Predictors of Failed Transcystic Laparoscopic Common Bile Duct Exploration: Analysis of Multicenter Integrated Health System Database

Abstract

Background:

The aim of this study is to report our experience with laparoscopic common bile duct exploration (LCBDE) and validate the experts' opinion about anatomical predictors of failed transcystic LCBDE (TLCBDE) approach.

Methods:

Patients undergoing LCBDE at Kaiser Permanente Southern California hospitals (2005–2015) were included. Predictors of failed TLCBDE were identified using bivariate analysis.

Results:

Of 115 LCBDE, 89.6% were TLCBDE and 10.4% through choledochotomy. Success rate, morbidity, and length of hospital stay were 83.5%, 6.1%, and 3.8 days respectively. Only stone size:cystic duct ratio >1 (35% versus 63%, P = .044) was associated with failure of TLCBDE. In accordance with experts' opinion, there was a suggestive association of stone size ≥6 mm, cystic duct ≤4 mm, multiple stones, and proximal stone location with failure; however, these did not reach statistical significance.

Conclusion:

LCBDE is an effective and safe mean of clearing common bile duct stones at community hospitals of an integrated health system. Previously cited contraindications for TLCBDE are not absolute, but rather predictors of failure.

Introduction

Common bile duct stones (CBDS) are encountered in 5%–20% of patients undergoing cholecystectomy for gallstones.¹ There is a consensus that timely management of CBDS is critical to avoid CBDS-associated complications^2,3 such as pancreatitis, cholangitis, and intrahepatic abscess.² Established management options for CBDS include: Laparoscopic common bile duct exploration (LCBDE), open common bile duct exploration (OCBDE), and endoscopic retrograde cholangiopancreatography (ERCP) before, during, or after cholecystectomy.^2–4 Previous studies have demonstrated comparable morbidity and effectiveness to clear CBDS between LCBDE and ERCP^4,5; however, LCBDE has the benefit of a one-stage procedure with reduced length of hospital stay and cost.⁵ Despite these appealing advantages of LCBDE over two-stage laparoscopic cholecystectomy with ERCP (LC+ERCP), it is a technically demanding and time-consuming procedure. As such, two-stage LC/ERCP for the management of CBDS remains more commonly practiced.⁶

Laparoscopic clearance of CBDS can be performed through one of two methods: through the cystic duct (transcystic LCBDE [TLCBDE]) or through laparoscopic choledochotomy (LCD).^7,8 TLCBDE is associated with 70% CBDS clearance rate,⁹ low postoperative morbidity that is comparable to LC, and short hospital stay.¹⁰ LCD is less favorable and technically more demanding than TLCBDE, because the choledochotomy incision compromises the integrity of the common duct. LCD is also associated with a high postoperative morbidity (17%) and stone clearance rate (98%) in randomized controlled studies.⁹

The choice of either method to clear CBDS laparoscopically depends on the available expertise and the feasibility of performing the least morbid procedure (TLCBDE).^7,8 Two previous studies specified the anatomical characteristics that would preclude performing TLCBDE or LCD. Puhalla et al.⁷ in accordance with Petelin et al.⁸ identified cystic duct diameter less than 4 mm, CBDS diameter >6 mm, multiple stones, intrahepatic stone, and posterior or distal cystic duct entrance into the CBD as contraindications to TLCBDE. Both authors also agree about the indications and contraindications for LCD, except for the CBD diameter cutoff to perform LCD. Petelin et al. considered CBD diameter <6 mm as a contraindication for LCD, whereas Puhalla et al. used a CBD diameter of <10 mm as a cutoff for contraindication to perform LCD.

Those indications and contraindications for TLCBDE and LCD were defined by experts based on previous literature^7–13; however, since the publication of those criteria, no study has put them to the test. The purpose of this study is to report cases of LCBDE from a large integrated community-based health system and identify anatomical predictors of failed TLCBDE.

