the final statistical report, two blinded conclusions will be drawn by the steering group.

Discussion

We present a pre-defined statistical analysis plan and data monitoring scheme for the preGallstep pilot and feasibility trial, which, limits bias, p-hacking, and data-driven interpretations.

Trial registration: ClinicalTrials.gov, NCT04801238. Registered on March 16, 2021 https://clinicaltrials.gov/ct2/show/NCT04801238

In Denmark alone, more than 7,500 cholecystectomies are conducted each year making it one of the most common general surgical procedures (1). Common bile duct gallstones (CBDS) are found in up to 18% of patients undergoing cholecystectomy (2). CBDS are associated with the risk of pain and jaundice and are the leading cause of acute pancreatitis and acute cholangitis (3, 4). The two-step approach including endoscopic retrograde cholangiography (ERC) with stone extraction and papillotomy plus a subsequent laparoscopic cholecystectomy has become standard treatment of CBDS (5). However, ERC is associated with a high risk of postoperative pancreatitis, and more than half of patients may require multiple ERCs due to retained stones (2, 6). Furthermore, long term follow-up indicates that ERC with papillotomy is associated with a higher risk of developing bile duct cancer (7)

Recent randomised clinical trials have shown comparable proportions of successful CBDS clearance, risk of short-term postoperative complications (perioperative bleeding, postoperative infections, and damage to the biliary structures), and mortality between the two-step versus the one-step approach (8–13). The one-step approach entails laparoscopic cholecystectomy with intraoperative common bile duct exploration and stone clearance. A one-step approach may seem beneficial compared with the two procedures separated by a couple of days or weeks. However, the one-step procedure requires special equipment, special surgical training, and often longer duration of the operation. Furthermore, anatomical variations in the biliary system, number of stones, or the size of stones may influence on the possibility of retrieving the stones during the one-step approach. Moreover, recently published meta-analyses and a systematic review with meta-analyses find that the one-step approach may be superior to the two-step approach in terms of safety, including perioperative complications, conversion to other procedures, CBDS clearance, hospital stay, operative time, in-hospital costs, and stone recurrence (14–16). However, these meta-analyses selected only English language studies; included only fully published articles; included trials with questionable randomisation, lack of confirmation of CBDS on preoperative imaging, lack of blinding, lack of follow-up, and did not assess patient-reported outcome measures (PROMs) (14–16). Moreover, only 1/13 randomised clinical trials (8–13, 17–23) assessed systematically post-operative complications by the Clavien-Dindo classification (23).

The Clavien-Dindo score grades the post-operative complications according to requirement of treatment needed (24). The five-grade scale contains grade I defined by a mild deviation during the post-operative cause; grade II as requiring pharmacological drugs or blood transfusion; grade III as requiring surgical, endoscopic, or radiological intervention; grade IV as organ failure requiring intensive care unit (ICU) treatment; and finally grade V as death. Furthermore, Trial Sequential Analysis (TSA) (25–28) was only carried out in one of the meta-analyses (15). The TSA focused on proportion of successful CBDS clearance and excluded trials with only clinical suspicion of CBDS without radiology confirmation. The overall conclusion of the systematic review suggests a potential for a true superiority of the one-step approach, but the TSA is still underpowered.

To date, no large-scale randomised clinical trial has been conducted, presumably because of the complexity of the trial setup. More importantly, heterogenicity of previous trials is present in the form of exclusion of randomised patients without CBDS at intraoperative cholangiography (11, 20, 22), exclusion of patients that did not complete the protocolled treatments (10–11, 18), randomisation of patients with only a clinical suspicion of CBDS (11, 17–18, 20, 22–23), and only few trials randomised patients with CBDS proven by magnetic resonance cholangio-pancreaticography (MRCP) or endoscopic ultrasound (EUL) (8–11, 21). Before conducting a large-scale pragmatic trial, it is crucial to investigate the feasibility and the practical approaches needed. Therefore, we launched a pilot and feasibility trial – the preGallstep trial – prior to conducting a large pragmatic randomised clinical (29). Here, we present our detailed statistical analysis and central data monitoring plan for the preGallstep trial.

