A Novel Technique for the Intraoperative Identification of Biliary Drainage Areas in the Liver after Hepatobiliary Resection for Perihilar Cholangiocarcinoma

Accepted Manuscript A Novel Technique for the Intraoperative Identification of Biliary Drainage Areas in the Liver after Hepatobiliary Resection for Perihilar Cholangiocarcinoma Takumi Fukumoto, MD, PhD, Takeshi Urade, MD, PhD, Masahiro Kido, MD, PhD, Atsushi Takebe, MD, PhD, Tetsuo Ajiki, MD, PhD, Hirochika Toyama, MD, PhD, Ippei Matsumoto, MD, PhD, Yonson Ku, MD, PhD PII:

S1072-7515(16)00005-3

DOI:

10.1016/j.jamcollsurg.2015.12.030

Reference:

ACS 8144

To appear in:

Journal of the American College of Surgeons

Received Date: 11 November 2015 Revised Date:

25 December 2015

Accepted Date: 30 December 2015

Please cite this article as: Fukumoto T, Urade T, Kido M, Takebe A, Ajiki T, Toyama H, Matsumoto I, Ku Y, A Novel Technique for the Intraoperative Identification of Biliary Drainage Areas in the Liver after Hepatobiliary Resection for Perihilar Cholangiocarcinoma, Journal of the American College of Surgeons (2016), doi: 10.1016/j.jamcollsurg.2015.12.030. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

ACCEPTED MANUSCRIPT

A Novel Technique for the Intraoperative Identification of Biliary Drainage Areas in the Liver after Hepatobiliary Resection for Perihilar Cholangiocarcinoma

RI PT

Takumi Fukumoto, MD, PhD＊, Takeshi Urade, MD, PhD＊, Masahiro Kido, MD, PhD＊, Atsushi Takebe, MD, PhD＊, Tetsuo Ajiki, MD, PhD＊, Hirochika Toyama, MD, PhD＊, Ippei Matsumoto, MD, PhD† and Yonson Ku, MD, PhD＊

Department of Surgery, Division of Hepato-Biliary-Pancreatic Surgery, Graduate School of Medicine, Kobe University, Kobe, Japan

＊

Department of Surgery, Kinki University Faculty of Medicine, Osaka-Sayama, Japan

Disclosure Information: Nothing to disclose.

SC

†

M AN U

Presented at the 27th Meeting of Japanese Society of Hepato-Biliary-Pancreatic Surgery, Tokyo, Japan, June 2015.

EP

TE D

Corresponding author: Takumi Fukumoto, MD, PhD Department of Surgery, Division of Hepato-Biliary-Pancreatic Surgery, Kobe University Graduate School of Medicine, 7-5-2 Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan Tel: (+81) 78-382-6302; Fax: (+81) 78-382-6307; E-mail: [email protected]

AC C

Brief title: Identification of Biliary Drainage Areas

2

ACCEPTED MANUSCRIPT Key words: biliary drainage area; hepatobiliary surgery; hepatectomy; intraoperative ultrasonic cholangiography; perihilar cholangiocarcinoma; Sonazoid; ultrasound contrast

Abbreviations and Acronyms

RI PT

agent

ISGLS: International Study Group for Liver Surgery IOUS: intraoperative ultrasonography MRCP: magnetic resonance cholangiopancreatography

M AN U

3D: three-dimensional

SC

CE-IOUSC: contrast-enhanced intraoperative ultrasonic cholangiography

AC C

EP

TE D

2D: two-dimensional

3

ACCEPTED MANUSCRIPT INTRODUCTION Surgical treatment of perihilar cholangiocarcinoma is technically demanding even for high-volume centers due to its complicated operative procedures (1-3). In addition to lobectomy, it typically requires bile duct resection and bile duct reconstruction using

RI PT

Roux-en-Y hepaticojejunostomy(4). Occasionally, multiple bile duct orifices emerge on the cutting surface of the hilar plate contrary to preoperative expectation in the pursuit of curative resection. Thus, a precise evaluation of the bile duct anatomy and appropriate provision for

SC

expected and unexpected bile duct orifices are critical for the success of the surgical procedure.