Methods

This was a retrospective study of the Southern California Kaiser Permanente database, which is collected from 13 hospitals that serve more than 3 million people. Patients admitted between January 2005 and September 2015 with diagnosis of CBDS, cholelithiasis, cholecystitis, gallstone pancreatitis, or cholangitis were identified using the International Classification of Disease Ninth Edition (ICD-9) codes. Those who underwent LCBDE were identified using the Current Procedural Terminology codes. Both approaches of LCBDE (either trancystic or choledochotomy) were included. Patients undergoing OCBDE as the initial intervention were excluded. Patient demographics, preoperative comorbidities, laboratory findings, intraoperative details, postoperative course, and complications were abstracted by reviewing the patients' electronic medical records. Cases were defined as emergent, urgent, or elective depending on the time from submitting the case request to time of operation as follows: emergent <24 hours, urgent 24–48 hours, or elective >48 hours. Stone size, number, and location, as well as characteristics and measurements of the biliary tract were obtained from the intraoperative cholangiography (IOC) images. Primary outcome of interest was successful clearance of the common bile duct that was determined by a post procedural IOC. Failure was defined as a persistent filling defect on cholangiogram, hospital readmission for retained stones within 1 year, or conversion to an open procedure. In a subset analysis to identify predictors of failed TLCBDE, failure was defined as inability to clear the CBD. Secondary outcomes were length of hospital stay, readmission within 30 days, 30-day postoperative morbidity, and return to the operating room. Preoperative characteristics, intraoperative details, and postoperative outcomes were compared between the two groups using chi-square test for categorical variables and Student's t-test for continuous variables. A P value of <.05 was used for statistical significance. All analyses were performed using STAT 11 (College Station, TX).

All patients underwent standard four-port laparoscopic cholecystectomy with an intraoperative cholangiogram. The LCBDE was performed mainly with transcystic approach. The choledochotomy approach was utilized if the former failed or if the case was deemed inappropriate for TLCBDE. First, saline was used to flush obstructive stones through the common bile duct after 1 mg injection of intravenous glucagon to relax the sphincter of Oddi. If obstruction persisted, a 4 F Fogarty balloon was inserted, inflated, and withdrawn to try to draw out the stones. If unsuccessful, the cystic duct was dilated with over-the-wire dilator or over-the-wire balloon dilator. Subsequently, a choledochoscope and retrieval basket was inserted under direct visualization in an attempt to retrieve the stone. If despite these attempts the obstruction remained, patients were subjected to conversion to LCD, an open procedure, or completion of the cholecystectomy and postoperative ERCP. LCD was performed with longitudinal choledochotomy and choledochoscope insertion. Most patients with LCD had a T-tube placement. All ERCPs were performed by GI endoscopists. At the end of the procedure, a completion occlusion cholangiogram was shot to confirm clearance of the ducts.

Results

Out of 3650 patients diagnosed with CBDS during the study period, 125 patients were eligible for inclusion. After chart review, 10 were excluded, because they had the stone flushed out (n = 8) or they were operated on for choledochoduodenal fistula (n = 2). The final included cohort was 115 patients. Of those, 15 (10.4%) cases were done after failed ERCP and 16 (13.9%) patients had Roux-en-y gastric bypass (RYGB). TLCBDE (89.6%) was more commonly performed than LCD (10.4%). Mean patient age was 53 years and the majority was females (73.9%) (Table 1).

Table 1.