The preGallstep trial is conducted according to the 2013 SPIRIT statement (30). The preGallStep trial is an investigator-initiated, multicentre randomised parallel group, pilot and feasibility clinical trial, with blinded outcome assessment comparing the one-step versus the two-step approach. The preGallstep trial is approved by the Regional Committee on Health Research Ethics in the Capital Region (H-20041609, 4 March 2021) and the Danish Data Protection Agency in the Capital Region (P-2020-1056, 13 November 2020) which among other things ensures adherence to the General Data Protection Regulation (29). Patients not included in the trial or withdrawing their consent will be offered the course of treatment deemed most appropriate by their attending surgeon. The trial was registered on ClinicalTrials.gov (identification no. NCT04801238) before inclusion of the first participant. The first participant was included and randomised on 22 April 2021 and recruitment is ongoing.

Objectives

The main objective of this randomised pilot and feasibility clinical trial is to estimate the proportions of participants in each intervention group with post-operative complications according to the Clavien-Dindo classification grade II or more. The aim is for the estimates to be used in a sample size estimation for a future pragmatic randomised clinical trial. Given the sample size of the current trial, any results will be purely hypothesis-generating. Secondary, we want to assess the feasibility of conducting a future larger pragmatic trial, by assessing several feasibility outcomes. Further, we aim to explore the effects of the interventions on a range of clinical outcomes.

Participant timeline

Eligible patients can enter the trial through a variety of ways. While the most frequent in-hospital entry is through the emergency department, a few patients with CBDS will also be referred to the out-patient clinic from other departments or from private practices.

Patients will be offered enrolment into the trial if all inclusion and no exclusion criteria are met (see ‘Criteria for eligibility’). Patients will be informed of the trial by the attending surgeon and offered participation. Written and oral informed consent shall be obtained, and baseline data collected. Central randomisation will be carried out by the trial site investigator. If no immediate complications occur during or after the interventions, the patients will be discharged within 24 hours. Blood samples including serum amylase will be drawn 24 to 36 hours postoperatively. A 90-day follow-up will be performed to assess patient-related outcomes. Radiographic imaging will be performed only if clinically indicated. Registration of lost to follow-up and reasons will also be assessed.

Criteria for eligibility

At inclusion participants must meet the following Inclusion criteria:

CBDS identified by MRCP;
age 18 years or older;
ability to perform both interventions within reasonable time; and
informed consent.

Furthermore, the participants must meet none of the exclusion criteria:

Common bile duct cysts;
pancreatic/biliary/hepatic malignancies;
prior cholecystectomy or sphincterotomy;
chronic pancreatitis;
cholangitis grade 3 according to the Tokyo Guidelines (cholangitis with organ dysfunction) (31);
previous gastric-bypass surgery or other previous surgery preventing ERC, LC, or LCBDE;
pregnancy; and
if the patient is unable to give an informed consent.

Randomisation

Participants will be randomised at the allocation ratio 1:1 performed centrally at the Copenhagen Trial Unit (Copenhagen, Denmark) using a computer-generated allocation sequence with a varying block size concealed from the investigators. The allocation sequence will be stratified by trial site only. Copenhagen Trial Unit will generate the allocation sequence, and participants are enrolled using a web-based system developed by the unit. The designated intervention will be carried out within 24 hours of randomisation or no longer than within twice weekly if local facilities are unable to provide operating facilities within 24 hours of randomisation.

Blinding

The obvious advantages of the one-step approach compared with the two-step approach are the fewer procedures required for CBDS clearance and removal of the gallbladder and thus, a shorter course of treatment. Due to the nature of the surgical and endoscopic interventions, blinding of patients or surgeon/endoscopist is not possible in this trial. To prevent dropout due to patient preferences, patients are blinded from assigned intervention until up to at most 72 hours prior to surgery. Most of the outcomes are dependent on the physician’s clinical assessment. However, we will engage a blinded adjudication committee of three independent experts who will examine medical charts from randomisation to 90-days after first surgical intervention for outcome assessment. The medical charts presented to the adjudication committee will be blinded for any phrases related to the intervention, and the committee will thereby be blinded to the intervention.

