PDF
Citation
Print
INTRODUCTION
Laparoscopic cholecystectomy (LC) is among the most frequently performed procedures in general surgery, both electively and in emergencies. In Ireland, approximately 6000 cholecystectomies performed annually, with this figure continuing to rise over recent years [1]. Bile duct injury during LC remains a potentially devastating complication, associated with considerable morbidity and mortality [2].
An initial increase in the incidence of bile duct injuries (BDIs) was documented during the early introduction of LC in the 1990s, with reported rates ranging from 0.4% to 0.6% [3]. Subsequent studies have demonstrated a marked reduction in BDI incidence, now estimated at around 0.2%–0.4%, which is comparable to the rate observed with open cholecystectomy [4,5]. Anatomic misidentification is recognised as the primary risk factor for bile duct injury, with evidence from the Japanese national clinical database showing that approximately 60% of such injuries are due to this cause [5]. Additional risk factors include anatomic variations and the presence of acute inflammation [6]. Minor BDIs are typically treated with endoscopic stenting. In contrast, major bile duct injuries (MBDIs) usually necessitate complex major bilio-enteric reconstruction. The World Journal of Emergency Surgery guidelines from 2020 recommend Roux-en-Y hepatico-jejunostomy for all MBDI post LC [7]. The timing of BDI detection is a critical prognostic indicator and influences management strategies [8]. Optimal timing of BDI repair is currently debated. Early repair may reduce length of stay (LOS) and costs, while certain centres recommend initial drainage followed by delayed repair to allow resolution of inflammation [9,10].
Long-term morbidity following BDIs is frequently attributed to the development of biliary strictures. These strictures often require additional interventions, including endoscopic procedures or revision of the bilioenteric anastomosis, and in some cases may necessitate liver transplantation. As a result, such injuries pose a significant economic burden on the healthcare system [11].
The aim of this study was to evaluate peri-operative and long-term outcomes in patients undergoing surgical management of BDI across the three hepatobiliary (HPB) centres in Ireland and to estimate the associated inpatient healthcare costs.
MATERIALS AND METHODS
Study design
This study was conducted as a retrospective multi-centre analysis involving patients who underwent surgical management of iatrogenic BDI following LC at the three HPB centres in Ireland between 2014 and 2024. Patients who underwent primary repair, hepatico-jejunostomy, and/or right hepatectomies for BDI were identified through theatre records, and corresponding patient charts were retrieved for review. Demographic characteristics, time to surgical management, peri-operative data, and follow-up information, including data on re-operations, were collected and analysed. Classification of BDIs was performed according to the Strasberg system using imaging and operative note records. While Strasberg classification is widely used for BDIs, it does not account for related vascular injuries, injury mechanisms, or timing of diagnosis. Therefore, all BDIs were also retrospectively classified using the more recent ATOM (Anatomic, Time Of detection, Mechanism) classification in addition to the Strasberg system. Post-operative complication severity was graded according to the Clavien-Dindo system. Peri-operative outcomes were compared across groups based on timing of BDI detection. Ethical approval was obtained from St. Vincent’s University Hospital research and ethics committee on the 10.12.2024 (reference number = RCR24-025).
Statistical analysis
Statistical analyses were performed using Jamovi (Version 2.62; The Jamovi Project [2024], https://www.jamovi.org). Continuous variables were compared between groups using the nonparametric Mann–Whitney U-test where appropriate. Frequencies of categorical variables were compared using Pearson’s chi-square or Fisher’s exact test, as indicated. p-value of < 0.050 was considered statistically significant.
Cost analysis
Costs were estimated using the Healthcare Pricing Office published admitted patient price list. This document provides estimated costs per admission for procedures categorised according to the diagnosis-related group (DRG) using Australian-Refined Diagnosis-Related Groups (AR-DRG) version 8.0. Surgically managed BDIs were classified under the major biliary tract procedures (H02A) AR-DRG. The direct cost estimate was based on the assigned DRG and LOS. Costs from subsequent admissions due to re-operations were incorporated using the relevant subsequent procedure DRG.
RESULTS
Demographics and pre-operative data
A total of 34 patients who underwent surgical management for iatrogenic BDI following LC were identified (Table 1). During the study period, 26 (76.4%) patients had bilio-enteric anastomosis using hepatico-jejunostomy. Five patients (14.7%) underwent a primary repair. In addition, three patients (8.8%) required a right hepatectomy in conjunction with hepatico-jejunostomy. Of these patients, 32 (94%) were referred from a non-HPB centre with suspected or confirmed BDI. The median age was 58 years (range = 24–79). The female to male ratio was 2.09. The median BMI was 29 (range = 19–44.8). The most common indication for LC was cholecystitis, representing 58.8% of cases (n = 20). Six patients (17.6%) underwent LC due to biliary colic, and five (14.7%) for either choledocholithiasis or gallstone-induced pancreatitis. In 30 (88.2%) cases, LC was performed electively, whereas four patients (11.7%) had emergency LC for cholecystitis. Conversion to open surgery occurred in six (17.6%) cases. The overall median time to surgical management was 6.5 days (range 0–685). There were 4 revision hepatico-jejunostomies performed for biliary strictures following initial primary repairs. For these patients, time to surgery was calculated from the date of revision hepatico-jejunostomy, which was considered the definitive surgical intervention.
