PDF
Citation
Print
INTRODUCTION
Cholelithiasis is a prevalent digestive disorder, accounting for up to one-third of general surgery procedures [1]. Approximately 5%–10% of these patients develop choledocholithiasis, although the true incidence may be underestimated. The disease is dynamic—gallstones may spontaneously migrate into the intestine—so imaging provides only a snapshot. It is not uncommon to observe disappearance of previously identified stones or the appearance of new stones during surgery [1].
Improvements in minimally invasive surgery and diagnostic modalities have contributed to earlier and often incidental detection. In conjunction with contemporary guidelines advocating treatment of all confirmed cases, this has notably shortened the disease’s natural history [1].
Endoscopic retrograde cholangiopancreatography (ERCP) has conventionally been the treatment of choice, either prior to or following cholecystectomy (two-stage approach). Laparoscopic bile duct exploration (LBDE), however, provides a one-stage alternative and has potential to alter standard practice [2,3]. Evidence indicates that stone clearance, morbidity, and mortality are comparable for ERCP and LBDE. ERCP is more frequently associated with pancreatitis, whereas bile leakage is the principal risk with LBDE [4-12].
There are two types of LBDE: choledochotomy and transcystic. The transcystic approach, employing 2.8–3.5 mm choledochoscopes and utilized in approximately 30% of reported cases, is associated with improved outcomes—including reduced incidence of bile leaks, shorter operative times, decreased length of hospital stay, and lower overall costs [12-18]. Therefore, it is essential that studies distinguish between these techniques; failure to do so may mask the specific benefits offered by minimally invasive approaches [17].
Study objectives
• To establish transcystic LBDE as the first-line treatment for choledocholithiasis in patients with an in situ gallbladder.
• To define criteria for appropriate patient selection for this technique.
• To determine risk factors associated with bile duct injury in individuals undergoing this procedure.
MATERIALS AND METHODS
Study design and sample
This prospective observational cohort study included 71 patients who underwent elective laparoscopic cholecystectomy with transcystic LBDE at our institution between June 2020 and June 2024. Diagnosis of choledocholithiasis was established through ultrasound, computerized tomography, magnetic resonance cholangiopancreatography (MRCP), or prior ERCP. All patients were followed for a minimum of 6 months, with clinical and laboratory assessments performed at 1 month and imaging (predominantly MRCP) at 6 months.
Two groups were compared in the analysis:
• Group A: Patients who developed bile duct injury
• Group B: Patients who did not experience bile duct injury
Definition of bile duct injury
Bile duct injury was defined as the occurrence of either postoperative bile leak or biliary stricture. Biliary stricture was diagnosed based on follow-up imaging, commonly using MRCP [19]. The diagnosis of bile leak was established if drain bilirubin measured at least 3x the serum level, or when additional intervention, such as drainage, reoperation, or procedures for biliary peritonitis, was required [20]. The severity of bile leaks was graded according to the International Study Group of Liver Surgery (ISGLS) criteria [21]:
• Type A: No change in management and no associated infection.
• Type B: Requires active therapeutic measures not involving laparotomy or laparoscopy, and is associated with mild to moderate infection.
• Type C: Necessitates surgical intervention and is accompanied by severe infection.
Exclusion criteria
The exclusion criteria included patients with neoplasms of the biliary tract and those initially managed with non-transcystic approaches (such as choledochotomy or transinfundibular techniques).
Data and ethics
The study adhered to STROBE guidelines and received approval from our hospital’s Ethics Committee (EO260-23_HIE), in accordance with the principles outlined in the Declaration of Helsinki.
Surgical technique
Fig. 1 and 2 depict the arrangement of the patient, surgical team, and instruments during the procedure. The operation utilized a laparoscopic approach with four trocars. Initially, an anterograde cholecystectomy was performed, after which the cystic–common bile duct angle was carefully dissected, aiming to limit the use to three trocars during this phase. Following this, an Endoloop (Ethicon, Johnson & Johnson) (Fig. 2A) was attached at the junction between the infundibulum and cystic duct, then brought out through the right hypochondrium using the assistance of an EndoClose (Covidien, Medtronic) (Fig. 2B). This enabled application of traction on the gallbladder to achieve a 90° angle between the cystic duct and common bile duct, as illustrated in Fig. 3. Next, a partial transverse incision was made on the anterior wall of the cystic duct 1–2 centimetres from the hepatic duct. At this stage, the fourth trocar was inserted, intended for introducing the choledochoscope, which should be advanced along the same traction line created by the Endoloop on the cystic duct. In selected cases, a Flexor Parallel access system (Cook Medical) (Fig. 2C) functioned as the introducer. The choledochoscopy was generally carried out using the SpyGlass Discover system (Boston Scientific Corporation) (Fig. 2D). If stones were visualized, extraction was attempted with the NGage device (Cook Medical) (Fig. 2E, 4). For stones that were large or impacted, lithotripsy was performed with the Auriga XL laser system (Boston Scientific Corporation) and a 270-micron Holmium:YAG fiber (Cook Medical) (Fig. 2F, 2G, 5). When appropriate, crushing maneuvers were employed. Management of microlithiasis or sludge involved flushing with pressurized saline. Intraoperative cholangiography (Fig. 6) was selectively performed when bile duct clearance remained uncertain, especially if any portion of the exploration—either distal or proximal to the cystic duct—could not be fully accomplished. Correlation between imaging findings and number of stones removed was not emphasized, as the pathological condition is dynamic. The cystic duct stump was closed with double non-absorbable clips. In cases where transcystic access was unsuccessful, an infundibular approach was considered. Choledochotomy was reserved for situations in which distal exploration or stone extraction via the cystic duct could not be achieved.