Baseline characteristics

We will present the following baseline characteristics stratified by group:

Age, sex, weight, height, WHO performance score
Preoperative diagnose
Days from MRCP to intervention
Baseline serum or plasma biochemistry: leucocytes; c-reactive peptide (CRP); alkaline phosphatases (ALP); bilirubin; and amylase
Number of CBDS on MRCP
Size of largest stone on MRCP
Anatomical position of CBDS.

Outcomes

Pilot outcome

The pilot outcome is:

The proportion of participants in both intervention groups with at least one post-operative complication during the 90 days follow-up, assessed according the Clavien-Dindo score grade II and above.

This outcome will be used to ascertain proportions in each group, and possible difference between the groups will be attributed type-1 error.

Feasibility outcomes

The quantitative feasibility outcomes are:

Proportion of eligible patients consenting to inclusion.

Eligible participants are to be persons who fulfil all inclusion criteria and none of the exclusion criteria, besides informed consent. If the number of participants randomised out of the number of eligible persons is 150 out of 200, the proportion will be 75%. The 95% confidence interval (CI) will be: 69% to 81 %. A randomisation proportion of 69% or more will be acceptable for a future trial, while a fraction below 69% will impose serious problems of recruitment for a future large pragmatic trial.

The qualitative feasibility outcomes are:

consumption of manpower;
difficulties getting the first participant randomised at each clinical site;
reasons for not being eligible for inclusion;
reasons for declining participation.
difficulties during the informed consent procedure;
difficulties with randomisation;
difficulties in data management;
difficulties with blinding patient charts and forms;
difficulties in maintaining blinding for the outcome assessors.

The qualitative outcomes will be assessed by discussions within the investigator group. The qualitative outcomes will be described in detail and used to assess the feasibility of conducting a later confirmatory trial.

Exploratory clinical outcomes

All clinical outcomes are assessed at 90 days after randomisation:

Proportion of participants with stone clearance failure, defined as participants in each intervention group with CBDS.
Mean number of additional ERC needed to obtain safe clearance of CBD.
Mean length of hospital stay (days).
Mean procedure time.
Quality of life assessed with Short-Form 36 (SF-36), total score.
Postoperative Liver biochemistry:
1. CRP
2. leucocytes
3. ALP
4. amylase
5. bilirubin
Proportion of participants with any complication, defined as Clavien-Dindo score (0; I; II, III; IV; V) in each group.

Subgroup analysis

We plan on assessing the primary outcome using subgroup analyses with the median stone sizes, number of stones, and grouped by anatomical position of the stones. These analyses will involve adding treatment-by-subgroup interaction terms to the same models as used to analyse the primary outcome and assessing the statistical significance of these interaction terms. If significant interaction is identified, we will carry out analysis for a subset of the data using each of the subgroups.

Sample size

This is a pilot and feasibility trial assessing the possibility of conducting a large-scale, pragmatic randomised clinical trial with the same primary outcome. Thus, no formal sample size estimation has been conducted (32). We pragmatically aim to include 150 participants in total, 75 in each group. With current numbers of procedures performed at each institution per year, the necessary trial inclusion time to include patients is expected to be 18 months.

Central data monitoring

Central data monitoring will be initiated no later than after inclusion of one third of the participants. Every month a central data monitoring report will be sent to the steering committee for review. Overall, the report will include a simplified trial status, missing data overview, and data deviation figures. The steering committee consists of experienced clinicians, statisticians, and trialists; thus, the necessary composition to ensure optimal monitoring (33). The aim of the central data monitoring is to optimise completeness, quality and minimise deviations through blinded evaluation of the data (34).

General analysis principles

Statistical analyses will be handled using the latest available stable version of R (R Core Team, Vienna, Austria) and/or Stata (StataCop LLC, Texas, USA). All randomised participants will be included in all analyses. The baseline characteristics will be presented for each group. The pilot outcomes and all clinical outcomes are exploratory and will be interpreted as such.

Statistical analysis

Analysis of feasibility outcomes

The feasibility outcomes were pragmatically decided based on consensus and agreement between the investigators and were based on clinical expertise (Kirkegaard-Klitbo A, Shabanzadeh DM, Sørensen LT) and trial experience from previous pragmatic and feasibility trials (Lindschou J, Gluud C, and Olsen MH). These are all seen as relevant for carrying out a definitive large-scale trial. The quantitative primary feasibility outcome which are fractions will be presented together with the confidence intervals using a 1-sample proportions test with continuity correction, with an adjusted maximum confidence limit of 100%. The qualitative primary feasibility outcomes will be interpreted primarily by merging and counting comparable feasibility issues and discuss these.