Classification of the BDIs including vascular injuries
Tables 2 and 3 provide a summary of the results. More than 90% (n = 31) of BDIs were classified as MBDIs, namely Strasberg E, and the majority were located within 2 cm of the confluence of the right and left hepatic ducts. Vascular injuries were identified in 13 patients (38.2%), predominantly involving the right hepatic artery (n = 12). Intra-operative diagnosis of BDI occurred in 15 patients (44.1%). Early postoperative diagnosis was made in 14 patients (41.1%), while in five cases, the diagnosis was delayed beyond 21 days.
Management of intra-operatively diagnosed BDIs
Fourteen of the 15 intra-operatively diagnosed BDIs were MBDI that occurred at non-HPB centres. In ten cases, initial management involved placement of an external drain followed by transfer to an HPB centre for definitive care. Median time to surgical intervention was 2 days (range 0–685). Eight patients ultimately underwent hepatico-jejunostomy, while two required both right hepatectomy and hepatico-jejunostomy. One patient needed abdominal washout for biliary peritonitis prior to hepatico-jejunostomy. In another case, a patient with confluence disruption and concurrent right hepatic artery injury received hepatico-jejunostomy to the left main duct with ligation of the right duct.
Five patients underwent primary repair intra-operatively. Four of these cases involved MBDIs, which were initially repaired at non-HPB centres before being referred. All four of these primary repairs ultimately required revision to hepatico-jejunostomy. The single non-MBDI addressed with primary repair was a cystic duct stump leak. Peri-operative outcomes are detailed in Table 4. The mean LOS was 27.4 days (range 5–90). Nine patients (60%) experienced post-operative complications, including four cases of surgical site infection and three bile leaks. Three patients had Clavien-Dindo grade I-IIb complications, while four had grade IIIa to IV events. There were two peri-operative deaths (Clavien-Dindo grade V): one patient developed respiratory failure secondary to lower respiratory tract infection and significant intra-abdominal collections, and another developed uncontrolled sepsis following re-laparotomy for biliary peritonitis washout.
Management of post-operatively diagnosed BDIs
Among the post-operatively diagnosed BDIs, 17 out of 19 were classified as major injuries. The two non-MBDI were lateral injuries to the common bile duct (Strasberg D). Of the 19 patients, 18 underwent hepatico-jejunostomy, while one required both right hepatectomy and hepatico-jejunostomy. Median time to surgical intervention was 6.5 days in early post-operative BDIs (<2 weeks) compared to 81 days for late-diagnosed cases.
The mean LOS was 17.2 days in the early post-operative diagnosis group and 14 days in the late diagnosis group. Nine patients (50%) experienced post-operative complications, including five bile leaks and one portal vein thrombosis. Intensive care unit (ICU) admission was required for three patients. One peri-operative mortality resulted from a combination of respiratory sepsis, liver failure, and cardiac tamponade.
Overall outcomes and statistical analysis
Peri-operative morbidity for the entire cohort was 52.9%, with a median LOS of 21.5 days (range 5–90). No statistically significant differences were observed for LOS (p = 0.278), post-operative complication rate (p = 0.73), or ICU admission (p = 0.169) across intra-operative diagnosis, post-operative diagnosis, and late diagnosis groups in this study. There was no significant association between vasculobiliary injury and the development of biliary strictures (p = 0.955).
Long term outcomes
The follow-up period was ten months (range 0–109). Ten patients required re-operation, including four who underwent hepatico-jejunostomy revision due to biliary strictures following failed primary repairs. Four patients underwent incisional hernia repairs. One patient ultimately required orthotopic liver transplantation for recurrent cholangitis due to a biliary stricture. Additional interventions included re-laparotomy and washout for bile leak in one patient, and adhesiolysis for small bowel obstruction in another. Four patients developed biliary strictures after hepatico-jejunostomy, which were managed with endoscopic stenting, while one patient is awaiting revision re-hepatico-jejunostomy for recurrent cholangitis.
