Efficacy and safety of laparoscopic bile duct exploration versus endoscopic sphincterotomy for concomitant gallstones and common bile duct stones: A meta-analysis of randomized controlled trials.

BACKGROUND: The purpose of this study was to compare the efficacy and safety of laparoscopic cholecystectomy (LC) plus laparoscopic common bile duct (CBD) stones exploration (LCBDE) with LC plus endoscopic sphincterotomy (EST) in the treatment of patients with gallstones and CBD stones.
METHODS: The authors searched PubMed, Web of Science, and Embase to identify relevant studies. Risk ratios (RRs) were pooled to compare stone clear, retained stone, conversion to other procedures, and complications. Weighted mean differences (WMDs) were pooled to compare operative time, and length of hospital stay. A fixed-effects model or random-effects model was used to pool the estimates, according to the heterogeneity among the included studies.
RESULTS: A total of 11 randomized controlled trials (RCTs) involving 1663 patients were included in this meta-analysis. The pooled estimate suggested that LC-LCBDE had comparable effects with LC-EST in terms of CBD stone clear rate (RR = 1.02, 95% CI: 0.95, 1.09; P = .583), retained stones rate (RR = 1.27, 95% CI: 0.51, 3.19; P = .607), and length of hospital stay (WMD = -0.96 days, 95% CI: -2.20, 0.28). In addition, LC-LCBDE was associated with significantly higher conversion rate (RR = 1.59, 95% CI: 1.08, 2.35; P = .019) and less operative time (WMD = -11.55 minutes, 95% CI: -16.68, -6.42; P < .001) than LC-EST. The incidence of complications was not significant difference between the 2 surgical approaches (RR = 1.07, 95% CI: 0.86, 1.34; P = .550).
CONCLUSION: Based on the current evidence, both LC-LCBDE and LC-EST were highly effective in detecting and removing CBD stones and were equivalent in complications. However, our results might be biased by the limitations. Large-scale well-designed RCTs are needed to confirm our findings.

Introduction

Concomitant gallstones and common bile duct stones (CBD) are detected in approximately 10% of patients, which are associated with serious complications, such as cholangitis and pancreatitis.[ The open CBD exploration, used as the conventional approach, has been proven as an effective treatment option.[ In the past decades, with the development of new surgical techniques, more and more options and alternatives have been used in the management of gallstones. These approaches include endoscopic retrograde cholangiopancreatography (ERCP) with sphincterotomy and laparoscopic cholecystectomy (LC) with the possibility to explore and clear the CBD.

There are several approaches that used for the treatment of preoperatively suspected choledocholithiasis. These options include preoperative ERCP, endoscopic sphincterotomy (EST), and stone extraction followed by LC[; transcystic cholangiography followed by transcystic or direct CBD exploration[; postoperative ERCP with EST and stone extraction[; LC plus intraoperative ERCP, EST, and stone extraction[; and conversion to open cholecystectomy and CBD exploration.[

The previous study reported that, single-session management of gallstones and CBD stones is effective, safe and less costly than staged procedures.[ And it is still controversial whether LC combined with laparoscopic CBD exploration (LC-LCBDE) is superior to LC with EST (LC-EST). Therefore, we conducted this meta-analysis to compare the efficacy and safety between the 2 minimally invasive techniques in the treatment of patients with gallstones and CBD stones.

Materials and methods

The ethical approval and patient consent was not applicable for meta-analysis.