Analysis of pilot outcome and exploratory clinical outcomes

We plan to analyse the exploratory clinical outcomes as we plan to analyse these outcomes in the planned larger pragmatic trial, by choosing the analyses which fulfils the assumptions. The results will be interpreted with caution as this trial is not powered to investigate clinical outcomes, but the signals will help inform which outcomes we might choose for the larger pragmatic trial.

Continuous outcomes

Continuous exploratory clinical outcomes will be presented as means and standard deviations (SD) or 95% confidence interval (CI) for each group, with an annotation in the tables of the percentage of missing data per group. As previously recommended, we will use linear regression analyses adjusted for the baseline value for the continuous exploratory clinical outcomes (32).

Count data outcomes

Count data exploratory clinical outcomes will be presented as medians and interquartile ranges for each group, with an annotation in the tables of the percentage of missing data per group. Count data exploratory clinical outcomes will be analysed using the van Elteren test from Stata or an equivalent in R (33,34). The results will be presented with median differences and Hodges-Lehmann confidence intervals to demonstrate the uncertainty of the results (35).

Dichotomous outcomes

Dichotomous exploratory clinical outcomes will be presented as proportions for each group with an annotation in the tables of the percentage of missing data per group. Dichotomous exploratory clinical outcomes will be analysed using logistic regression. We will estimate the marginal effects to obtain RRs and confidence intervals of the RRs (based on ‘nlcom’ from Stata (StataCorp LLC, Texas, USA)) or by bootstrapping in R.

Handling of missing data

No specific methodology, including multiple imputation, will be used to handle missing data, but missingness will be listed in detail in the tables in the statistical reports (see below) as a tool to adapt the design of a larger pragmatic randomised trial.

Assessments of underlying statistical assumptions

The chosen analyses have few assumptions, with the main assumptions being related to the linear and logistic regressions (35,39). The variables included in the linear regression models will be visually assessed for normal distribution using histograms and quantile-quantile plots of the residuals, and for homogeneity using residuals plotted against covariates and fitted values, with the possibility of a logarithmic transformation or applying robust standard errors to minimise deviations from the model (35). The confidence interval for count outcomes will be derived from the Mann-Whitney U test and if they do not converge with the p-value from the Van Elteren test, we will interpret any significance with caution.

The deviance divided by the degrees of freedom for logistic regression model will be calculated to assess relevant overdispersion. The logistic regression used will be univariable, i.e. with no covariates, and if few or zero events are identified (substantially lower than the rule of thumb of 10 events) the analyses will be carried out using Fisher’s exact test. The robustness of the confidence intervals and p-values might be affected by the small sample size and these will be interpreted with caution [32].

Early stopping

If the sample size of 150 participants is not reached, adjustment of the risk of type-1 should be carried out. In confirmatory trials, the statistical significance level would normally be adjusted according to the Lan-DeMets’ sequential monitoring boundaries based on O'Brien Fleming alpha-spending function (25-28). Since we do not have any set parameters for minimum relevant difference and relative risk reductions, we will adjust the significance level for all analyses to below 0.01. Given the trial is designed as a pilot and feasibility trial, we will interpret signal of significance cautiously.

Statistical reports

After completion of the trial and the final central data monitoring discrepancies are handled, blinded data will be analysed by two independent statisticians blinded to the intervention, where ‘A’ and ‘B’ refers to the two groups. The two statisticians will independently analyse all data and present the results in two independent reports. The coordinating investigator, the two statisticians and the Steering Committee will compare these reports and discrepancies will be discussed. The statistical reports will be published as supplementary material. Based on the final statistical report, two blinded conclusions will be drawn by the Steering Committee: One assuming ‘A’ is the experimental group and ‘B’ is the control group – and one assuming the opposite. These abstracts will utilise the results from the blinded reports, and when the blinding is broken, the ‘correct’ abstract will be chosen and the conclusions in this abstract will not be revised. This described process of analysing data and interpreting data will also be used in the future large randomised clinical trial.

