Variations in the cystic duct: frequency and the relationship among insertion sides and heights on the bile duct

Background Evaluation of the cystic duct anatomy prior to cholecystectomy is important, to decrease the risk of bile duct injury. This study aimed to clarify the frequency of cystic duct variations and the relationship between them. Methods Data of 205 patients who underwent cholecystectomy after imaging at Sada Hospital, Japan, were analyzed. The Chi-square test was used to analyze the relationships among variations.


Background
To perform a cholecystectomy, the anatomy of the bile duct is often evaluated preoperatively to decrease the risk of perioperative complications, such as bile duct injury [1].Two factors, anatomical variations in the bile duct and in ammation around the gallbladder, make intraoperative comprehension of the structure di cult.The cystic duct typically enters the common hepatic duct (CHD) from the right lateral aspect to form the common bile duct (CBD), but many variations in the insertion of the cystic duct have been reported [2][3] [4].Surgeons may misidentify the cystic duct as another tissue when it is encountered in atypical areas.In ammation, such as cholecystitis, causes brotic changes around the gallbladder, making detection of the cystic duct di cult during surgery [5][6].When variation and in ammation coexist, the risk of misidenti cation increases [7].Images of the bile duct obtained from preoperative examinations, such as magnetic resonance cholangiopancreatography (MRCP) or drip infusion cholangiography with computed tomography (DIC-CT), support intraoperative recognition [8].
Intraoperative imaging is a more direct detection method; however, its accuracy is not the same as MRCP [9].
Although preoperative evaluation with imaging has bene ts, cholecystectomy is often performed without evaluation of the biliary tract anatomy.CT exposes patients to radiation and biliary contrast media may cause hypersensitivity reactions, which are reasons for avoiding this imaging method [10].Meanwhile, MRCP is not as widely available as CT, is contraindicated in patients with cardiac pacemakers, has a limitation in the number of patients it can be performed on due to long examination time, and demands patient tolerability during long examinations[8] [11].In addition, the necessity of routine preoperative evaluation of bile duct anatomy has cast doubt in some cases such as uncomplicated cholecystolithiasis [12].Even if preoperative MRCP imaging is performed, it sometimes does not provide su cient information because of artifacts or pseudolesions [8].Mechanical obstruction or in ammation around the gallbladder can also interfere [13].
For cases with limited preoperative information about the bile duct anatomy, knowledge of these variations may support the recognition of the cystic duct.In this study, we analyzed the frequency of variations in cystic duct insertion using detailed imaging.In our facility, preoperative imaging with DIC-CT or MRCP is routinely performed in most cholecystectomy cases, and we utilized the data from these techniques for this project.We also analyzed the relationships among the variations in cystic duct insertion.Furthermore, we analyzed variations in the right hepatic duct (RHD), which includes a thin hepatic duct close to the cystic duct that carries a risk of accidental injury during surgery [14].

Patients' characteristics and data management
From December 2018 to July 2019, 227 cholecystectomies were performed at Sada Hospital in Japan (S1 Table ).To analyze the data containing three-dimensional (3D) images, we collected data of patients examined only by DIC-CT or MRCP and excluded 19 patients who were examined using other imaging methods, such as endoscopic retrograde cholangiopancreatography.We excluded three patients of non-Eastern Asian origin according to the United Nations geoscheme (https://unstats.un.org/unsd/methodology/m49/) because regional anatomical differences in the bile ducts between origins have been reported [3].
The insertion of the cystic duct was classi ed based on the insertion point into the CHD and its course, using Sarawagi et al.'s classi cation [4].The point of insertion was de ned based on both the vertical and horizontal aspects of the CBD in 3D view (Fig. 1a and 1b).The bile duct between the hepatic con uence and the ampulla of Vater was divided into three sections based on the point of insertion: the bile duct was observed from an angle that showed its longest side, and insertion of the cystic duct into the upper, middle, and lower thirds of the bile duct were termed as high, middle, and low insertion, respectively.Similarly, based on a 45° diagonal angle for horizontal division, the points of insertion were termed lateral (to the right of the CHD), anterior, posterior, and medial (to the left of the CHD, also called the spiral course) insertion.When the cystic duct ran parallel to the CHD for > 20 mm, it was considered a parallel course.We also evaluated variations in the RHD using the classi cation proposed by Huang et al. [3][15].RHD is classi ed based on the right anterior segmental duct (RASD) insertions, which drain segments 5 and 8, and right posterior segmental duct (RPSD) insertions, which drain segments 6 and 7.The images were reviewed retrospectively by a board-certi ed radiologist and abdominal surgeon.