Cost analysis
The direct cost of a BDI was calculated using formulae adopted from the Admitted Patient Price List for 2019 [12]. The base price corresponds to the monetary value reflecting the average resource use for the DRG procedure. The relative value (RV) is established from the projected LOS for each DRG [12]. For complex biliary interventions, the base price is €21,735 with an inlier RV of 3.955 and an anticipated LOS between 9 and 43 days. Accordingly, the mean expected cost for a BDI repair falling within the expected LOS was €85,961. When applying our LOS findings to this calculation, we observed a median cost of €85,961 for the initial repairs, ranging from €52,663 to €278,012. Table 5 provides a summary of these results. We then incorporated costs for admissions related to re-operations by using the corresponding DRG and calculating the RV based on the actual LOS. The overall expenditure for the 34 BDIs, inclusive of subsequent procedures, was €7,686,245, with individual BDI costs ranging from €52,663 to €4,656,064. The most significant outlier was a patient who underwent an orthotopic liver transplant and had a 177 day hospitalisation following the procedure.
DISCUSSION
This study presents a comprehensive overview of the surgical management and outcomes of BDIs occurring after LC in Ireland. Hepaticojejunostomy was the predominant intervention (76.4%), consistent with recommendations for MBDI treatment [7], while a proportion of patients underwent right hepatectomy, mainly due to extensive confluence disruption or significant vascular injury. Notably, severe vasculobiliary injuries were identified in 13 patients (38.2%), highlighting the need for precise classification and documentation of BDIs referred to HPB centres. Although the Strasberg classification is widely used, it omits important information. The new ATOM classification published by the European Association for Endoscopic Surgery should be utilised for its increased granularity on important prognostic factors including associated vascular injury, timing and mechanism of injury [13,14]. Within this study population, at the time of diagnosis BDI categorisation was reported either using the Strasberg system or was not formally classified; none of the cases were assessed with the new ATOM classification.
Our median time to definitive surgical management was 6.5 days, closely aligning with a recent 2024 national cohort multi-centre Danish study by Stilling et al. [15], which evaluated surgical management of BDIs and reported a median time to surgery of 5 days. In contrast, these intervals differ from an older large study on surgically managed BDIs by Lillemoe et al. [16], where the median time to surgery was 3 months, highlighting a significant shift towards aggressive early repair in contemporary practice. Several factors likely explain these temporal differences. Over the past decade there has been a trend towards prompt imaging and expedited referral to specialist HPB services, driven by improved access to cross-sectional imaging (including magnetic resonance cholangiopancreatography [MRCP]), more rapid inter-hospital transfer pathways, and a contemporary preference in many centres for early definitive repair in appropriately selected patients. In contrast, older series frequently included patients managed with delayed elective reconstruction and often combined minor and MBDIs, resulting in longer median times to repair.
The predominance of referrals from non-HPB centres (94%) emphasises the need for early recognition and prompt referral of such injuries. In Ireland, most LCs are performed outside HPB centres, as the HPB centres alone lack the capacity to perfrom a high volume of LCs. Our data strongly supports the management of MBDI repairs in HPB centres. Every primary repair of MBDIs performed outside an HPB setting subsequently required revision with hepatico-jejunostomy, indicating that definitive management of complex BDIs is best undertaken in specialised units. This is supported by a single-centre review of 200 patients, which demonstrated that early repair by non-specialist HPB surgeons is an independent risk factor for recurrent cholangitis and repeat reconstructions [17]. Additionally, a recent meta-analysis comparing early and delayed surgical repair reported that non-specialist early repair may compromise future revisions by subspecialist HPB surgeons [18].
The overall peri-operative morbidity in our cohort was 52.9%. This rate is somewhat higher than figures reported in several large series of surgically managed BDIs, where complication rates generally range from 37% to 50%. Sicklick et al. [19] reported a morbidity rate of approximately 42% in a cohort of 200 patients, while Martin et al. [20] observed complication rates nearing 45% in an 11-year single-centre study; Zidan et al. [21] documented 40% morbidity in a recent cohort [19], and Reinsoo et al. [22] described rates from 37%–43% in a population-based study. The slightly elevated morbidity rate in our series is likely due to the case-mix, which included a high proportion of major injuries (over 90% Strasberg E), a substantial rate of associated vascular injuries (38%), and a referral pattern favouring complex cases, including four with failed primary repair prior to transfer from non-HPB centres. These are recognised factors that contribute to increased peri-operative complication risk and may account for the higher morbidity observed in our national cohort.