The results of the preGallstep trial are presented in the following mock tables. The quality feasibility outcomes will be presented in text.

Table 1: Baseline characteristics

	LC +LCBDE	ERC +LC
N	75	75
Age (mean (SD))	54 (7)	58 (8)
Gender, female (n, %)	50 (67%)	32 (43%)
BMI (mean (SD))	29(3)	27(2)
WHO Performance score (mean (SD))	1(1)	1(1)
Preoperative diagnosis (n, %) CBDS jaundice cholecystitis cholangitis pancreatitis other	42 (56%) 4 (5%) 4 (5%) 11 (14%) 14 (19%) 0 (0%)	36 (48%) 6 (8%) 1 (1%) 17 (23%) 15 (20%) 0 (0%)
Days from MRCP to intervention (mean (SD))	22 (6)	15 (7)
Baseline biochemistry (mean, SD)) leukocytes CRP ALP Bilirubin Amylase	11 (6) 43 (7) 250 (20) 42 (8) 68 (9)	12(6) 48 (8) 312 (16) 39 (7) 72 (5)
Number of CBDS on MRCP (mean (SD))	3 (2)	5 (2)
Size of Largest stone on MRCP (mm) (mean (SD))	5 (1)	8 (2)
Anatomical position of CBDS (n, %) Cystic duct Common bile duct Common hepatic duct intrahepatic	2 (2%) 64 (86%) 8 (11%) 1 (1%)	3 (4%) 56 (75%) 14 (19%) 2 (2%)

Table 2: Pilot outcome

LC + LCBDE

ERC + LC

Estimate

RR (95%CI)

p-value

N

75

Proportion of patients with clavien-dindo >2

10 (13%)

12 (16%)

1.05 (0.80-1.25)

0.43

Table 3: Exploratory clinical outcome

	LC +LCBDE	ERC + LC	Estimate (95%CI)	p-value
N	75	75
Stone clearance failure (n, %)	6 (8%)	9 (12%)	RR 1.03 (0.70-1.36)	0.56
Number of additional ERC (n, %))	4 (5%)	42 (56%)
Length of stay (mean (SD))	4 (1)	5 (1)	0.7 (-0.3;2.1)	0.94
Procedure time (mean (SD))	104 (12)	38 (4)
Postoperative biochemistry (mean (SD)) leukocytes CRP BASP Bilirubin Amylase	11 (6) 43 (7) 250 (20) 42 (8) 68 (9)	12(6) 48 (8) 312 (16) 39 (7) 72 (5)
Proportion of complications Clavien-Dindo (n,%) 0 1 2 3 4 5	26 (35%) 24 (32%) 16 (21%) 8 (11%) 1 (1%) 0	12 (16%) 16 (21%) 26 (36%) 19 (25%) 2 (2%) 0

The preGallstep trial is the first randomised clinical trial with blinded outcome assessment comparing two different approaches to CBDS. Our focus is not just to compare the two different approaches to CBDS, but also to focus on patient-reported outcomes. Only in 1 of the 13 trials presented in the introduction (18) have a focus on patient-reported outcomes and this is only loosely described as patient satisfaction. We introduce the SF-36 questionnaire prior to initial intervention and 90 days postoperatively.

This randomised pilot and feasibility clinical trial has a pragmatically chosen sample size, with associated risks of random errors in the assessment of complication estimates. Yet, we have based our estimates for this feasibility and pilot trial on previous trials and on the flow of relevant patients through the surgical departments in Denmark.

The variations in the inclusion procedures of each trial site due to patients being both acute and ambulant could cause one of the bigger challenges in estimating feasibility issues and implementing changes in procedure in a larger pragmatic trial. We will implement a central data monitoring scheme designed to illustrate ‘inter-site’ differences and the causes of these (33). These differences are both regarding screening and randomisation, but also executing the trial interventions and follow-up. This monitoring plan will be published separately to ensure complete transparency.

The data collected in the present trial will not be pooled with data for a subsequent larger pragmatic trial since there will be differences in procedures. However, the data from the present trial will be available for later systematic reviews with meta-analysis.