M AN U

Recent advances in medical image analyses provide accurate information regarding the hepatic vasculature. At present, we can evaluate the 3D relationship of the portal vein, hepatic artery, hepatic vein and bile duct with a 3D image workstation using multi-detector CT and/or magnetic resonance imaging data (5-7). In addition, hepatectomy simulation software

TE D

programs enable us to calculate the liver resection volume and liver remnant volume preoperatively. They can even visualize the drainage area of the hepatic vein and perfusion area of the portal vein (8, 9). During surgery, hepatic vasculatures are typically detected by

EP

intraoperative ultrasonography (IOUS). The drainage area of the hepatic vein and perfusion area of the portal vein can also be visualized by the dye injection method (10, 11) or clamp

AC C

methods (12). However, there are currently no methods for identifying the drainage area of bile ducts in the liver.

We recently developed contrast-enhanced intraoperative ultrasonic cholangiography (CE-IOUSC) with intrabiliary injection of an ultrasound contrast agent for real-time biliary navigation (13). Besides the biliary tree, this cholangiography technique enables us to identify the biliary drainage areas of each bile duct on the cut surface of the hilar plate. In this study, we demonstrate the usefulness of CE-IOUSC as a tool for the identification of biliary

4

ACCEPTED MANUSCRIPT drainage areas in the liver before biliary reconstruction for perihilar cholangiocarcinoma. METHODS Patients Between September 2012 and May 2015, 22 consecutive patients who underwent

RI PT

intrahepatic cholangiojejunostomy following hepatobiliary resection for perihilar

cholangiocarcinoma at Kobe University Hospital were enrolled in this study. The term

“perihilar cholangiocarcinoma” is used here for all tumors involving the hepatic confluence

SC

(14).

This study was approved by the Institutional Review Board of Kobe University

M AN U

Hospital. A summary of the study was submitted (registration ID: UMIN000006259) to the Clinical Trials Registry managed by the University Hospital Medical Information Network in Japan (http://www.umin.ac.jp/ctr/index-j.htm). Preoperative imaging analysis

TE D

Conventional multidetector-row CT (MDCT), endoscopic retrograde cholangiography (ERC), magnetic resonance cholangiography and 3-dimensional CT cholangiography

of the bile ducts.

EP

(3D-CTC)(15) were performed to provide bile duct information and to determine the cut line

Ultrasound imaging system and contrast agent

AC C

CE-IOUSC was performed using an Aplio XG or Aplio 500 ultrasound imaging

system (Toshiba Medical Systems Co, Tokyo, Japan) with a T-shaped linear probe (PLT-705BTH, 7 MHz) or a micro-convex probe (PVT-745BTH, 7 MHz). Perfluorobutane microbubbles (Sonazoid; Daiichi-Sankyo Pharmaceutical Co, Tokyo, Japan) were used as the contrast agent for intrabiliary injection in this study. A vial of Sonazoid was reconstituted with 2 mL of water for intravenous injection. The original Sonazoid solution was diluted 1,000-fold with saline for intrabiliary injection. This concentration was determined in our

5

ACCEPTED MANUSCRIPT previous study (13). Identification of biliary drainage areas in the remnant liver using contrast-enhanced intraoperative ultrasonic cholangiography After hepatobiliary resection, a 4Fr balloon catheter for cholangiography (Teleflex,

RI PT

Tokyo, Japan) was inserted into all bile duct orifices on the cut surface and the balloon was inflated on the central side of the bile duct. A linear probe was attached to the remnant liver surface and scanning was started (Fig. 1). After temporary occlusion with the balloon, a

SC

maximum of 5 mL of the Sonazoid dilution was slowly injected into each bile duct until it was filled to the peripheral branches from ventral to dorsal. All scans were performed in the

M AN U

contrast harmonic imaging mode, which is set at a low mechanical index of 0.13–0.22 at 15 frames/s to avoid microbubble disruption. The gain was properly selected at 70–90%. All imaging data were stored as moving images in the ultrasound imaging system. Any adverse contrast reactions were closely monitored and recorded during and after the operation for all

TE D

patients. Biliary reconstruction procedure

Biliary reconstruction was performed by intrahepatic cholangiojejunostomy in the

EP

typically Roux-en-Y fashion using interrupted sutures of 5-0 or 6-0 polydioxanone (PDS®; Ethicon, Tokyo, Japan). Multiple intrahepatic bile ducts to be anastomosed were generally

AC C

grouped to one to three orifices. When the drainage area of the bile duct was categorized as a portion of the segment and sufficiently small (smaller than the remaining caudate lobe) by IOUCS, this bile duct was abandoned and the orifice was closed with ligation or a sufficient running suture. Each bile duct group was drained externally and stented at anastomosis using 2 mm or 2.5 mm polyvinyl chloride tubes (RTBD tube; Sumitomo Bakelite Co., Tokyo, Japan). All postoperative complications were recorded until the patient was discharged from the hospital.