Preoperative Cohort Characteristics Stratified by Method of Laparoscopic Common Bile Duct Exploration

	All n = 115, (%)	Transcystic n = 103, (%)	Choledochotomy n = 12, (%)	P
Age (year), mean ± SD	53.0 ± 22.5	50.6 ± 21.9	74.0 ± 15.6	<.01
Gender (male)	30 (26.1%)	30 (29.1%)	0 (0%)	.030
ASA class, 3–4 versus 1–2	35 (30.4%)	31 (30.1%)	4 (33.3%)	.818
BMI, mean ± SD	30.1 ± 8.5	30.4 ± 8.4	26.9 ± 9.7	.218
Hypertension (yes)	47 (40.9%)	42 (40.8%)	5 (41.7%)	.953
Diabetes (yes)	22 (19.1%)	20 (19.4%)	2 (16.7%)	.819
Coronary artery disease (yes)	5 (4.4%)	5 (4.9%)	0 (0%)	.435
Congestive heart failure (yes)	10 (8.7%)	10 (9.7%)	0 (0%)	.259
Hyperlipidemia (yes)	37 (32.2%)	35 (34.0%)	2 (16.7%)	.224
Renal failure	13 (11.3%)	13 (12.6%)	0 (0%)	.191
Total bilirubin (mg/dL), mean ± SD	3.2 ± 3.7	3.2 ± 3.7	3.4 ± 3.8	.859
Alkaline phosphatase, mean ± SD	50.3 ± 31.0	49.9 ± 30.7	54.0 ± 34.4	.678
RYGB (yes)	16 (13.9%)	14 (13.6%)	2 (16.7%)	.771
Preoperative MRCP (yes)	34 (29.6%)	29 (28.2%)	5 (41.7%)	.332
Urgent/emergent versus elective	94 (81.7%)	87 (84.5%)	7 (58.3%)	.027
Diagnosis				.286
Choledocholithiais	106 (92.2%)	94 (91.3%)	12 (100%)
Gallstone pancreatitis or cholangitis	9 (7.8%)	9 (8.7%)	0 (0%)

ASA, American Society of Anesthesiologists; BMI, body mass index; MRCP, magnetic resonance cholangiography; RYGB, Roux-en-Y gastric bypass; SD, standard deviation.

Those in the TLCBDE group were younger (mean ± standard deviation [SD]; 50.6 ± 21.9 years versus 74.0 ± 15.6 years, P < .01) and had more males (29.1% versus 0.0%, P = .030) than those in the LCD group. Preoperative comorbidities, preoperative diagnosis, and laboratory values were not significantly different between the two groups (Table 1).

With regard to intraoperative details, T-tube use (75.0% versus 2.9%, P < .01), larger CBD diameter (mean ± SD; 12.9 ± 3.4 mm versus 9.6 ± 3.2 mm, P = .001), and larger stone size (mean ± SD; 9.1 ± 4.8 mm versus 4.9 ± 3.3 mm, P < .01) were more common in the LCD group than TLCBDE. Cystic duct diameter (mean ± SD; 5.1 ± 2.7 mm versus 4.4 ± 2.3 mm, P = .358), cystic duct structure (straight versus tortuous), and stone number (multiple versus single) were not different (P > .05) between the two groups. There was a suggestive association of LCD with proximal CBD (16.7% versus 2.9%, P = .074), but that did not reach statistical significance (Table 2).

Table 2.

Intraoperative Cohort Characteristics Stratified by Method of Laparoscopic Common Bile Duct Exploration

	All n = 115, (%)	Transcystic n = 103, (%)	Choledochotomy n = 12, (%)	P
Resident involvement (yes)	54 (47.0%)	47 (45.6%)	7 (58.3%)	.404
EBL (mL), mean ± SD	37.3 ± 48.7	33.6 ± 42.8	71.4 ± 81.9	.014
T-tube (yes)	12 (10.4%)	3 (2.9%)	9 (75.0%)	<.01
CBD diameter (mm), mean ± SD	10 ± 3.4	9.6 ± 3.2	12.9 ± 3.4	.001
Cystic duct diameter (mm), mean ± SD	4.5 ± 2.3	4.4 ± 2.3	5.1 ± 2.7	.358
Tortuous cystic duct	7 (6.2%)	7 (6.9%)	0 (0%)	.346
Largest stone diameter (mm), mean ± SD	5.4 ± 4.8	4.9 ± 3.3	9.1 ± 4.8	<.01
Multiple stones versus single stone	84 (76.4%)	76 (77.6%)	8 (66.7%)	.402
Proximal CBD stone^a versus distal	5 (4.4%)	3 (2.9%)	2 (16.7%)	.074

Including common hepatic duct stone.