Statistical analysis
Relationships between variations were analyzed using the Chi-squared test.Statistical analyses were performed using JMP Pro 15.1.0(SAS Institute Inc., Cary, NC, USA).Statistical signi cance was set at P < 0.05.

Results
The percentages of the anatomical variations of cystic duct insertion A total of 205 patients were enrolled in this study, and their demographic characteristics are shown in Table 1.The most common insertion sites in the horizontal direction were the lateral (right) (92 patients, 44.9%) and posterior (92 patients) sides (Table 2).The most common insertion height in the vertical plane was middle (135 cases, 65.9%), followed by high (55 cases, 26.8%).In Fig. 2, we presented images of these variations.Sarawagi et al. de ned some variations of the cystic duct as clinically important because of the high risk of complications: spiral (medial insertion) course, parallel course, low-height insertion, and draining into the RHD [4].In our study, 24 cases (11.7%) t these criteria (Table 3).These variations sometimes overlapped, and we analyzed the relationships among them (Fig. 3a).The parallel course overlapped mainly with low-height insertions (4 7, 57.2%).The spiral courses barely overlapped with the low insertions (one of nine cases, 11.1%).The overlap between a spiral course and a parallel course was only found in one low-insertion case, indicating that the percentage of triple overlap of these clinically important variations was 0.5% (Fig. 3b).None of the 205 patients experienced any major complications during surgery, such as bile duct injury.

Relationship between heights and sides of the insertion point
Next, we analyzed the relationship between the height and side of the insertion points of the cystic duct (Fig. 4).Posterior insertion was signi cantly more frequent in low-height (75.0%) than in high-height insertions (12.7%) (P < 0.001).Conversely, the anterior insertion was signi cantly more frequent in cases of high-height insertion (10.9%,P < 0.001) than in others.The frequency of lateral insertion decreased when the height of the insertion decreased (69.1%, 37.8%, and 12.5% in the high, middle, and low insertion groups, respectively).A spiral course with a medial insertion has an anterior course running in front of the CHD (Fig. 2b) and a posterior course running behind it (Fig. 2c) [3][4].The six cases with anterior spiral courses consisted of four high-height insertions and two middle-height insertions, and the three posterior spiral course cases consisted of two middle-height insertions and one low-height insertion.

Variations in the RHD, RASD, and RPSD
We analyzed the RHD variations in our cases using Huang et al.'s classi cation, classifying them into ve types according to RASD and RPSD insertions [3][15][16] (Fig. 5).Types A4 (RPSD drains into the CHD, sometimes called the aberrant hepatic duct) and A5 (RPSD drains into the cystic duct) are related to injury or inappropriate ligation during surgery due to misidenti cation by surgeons [17].There were six cases of A4 (2.9%) and two cases of A5 (1.0%), while the majority of RHD variants were type A1 (RPSD drains into the RASD, 77.6%) (Table 4).The proximity between the RPSD of type A4, called the aberrant hepatic duct, and cystic duct was then measured [18]; in the six type A4 cases, the distance between the aberrant hepatic duct and cystic duct ranged from 2.5-16.9mm (median, 5.3 mm), and the closest point to the aberrant hepatic duct was the cystic duct or neck of the gallbladder.A thin aberrant hepatic duct that drains a segmental or smaller hepatic area may be encountered [17][19] [20].Surgeons may injure or incorrectly ligate these ducts, resulting in a stula or liver abscess [21].In our study, we found nine cases of these thin aberrant hepatic ducts (4.4%), including six that drained segment 5 and three that drained segment 6 (Fig S1).Among the 9 cases, the distance between these thin aberrant hepatic ducts and cystic ducts or gallbladder ranged from 0.7-10 mm (median, 3.2 mm).
A relationship, albeit controversial, has been suggested in some reports, we also analyzed the relationship between the insertion type and the occurrence of choledocholithiasis [22][23].Among the 12 choledocholithiasis cases in our study, middle-height (10 cases, 83.3%) and posterior-side insertions (8 cases, 66.7%) were the most frequent (S2 Table ).