6

ACCEPTED MANUSCRIPT Assessment The actual bile duct orifices on the cut surface of the remnant liver were compared with the expected bile duct orifices before surgery according to the preoperative images.

The location of the biliary drainage areas of each bile duct orifice in the remnant liver for

RI PT

each patient was assessed according to Couinaud’s classification. The size of the biliary

drainage areas of each bile duct orifice detected by CE-IOUSC were also classified into the hemiliver, between the hemiliver and the section, the section, between the section and the

SC

segment, the segment and a portion of the segment according to the Brisbane 2000

terminology of liver anatomy and resections (16). Bile leakage was defined according to the

M AN U

International Study Group for Liver Surgery (ISGLS) definition of bile leakage after liver resection (17).

To evaluate liver damage after operation and CE-IOUSC, the serum aspartate aminotransferase (AST) and γ-glutamyl transferase (GTP) levels before operation and the

TE D

maximum level after operation and one week after operation were measured. Liver failure was defined according to the ISGLS definition of posthepatectomy liver failure (18). RESULTS

EP

Twenty-two patients were eligible for this study. The surgical data of the patients are summarized in Table 1. There were 14 men and 8 women, with a median age of 71.5 years

AC C

(range 54 to 80 years). According to the Bismuth-Corlette classification (19), there were 3 patients with type II, 8 patients with type IIIa, 5 patients with type IIIb, and 6 patients with type IV. Nine patients underwent right hepatectomy, 11 underwent extended left hepatectomy, 1 underwent extended right hepatectomy and 1 underwent hilar bile duct resection. The median operation time was 577.5 min (range 424 to 802 min) and the median blood loss was 375 mL (range 80 to 1,280 mL). The median time required for CE-IOUSC was 295 seconds (range 105 to 1555 seconds). Two patients developed anastomotic leaks and another two

7

ACCEPTED MANUSCRIPT patients developed abdominal abscesses not related to the anastomotic leaks. They were controlled by percutaneous drainage. Postoperative cholangitis was suspected in five patients (22.7%), which was controlled by antibiotics. No intraoperative or postoperative

complications related to intrabiliary injection of Sonazoid occurred. The median postoperative

RI PT

hospital stay was 26 days (range 14 to 61 days).

The median serum AST level before operation, maximum level after operation and one week after operation were 33.5 U/L (range 17 to 373 U/L), 361.5 U/L (range 164 to 716 U/L)

SC

and 30.5 U/L (range 18 to 87 U/L), respectively. The median serum γ-GTP level before

operation, maximum level after operation and one week after operation were 153 U/L (range

M AN U

36 to 658 U/L), 108.5 U/L (range 40 to 528 U/L) and 100.5 U/L (range 48 to 266 U/L), respectively. There was no unexpected elevation of the AST, ALT, or γ-GTP levels after operation.

Posthepatectomy liver failure according to the ISGLS definition was recognized in 11 of

management.

TE D

22 patients; however, all patients were grade A and did not require changes in their clinical

Identification of biliary drainage areas in the remnant liver

EP

After injection of the contrast agent, the biliary drainage area of each bile duct orifice

AC C

was detected as a spotty to diffuse hyperenhancement of the liver on 2D echo images by CE-IOUSC (Fig. 2a-2c and Supplementary Video 1) with well-defined borders. We referred to these hyperenhancements of the liver as “pseudo-staining.” Table 2 summarizes the expected bile duct orifices according to the preoperative images and the actual bile duct orifices on the cut surface of the remnant liver. In nine of 22 patients, we observed 10 unexpected bile duct orifices according to the preoperative diagnosis during operation due to an inaccurate diagnosis of the bile duct tree (n=4), the pursuit of curability (n=5) or/and preoperatively undetectable orifices (n=5) (includes multiple responses). Four of

8

ACCEPTED MANUSCRIPT

10 unexpected bile duct orifices were bile duct orifices of the caudate lobe, but the remaining 6 were those of other segments. The total number of bile duct orifices according to the preoperative images in all 22 patients was 41, while the total number of bile duct orifices on the cut surface of the remnant liver was 51. Fifty-one biliary drainage areas were classified

RI PT

into the hemiliver (n=10), in-between the hemiliver and section (n=4), the section (n=9),

in-between the section and segment (n=4), the segment (n=15) and a portion of the segment (n=9) by CE-IOUSC (Table 2). Finally, 45 bile duct orifices (88.2%) were reconstructed. The

SC

remaining 6 bile duct orifices were closed with running sutures. The diameter of the two

closed bile ducts was 2 mm (B5 and B1). The four remaining closed ducts were estimated to

M AN U

measure 1.4 mm in diameter because a 4Fr balloon catheter (approximately 1.4 mm) could be inserted into the ducts for cholangiography. One bile duct draining a small area (patient No. 2, B1, 1.4 mm in diameter) was reconstructed because the remnant liver volume was small after right hepatectomy.