We present a detailed predefined description of the statistical analysis of the preGallstep feasibility trial. The primary aim of this statistical analysis plan is to limit bias, p-hacking, and data-driven interpretations.

Strengths

The primary strengths are the predefined statistical analysis plan and publication of a version-controlled pre-programmed statistical report before any data were available. This secures methodological transparency and enables reproducibility of our results. Completion of a feasibility trial with three independent feasibility outcomes and multiple exploratory clinical outcomes will contribute with important data for the future randomised clinical trial we have planned.

‘Good clinical practice data monitoring’ of clinical trials is a tool to ensure high quality and accuracy of the data, and adherence to the trial protocol. Quality and accuracy of the data is threatened by random and systematic errors. Inspired by the SafeBoosC III trial, we introduce a data monitoring plan. Central data monitoring may optimise concurrent data completeness and may help timely detection of data deviations due to misunderstandings or fabricated data (33).

Limitations

Since no correction for multiplicity will be applied to the exploratory outcomes, any significance must be interpreted with caution. We assess multiple outcomes which increases the risk of false positive results (type I error); any difference between the groups might be explained by random errors (‘play of chance’).

This being a feasibility trial sample size is defined pragmatically and is not seen as large enough to decide whether to recommend the one-step procedure over two-step. The results are applicable to a later TSA analysis regarding this recommendation.

We present a pre-defined statistical analysis plan for the preGallstep feasibility trial, which limits bias, p-hacking, and data-driven interpretations. This statistical analysis plan is, furthermore, accompanied by a pre-programmed version-controlled statistical report with simulated data, which increases transparency and reproducibility.

AHH

Amager-Hvidovre Hospital

RHG

Regional Hospital Gødstrup

BFH

Bispebjerg-Frederiksberg Hospital

CBDS

Common bile duct stones

CRP

C-reactive protein

CT

Computed Tomography

ERC

Endoscopic retrograde cholangiography

EUS

Endoscopic Ultrasound

LC

Laparoscopic cholecystectomy

LCBDE

Laparoscopic common bile duct exploration with stone clearance

MRCP

Magnetic resonance cholangio-pancreaticography

RCT

Randomised clinical trial

US

Ultrasound

Ethics approval and consent to participate: Not applicable

Consent for publication: Not applicable

Availability of data and materials: The datasets generated and/or analyzed during the current study are available in at

Competing interests: The authors declare that they have no competing interests.

Funding: The trial has been fully funded by the Novo Nordisk Foundation in the call “Project Grants in Surgical Research 2019” Grant reference NNF19OC0058479.

Authors' contributions: MHO and AKK drafted the first version of the manuscript and the statistical report. All other authors revised the manuscript. All authors approved the final version.