Literature search for identifying related studies

We conducted this meta-analysis according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.[ A comprehensive systematic search was performed for relevant literature in PubMed, Web of Science, and Embase from inception through May 10, 2017. The search terms included: (concomitant[All Fields] AND (“common bile duct”[MeSH Terms] OR (“common”[All Fields] AND “bile”[All Fields] AND “duct”[All Fields]) OR “common bile duct”[All Fields]) AND (“calculi”[MeSH Terms] OR “calculi”[All Fields] OR “stones”[All Fields])) AND (“cholecystectomy, laparoscopic”[MeSH Terms] OR (“cholecystectomy”[All Fields] AND “laparoscopic”[All Fields]) OR “laparoscopic cholecystectomy”[All Fields] OR (“laparoscopic”[All Fields] AND “cholecystectomy”[All Fields])) AND (“sphincterotomy, endoscopic”[MeSH Terms] OR (“sphincterotomy”[All Fields] AND “endoscopic”[All Fields]) OR “endoscopic sphincterotomy”[All Fields] OR (“endoscopic”[All Fields] AND “sphincterotomy”[All Fields])) AND (“sphincterotomy, endoscopic”[MeSH Terms] OR (“sphincterotomy”[All Fields] AND “endoscopic”[All Fields]) OR “endoscopic sphincterotomy”[All Fields] OR (“endoscopic”[All Fields] AND “sphincterotomy”[All Fields])). The search was limited to human subjects and no language restriction was applied. We also identified additional references by manually searching for publications that cited in the included articles and related reviews.

Study selection

Two reviewers independently assessed the eligibility of each article. After screening the title/abstract and the full-text information, these studies were determined as inclusion or exclusion in this study. Disagreements between the reviewers were resolved by discussion and consensus. To be included in this meta-analysis, the study must meet the following inclusive selection criteria: study design: randomized controlled trials (RCTs); study population: patients with confirmed or suspected CBD stones with gallstones; intervention: LC + LCBDE, or LC + EST; outcomes measure: stone clear, retained stone, conversion to other procedures, operative time, length of hospital stay, and complications.

Data extraction

Data extraction was conducted by 2 independent investigators. For each included study, the following information was collected: first author's name, year of publication, country, number of patients in each group, median age of patients, gender composition, stone clear rate, retained stone rate, conversion to other procedures, operative time, length of hospital stay, and complications.

Assessment for risk of bias and grading the quality of evidence

We conducted the assessment for risk of bias in accordance with guidelines outlined in the Cochrane handbook for systematic reviews of interventions (version 5.1.0).[ Each study was assigned a value of “high,” “low,” or “unclear” according to the following domains: random sequence generation; allocation concealment; blinding of participants and personnel; blinding of outcome assessment; incomplete outcome data; selective reporting; and other bias.[

The quality of evidence for the outcome measures was evaluated using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach.[ A summary table was established with the GRADE profiler (GRADE pro, version 3.6).

Statistical analysis

We estimated the pooled risk ratio (RR) with 95% confidence intervals (CIs) for dichotomous outcomes and the pooled weight mean difference (WMD) with 95% CI for continuous outcomes. Heterogeneity across the studies was tested using the Cochran Q statistic and quantified with the I2 statistic, in which I2 > 50% indicated significant heterogeneity.[ A fixed-effects model[ was used to pool the results when heterogeneity was <50%, while a random-effects model[ was chosen when heterogeneity was >50%. We also investigated the influence of a single study on the overall pooled estimate by deleting one study in each turn. If one study reported the result that was significantly different with others or out of the range of others, we would search for the likely reasons and perform a sensitivity analysis by deleting this study. The presence of publication bias was evaluated using the Begg and Egger test.[ Results were considered as statistically significant for P < .05, except otherwise stated. All statistical analyses were performed using STATA, version 12.0 (Stata Corporation, College Station, TX).

Results

The initial search yielded 764 publications, of which 528 were eliminated because of duplicate records. Then 236 were left for title/abstract and full-text information review, and 214 and 11 were excluded, respectively, because they were case reports, non-RCTs, sing-arm trials, or unrelated with our topic. Finally, 11 RCTs[ studies met the inclusion criteria and were included in this meta-analysis. The flow chart for study selection is shown in Fig. 1.

Figure 1

Eligibility of studies for inclusion in meta-analysis.

Study characteristics

The baseline characteristics of these included studies are summarized in Table 1. The trials were published between 1996 and 2016. And 2 of the 11 trials were conducted in multicenter countries,[ and the others were conducted in Egypt,[ China,[ UK,[ India,[ USA,[ Australia,[ and Turkey.[ The sample size of the enrolled patients ranged from 30 to 300, with a total number of 1663. Among the included studies, 3 trials[ used 1-stage EST and LC for patients with gallstones and CBD stones; whereas the remaining 8 trials[ used 2-stage EST followed by LC. Notably, in the trial conducted by Noble et al[ patients enrolled were with higher medical risk; whereas, in other studies, patients were with a good-risk.