CBD, common bile duct; EBL, estimated blood loss; SD, standard deviation.

Postoperative outcomes were comparable between the two groups. LCBDE was successful in 83.5% of cases (83.5% TLCBDE versus 83.3% LCD, P = .989). Two patients failed LCD (15.4%); 1 was converted to open for safety due to aberrant biliary anatomy, and 1 was due to failure to pass the choledochoscope through the ampulla with subsequent presentation of retained stone. Seventeen (16.5%) patients failed TLCBDE; 6 were due to failed stone extraction; 1 was due to common hepatic duct stone; 2 were due to equipment failure; 1 was due to the surgeon's decision to abort the procedure given the long anesthesia time and unstable patient; 5 were due to a stuck basket through the cystic duct, which required conversion to open CBD exploration; and 2 were due to inability to visualize the proximal common bile duct with intraoperative ERCP to ensure there is no CBD injury. Failed TLCBDE were either converted to open (n = 5) or were addressed with ERCP (n = 12) (Table 3).

Table 3.

Postoperative Cohort Characteristics Stratified by Method of Laparoscopic Common Bile Duct Exploration

	All n = 115, (%)	Transcystic n = 103, (%)	Choledochotomy n = 12, (%)	P
Successfully completed procedure^a	96 (83.5%)	86 (83.5%)	10 (83.3%)	.989
Failure to clear CBDS^b	13 (11.3%)	12 (11.7%)	1 (8.3%)
Other reasons of failure^c	6 (5.2%)	5 (4.8%)	1 (8.3%)
Length of hospital stay mean ± SD	3.8 ± 2.6	3.6 ± 2.5	5.0 ± 3.2	.081
Conversion to open	6 (5.2%)	5 (4.9%)	1 (8.3%)	.608
Any 30-day complication^d	7 (6.1%)	6 (5.8%)	1 (8.3%)	.485
Type of complications
Postoperative bleeding^e	1 (0.9%)	1 (1.0%)	0 (0%)	.732
Duodenal perforation^f	1 (0.9%)	0 (0.0%)	1 (8.3%)	.003
Pancreatitis^g	2 (1.7%)	2 (1.9%)	0 (0%)	.626
Postoperative wound infection/abscess	2 (1.7%)	2 (1.9%)	0 (0%)	.626
CHF exacerbation	1 (0.9%)	1 (1.0%)	0 (0%)	.732
Ileus	1 (0.9%)	1 (1.0%)	0 (0%)	.732
Blocked T-tube	1 (0.9%)	0 (0.0%)	1 (8.3%)	.003
Bile leak	0 (0.0%)	0	0	1
Mortality	0 (0.0%)	0	0	1
Return to OR (yes)	1 (0.9%)	1 (1.0%)	0 (0%)	.732
Readmission (yes)	6 (5.2%)	4 (3.9%)	2 (16.7%)	.059

Successful clearance of the common bile duct with no conversion to open, no retained stones, and no need for adjunct ERCP procedure to resolve the primary pathology.

Including retained stones within 1 year of the procedure.

Conversion to open for reasons other than failure to clear the duct or equipment failure.

Not including ERCP-related complications.

Port-site bleeding with return to the operating room.

ERCP-related duodenal perforation before laparoscopic common bile duct exploration.

One is related to the procedure itself and one is related to preoperative ERCP.

CBDS, common bile duct stones; CHF, congestive heart failure; ERCP, endoscopic retrograde cholangiopancreatography; OR, operating room; SD, standard deviation.