TE D

CASE PRESENTATION

A 54-year-old man with perihilar cholangiocarcinoma of Bismuth type IIIb (patient No. 7) was scheduled to undergo extended left hepatectomy with extrahepatic bile duct resection.

EP

Three bile duct orifices were predicted to appear after hilar dissection according to preoperative MDCT (Fig. 3a) and 3D-CTC (Fig. 3b). In fact, four bile duct orifices were

AC C

exposed on the cut surface of the remnant liver (Fig. 3c). Before biliary reconstruction, CE-IOUSC was performed to confirm the biliary drainage area of each bile duct orifice. Four orifices drained a small portion of segment 5 and a large portion of segment 5, segment 8 and segment 6+7 (Fig. 4a-d and Supplementary Video 2), indicating that the four bile duct orifices consisted of B5, B5, B8 and B6+7. We elected to close a part of B5 that drained a small portion of segment 5 for the facilitation of anastomotic sutures. After two orifices in the remaining B5 and B8 were grouped to form one orifice, biliary reconstruction was performed.

9

ACCEPTED MANUSCRIPT The postoperative course was uneventful. DISCUSSION

We recently developed CE-IOUSC as a tool for biliary navigation during hepatectomy to identify the biliary tree and facilitate the surgeon’s understanding of spatial relationships

RI PT

between the bile ducts and surrounding tissues (13). Along with the experience of this

technique, we found that CE-IOUSC can also delineate the biliary drainage areas of the liver parenchyma belonging to a bile duct orifice with a clear dividing line as pseudo-staining using

SC

a sufficient volume of the contrast agent. We believe that this novel technique will allow surgeons to minimize the biliary complications after hepatobiliary resection in perihilar

M AN U

cholangiocarcinoma for the following reasons.

During hepatectomy, the surgeon may encounter an unexpected orifice of the bile duct on the cut surface of the liver. In such cases, s/he is forced to make a decision whether these orifices will be reconstructed or closed. To maintain the liver function, all orifices on the cut

TE D

surface of the liver must be reconstructed. However, multiple and small bile duct anastomoses are technically demanding and can result in bile duct complications (20, 21). If a biliary drainage area is small, then closure of the small bile ducts without biliary reconstruction may

EP

be an excellent option, especially when the function of the remnant liver is well preserved. Currently, there is no way of knowing the biliary drainage area belonging to a specific bile

AC C

duct orifice. In this regard, the information obtained from CE-IOUSC is beneficial for the surgeon’s decision-making during the operation. By performing CE-IOUSC from one orifice of the bile duct on the cut surface of the remnant liver, a surgeon can at a glance recognize the location and size of the biliary drainage areas that belong to the specified orifice. In this study, we encountered 10 unexpected bile duct orifices by CE-IOUSC after hepatic resection and sutured 6 of them, whose drainage volumes were estimated to be smaller than the remaining caudate lobe according to CE-IOUSC. Although, there are no data regarding the safety of bile

10

ACCEPTED MANUSCRIPT duct ligation, we believe that the bile duct orifices with similar drainage volumes can be

safely closed because we occasionally safely ligated the bile duct branch of the caudate lobe during hemi-hepatectomy for liver cancer or graft harvesting. These patients had no complications associated with a bile duct obstruction.

RI PT

The actual number of bile ducts during surgery was ten more than the estimated number of bile ducts before operation. This increase may primarily be due to the pursuit of curability and/or preoperatively undetectable orifices. The pursuit of curability may increase the number

SC

of bile duct orifices on the cut surface of the liver because it sometimes requires extended resection. In addition, we speculate that bile ducts measuring approximately 1-2 mm in

M AN U

diameter, which drained into the bile duct stricture due to tumor invasion, are difficult to detect by any modalities before operation.