Table 1

Baseline characteristics of patients in the trials included in the meta-analysis.

Risk of bias and quality assessment of outcomes

The details of risk of bias assessment are presented in Fig. 2. Among the 11 trials, 8[ adequately reported the methods for randomized sequence and allocation sequence concealment. However, none of the included studies reported in detail the methods for blinding of participants and personnel, and blinding of outcome assessment. Although the double-blinding was difficult to implement due to the nature of the surgeries, assessment for surgical outcomes was unlikely to be influenced by the knowledge of trial allocations. Thus, we considered the 2 items to be unclear risk of bias. For other 3 items (incomplete outcome data, selective reporting, and other bias), we did not find these bias in all the included studies. Therefore, all the included trials were judged to be at unclear risk of bias.

Figure 2

Risk of bias summary.

The GRADE evidence profiles for the outcomes are shown in Table 2. Because there was substantial heterogeneity for CBD stone clear rate and length of hospital stay, the inconsistency of these 2 outcomes was regarded as serious and the quality of evidence was downgraded. Then since all the studies reported data for the 2 outcomes, and the pooled estimates were calculated based on a large sample size. The quality of evidence was upgraded. As shown in Table 2, the GRADE level of evidence was high for all these outcomes, which indicated that the results were reliable.

Table 2

GRADE evidence profile.

CBD stone clear rate

All the studies reported the data of CBD stone clear rate.[ The success rates of CBD clearance in the LC-LCBDE and LC-EST groups were 88.61% and 87.1%, respectively. Pooled estimates using a random-effects model showed that, there was no significant difference between the 2 groups (RR = 1.02, 95% CI: 0.95, 1.09; P = .583) (Fig. 3). There was significant heterogeneity (I2 = 71.9%, P < .001) among the included studies. Thus, we conducted sensitivity to explore the potential sources of heterogeneity. When we excluded the trial conducted by Noble et al[ the pooled estimate did not change substantially (RR = 0.99, 95% CI: 0.95, 1.04, P = .756), but no significant heterogeneity was found among the remaining studies (I2 = 46.8%, P = .050).

Figure 3

Forest plot showing the comparison between LC-LCBDE and LC-EST in CBD stone clear rate.

Retained stones rate

Six studies reported the data on retained stones rate.[ The rates of retained stones in the LC-LCBDE and LC-EST groups were 8.43% and 6.98%, respectively. The summarized estimate suggested that the rate of retained stones between the 2 groups was not significantly different (RR = 1.27, 95% CI: 0.51, 3.19; P = .607) (Fig. 4). There was significant heterogeneity among the included studies (I2 = 58.5%, P = .034).

Figure 4

Forest plot showing the comparison between LC-LCBDE and LC-EST in retained stones rate.

Conversion to other procedure

Eight studies reported the data on conversion.[ The conversion rate in the LC-LCBDE and LC-EST groups were 8.95% and 5.51%, respectively. The aggregated results showed that conversion rate was significantly higher in LC-LCBDE group than in the LC-EST group (RR = 1.59, 95% CI: 1.08, 2.35; P = .019) (Fig. 5). There was no significant heterogeneity among the included studies (I2 = 0.0%, P = .648).

Figure 5

Forest plot showing the comparison between LC-LCBDE and LC-EST in conversion rate.

Operative time

Eight studies reported data on operative time.[ The mean operative time in patients treated with LC-LCBDE was 119.5 minutes, and 129.0 minutes for patients treated with LC-EST. LC-LCBDE was associated with a significant less operative time than LC-EST (WMD = −11.55 minutes, 95% CI: −16.68, −6.42; P < .001) (Fig. 6). No significant heterogeneity was observed among the included studies (I2 = 12.6%, P = .332).