Committee on Health Statistics in the Northern Countries. 2017. Http://norden.diva-portal.org/smash/get/diva2:1148509/FULLTEXT05.pdf.
Sandblom G, Enochsson L. Yearly report for the Swedish quality register for gall stone surgery and ERCP 2014 [Årsrapport för Svenskt kvalitetsregister för gallstenkirurgi och ERCP 2014]. GallRiks; 2015. https://www.ucr.uu.se/gallriks/images/stories/%C3%85rsrapport_nationellt_kvalitetsregister_f%C3%B6r_gallstenskirurgi_och_ERCP_gallriks_2016.pdf.
Matta B, Gougol A, Gao X, Reddy N, Talukdar R, et al. Worldwide Variations in Demographics, Management, and Outcomes of Acute Pancreatitis. Clin Gastroenterol Hepatol. 2020 June;18(7):1567–75.
Alizadeh AHM, Cholangitis. Diagnosis, Treatment and Prognosis. J Clin Transl Hepatol. 2017;5(4):404–13.
Manes G, Paspatis G, Aabakken L, Anderloni A, Arvanitakis M, et al. Endoscopic management of common bile duct stones: European Society of Gastrointestinal Endoscopy (ESGE) guideline. Endoscopy. 2019;51(05):472–91.
Nielsen LBJ, Shabanzadeh DM, Aaresøn A, Sørensen LT. The clinical course of common bile duct stone clearance with endoscopic retrograde cholangio-pancreaticography. Scand J Gastroenterol. 2019;54(9):1166–71.
Shabanzadeh DM, Martinussen T, Sørensen LT. Development of upper gastrointestinal cancer in patients with symptomatic gallstones, cholecystectomy and sphincterotomy; A nationwide cohort study. Scand J Gastroenterol. 2022;111(3):39–47.
Bansal VK, Misra MC, Rajan K, Kilambi R, Kumar S, Krishna A, et al. Single-stage laparoscopic common bile duct exploration and cholecystectomy versus two-stage endoscopic stone extraction followed by laparoscopic cholecystectomy for patients with concomitant gallbladder stones and common bile duct stones: a randomized controlled trial. Surg Endosc. 2014;28(3):875–85.
Ding G, Cai W, Qin M. Single-stage vs. two-stage management for concomitant gallstones and common bile duct stones: a prospective randomized trial with long-term follow-up. J Gastrointest Surg. 2014;18(5):947–51.
Lv F, Zhang S, Ji M, Wang Y, Li P, Han W. Single-stage management with combined tri-endoscopic approach for concomitant cholecystolithiasis and choledocholithiasis. Surg Endosc. 2016.
Koc B, Karahan S, Adas G, Tutal F, Guven H, Ozsoy A. Comparison of laparoscopic common bile duct exploration and endoscopic retrograde cholangiopancreatography plus laparoscopic cholecystectomy for choledocholithiasis: a prospective randomized study. Am J Surg. 2013;206(4):457–63.
Ke-Yue L, Cheng-Xian S, Ke-Li T, J ian-Zhao H, De-Lin Z. Advantages of laparoscopic common bile duct exploration in common bile duct stones. Wien Klin Wochenschr. 2018;130(3–4):100–4.
Salem MM, Esmat ME, Hassan AMA, Amer Y, Abdelaziz H, Rady M. Comparative study between laparoscopic common bile duct exploration and endoscopic retrograde cholangiopancreatography plus laparoscopic cholecystectomy for choledocholithiasis. Int Surg J. 2019;6(7):2250–7.
Pan L, Chen M, Ji L, Zheng L, Yan P, Fang J, et al. The safety and efficacy of laparoscopic common bile duct exploration combined with cholecystectomy for the management of cholecysto-choledocholithiasis: An up-to-date meta-analysis. Ann Surg. 2018;268(2):247–53.
Singh AN, Kilambi R. Single-stage laparoscopic common bile duct exploration and cholecystectomy versus two-stage endoscopic stone extraction followed by laparoscopic cholecystectomy for patients with gallbladder stones with common bile duct stones: systematic review and meta-analysis of randomized trials with trial sequential analysis. Surg Endosc. 2018;32(9):3763–76.
Ricci C, Pagano N, Taffurelli G, Pacilio CA, Migliori M, Bazzoli F, et al. Comparison of efficacy and safety of 4 combinations of laparoscopic and intraoperative techniques for management of gallstone disease with biliary duct calculi: A systematic review and network meta-analysis. JAMA Surg. 2018;153(7):e181167.
Cuschieri A, Lezoche E, Morino M, Croce E, Lacy A, Toouli J, et al. E.A.E.S. multicenter prospective randomized trial comparing two-stage vs single-stage management of patients with gallstone disease and ductal calculi. Surg Endosc. 1999;13(10):952–7.
Rogers SJ, Cello JP, Horn JK, Siperstein AE, Schecter WP, Campbell AR, et al. Prospective randomized trial of LC + LCBDE vs ERCP/S + LC for common bile duct stone disease. Arch Surg. 2010;145(1):28–33.
