- Email: [email protected]
Is extended hemihepatectomy plus pancreaticoduodenectomy justified for advanced bile duct cancer and gallbladder cancer?
Is extended hemihepatectomy plus pancreaticoduodenectomy justified for advanced bile duct cancer and gallbladder cancer? Yoshihiro Sakamoto, MD,a Satoshi Nara, MD,b Yoji Kishi, MD,b Minoru Esaki, MD,b Kazuaki Shimada, MD,b Norihiro Kokudo, MD,a and Tomoo Kosuge, MD,b Tokyo, Japan
Background. Major hepatopancreaticoduodenectomy (HPD) is an extensive surgical procedure offering the highest curability for patients with advanced biliary cancer. However, surgical morbidity associated with major HPD is high, and optimal indications for this procedure remain unclear. Methods. Between 1989 and 2010, 14 patients with widespread bile duct cancer and 5 with gallbladder cancer having biliary infiltration underwent major HPD at our hospital. Preoperative portal vein embolization was performed in 17 patients undergoing right HPD. Clinicopathologic factors and survivals following HPD were compared between patients with bile duct cancer and those with gallbladder cancer. Results. One patient who underwent right HPD for gallbladder cancer died of hepatic failure (5.3%) and 18 of the 19 patients (95%) developed postoperative pancreatic fistulas. The median hospital stay was 47 days. Depth of invasion was T3 in 1 patient and T4 in 2 patients with bile duct cancer and was T4 in all 5 patients with gallbladder cancer (P = .002). The clinical stage was IV in 3 patients (21%) with bile duct cancer and in all 5 patients with gallbladder cancer (P = .002). The 5-year survival rates and median survival rates of patients with bile duct cancer and gallbladder cancer were 45% vs 0 and 3.3 years vs 8 months, respectively (P < .001). Conclusion. HPD can be an acceptable treatment option for widespread bile duct cancer. However, the indication for HPD in advanced-stage gallbladder cancer should be considered carefully, considering the high morbidity rate and the advanced stage of the disease. (Surgery 2013;153:794-800.) From the Hepato-Biliary-Pancreatic Surgery Division,a Department of Surgery, Graduate School of Medicine, The University of Tokyo, and the Hepatobiliary and Pancreatic Surgery Division,b National Cancer Center Hospital, Tokyo, Japan
SURGICAL RESECTION for advanced biliary cancer remains challenging. Patients with widespread bile duct cancer and advanced gallbladder cancer having biliary and peripancreatic infiltration are possible candidates for hepatopancreaticoduodenectomy (HPD). HPD, especially extended hemihepatectomy plus pancreaticoduodenectomy, offers the highest curability to obtain negative ductal or radial
Supported in part by a grant-in-aid for scientific research from the Ministry of Health and Welfare of Japan and the National Cancer Center Research and Development Fund. Accepted for publication November 28, 2012. Reprint requests: Yoshihiro Sakamoto, MD, Hepato-BiliaryPancreatic Surgery Division, Department of Surgery, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2013 Mosby, Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2012.11.024
794 SURGERY
margins for advanced biliary cancer.1-4 However, this procedure is associated with high morbidity and mortality rates. The mortality rate for HPD has been reported to be as high as 33%,5-10 and the major complications associated with this procedure include pancreatic fistula, hepatic failure, and other infectious complications. Therefore, surgical indications for major HPD should be evaluated carefully after considering the balance between the risks and the expected prognosis of the patient. We have performed 19 HPDs in the past 20 years for biliary malignancies. The objective of the present article is to discuss optimal indications for HPD in patients with advanced biliary cancer by reviewing our experiences and other reported case series. METHODS Between 1989 and 2010, 283 patients with bile duct cancer (excluding intrahepatic cholangiocarcinoma) and 204 patients with gallbladder cancer
Surgery Volume 153, Number 6
underwent surgical resection at the Hepatobiliary and Pancreatic Surgery Division, National Cancer Hospital, Tokyo. Of these, 14 patients with widespread bile duct cancer (4.9%) underwent right or left HPD, and 5 patients with gallbladder cancer having biliary infiltration (2.5%) underwent right HPD. One patient underwent right HPD on the preoperative diagnosis of diffuse-type bile duct cancer, but the postoperative pathologic examination revealed primary sclerosing cholangitis without evidence of malignancy. This patient was therefore excluded from the present series. Imaging studies for biliary cancer. A preoperative diagnosis of biliary cancer was made on the basis of multi-detector-row computed tomography, magnetic resonance imaging, and abdominal ultrasonography. In patients with obstructive jaundice, a percutaneous