In this study, two patients (9.1%) were complicated with bile leakage from cholangiojejunostomy after operation. The bile leakages were not associated with ligation of

TE D

the small bile ducts indicated by CE-IOUSC because the breakdown of the ligated bile ducts resulted in excluded segmental duct bile leakage (22-24). Both patients with bile leakage required additional bile duct resection due to positive margins by the intraoperative rapid

EP

diagnosis. Nevertheless, we could not achieve R0 resection in these patients. We speculate that additional bile duct resection and macroscopic residual cancer may be associated with the

AC C

breakdown of cholangiojejunostomy. By performing CE-IOUSC from all orifices of the bile duct on the cut surface of the liver, a surgeon can easily identify a liver parenchyma that does not connect to the detected orifices on the cut surface of the liver as the missing piece of pseudo-staining of the liver parenchyma. Theoretically, the missing piece of pseudo-staining of the liver parenchyma under this condition is caused by inadvertent ligation of the bile duct branch during hepatectomy or overlooking the bile duct orifice on the cut surface after hepatobiliary resection. Both

11

ACCEPTED MANUSCRIPT

conditions of the bile duct have the potential for causing serious bile duct complications after operation. Inadvertent ligation of the major bile duct branch is uncommon, but may impair the liver function, resulting in postoperative liver failure. Overlooking the bile duct orifice may lead to postoperative complications, excluding bile leakage, which is also not common but

RI PT

difficult to diagnose and treat (22-24). Currently, there is no method to detect these conditions during operation besides meticulous inspection. CE-IOUSC can easily detect both conditions of the bile duct branch during operation, although it cannot distinguish them.

SC

Currently, to detect the location and extent of the bile duct belonging to the specified orifice on the cut surface of the liver after hepatectomy, surgeons generally use a surgical

M AN U

probe or the radiographic IOC (25-27). A surgical probe is a conventional device found in any center but cannot show the extent or depth of the bile duct accurately. It can only indicate the direction of the bile duct. Radiographic IOC can also be performed in most centers, however, it is difficult for surgeons to recognize the location of the bile duct because it is projected only

TE D

on the 2D fluoroscopic image. In addition, it has a risk of radiation exposure to the medical staff and patients (28). On the other hand, CE-IOUSC has several distinct advantages over these classical methods. CE-IOUSC can show the biliary drainage area and bile ducts as well

EP

as the surrounding tissues including vessels that provide a clue to distinguish the segments of the liver on 2D echo imaging. Therefore, by performing CE-IOUSC, surgeons can at a glance

AC C

and accurately recognize the size and location of the biliary drainage area of each bile duct orifice. Needless to say, it has no risk of radiation exposure. These aspects may minimize the biliary complications after hepatobiliary resection. With an increased injection volume of the ultrasonic contrast agent, hyperenhancement of the liver parenchyma changes from spotty to diffuse. We refer to this hyperenhancement of the liver parenchyma as “pseudo-staining.” Pseudo-staining indicates that the biliary drainage area belongs to a specific bile duct orifice. As the term implies, this is not real staining of the

12

ACCEPTED MANUSCRIPT liver parenchyma. The contrast agent with microbubbles injected into the bile duct cannot diffuse into the extracellular and intravascular spaces of the liver. Therefore, it may be retained within the bile duct for a prolonged period. We speculate that the ultrasonography

machine depicts a persistent, high concentration of contrast agent in the peripheral small bile

RI PT

duct as pseudo-staining of the liver parenchyma due to its limited resolution.

Pseudo-staining of the liver has well-defined borders. This observation is consistent with the previous observation that peripheral bile ducts have no communication in the liver.

SC

Clinically, well-defined borders with pseudo-staining provide an excellent opportunity to estimate the size of the biliary drainage area. In this study, we roughly estimated the size of

M AN U

the biliary drainage area on 2D echo images (Fig. 4a). To accurately measure the volume of the biliary drainage area, the optimization of software programs for 3D images specifically developed for use with ultrasonography machines is imperative.

One drawback of this technique is that Sonazoid is not presently available in all countries.

TE D

However, other ultrasound contrast agents (e.g., SonoVue) are available in Europe, the USA, and China. We believe that CE-IOUSC can be successfully performed using these agents if the optimum conditions can be determined.

EP

The final goal of CE-IOUSC was reducing the bile duct-associated complication in the treatment of perihilar cholangiocarcinoma. However, this study can only demonstrate the

AC C

feasibility of the identification of the biliary drainage areas during operation by CE-IOUSC. The validity of CE-IOUSC in the treatment of perihilar cholangiocarcinoma should be determined by a prospective and randomized study with and without this new technique. CONCLUSIONS CE-IOUSC can visualize the biliary drainage area of each bile duct as pseudo-staining of the liver parenchyma. This novel technique may be useful for selecting the small duct to be sacrificed in the treatment of perihilar cholangiocarcinoma.