Bandeh-Moghadam H, Silva JP, Carmona J, Franky R, Pinto H. Laparoscopic management of patients with suspect common bile duct stones. J Venezuelan Soc Surg. 2010;63(1):20.
Sgourakis G, Karaliotas K. Laparoscopic common bile duct exploration and cholecystectomy versus endoscopic stone extraction and laparoscopic cholecystectomy for choledocholithiasis. A prospective randomized study. Minerva Chir. 2002;57(4):467–74.
Bansal VK, Misra MC, Garg P, Prabhu M. A prospective randomized trial comparing two-stage versus single-stage management of patients with gallstone disease and common bile duct stones. Surg Endosc. 2010;24(8):1986–9.
Barreras Gonzalez JE, Torres Pena R, Ruiz Torres J, Martinez Alfonso MA, Brizuela Quintanilla R, Morera Perez M. Endoscopic versus laparoscopic treatment for choledocholithiasis: a prospective randomized controlled trial. Endosc Int Open. 2016;4(11):E1188–E93.
Noble H, Tranter S, Chesworth T, Norton S, Thompson M. A randomized, clinical trial to compare endoscopic sphincterotomy and subsequent laparoscopic cholecystectomy with primary laparoscopic bile duct exploration during cholecystectomy in higher risk patients with choledocholithiasis. J Laparoendosc Adv Surg Tech A. 2009;19(6):713–20.
Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240(2):205–13.
Thorlund K, Engstrøm J, Wetterslev J, Brok J, Imberger G, Gluud C. User manual for Trial Sequential Analysis (TSA), 2011. ctu.dk/tsa/files/tsa_manual.pdf (accessed 23 April 2020).
Thorlund K, Devereaux PJ, Wetterslev J, Guyatt G, Ioannidis JP, Thabane L, et al. Can trial sequential monitoring boundaries reduce spurious inferences from meta-analyses? Int J Epidemiol. 2009;38(1):276–86.
Wetterslev J, Jakobsen JC, Gluud C. Trial Sequential Analysis in systematic reviews with meta-analysis. BMC Med Res Methodol. 2017;17(1):39.
Wetterslev J, Thorlund K, Brok J, Gluud C. Estimating required information size by quantifying diversity in a random-effects meta-analysis. BMC Med Res Methodol. 2009;30:9–86.
Kirkegaard-Klitbo A, Shabanzadeh DM, Olsen MH, Lindschou J, Gluud C, et al. One-step laparoscopic cholecystectomy with common bile duct exploration and stone extraction versus two-step endoscopic retrograde cholangiography with stone extraction plus laparoscopic cholecystectomy for patients with common bile duct stones: a randomised feasibility and pilot clinical trial—the preGallStep trial. Pilot Feasibility Stud. 2023;9:21. Published online 2023 Feb 6.
Chan AW, Tetzlaff JM, Altman DG, Laupacis A, Gøtzsche PC et al. February,. The SPIRIT statement 2013. https://www.spirit-statement.org/. Accessed 28 2022.
Okamoto K, Takada T, Strasberg SM, Asbun HJ, Pitt HA, et al. Tokyo Guidelines 2018: initial management of acute biliary infection and flowchart for acute cholangitis. J Hepatobiliary Pancreat Sci. 2018;25(1):31–40.
Julious SA. Sample size of 12 per group rule of thumb for a pilot study. Pharm Stat. 2005;4:287–91.
Olsen MH, Hansen ML, Safi S, Jakobsen JC, Greisen G, Gluud C. SafeBoosC-III Trial Group. Central data monitoring in the multicentre randomised SafeBoosC-III trial - a pragmatic approach. BMC Med Res Methodol. 2021;21(1):160.
Visalakshi J, Jeyaseelan L. Confidence Interval for skewed distribution in outcome of change or difference between methods. Clin Epidemiol Glob Heal. 2014;2:117–20.
Nørskov AK, Lange T, Nielsen EE, Gluud C, Winkel P, Beyersmann J, et al. Assessment of assumptions of statistical analysis methods in randomised clinical trials: the what and how. BMJ evidence-based Med. 2021;26(3):121–6.
Kawaguchi A, Koch GG. Sanon: An R package for stratified analysis with nonparametric covariable adjustment. J Stat Softw. 2015;67.
Jakobsen JC, Tamborrino M, Winkel P, Haase N, Perner A, Wetterslev J et al. Count data analysis in randomised clinical trials. J Biom Biostat. 2015;06.??????????????.
Ehsanes Saleh AKM. Hodges-Lehmann estimate of the location parameter in censored samples. Ann Inst Stat Math [Internet]. 1976;28:235–47.
Nielsen EE, Nørskov AK, Lange T, Thabane L, Wetterslev J, Beyersmann J, et al. Assessing assumptions for statistical analyses in randomised clinical trials. BMJ evidence-based Med. 2019;24:185–9.

Status:

Version 1

Abstract

Introduction