Study variables
A comprehensive range of variables was examined to determine possible risk factors for bile duct injury.
Epidemiological and preoperative variables
We collected data on sex, age, American Society of Anestheologist (ASA) classification [21], and Charlson Comorbidity Index, which estimates 1-year mortality risk from comorbidities and age [22]. Nutritional status was assessed using serum albumin, in addition to evaluating diabetes, international normalized ratio (INR), and imaging characteristics such as bile duct diameter and number/size of stones. Preoperative interventions—including ERCP, percutaneous transhepatic cholangiography (THPC), or radiological drainage (e.g., cholecystostomy)—were documented. Laboratory results from within 48 hours before surgery included amylase, bilirubin, aspartate aminotransferase (AST), alamine aminotransferase (ALT), gammaglutamyl transferase (GGT), alkaline phosphatase (ALP), white blood cell (WBC) count, and hemoglobin.
Intraoperative variables
We examined potential effects of cholecystitis, cystic/common bile duct diameter, and characteristics of stones, as well as the surgical approach utilized. Additional intraoperative aspects included the type of introducer, instruments, cranial exploration, flushing or crushing techniques, details on lithotripsy use (including frequency and energy settings), duodenoscopy, duration of surgery, and intra-abdominal drainage placement.
Postoperative variables
Complications were classified according to the Clavien-Dindo system [23], including bleeding, pancreatitis, and cholangitis. We documented admissions to the intensive care unit (ICU), need for ERCP or THPC, and detected residual stones or sludge. Duration of hospitalization and early postoperative laboratory values (amylase, bilirubin, AST, ALT, GGT, ALP, WBC, hemoglobin) were also analyzed.
Post-discharge follow-up
We recorded data on hospital readmissions, additional procedures, mortality, and follow-up laboratory values (at 1 and 6 months). Associations with bile duct injury were analyzed. MRCP findings at 6 months were included when obtained.
Statistical analysis
Categorical variables, including sex and readmission, were summarized using frequencies and percentages, while quantitative variables, such as age, stone size, and days until reintervention, were reported as median and interquartile range. The assumption of normality for numeric variables was evaluated using the Anderson-Darling test. Associations between biliary injury and clinical or sociodemographic variables were examined with the chi-squared test and Fisher’s exact test when appropriate. Comparisons of quantitative variables between groups with and without biliary injury were conducted using the Mann-Whitney U test. Cramer’s V was calculated to assess strength of association (classified as small if <0.2, moderate if 0.2–0.6, and large if >0.6, per Cramer, 1946), and Cohen’s d was used for effect size (small if <0.5, moderate if 0.5–0.8, and large if >0.8, in accordance with Cohen, 1988). For variables demonstrating statistical significance in the preceding analyses, a univariate logistic regression model was applied to estimate the risk associated with each in relation to biliary injury. A type I error of 0.05 was used for all tests. Statistical analyses were performed using R software (version 4.4.1).
RESULTS
Bile sludge (n = 5) or choledocholithiasis (n = 52) was identified in 57 patients (80.2%), while 12 patients (16.9%) had a clear bile duct. Malignant biliary dilation was observed in 2 patients (2.8%), including one distal cholangiocarcinoma and one gallbladder carcinoma. The mean operative time was 209 ± 92.5 minutes, and the transcystic approach was successful in 92.9% of cases. In cases where this approach failed (7.1%), incomplete stone fragmentation necessitated conversion either to a transinfundibular approach (1.4%) or to choledochotomy (5.7%). The transinfundibular route was chosen in a single case due to significant cystic duct distortion from scleroatrophic cholecystitis. The mean bile duct diameter was 10.1 ± 2.9 mm, with laser lithotripsy utilized in 51.4% of patients. Complete bile duct clearance was obtained in 94.4% of cases. The mean postoperative hospital stay was 5.3 ± 4.2 days.