Figure 6

Forest plot showing the comparison between LC-LCBDE and LC-EST in operative time.

Subgroup analysis based on stage procedure of LC-EST showed that, LC-LCBDE had a less operative time than 1-stage (WMD = −9.42 minutes, 95% CI: −15.57, −3.27; P = .003) and 2-stage LC-EST (WMD = −14.46 minutes, 95% CI: −21.14, −7.79; P < .001).

Length of hospital stay

Length of hospital stay was reported in all the included studies.[ The mean hospital stay was 4.13 days in the LC-LCBDE group and 5.11 days in the LC-EST group, respectively. LC-LCBDE had a shorter length of hospital stay than LC-EST (WMD = −0.96 days, 95% CI: −2.20, 0.28) (Fig. 7), however, the difference did not reach statistical significant (P = .129). There was substantially significant heterogeneity among the included studies (I2 = 97.2%, P < .001).

Figure 7

Forest plot showing the comparison between LC-LCBDE and LC-EST in length of hospital stay.

Subgroup analysis based on stage procedure of LC-EST showed that, LC-LCBDE was associated with a similar length of hospital stay with 1-stage LC-EST (WMD = 0.14 days, 95% CI: −0.16, 0.44; P = .368), and a shorter length of hospital stay than 2-stage LC-EST (WMD = −1.41 days, 95% CI: −1.68, −1.15; P < .001).

Among these included studies, the trial conducted by Cuschieri et al[ reported 3 days more of length of hospital stay in the LC-LCBDE group than in the LC-EST group. Therefore, we conducted sensitivity analysis by excluding this trial. After excluding this trial, the pooled data did not change substantially (WMD = −0.74 days, 95% CI: −1.95, 0.47; P = .231), but significant heterogeneity was still present (I2 = 96.2%, P < .001). We further excluded a single trial at one time. However, the heterogeneity was still present, and the overall combined WMD altered slightly (data not shown).

Complication

All the studies reported the data on complications.[ The complication rates in the LC-LCBDE and LC-EST groups were 10.25% and 9.15%, respectively. There was no significant difference between the 2 groups (RR = 1.07, 95% CI: 0.86, 1.34; P = .550). The test for heterogeneity was not significant (I2 = 0.0%, P = .771).

LC-LCBDE was associated with a significantly higher rate of bile leakage than LC-EST (RR = 2.66, 95% CI: 1.23, 5.77; P = .013). Whereas, for other complications, including pancreatitis, pneumonia, hemorrhage, surgical-site infection, and cholangitis, the rates between the 2 treatment procedures were not significant (Table 3).

Table 3

Summary of the risk ratio (RR) of complications in patients with concomitant gallstones and CBD stones.

Publication bias

Assessment of publication bias using Begg and Egger tests showed that, there was no potential publication bias among the included studies (Begg test, P = .350; Egger test, P = .210).

Discussion

The purpose of this meta-analysis was to compare the efficacy and safety of LC-LCBDE with LC-EST in patients with gallstones and concomitant CBD stones. Our study showed that, LC-LCBDE was associated with comparable effects with LC-EST in terms of CBD stone clear rate, retained stones rate, and length of hospital stay. In addition, LC-LCBDE resulted in significantly higher conversion rate and less operative time than LC-EST. The incidence of complications between the 2 surgical approaches was not significant difference.

To the best of our knowledge, this study is the first meta-analysis to compare the effects and safety of LC-LCBDE with LC-EST in patients with gallstone and CBD stones. LC-LCBDE and LC-EST are the most ideal of mini-invasive procedures, which are widely used in patients with cholelithiasis and choledocholithiasis.[ There has been one recently published systematic review and meta-analysis,[ which compared ERCP with open or laparoscopic surgery. In that study, the authors reported of 8 trials with 760 patients that compared ERCP with open surgical clearance.[ They found that ERCP was less effective than open surgery in CBD stone clear (odd ratio, OR = 2.89, 95% CI: 1.81, 4.61).[ The mortality seemed to be higher (risk difference, RD = 1%, 95% CI: −1%, 4%), but the difference did not reach a statistical significance.[