13

ACCEPTED MANUSCRIPT REFERENCES 1. Neuhaus P, Jonas S, Bechstein WO, et al. Extended resections for hilar cholangiocarcinoma. Ann Surg. 1999 Dec;230(6):808-18; discussion 19. 2.

Nakeeb A, Pitt HA, Sohn TA, et al. Cholangiocarcinoma. A spectrum of

RI PT

intrahepatic, perihilar, and distal tumors. Ann Surg. 1996 Oct;224(4):463-73; discussion 73-5. 3.

Nagino M, Ebata T, Yokoyama Y, et al. Evolution of surgical treatment for

resections. Ann Surg. 2013 Jul;258(1):129-40.

Nagino M, Kamiya J, Kanai M, et al. Hepaticojejunostomy using a Roux-en-Y

M AN U

4.

SC

perihilar cholangiocarcinoma: a single-center 34-year review of 574 consecutive

jejunal limb via the retrocolic-retrogastric route. Langenbecks Arch Surg. 2002 Jul;387(3-4):188-9. 5.

Wigmore SJ, Redhead DN, Yan XJ, et al. Virtual hepatic resection using

TE D

three-dimensional reconstruction of helical computed tomography angioportograms. Ann Surg. 2001 Feb;233(2):221-6.

Marescaux J, Clement JM, Tassetti V, et al. Virtual reality applied to hepatic

EP

6.

surgery simulation: the next revolution. Ann Surg. 1998 Nov;228(5):627-34. Saito S, Yamanaka J, Miura K, et al. A novel 3D hepatectomy simulation based on

AC C

7.

liver circulation: application to liver resection and transplantation. Hepatology. 2005 Jun;41(6):1297-304. 8.

Hiroshige S, Shimada M, Harada N, et al. Accurate preoperative estimation of

liver-graft volumetry using three-dimensional computed tomography. Transplantation. 2003 May 15;75(9):1561-4.

14

ACCEPTED MANUSCRIPT 9.

Takamoto T, Hashimoto T, Ogata S, et al. Planning of anatomical liver

segmentectomy and subsegmentectomy with 3-dimensional simulation software. Am J Surg. 2013 Oct;206(4):530-8. 10.

Sakairi T, Makuuchi M. Identification of the intersegmental or subsegmental

11.

RI PT

plane in the liver with a surgical clip. Surgery. 1991 Nov;110(5):903-4.

Takayama T, Makuuchi M, Watanabe K, et al. A new method for mapping hepatic

subsegment: counterstaining identification technique. Surgery. 1991 Feb;109(2):226-9. Sano K, Makuuchi M, Miki K, et al. Evaluation of hepatic venous congestion:

SC

12.

proposed indication criteria for hepatic vein reconstruction. Ann Surg. 2002

13.

M AN U

Aug;236(2):241-7.

Urade T, Fukumoto T, Tanaka M, et al. Contrast-enhanced intraoperative

ultrasonic cholangiography for real-time biliary navigation in hepatobiliary surgery. J Am Coll Surg. 2014 Feb;218(2):e43-50.

Ebata T, Kamiya J, Nishio H, Nagasaka T, Nimura Y, Nagino M. The concept of

TE D

14.

perihilar cholangiocarcinoma is valid. Br J Surg. 2009 Aug;96(8):926-34. Ajiki T, Fukumoto T, Ueno K, Okazaki T, Matsumoto I, Ku Y. Three-dimensional

EP

15.

computed tomographic cholangiography as a novel diagnostic tool for evaluation of bile

AC C

duct invasion of perihilar cholangiocarcinoma. Hepatogastroenterology. 2013 Nov-Dec;60(128):1833-8. 16.

Pang YY. The Brisbane 2000 terminology of liver anatomy and resections. HPB

2000; 2:333-39. HPB (Oxford). 2002;4(2):99; author reply -100. 17.

Brooke-Smith M, Figueras J, Ullah S, et al. Prospective evaluation of the

International Study Group for Liver Surgery definition of bile leak after a liver resection and the role of routine operative drainage: an international multicentre study. HPB (Oxford). 2015 Jan;17(1):46-51.

15

ACCEPTED MANUSCRIPT 18.

Rahbari NN, Garden OJ, Padbury R, et al. Posthepatectomy liver failure: a

definition and grading by the International Study Group of Liver Surgery (ISGLS). Surgery. 2011 May;149(5):713-24. 19.

Bismuth H, Nakache R, Diamond T. Management strategies in resection for hilar

20.

RI PT

cholangiocarcinoma. Ann Surg. 1992 Jan;215(1):31-8.