Among the 71 patients, 15.4% (n = 11: 9 leaks, 2 strictures) experienced biliary injuries. Reintervention was necessary in 3 patients (4.3%). Malignancy accounted for 2 of these cases: one patient with high-grade dysplasia required pancreaticoduodenectomy and another, with pT2 gallbladder carcinoma and a positive cystic margin, was treated with margin re-excision with laparoscopic resection of segments IVb and V and hilar lymphadenectomy.
Bile leakage was recorded in 12.7% (n = 9) and represented the most common complication, with a range of severities observed: 1 type A (managed conservatively), 7 type B cases (treated using percutaneous/ERCP/THPC drainage), and 1 type C that required surgery and resulted in death from biliary sepsis and biliohemorrhage. In this case, choledochotomy with T-tube drainage was performed. Biliary stricture was observed in 4.2% (3 patients).
Postoperative factors
Comparisons of postoperative outcomes between the groups are outlined in Table 4.
Post-discharge follow-up
A comparative assessment of post-discharge findings between groups is shown in Table 5.
DISCUSSION
To our knowledge, this represents the largest cohort study of transcystic LBDE in our country. Our data support the practical feasibility of a one-stage transcystic approach for the majority of patients, irrespective of stone burden or bile duct diameter—even in cases involving nondilated ducts [24]. The strategy described here, which avoids both papillary manipulation and choledochotomy—and therefore reduces the associated complications [14]—was successfully completed in 92.3% of cases, with a bile duct clearance rate of 94.4%, closely matching results reported in larger cohorts [25]. We recommend the transcystic approach as the first-line option for patients with choledocholithiasis and cholecystolithiasis and suggest that strict selection based on cystic duct diameter is unnecessary. Noteworthy, the conversion rate to choledochotomy remained low, despite our experience stemming from an earlier series, and these conversion rates are comparable to those reported elsewhere [24].
The incidence of postoperative pancreatitis was slightly higher than previously described in other series [17,25], but this did not result in adverse clinical consequences. This finding may be attributable to our protocol of systematically measuring amylase levels during the first 24 hours postoperatively, which allowed for the identification of hyperamylasemia and for distinguishing true pancreatitis from postoperative abdominal pain.
The incidence of biliary stricture following laparoscopic common bile duct exploration remains low, consistently falling within the 0%–2% range in recent literature, regardless of prior abdominal operations [26]. Reported bile leak rates after transcystic LBDE are generally 1%–3% in studies using the ISGLS definition [27]. This value corresponds to those from large cohorts of LBDE, for which bile leak constitutes the most frequent complication; most episodes are low-grade and resolve with conservative or endoscopic measures [27]. In our series, bile leak was also the most frequent complication. The marginally increased rates of leaks and strictures may be related to a higher proportion of individuals with a history of biliary interventions (primarily ERCP), which could contribute to heightened local inflammation and technical difficulty. Additionally, our sample included a greater number of patients with higher ASA physical status scores, reflecting an increased burden of systemic comorbidities, previously identified as a risk factor for biliary complications after surgery [25,27].
Our study demonstrates that bile duct injury is associated with increased complications, prolonged hospitalization, higher rates of ICU admission and readmission, and a greater reliance on postoperative THPC and ERCP. These findings highlight the substantial clinical and economic consequences of bile duct injury. One of the primary aims was to assess, specifically through evaluation of preoperative and intraoperative factors, the potential for optimizing patient selection and identifying predictors of bile duct injury risk.
In our cohort, the transcystic approach, irrespective of cystic duct or common bile duct diameter, was linked to a reduced risk of bile duct injury. Consistent with prior research [28,29], the identification of residual choledocholithiasis or biliary sludge—diagnosed postoperatively using ERCP, THPC, or MRCP—correlated with a higher occurrence of bile duct injury. This relationship may clarify the significant differences seen in preoperative liver enzyme profiles. Normal transaminase results may cause clinically significant choledocholithiasis to be overlooked, increasing the likelihood of incomplete bile duct clearance and subsequent bile duct injury. Notably, lower preoperative levels of AST, ALT, and GGT were significantly correlated with bile duct injury, with affected patients displaying lower median values than those without injury. Conversely, this correlation was not observed for other markers of cholestasis, such as bilirubin or ALP, which did not show significant variation between groups.
We did not detect significant differences with respect to intraoperative cholecystitis. Nonetheless, higher preoperative leukocyte counts might be linked to the occurrence of bile duct injury. An unexpected observation, given that none of the patients underwent surgery with coagulopathy, was the association between elevated INR values and a heightened risk of bile duct injury.