There was no mortality and 7 patients (6.1%) had LCBDE-related complications, which was not different between the two procedures. Retained stones were not considered postoperative complications. In the LCD group, 1 (8.3%) patient was readmitted with decreased T-tube output; however, cholangiogram did not show an obstruction. Another patient was readmitted with a retained stone that was cleared by ERCP. In TLCBDE group, six complications (5.8%) were related to LCBDE. Those complications included: Port-site bleeding that was taken back to the OR (n = 1), superficial wound infection (n = 1), intra-abdominal abscess that was drained percutaneously (n = 1), ileus (n = 1), pancreatitis (n = 1), and congestive heart failure exacerbation (n = 1). Two patients had ERCP-related complications before LCBDE. One developed pancreatitis after ERCP and the other had a contained duodenal perforation that was treated conservatively before LCBDE. Average length of hospital stay was 3.8 days, which was shorter with the TLCBDE than LCD (mean ± SD; 3.6 ± 2.5 days versus 5.0 ± 3.2 days, P = .081) (Table 3).

For the analysis of predictors of failed TLCBDE, we included all cases, where transcystic approach was attempted (intention to treat). Also failure was defined as the inability to clear the CBD. Therefore, patients with planned laparoscopic choledochotomy (n = 7) and those with failure due to use of intraoperative ERCP (n = 2), equipment failure (n = 2), or aborting the procedure due to patient comorbidities (n = 1) were excluded. Among the remaining 103 patients with attempted TLCBDE, 17 (16.5%) failed and had successful postoperative ERCP, conversion to OCBDE or LCD. Stone to cystic duct ratio >1 (35.4% versus 62.5%, P = .044) was more common in the group that failed TLCBDE. There was a suggestive association of stone size (≥6 mm), presence of multiple stones, and proximal stone location with failure; however, these did not reach statistical significance (P > .05). Patients with failed TLCBDE had longer hospital stay than those with successful procedure (4.9 days versus 3.4 days, P = .016) (Table 4).

Table 4.

Factors Associated with Failed Transcystic Laparoscopic Common Bile Duct Exploration

	Success		Failure
	n = 86	%	n = 17	%	P
Age (year), mean ± SD	49.9 ± 21.8		56.4 ± 22.3		.263
Gender (male)	25	29.1	3	17.7	.333
ASA class, 3–4 versus 1–2	26	30.2	3	17.7	.292
BMI, mean ± SD	30.2 ± 8.0		32.1 ± 10.7		.431
Bilirubin (mg), mean ± SD	3.5 ± 3.9		2.7 ± 3.4		.484
Alkaline phosphatase, mean ± SD	47.1 ± 30.2		60.0 ± 33.7		.116
RYGB	12	14.0	2	11.8	.810
Diagnosis (gallstone pancreatitis or cholangitis)	7	8.1	0	0.0	.223
Resident involvement (yes)	40	46.5	9	52.9	.628
Cystic duct diameter, mean ± SD	4.5 ± 2.3		4.2 ± 2.2		.611
Cystic duct diameter ≤4 mm	48	57.1	11	64.7	.564
Tortuous cystic duct	4	4.8	1	5.9	.846
Long cystic duct	4	4.8	1	5.9	.846
Largest stone diameter (≥6 mm)	22	27.5	8	50	.076
Stone size, mean ± SD	4.9 ± 3.2		6.4 ± 4.6		.101
Stone: cystic duct ratio (>1)	28	35.4	10	62.5	.044
Proximal stone location	2	2.4	2	11.8	.133
Multiple stones	59	72.8	15	88.2	.180
Size of CBD (mm), mean ± SD	9.4 ± 3.0		10.6 ± 3.8		.155
Length of hospital stay (days), mean ± SD	3.4 ± 2.2		4.9 ± 3.3		.016

ASA, American Society of Anesthesiologists; BMI, body mass index; CBD, common bile duct diameter; RYGB, Roux-en-Y gastric bypass; SD, standard deviation.