Hirano S, Tanaka E, Tsuchikawa T, Matsumoto J, Shichinohe T, Kato K.

Hepatobiliary Pancreat Sci. 2012 May;19(3):203-9. 21.

SC

Techniques of biliary reconstruction following bile duct resection (with video). J

Hasegawa K, Kokudo N, Imamura H, et al. Bilioenteric reconstruction for small

M AN U

bile ducts without mucosa-to-mucosa alignment: long-term results. Arch Surg. 2004 Oct;139(10):1050-4. 22.

Honore C, Vibert E, Hoti E, Azoulay D, Adam R, Castaing D. Management of

Jun;11(4):364-9. 23.

TE D

excluded segmental bile duct leakage following liver resection. HPB (Oxford). 2009

Patrono D, Tandoi F, Romagnoli R, Salizzoni M. Excluded segmental duct bile

24.

EP

leakage: the case for bilio-enteric anastomosis. Updates Surg. 2014 Jun;66(2):115-9. Conzen KD, Lowell JA, Chapman WC, et al. Management of excluded bile ducts in

AC C

paediatric orthotopic liver transplant recipients of technical variant allografts. HPB (Oxford). 2011 Dec;13(12):893-8. 25.

Lo CM, Fan ST, Liu CL, Lai EC, Wong J. Biliary complications after hepatic

resection: risk factors, management, and outcome. Arch Surg. 1998 Feb;133(2):156-61. 26.

Puente SG, Bannura GC. Radiological anatomy of the biliary tract: variations and

congenital abnormalities. World J Surg. 1983 Mar;7(2):271-6. 27.

Kokudo N, Aoki T. Hepatic hilar transection method for liver surgery (with

video). J Hepatobiliary Pancreat Sci. 2012 Jan;19(1):9-14.

16

ACCEPTED MANUSCRIPT 28.

Moskovitz AH, Bush WH, Jr., Horvath KD. Anaphylactoid reaction to

intraoperative cholangiogram. Report of a case, review of literature, and guidelines for

AC C

EP

TE D

M AN U

SC

RI PT

prevention. Surg Endosc. 2001 Oct;15(10):1227.

ACCEPTED MANUSCRIPT

17

IIIa

Extended left hepatectomy* Right hepatectomy*

RI PT

Table 1. Surgical Data of the Patients

Operation time, min 802 566

Blood loss, mL 720 660

Postoperative complication Cholangitis -

75 67 72 74 54 71

II IIIa IIIa IIIb IIIb IV

Bile duct resection Right hepatectomy Right hepatectomy* Extended left hepatectomy Extended left hepatectomy Extended left hepatectomy*

424 515 639 506 560 571

200 245 845 260 410 960

M

73

IIIa

Right hepatectomy

606

290

M M F M M F M M M M M M M

66 66 72 72 65 66 69 75 57 68 80 66 76

IIIa IIIa IIIa II IIIb IIIa II IIIb IV IV IIIb IV IV

Right hepatectomy Right hepatectomy Right hepatectomy Extended right hepatectomy Extended left hepatectomy† Right hepatectomy Right hepatectomy Extended left hepatectomy Extended left hepatectomy Extended left hepatectomy Extended left hepatectomy Extended left hepatectomy Extended left hepatectomy

548 584 474 624 657 689

610 310 80 330 340 290

502 503 686 591 559 617 691

490 335 530 660 330 590 1280

Anastomotic leak Cholangitis Cholangitis Cholangitis, Abdominal abscess Pancreatic fistula Cholangitis Ileus Anastomotic leak Abdominal abscess

Age, y

Bismuth classification

Operative procedure**

F F

73

IV

72

3 4 5 6 7 8

M M F F M F

9 10 11 12 13 14 15 16 17 18 19 20 21 22

AC C

EP

TE D

M AN U

SC

Sex

Patient No. 1 2

ACCEPTED MANUSCRIPT

18 *

Combined portal vein resection and reconstruction.

**

EP

TE D

M AN U

SC

Combined right hepatic artery resection and reconstruction.

AC C

†

RI PT

All hepatectomies were combined with extrahepatic bile duct resection.

ACCEPTED MANUSCRIPT

19

Table 2. Size and Location of the Detected Bile Duct Orifices on the Cut Surface of the Liver Compared with Those of the Expected Bile Duct Orifices According to the Preoperative Images. Expected orifices (n)

Detected orifices (n)

Reconstructed orifices

Closed orifices

4

-

3

-

2

-

1

B1

RI PT

Patient No.