In this met-analysis, we found the success rates of CBD clearance was 88.6% for LC-LCBDE and 87.1% for LC-EST, which was in consistent with the previous studies.[ There was no significant difference in success rate between the 2 surgical approaches. Among the included studies, LC-LCBDE was reported to have better CBD stone clearance than LC-EST, despite the difference was not significant. However, in the trial conducted by Elgeidie et al,[ a higher success rate of CBD clearance in LC-EST group was reported. In that study, 112 and 107 patients were randomly assigned to LC-LCBDE and LC-EST groups, respectively; and the success rate of stone clearance in the 2 groups was 92.0% and 97.2%, respectively.[ Similarly, Rogers et al[ also reported a higher stone clearance for LC-EST. In that study, the stone clearance rate for patients undergoing LC-LCBDE was 88%, compared to 98% for patients with LC-EST.[ The authors assumed that the higher stone clearance rate for LC-EST in their studies might be the result of a type II error.[ Given the stone detected in 17 of 57 LC-LCBDE patients and 31 of 55 LC-EST patients, the power to detect a significant difference (α = 0.05) in clearance should be approximately 0.82.[

Regarding to the conversion to an open procedure, patients undergoing LC-LCBDE had a significantly higher conversion rate than those with LC-EST. The corresponding values for these patients were 8.95% and 6.98%, respectively. The time interval between endoscopic papillotomy and LC remains controversial. In the multicenter trial of European Association of Endoscopic Surgery (E.A.E.S.), the time interval for the 2-stage procedures was not specified. While in the trial conducted by Sgourakis and Karaliotas,[ LC was performed within 2 days after the stones were removed by endoscopic surgery. de Vries et al[ reported a higher conversion rate when LC was performed more than 2 weeks after endoscopic papillotomy. They found that the conversion rate was 31% when cholecystectomy was done between 2 and 6 weeks, 4% within 2 weeks, and 16% after 6 weeks.[

In this study, the mean operative time was significantly less for patients undergoing LC-LCBDE than those with LC-EST. Our result was in consistent with the findings from the included studies. It has been reported that, the prolonged operative time for LCBDE may increase the incidence of complications for higher risk patients.[ However, this result was not found in the trial conducted by Paganini et al,[ who reported a similar morbidity or mortality between patients over 70 and less than 70 years. Mover, in the trial of Noble et al[ they did not find any increased complications because of longer operative time. In that study, 91 high-risk patients with median age of 74.56 years were randomly assigned into LC-LCBDE or LC-EST groups. The authors concluded that, the 2 approaches had comparable results in terms of postoperative hospital stay, conversion, and complications in high-risk patients, but LC-LCBDE was more effective and efficient, and avoided unnecessary procedures.[ Since there was only one study reporting LCBDE in high-risk patients, we could not perform subgroup analysis to identify whether LCBDE would lead to increased complications in these patients. Further well-conducted RCTs are needed to address this issue.

There were several potential limitations in this study which should be considered. First, there was moderate heterogeneity among the included studies. However, one should not be surprising given the variation in characteristics of populations (age, female/male, stone number, stone size), time intervention between endoscopic papillotomy and LC, laparoscope device and operator experience between the 2 approaches. All these factors contributed to the heterogeneity and had potential impacts on our results. Second, all included trials were not performed with double-blinded method, which might lead to performance and detection bias. Third, our meta-analysis is based on 11 RCTs, and 3 of them have a relatively small sample size (less than 100). Studies with small sample size are more likely to result in an overestimated treated effect than large studies. Fourth, we were unable to assess the effects and safety of LCBDE in high-risk patients due to the sparse reporting across studies.

In conclusion, the current meta-analysis suggested that both LC-LCBDE and LC-EST were highly effective in detecting and removing CBD stones and were equivalent in complications. However, the conversion rate was higher and operative time was less for LC-LCBDE. More large-scale, well-designed RCTs are needed to confirm our findings. In the future studies, the effects and safety of LCBDE in patients with high-risk should be given more attention.