Interestingly, there was no significant difference in post-operative complication rate, LOS, or ICU admissions between intra-operative and post-operative diagnosis groups. These results align with those reported by Stilling et al. [15]. In their population-based study of 139 BDIs, peri-operative outcomes did not differ between intra-operative and post-operative diagnosis of BDI or between early and late repair with hepatico-jejunostomy. This is in contrast to an increasing number of studies suggesting early BDI diagnosis improves peri-operative outcomes [14,23]. Törnqvist et al. [24] examined the impact of early BDI detection on survival in 747 BDIs using data from the GallRik’s national Swedish registry for gallbladder surgery. They found intra-operative diagnosis of BDIs resulted in 1 year survival rates comparable to patients post-LC with no BDI. Conversely, patients with post-operatively diagnosed BDIs had a twofold increase in mortality risk compared to those undergoing LC without BDI, according to Cox regression analysis. However, it is important to recognise that the GallRik’s database includes all minor BDIs, while both our study and that by Stilling et al. [15] focused exclusively on MBDIs.
In our study, one (2.9%) patient with end stage liver disease following a revision hepatico-jejunostomy for failed primary repair required an orthotopic liver transplant, and three additional patients underwent a right hepatectomy. Nordin et al. [25] observed a 22% liver transplantation rate in a small BDI case series from a tertiary referral centre. In that investigation, most patients had received prior primary repair before referral. de Santibañes et al. [26] reported a 7.3% incidence of liver transplantation after BDI with a median follow-up of 62 months, while Schmidt et al. [27] identified an overall liver transplantation rate of 4.3% along with an additional 4.3% of patients requiring liver resection.
With a median follow-up of 12 months, the overall stricture rate in our cohort was 23.5%, and four patients subsequently required a revision hepatico-jejunostomy. These results are consistent with previously published findings. Stilling et al. [15] documented a 30% stricture rate, and Schmidt et al. [27] found a 19.6% stricture rate in their cohort. Interestingly, we found no significant correlation between vasculobiliary injuries and later strictures (p = 0.95). This was comparable to findings from the Danish cohort study [15]. Previous studies demonstrated higher rates of strictures in associated vascular injuries and have theorised that stricture formation post BDI is due to ischaemic consequences of arterial disruption [28,29]. The retrospective design of our study means vascular injuries may have gone unrecorded if not documented by the operating surgeon or on pre-operative imaging. This limitation may have led to underreporting of vascular injuries in patients who later developed strictures.
Several recent studies in Ireland have employed the Activity Based Funding framework [12] to estimate healthcare costs, including a recent report on testicular torsion [30] and previously on the costs associated with hip fracture [31]. Our study found a median cost of €85,961, which is substantially higher than the €26,569 per MBDI estimated in a recent study by Rystedt et al. [32]. In that study, direct costs were derived from a regional debit price list, and repeat hospital admissions for re-operations were excluded. Elser et al. [33]’s comprehensive US review using private healthcare claims data recorded estimated costs of $63,524 for same day repairs and $122,055 for early repairs, with significant cost increases observed with longer delays to definitive surgical intervention.
Preventive strategies remain central in reducing the occurrence of IBDIs. The “critical view of safety” technique is widely advocated, and the 2020 World Society of Emergency Surgery guidelines advocate its routine use, particularly with selective intraoperative cholangiography when the anatomy may be unclear [7]. In the Irish setting, intraoperative cholangiography is not routinely implemented but may be used on a case-by-case basis. MRCP is commonly adopted in the early postoperative period when IBDI is suspected, offering high diagnostic precision before patients are referred to a HPB centre. The integration of these preventive measures, together with new approaches such as artificial intelligence–assisted anatomical recognition, reflects the evolving international focus on prevention and early detection [5].
This study’s strengths include the collection of data from all three HPB referral centres in Ireland, supporting its validity as a comprehensive representation of the Irish experience with BDI. Compared to a recent national registry study that identified only 4 MBDI among 3,500 emergency cholecystectomies, our study found approximately the expected number of BDIs [6]. However, it is important to consider that some BDIs may have been overlooked if repaired outside of HPB centres or in private hospitals. There are also several limitations. First, the sample size is relatively small (n = 34). This limits statistical power and subgroup analysis. Second, the retrospective design and referral bias (the majority of cases were referred from non-HPB centres) may overrepresent more complex injuries, including those managed endoscopically or treated definitively in non-HPB settings. In addition, our cost analysis is necessarily specific to the Irish publicly funded healthcare system, and we acknowledge that direct extrapolation to other international contexts is limited.
In conclusion, in this study of BDIs in Ireland we found a significant rate of morbidity (52.9%) and mortality (8.8%) post-surgical management with significant cost associated to the healthcare system. Subspecialist HPB referral is key to successful management, which typically involves bilio-enteric anastomosis with occasional hepatic resection reserved for selected cases of concomitant vascular injury. These findings complement international, prevention-oriented research by providing national outcome and cost data, reinforcing the global consensus that early recognition, timely referral, and prevention strategies are central to care.
XML Download