Discussion
In our study, 11.7% of the cases had clinically signi cant anatomical variations, which is lower than that reported in previous studies (S3 Table )[4] [13][17] [23].One plausible reason for the lower percentage in this study may be the difference in the de nition of the insertion points.Gündüz et al. summarized the confusion and problems encountered in classifying insertion points in previous studies [23].We did not nd a clear de nition of the transverse side, except in Gündüz's report; however, their clock method (dividing sides with a combination of 60° and 120°) was different from ours (diagonal division at 90°).
For the insertion height, we used 3D images to divide the CBD and CHD into three sections by changing the angles, which might have affected the classi cation.By observing the skew angle, we noticed that the dividing point differed in the classi cation based on the front view.Another reason could be that the detailed views from DIC-CT show the insertion site more clearly than those from MRCP[8].Our data mainly came from DIC-CT (85.4%), and the images clearly showed the bile duct anatomy.For example, the seemingly lateral cystic duct eventually joins posteriorly after running parallel to a thin septum.In addition, regional differences in the anatomy of the biliary system have been reported.Therefore, our results may show a variation rate speci c to Eastern-Asian regions [3].Our data came mainly from residents living in Japan, which is one of the limitations of our study.Regardless of the superiority of the classi cation method, clinically signi cant variations of 11.7% indicated that variations in the cystic duct were common.Preoperative examinations, especially in cases of in ammation, are important, and surgeons should consider the possibility of variants when preoperative imaging is not available [5].
Asides the ambiguity of the classi cation, it is important to con rm that there is a correlation between the height and side of the insertion point.High-insertion cases tended to have anterior or lateral insertions.
With the decreased height of insertion, the number of posterior insertion cases increased, and anterior insertion became infrequent.We did not nd any reports of or contradictions to this nding in previous studies, and our ndings may provide medical professionals with knowledge that will aid them during surgery.Regarding medial insertion, we found a tendency for anterior and posterior spiral courses in cases of high and low insertion, respectively, although the number of cases was small.Further evaluation in future studies may con rm or refute this nding.It seems natural that a parallel course would occur more frequently in cases of low, rather than high insertion because the cystic duct of the parallel course joins the CHD after a downward and parallel route of ≥ 2 cm.
High-risk RHD variations were found in 3.9% of the cases, and thin aberrant hepatic ducts draining a segmental or smaller hepatic area were more common in 4.4% of the cases.These results are similar to those previously reported (S4 Table) [3][4][17] [24].These ducts are commonly located near the neck of the gallbladder or cystic duct [24].Hence, the resection line during surgery should be kept as close to the gallbladder as possible to achieve a critical view of safety during the dissection of Calot's triangle and avoid complications[6] [24].
We acknowledge that our data could have some biases because the examination data only came from patients undergoing surgeries and did not include patients undergoing only examinations.Our data did not contain bile duct injury cases maybe due to low case numbers because bile duct injuries are reported in 0.23% of bile duct surgeries [25][26].The collection of more cases of bile duct injuries may reveal highrisk of injuries with variations.In our study, the number of spiral course cases was low (4.4%); therefore, further research on anterior or posterior spiral courses must be conducted with more cases.The relationship between the types of choledocholithiasis and their variation requires further evaluation.
Finally, we mainly used DIC-CT in this study, and the indications for DIC-CT should be considered with caution because of the high occurrence of allergy [10].

Conclusions
This study con rmed that clinically signi cant variations in the cystic duct, RHD, and accessory hepatic ducts are common.Surgeons should be aware of these variations pre-and peri-operatively.The insertion point of the cystic duct tends to be anterior or lateral, superiorly and moves from a lateral to a posterior position, inferiorly.We believe that these data will be of assistance to clinicians performing bile duct surgery.The relationship between the heights and sides of the insertion point of the cystic duct.

Figure 2 Three
Figure 2

Table 3
Frequency of clinically signi cant anatomical variations in the cystic duct