Once choledocholithiasis was identified, 67.6% of patients underwent immediate surgery. For the remaining patients, factors such as healthcare logistics, staffing constraints, resource limitations, or initial management by other departments necessitated preoperative interventions, as outlined in Table 1. All forms of prior interventions were independently associated with an increased risk of bile duct injury. The notable frequency of these interventions in our cohort likely reflects the initial absence of clear, standardized protocols. ERCP was most frequently performed, but was unsuccessful in cases involving residual stones (n = 6), papillary ectropion (n = 3), procedural difficulties or intolerance (n = 2), and a history of subtotal gastrectomy (n = 1). THPC was the next most common and was mainly used for emergent biliary drainage in patients with cholangitis. In four patients, THPC was conducted following unsuccessful ERCP. Two patients at high surgical risk received percutaneous cholecystostomy before undergoing definitive surgery. We propose that both endoscopic and percutaneous interventions contribute to local inflammation and increase the likelihood of stricture formation, complicating subsequent bile duct exploration and elevating the risk of injury. As clinical experience increased and adherence to protocols improved, the necessity for preoperative procedures decreased. LBDE—particularly via the transcystic route—offers the advantage of definitive management in a single procedure, reducing the need for additional interventions and their associated risks. Implementation of this approach has been associated with superior outcomes, shorter hospitalization, and reduced healthcare expenditures. Our results support routine transcystic exploration in patients with suspected choledocholithiasis and gallbladder in situ, including in acute situations.
Examination of demographic variables showed no significant associations between sex, ASA status, Charlson score, nutritional assessment, or presence of diabetes and the occurrence of bile duct injury. Notably, patient age was significantly correlated with bile leakage, with those sustaining duct injuries exhibiting a higher median age than their counterparts. While aging is frequently associated with bile duct dilation, duct diameter was not significantly related to injury rates in our cohort. This age effect may indicate that older patients experience delays in diagnosis and intervention, possibly due to their initial management by specialized departments such as geriatrics. Additionally, atypical clinical presentations in these patients may contribute to the challenges of early diagnosis.
Other than the choice of surgical technique, analysis revealed no significant differences between patient groups concerning the particular intraoperative methods employed. The risk for bile duct injury was unaffected by maneuvers such as cystic duct stenting, the crush and flush method, proximal stone retrieval, laparoscopic irrigation, or duodenoscopy. Similarly, neither the auxiliary devices utilized, type or frequency of lithotripsy, drain use, nor operative duration showed any measurable influence on the risk of bile duct injury.
In the analysis of postoperative laboratory values obtained 24–48 hours after surgery, no significant differences were identified between groups. Therefore, early postoperative indicators—such as amylase, hemoglobin, bilirubin, transaminases, ALP, and leukocyte counts—do not seem to predict bile duct injury. Nonetheless, significant differences in bilirubin at one month and GGT at six months were observed postoperatively. The finding of lower bilirubin values at one month in the bile injury group is probably a methodological artifact, as our laboratory does not provide precise bilirubin measures in outpatient settings when values fall within the normal range. Consequently, we assumed these as 1 mg/dL, which may have led to an overestimation of levels in the non-injury group. The observation of lower GGT levels at six months in the injury group lacks a credible clinical basis and most likely represents a spurious association.
Despite substantial loss to follow-up, no radiological differences were identified between groups on MRCP performed at six months. Importantly, patients experiencing early postoperative biliary strictures did not exhibit lasting abnormalities on follow-up imaging, supporting the interpretation that these were likely transient inflammatory events rather than established lesions.
Our study is subject to several limitations. Being a retrospective observational investigation, it carries inherent selection bias and reflects the clinical practice of a single primary-level institution. As such, the management strategy for choledocholithiasis with gallbladder in situ described herein may not be applicable to other clinical environments. A notable limitation is that the intervention could not always be conducted within 72 hours after diagnosis, which may permit progression of inflammation and complicate surgical dissection. Additionally, our series included a high proportion of patients who had undergone prior interventional procedures, as previously discussed. Moreover, the study comprises the first 71 cases performed, representing the complete learning curve. This factor may explain certain outcomes, such as increased operative times, higher bile leak rates, or longer hospitalizations, compared to other published series [12,25].
In summary, the transcystic approach represents a valid, reproducible, and safe first-line technique for patients presenting with choledocholithiasis and gallbladder in situ—regardless of stone number or size, cystic duct or bile duct diameter, clinical or biochemical presentation, or individual patient characteristics. Based on these results, we propose that routine preselection of patients is unnecessary; only factors such as patient age and preoperative transaminase and leukocyte levels require closer monitoring. The intraoperative techniques employed and devices used do not seem to impact the incidence of bile duct injury. Nevertheless, there is a need for prospective randomized studies to evaluate long-term outcomes in comparison with alternative techniques.
XML Download