B5+8, B6+7

2

B1, B2+4, B3 (3)

B1, B2+4, B3 (3)

3

BL, BR (2)

BL, BR (2)

4

BL (1)

B1, B2+3+4 (2)

5

BL (1)

BL (1)

1

-

6

BR (1)

BR (1)

1

-

7

B5, B8, B6+7 (3)

B5, B5, B8, B6+7 (4)

3

B5

8

B5+8, B6+7 (2)

B5+6+7, B8 (2)

2

-

9

BL (1)

BL (1)

1

-

10

BL (1)

BL (1)

1

-

11

B1+2, B3+4 (2)

B1+2, B3+4 (2)

2

-

12

B2+3+4 (1) B1, B2+3, B4 (3) B5+8, B6+7 (2) B1, B2+3, B4a, B4b* (4) BL (1) BR (1) B5+8, B6+7 (2) B5+8, B6+7 (2) B5+6, B7+8 (2)

1

B4 -

EP

TE D

M AN U

B5, B6, B7, B8 (4)

AC C

13 14 15 16 17 18 19 20

(2)

SC

1

B2+3+4 (1) B2+3, B4, B4 (3) B5+8, B6+7 (2) B2+3, B4a, B4b* (3) BL (1) BR (1) B5, B6+7, B7, B8 (4) B5+8, B6+7 (2) B5+6, B7, B8 (3)

2 2

3 1

1 4 2 3

ACCEPTED MANUSCRIPT

20

B5+8, B6+7 (2) B5+8, B6+7 (2)

B1, B5, B8, B6+7 (4) B1, B1, B5+B6+B7, B8 (4)

3 2

RI PT

21 22

*

B4a is a caudal side branch and B4b is cranial side branch.

AC C

EP

TE D

M AN U

SC

B, bile duct branch; BL, left hepatic duct; BR, right hepatic duct

B1 B1, B1

21

ACCEPTED MANUSCRIPT Figure Legend

Figure 1. Photograph of the operative field during the intraoperative identification of biliary drainage areas by contrast-enhanced intraoperative ultrasonic cholangiography (CE-IOUSC).

RI PT

Figure 2. (A) Pseudo-staining of segment 1 delineated by CE-IOUSC in patient No. 11 (see also Supplementary Video 1). (B) Pseudo-staining of segment 2 delineated by CE-IOUSC in patient No. 11 (see also Supplementary Video 1) (C) Pseudo-stained regions of the right

SC

anterior section show a clear border between the sections by CE-IOUSC in patient No. 19.

M AN U

Figure 3. A case of extended left hepatectomy with bile duct resection for perihilar cholangiocarcinoma of Bismuth type IIIb (patient No. 7) (see also Supplementary Video 2). (A) Preoperative multidetector-row CT before endoscopic nasobiliary drainage showed that the bilobar intrahepatic bile ducts were dilated by an obstruction around the perihilar bile

TE D

ducts. (B) A three-dimensional CT cholangiography (3D-CTC) image on the ventral side shows intrahepatic bile ducts of the right lobe and an endoscopic nasobiliary drainage tube. According to this image, the planned resection line was decided to predict three bile duct

EP

orifices on the cut surface of the remnant liver. (C) A photograph of the cut surface of the remnant liver. Three white arrowheads indicate the predicted bile duct orifices (B5, B8, B6+7).

AC C

A thick white arrow indicates an unexpected orifice (another B5). B, bile duct branch; RPV, right branch of the portal vein; RHA, right branch of the hepatic artery.

Figure 4. Pseudo-staining of the liver parenchyma delineated by contrast-enhanced intraoperative ultrasonic cholangiography (CE-IOUSC) through the four bile duct orifices on the cut surface of the liver in patient No 7. Using CE-IOUSC, the size and location of the drainage area of each bile duct orifice were classified into a small portion of (A) Segment 5,

22

ACCEPTED MANUSCRIPT (B) a large portion of Segment 5, (C) Segment 8, and (D) Segment 6+7 (right posterior section). In this patient, one bile duct orifice whose drainage area comprised a small portion of Segment 5 was closed for the facilitation of anastomotic sutures. The volume of this area

AC C

EP

TE D

M AN U

SC

depth (3.0.cm, data not shown) of it on two 2D echo images.

RI PT

was roughly estimated to be 11.3 ml according to the length (1.5 cm), width (2.5cm) and

23

ACCEPTED MANUSCRIPT List of Supplemental video files Supplemental video file 1.wmv

AC C

EP

TE D

M AN U

SC

RI PT

Supplemental video file 2.mp4

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT