Presence of Aberrant Anatomy Is an Independent Predictor of Bile Duct Injury During Cholecystectomy

Bile duct injury (BDI) during cholecystectomy has severe consequences for the health of patients, in terms of mortality and morbidity. If considerable disability results, those can be costly for individual patients and the health care system.^1–3 The incidence of BDI is reported to range between 0.15% and 1.5%.^4–7 Various potential predisposing factors have been described for BDI, including, but not limited to, inflammation, biliary anomaly, surgeon's experience, older age, and male sex.^4–12 However, previous studies addressing the influence of a preexisting biliary anomaly have been limited for 2 reasons. First, as many studies included anomalies that did not influence the occurrence of BDI, the results were not pertinent to preventing BDI during cholecystectomy. For instance, a diagnosis of triple (right anterior and posterior segmental ducts and left hepatic duct) confluence pattern or right posterior segmental duct draining into the left hepatic duct had no impact on the occurrence of BDI, although it is crucial for the success of donor hepatectomy.^13–15 To prevent BDI during cholecystectomy, we must understand the incidence and pattern of aberrant bile ducts that run near Calot's triangle, an important anatomical landmark. Second, there have been no reports that have compared the operative outcome, including BDI, in the presence and absence of aberrant anatomy or have examined the impact of these anomalies on the occurrence of BDI, while controlling for potential confounding factors. Although there was only one paper which reported no differences in the incidence of BDI between cases with or without preoperative knowledge of bile duct anatomy,¹⁶ they have not examined the impact of preoperative knowledge of “aberrant anatomy” on the occurrence of BDI.

Therefore, the aim of this study was to evaluate detailed information regarding aberrant bile ducts that might influence the occurrence of BDI during cholecystectomy and to investigate whether bile duct anomaly is an independent predictor for BDI.

Materials and Methods

Patients

Between January 2006 and December 2012, a total of 1479 consecutive patients with benign disease of the gallbladder, including gallstones, adenomyomatosis, acute or chronic cholecystitis, and polyps, underwent cholecystectomy at Toyohashi Municipal Hospital. Although magnetic resonance (MR) cholangiography was routinely performed before cholecystectomy for prevention of bile duct injury at our institution, it was not performed in 190 patients preoperatively. Among these 190 patients, 136 patients underwent endoscopic retrograde cholangiography (ERC) for the treatment of bile duct stones. The remaining 54 patients who refused to be examined did not receive either of these imaging studies. Thus, 1289 patients who received MR cholangiography preoperatively formed the cohort for this study. We retrospectively collected the clinical characteristics of these patients, including sex, age, surgeon's experience, operative procedure (laparoscopic or open surgery), maximum white blood cell (WBC) count, emergency or elective operation, and occurrence of BDI. Operating surgeons were classified in 2 groups based upon having performed more than 50 cholecystectomies. Emergent operations were those performed within 48 hours after admission to our hospital for acute cholecystitis. Moreover, all MR cholangiography images were reviewed by an experienced abdominal imaging radiologist (AT) and digestive surgeon (SN) independently. The findings, including the presence of aberrant anatomy, thickening of the gallbladder wall (TGW), and cystic duct stones, which may be related to the occurrence of BDI, were recorded by consensus of 2 reviewers. The type of aberrant anatomy was also noted according to the definition described below (see “Definition of type of aberrant anatomy” and Fig. 1).

View Figure

• Download as Powerpoint
• Download Figure

Results

Among the 1289 patients, an aberrant bile duct anatomy was identified in 144 (11.2%), including 123 type A, 15 type B, and 6 type C. No difference was observed in the distribution of cases among sex or surgical procedures (LC or OC). The age of patients with aberrant anatomy was significantly higher than that of patients without [64 (range, 50.25–72) years versus 60 (range, 47–70.5) years, P = 0.042]. With regard to the surgical outcome, the volume of blood loss and operative time were similar between the two groups, although the rate of conversion to OC was significantly higher in patients with aberrant anatomy (11.8% versus 7.0%, P = 0.044). The incidence of BDI was significantly higher in patients with aberrant anatomy compared with those with normal anatomy (3.5% versus 0.3%, P = 0.001) (Table 1).

Table 1 Comparison between patients without and with aberrant bile duct

View Fullscreen

By univariate analysis, OC, the presence of aberrant bile duct, and TGW were significantly associated with the occurrence of BDI (Table 2). By multivariate analysis, the presence of an aberrant bile duct and TGW were independent predictors for BDI occurrence (OR = 16.56, 95% CI = 3.28–83.51, P = 0.001 and OR = 10.96, 95% CI = 1.98–60.80, P = 0.006, respectively; Table 3).

Table 2 Univariate analysis for predisposing factors for BDI during cholecystectomy

View Fullscreen

Table 3 Multivariate analysis for predisposing factors for BDI during cholecystectomy

View Fullscreen

BDI, injury site, severity, and clinical outcome

The characteristics of 8 patients with BDI are summarized in Table 4, and diagrams of the injuries are presented in Figs. 2 and 3. Bile duct injury was diagnosed on the seventh (POD) in 1 case (12.5%; case 1), and the remaining 7 injuries were diagnosed intraoperatively. In case 1, thermal injury with electrocautery was suspected because a small perforation at the confluence of the cystic duct and the aberrant right posterior segmental duct was found on reoperation. An aberrant anomaly was present in 5 (62.5%) of the 8 cases and the aberrant bile duct was injured in 4 cases (cases 1–4). In case 6, laceration occurred at the common hepatic duct (CHD), although the patient had an aberrant bile duct. In one case (Case 5), CHD was transected, while laceration occurred in the CHD or common bile duct (CBD) in 3 cases (6–8). In all cases, a repair procedure was carried out with laparotomy, the injured site was sutured, and a stent put in place to prevent stenosis. However, because the injured bile duct was too thin, reconstruction was stopped in case 2 in which the small branch of the right anterior segmental duct draining into the common hepatic duct (BDI, type A) was severed entirely. Both of the cut stumps were ligated. As a result, bile leakage occurred and fortunately it was treated conservatively with technique of interventional radiology. In case 5, the circumference of the common hepatic duct was severed almost entirely. Postoperatively, bile leakage and stenosis of the CHD was complicated. Although he was recommended to undergo repair surgery, he refused. Therefore, even now, 7 years after cholecystectomy, he is treated with biliary stenting.

Table 4 Characteristics of 8 cases of BDI

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Discussion

There are many reports identifying risk factors for BDI. Older age, male sex, acute or chronic inflammation, surgeon's experience, and laparoscopic approach have all been identified as predictors of BDI in previous studies.^4–12 However, none of these earlier studies investigated whether a link existed between BDI and aberrant anatomy. To our knowledge, there have been no reports that have shown aberrant anatomy as an independent predictor for BDI while controlling for potential confounding factors. Using multivariate analysis, we showed that the presence of this anomaly is associated with a 16.56-fold increase in the risk of BDI occurrence. Several studies—including those from Kitami et al¹³ (2006, n = 202); Sharma et al¹⁴ (2008, n = 253); and Bageacu et al¹⁵ (2011, n = 124)—have reported the frequency of bile duct anomaly as 6.9%, 12.3%, and 24.1%, respectively, according to our definition of an aberrant bile duct. We reviewed 1289 MR cholangiography images and found an incidence of 11.2% in our patient cohort. Differences among the reported frequency of this anomaly may be attributed to the different modalities used for diagnosis. We may have underestimated the frequency in our series because drip-infusion cholangiography–computed tomography has been reported to provide higher quality images than MR cholangiography (as used in our study).^17,18 Although we did not experience aberrant bile duct injury cases in which anomalies had not been diagnosed by MR cholangiography, attention should be paid to not injure aberrant bile ducts too thin to be diagnosed preoperatively.

We classified our cases with aberrant anatomy into 3 types according to the draining site of the aberrant bile duct or cystic duct. Type A was the most common, a finding that is in agreement with earlier studies.^13–15 Due to the small sample size, we could not identify which type of aberrant anomaly was associated with the greatest risk for BDI occurrence. However, it is possible that there is no association between anomaly type and the occurrence of BDI. Because the important factor for the occurrence of BDI is not the location of the aberrant bile duct or where the cystic duct drains, but whether the aberrant bile duct runs close to Calot's triangle.

Leukocytosis and emergency surgery did not show any significant impact on the occurrence of BDI, whereas thickening of the gallbladder wall in MRCP was identified as a predictor of BDI by multivariate analysis. This result may be because chronic inflammation (in contrast to acute inflammation): is strongly associated with the occurrence of BDI, leukocytosis reflects acute inflammation, and emergency operations have been selected for cases of acute cholecystitis. As described by some authors, chronic inflammation with fibrosis may cause difficulties during cholecystectomy.^7,12 Lukas et al⁷ indicated that in cases of shrunken gallbladder with severe chronic inflammation, the BDI rate increased significantly to 3% because of the presence of a short cystic duct and difficult surgical dissection.

By multivariate analysis, the open surgical procedure was associated significantly with the occurrence of BDI compared with the laparoscopic approach. However, this result should be interpreted with caution as it is open to selection bias. In our institution, in principle, all patients with benign disease including cholelithiasis, adenomyomatosis, polyps, and acute or chronic inflammation were operated with the laparoscopic approach, whereas serious cases in poor general condition due to severe inflammation or adhesion from a previous operation were operated using the open approach from the start of the operation.

Surgeon's inexperience has been reported as one of the risk factors,^6–10,12 but was not associated with BDI occurrence in our study. This result may be because all operations performed by inexperienced surgeons were assisted by experienced staff surgeons. In our institute, should the initiating and responsible staff surgeon recognized the risk of BDI, he must order the inexperienced operating surgeon to be replaced by an experienced staff surgeon or to convert the operative procedure to OC if the cholecystectomy was started laparoscopically.

A repair procedure was carried out with laparotomy in all cases. One patient whose injury was diagnosed 7 days postsurgery underwent emergency laparotomy, whereas the remaining 7 patients whose injuries were detected during the initial surgery underwent repair surgery consecutively. Recently, several authors reported cases of right segmental bile duct injuries that were successfully treated with nonoperative management including percutaneous drainage and endoscopic stenting.^19–21 Although this management was used mainly for patients with delayed diagnosis of BDI, this repair strategy could be applied for injury cases detected intraoperatively. That is, if the surgeon suspected injury of the segmental bile duct during LC, drainage of the bile leakage could be performed intraoperatively. It is unnecessary to convert the procedure to OC and stenting of the injured bile duct can be performed later endoscopically if needed.

Because BDI during cholecystectomy is an important clinical problem resulting in serious morbidity, all efforts should be made to reduce its incidence. It is very important to achieve a “critical view of safety” in which the triangle of Calot is dissected free of all tissue, except for the cystic duct and artery, and the base of the liver bed is exposed.²² No ambiguous funicular structure must be occluded or divided before confirming this anatomy. However, it is often very difficult to achieve a critical view of safety due to severe inflammation or fibrosis in Calot's triangle. We have to consider how to proceed surgically if aberrant anatomy, a strong predictor of BDI occurrence, is diagnosed preoperatively and if severe fibrosis or inflammation is present in Calot's triangle. In such cases, subtotal cholecystectomy, in which Calot's triangle is not dissected, may be one option, although adverse effects of this operative procedure have been reported, including bile leakage from the stump of the gallbladder neck or a risk of residual malignancy.^23,24

The effect of IOC on the prevention of BDI is still under debate.^5,6,9,12,25 In the current study, the association between the performance of IOC and the occurrence of BDI was not investigated because at our institution, IOC is performed according to the surgeon's preference. Even if IOC was performed, the opacification was not enough to thoroughly reveal intrahepatic bile duct including an aberrant bile duct because this examination was performed mainly for the purpose of identifying the positional relationship between the cystic duct and the common bile duct or the common hepatic duct.

A total of 136 patients who underwent ERC but not MR cholangiography were excluded from our study. This exclusion occurred because most ERC imaging for these patients was unsatisfactory to establish whether an aberrant anomaly was present; all of these examinations were performed for the treatment of bile duct stones and not for a detailed examination of the biliary tree.

A limitation of this study is that we could not precisely know whether aberrant anatomy of 8 BDI cases was diagnosed before cholecystectomy, which should be the essential concern about this topic. According to operative records and medical chart, it was possible that aberrant anatomy which was diagnosed by authors who reviewed all images of MR cholangiography was not recognized in most BDI cases. Therefore, little understanding of aberrant anatomy might have resulted in the occurrence of BDI.

In conclusion, the incidence of an aberrant bile duct detected by MR cholangiography was 11.2% in our series. The presence of an aberrant bile duct and thickness of gallbladder wall was an independent risk factor for the occurrence of BDI. All efforts should be made to prevent BDI in cases where an aberrant bile duct and thickness of gallbladder wall have been identified.