Discussion

Two established techniques can be used to clear CBDS laparoscopically: through the cystic duct (TLCBDE) or through a choledochotomy (LCD). TLCBDE is more favorable than LCD; however, it is not always feasible.¹⁴ The feasibility and success of TLCBDE is largely dependent on the anatomical characteristics of the biliary tree and stone size and characteristics. Anatomical parameters as indications and contraindications of each approach are described in the literature^7,8; however, no studies have validated those parameters. We reviewed the practice of LCBDE by many surgeons with different levels of experience at multiple community medical centers of an integrated healthcare system. We also report anatomical predictors of failed TLCBDE in our patient population.

In our study, CBDS clearance was successful in 83.5% of cases, which is consistent with the rates reported in the literature and ranging from 75% to 100%.¹⁴ The complication rate among our patient population was 6.1% and mean length of hospital stay was 3.8 ± 2.6 days, which were comparable to those reported in other studies.^11,13,14 The safety and effectiveness of LCBDE is well established in the literature; however, the majority of previous studies reporting the outcomes of LCBDE were from a single surgeon or a single center.^14,15 In multicenter studies or randomized studies, most reported outcomes were from surgeons with expertise in performing these procedures.¹⁵ These factors limit the generalizability of many of these experiences to the surgeon at a community hospital. A major strength of our study is that we achieved good outcomes utilizing LCBDE from multiple surgeons with different levels of experience at multiple community hospitals. Our population and the surgeons managing these patients reflect the community clinical practice of managing CBDS.

The ability to perform TLCBDE depends upon whether one can traverse the scope through the cystic duct into the common bile duct and whether or not stone size permits retrieval through the cystic duct. Thus, cystic duct anatomical features (cystic duct diameter, bifurcation angle of the cystic and hepatic ducts) and stone characteristics (location, size, number of stones) are paramount to the rate of success of TLCBDE. Our study findings were consistent with the guidelines regarding the indication and contraindication of TLCBDE^7,8; however, the studied contraindications were not absolute, but were rather indicators of a higher failure rate.

Cystic duct diameter less than 4 mm was cited as contraindication to TLCBDE,^7,8 because the minimum choledochoscope diameter is 3 mm and stone size is often larger than 3 mm.⁸ In our study, average cystic duct diameter was more than 4 mm and it was not different between those who had failure or success of TLCBDE. Cystic duct diameter ≤4 mm was more common in the group that failed TCLCBDE; however, smaller cystic duct diameter was not associated with absolute failure. In some instances where the cystic duct was too small, a 7.5 F ureteroscope or ultrathin choledochoscope were used. We also performed dilatation of the cystic duct in some cases. Dilatation of the cystic duct is not uncommon during the performance of TLCBDE, which can increase the diameter of the cystic duct to allow the passage of the choledochoscope and stone removal.^8,10 Dilatation of the cystic duct is not recommended by one group of experts,⁷ but our case series, as well as other studies^8,9 do not demonstrate an increase in complications with the use of cystic duct dilatation. It is important to note that we avoid dilation when there is excessive inflammation and friable tissue. Accordingly, cystic duct diameter <4 mm might not be an absolute contraindication if the duct can be safely dilated to fit the choledochoscope. Stone size >6 mm was also cited as a contraindication for TLCBDE.^7,8 In accordance with prior studies, our results show a suggestive association of larger stone diameter (≥6 mm) with failure of TLCBDE; however, 27.5% of patients with successful TLCBDE had stone diameter ≥6 mm. Other studies have reported that stone size up to 10 mm were managed successfully with TLCBDE.^10,13 As such, this criterion should not be used as an absolute contraindication to TLCBDE as proposed by prior studies.^7,8 In the light of these findings, it is perhaps the stone size relative to the cystic duct diameter (SS:CD ratio) that matters the most, rather than an absolute cutoff of the cystic duct diameter or stone size. Indeed, SS:CD >1 was the only significant predictor of failure of TLCBDE in our study. However, we once again found that up to 35% of patients with successful TLCBDE had SS:CD ratio >1. This finding might be related to our ability to dilate the cystic duct to accommodate larger stone size and the ability to fragment some of the larger stones before retrieval.

There was a suggestive association of failed TLCBDE with multiple stones and proximal stone location in our study; however, neither of those characteristics was associated with absolute failure of TLCBDE. About 73% of patients with successful TLCBDE had multiple stones and very few (2.4%) had a proximal stone location. The presence of multiple stones or proximal stones can be associated with more technical challenges to clear the CBD; however, it is not necessarily associated with absolute failure.⁸

Based on our study results, none of the abovementioned criteria^7,8 represents an absolute contraindication to TLCBDE. We also believe that those criteria are important to standardize algorithms for the management of CBDS, as we clearly see some association between failed TLCBDE with cystic duct diameter, stone size, SS:CD ratio, as well as number and location of stones. We acknowledge that our study is underpowered to identify significant association of those characteristics with failure, but all associations in our study were consistent with the recommendations in the literature. We also recognize that the operator factor is difficult to standardize, and experience might play an important role in the success of those operations; however, the success rate in our study is comparable to that reported in the literature and failures were not restricted to a specific surgeon(s) in the group.

TLCBDE keeps the integrity of the CBD and it is associated with lower incidence of complications than LCD.¹⁴ This approach also obviates the need for T-tube drainage and its potential complications.¹⁰ In our study, patients undergoing LCD were older than those undergoing TLCBDE. While this is not likely to affect which procedure to be performed, it suggests in accordance with previous literature¹⁶ that TLCBDE and the more complex LCD can be safely performed in elderly patients. About 82% of the cases in our series were performed in an urgent/emergent fashion suggesting that these procedures can be performed successfully in these settings. LCD was also less likely than TLCBDE to be performed in an urgent/emergent fashion (58.3% versus 84.5%, P = .027). These results might be related to the technical complexity associated with LCD and the need for the available expertise, which is not always available under such circumstances.

As expected, T-tube placement and larger common bile duct diameter were more common in the LCD group than TLCBDE group. In addition, mean length of hospital stay was longer (5.0 ± 3.2 days versus 3.6 ± 2.5 days, P = .081) and readmission rate was higher (16.7% versus 3.9%, P = .059) with LCD. Those who have failed TLCBDE had similar biliary tract characteristics as those who underwent LCD (CBD diameter and cystic duct diameter) suggesting that they were potentially eligible candidates for LCD, but not all surgeons were comfortable with proceeding to LCD when TLCBDE fails. We suggest that in such scenarios, surgeons should thoroughly evaluate their options of switching to OCBDE versus ERCP depending on their level of experience, ERCP availability, CBD size, CBD condition, stone size, and other anatomical characteristics that might preclude postoperative ERCP. In such scenario, surgeons fear ERCP failure and the need to return to the operating room for OCBDE. In a recent multicenter study, first ERCP failure rate under such circumstances was 2%, with only 1 out of 195 cases (0.05%) requiring OCBDE after multiple ERCP failures.¹⁷

The sample size and event (failure) rate were small in our study, which might be associated with high rate of type II error. Thus, there is a chance that we have missed small differences in the outcome. We also did not account for the surgeon level of expertise as an indicator of success/failure of TLCBDE; however, our overall success rate is comparable to that reported in the literature.

Conclusion

TLCBDE and LCD are equally effective and safe in managing CBDS at community hospitals of an integrated healthcare system. Some anatomical characteristics might preclude performing TLCBDE. We found that many previously cited anatomical contraindications to TLCBDE were not absolute, but they represent factors with higher rate of procedure failure. SS:CD ratio >1 was significantly associated with TLCBDE failure, whereas stone diameter ≥6 mm, number, and proximal location have a suggestive association with failure.

Author Contributions

Study conception and design: M.H.A., D.A.S., and S.D.J.; Acquisition of data: M.H.A., S.R., B.C., E.G.K., C.N.T., and A.F.M.; Analysis and interpretation of data: M.H.A., D.A.S., and S.D.J.; Drafting of article: M.H.A., S.R., D.A.S., and S.D.J.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Collins

, Maguire

, Ireland

, et al. A prospective study of common bile duct calculi in patients undergoing laparoscopic cholecystectomy: Natural history of choledocholithiasis revisited. Ann Surg, 2004; 239:28–33.

Williams

, Green

, Beckingham

, et al. Guidelines on the management of common bile duct stones (CBDS). Gut, 2008; 57:1004–1021.

ASGE standards of Practice

Committee

, Maple

, Ben-Menachem

, Anderson

, et al. The role of endoscopy in the evaluation of suspected choledocholithiasis. Gastrointest Endosc, 2010; 71:1–9.

Dasari

, Tan

, Gurusamy

, et al. Surgical versus endoscopic treatment of bile duct stones. Cochrane Database Syst Rev, 2013; 12:CD003327.

Zhu

, Xu

, Shen

, et al. A meta-analysis of single-stage versus two-stage management for concomitant gallstones and common bile duct stones. Clin Res Hepatol Gastroenterol, 2015; 39:584–593.

Wandling

, Hungness

, Payey

, et al. Nationwide Assessment of Trends in Choledocholithiais Management in the United States From 1998 to 2013. JAMA Surg, 2016; 151:1125–1130.

Puhalla

, Flint

, O'Rourke

. Surgery for common bile duct stones—A lost surgical skill; still worthwhile in the minimally invasive century?. Langenbecks Arch Surg, 2015; 400:119–127.

Petelin

. Laparoscopic common bile duct exploration. Lessons learned from >12 years' experience. Surg Endosc, 2003; 17:1705–1715.

Nathanson

, O'Rourke

, Martin

, et al. Postoperative ERCP versus laparoscopic choledochotomy for clearance of selected bile duct calculi: A randomized trial. Ann Surg, 2005; 242:188–192.

10.

Martin

, Bailey

, Rhodes

, et al. Towards T-tube free laparoscopic bile duct exploration: A methodologic evolution during 300 consecutive procedures. Ann Surg, 1998; 228:29–34.

11.

Cuschieri

, Lezoche

, Morino

, 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; 13:952–957.

12.

Tokumura

, Umezawa

, Cao

, et al. Laparoscopic management of common bile duct stones: Transcystic approach and choledochotomy. J Hepatobiliary Pancreat Surg, 2002; 9:206–212.

13.

Lyass

, Philips

. Laparoscopic transcystic duct common bile duct exploration. Surg Endosc. 2006; 20:S441–S445.

14.

Feng

, Huang

, Wang

, et al. Laparoscopic transcystic common bile duct exploration: Advantages over laparoscopic choledochotomy. PLoS One, 2016; 11:e0162885.

15.

Zhu

, Xu

, Shen

, et al. A meta-analysis of single-stage versus two-stage management for concomitant gallstones and common bile duct stones. Clin Res Hepatol Gastroenterol, 2015; 39:584–593.

16.

Zheng

, Huang

, Xie

, et al. Laparoscopic common bile duct exploration: A safe and definitive treatment of elderly patients. Surg Endosc, 2017; 31:2541–2547.

17.

Al-Temimi

, Kim

, Chandrasekaran

, et al. Laparoscopic common bile duct exploration versus endoscopic retrograde cholangiopancreatography for choledocholithiasis found at time of laparoscopic cholecystectomy: Analysis of a large integrated health care system database. Am J Surg, 2017; 214:1075–1079.