biliary drain was placed and cytologic examination of the bile juice was performed. Endoscopic biliary drainage, endoscopic ultrasonography, and intraductal ductal ultrasonography were not performed. On the basis of multiple preoperative imaging study results, 19 patients with the following conditions were considered as candidates for major HPD: (1) diffuse-type bile duct cancer ranging from the intrahepatic bile duct to the bifurcation of the hilar bile duct (n = 6); (2) hilar bile duct cancer with possible involvement of the intrapancreatic bile duct or with apparent nodal metastasis in the region of the pancreatic head (n = 7); (3) middle to lower bile duct cancer with involvement of the right hepatic artery, or a possible extension to the hilar bile duct (n = 2); and (4) gallbladder cancer with biliary involvement and pancreatic or duodenal infiltration (n = 4). Preoperative portal vein embolization. Preoperative portal vein embolization (PVE) was indicated if the estimated resectional volume exceeded 50% of the whole liver, taking into consideration the hepatic functional reserve or invasiveness of major HPD, as described elsewhere.11 In 17 patients scheduled to undergo extended right hemihepatectomy plus pancreaticoduodenectomy, right PVE was performed in 24 (range, 19 to 35) days before the surgery. Preoperative PVE was performed via the ileocolic vein (transileocolic portal vein embolization) under general anesthesia in 8 patients until 2002 and percutaneously (percutaneous transhepatic portal vein embolization) in 9 patients since 2003. Procedures of hepatopancreaticoduodenectomy. Details of the surgical procedures of extended hemihepatectomy plus HPD11 and pancreaticoduodenectomy12 in our institute have
Sakamoto et al 795
been described elsewhere. Extended right hemihepatectomy involves resection of the right hemiliver, the caudate lobe, and the inferior portion of segment 4. Extended left hemihepatectomy involves resection of the left hemiliver, the caudate lob, and the inferior portion of segment 5. In brief, pancreaticoduodenectomy is performed before hepatectomy during conventional major HPD. Lymph nodes around the common hepatic artery or superior mesenteric artery are dissected. In the extended hemihepatectomy that follows, the hepatic artery and portal vein are ligated and divided prior to hepatic transection. Portal vein resection and reconstruction are performed at the end of hepatic transection if necessary. The caudate lobe is removed in patients with bile duct cancer but is preserved in patients with gallbladder cancer. The bile duct is transected at a point as far from the tumor as possible at the end of hepatic transection. Frozen sections of the cut end of the bile duct are examined in all patients.13,14 If the hepatic-side ductal margin is positive for cancer, additional resection is performed if possible. When extended hemihepatectomy precedes pancreaticoduodenectomy, if the duodenal-side ductal margin is positive even after additional resection, then pancreaticoduodenectomy is also performed. Pancreaticojejunostomy is performed with 2-layer anastomosis. The stump of the gastroduodenal artery is wrapped using the falciform ligament to prevent massive bleeding from the stump in case of pancreatic fistula.15 Postoperatively, drains placed beside the pancreaticojejunostomy are suctioned intermittently during the first 3 to 5 days after surgery. Definition of surgical complications. Postoperative pancreatic fistula and bile leakage were defined according to the definition of international study group of liver surgery.16,17 Liver failure was defined also according to the international study group of liver failure.18 Other complications were classified according to the grading system of Dindo et al,19 and grade III or further complications were considered to be significant. Comparison of clinicopathological factors and survival in patients with bile duct cancer and gallbladder cancer. Clinicopathologic factors and survival were compared in patients with bile duct cancer and gallbladder cancer using the following categorized variables: operative period (1989–1999 vs 2000–2010); age (>65 years vs <65 years); gender; operative time (>14h vs <14h); blood loss (>2,000 mL vs <2,000 mL); disease (bile duct cancer vs gallbladder cancer); surgical margins; and disease stage (I–III vs IV). Each threshold
796 Sakamoto et al
value was determined on the basis of the median value of each category. Statistic analysis. Results are reported as the median and range unless otherwise specified. A parametric statistical analysis was performed using the chi-square test analysis or the Fisher exact test. The cumulative survival rates were generated using a Kaplan-Meier method, and the differences among the rates of the groups were assessed using the log-rank test. Statistical significance was defined as a P value of < .05. All statistical analyses were performed using software package SPSSII 19.0 (IBM, Chicago, IL). RESULTS Surgical outcomes. Right HPD was performed in 17 patients, and left HPD was performed in 2 patients. Of these, HPD was scheduled at the beginning of surgery in 12 patients (7 with bile duct cancer and 5 with gallbladder cancer), according to imaging or intraoperative findings, including palpation of the hepatoduodenal ligament in 4 patients with bile duct cancer. In the remaining 3 patients, HPD proceeded based on the findings in the ductal margins; in 2 patients, extended right hemihepatectomy was performed when the duodenal-side ductal margin was positive for carcinoma in situ in 1 patient and the ductal wall was positive in the other. The hepatic-side ductal margin was negative in both. Pancreaticoduodenectomy was performed in these 2 patients, and the final surgical margins were determined to be negative on the basis of pathologic examination. In 1 patient, pancreaticoduodenectomy was performed, and the hepatic-side ductal margin was positive for carcinoma in situ. In this patient, PVE had been performed before surgery, and an extended right hemihepatectomy was subsequently completed. The final ductal margins on the left hepatic duct were positive for carcinoma in situ in this patient. Portal vein resection was performed in 4 of the 14 patients (29%) with bile duct cancer and in 4 of 5 patients (80%) with gallbladder cancer (P = .046). The median operation time was 13.5 h (range, 8.2 to 22.7 h). The median blood loss was 2,300 mL (range, 900 to 7,760 mL). A postoperative pancreatic fistula was found in 18 patients--grade A in 1, grade B in 14, and grade C in 3. Grade B bile leakage was found in 4 patients (21%). Other complications included grade III infectious abscess in 5 patients, intestinal bleeding in 2, and wound infection in 1. According to the international definition, posthepatectomy liver
Surgery June 2013
failure was found in 14 patients (74%), but the bilirubin value decreased without any specific treatment in 1 patient (7%, grade A) or after conservative treatments, including administering fresh-frozen plasma in 12 of the 14 patients (86%, grade B). The remaining patient, who underwent right HPD after percutaneous transhepatic portal vein embolization, died of bleeding caused by a pancreatic fistula and subsequent hepatic failure on day 8 (7%, grade C). Thus, the mortality rate in the 19 patients included in the study was 5%, and the median hospital stay was 47 days (range, 8 to 390days). Comparison of clinicopathologic factors and survival between patients with bile duct cancer and gallbladder cancer. Table I shows a comparison of clinicopathologic factors in patients with bile duct cancer and gallbladder cancer. The radial margins were positive in 1 patient (7%) with bile duct cancer but in 4 (80%) with gallbladder cancer (P = .001). The final hepatic-side ductal margin was positive for carcinoma in situ in 4 patients with bile duct cancer. The T stage in TNM classification was T3 in 1 patient and T4 in 2 with bile duct cancer, but it was T4 in all 5 patients (100%) with gallbladder cancer (P = .002). The clinical stage was IV in 3 patients (21%) with bile duct cancer and IV in all 5 patients (100%) with gallbladder cancer (P = .002). All 5 patients with gallbladder cancer had microscopically biliary infiltration; hepatic metastases were found in the resected livers of 2 patients, and the para-aortic nodes were positive in 1 patient. The overall survival of patients undergoing HPD are shown in the Figure. The 5-year survival rates and median survival rates of patients with bile duct cancer and gallbladder cancer were 45% vs 0 and 3.3 years vs 8 months, respectively (P < .001). DISCUSSION The present results showed a remarkable contrast with regard to the prognoses of patients undergoing HPD for bile duct cancer and gallbladder cancer. In 14 patients with bile duct cancer, 11 (79%) had T1 or T2 tumors, and the final stage was I or II in 11 other patients. The radial margin was negative in 93% of these patients, and the overall 5-year survival rate was 45%. In contrast, in 5 patients with gallbladder cancer, all had T4 tumors, and the final stage was IV in all. Biliary infiltration was also positive in all 5 patients, and the radial margin was positive in 4 (80%). The 5-year survival rate was 0. This report details a
Sakamoto et al 797
Surgery Volume 153, Number 6
Table I. Comparison of clinicopathologic factors and survivals in 14 patients with bile duct cancer and 5 patients with gallbladder cancer undergoing extended hemihepatectomy plus pancreaticoduodenectomy between 1989 and 2009 Bile duct cancer (n = 14) Patient characteristics Age =/<65 >65 Gender Male Female Operative period 1989–1999 2000–2010 Operative parameters Resection side Right Left Portal vein resection Not performed Performed Operative time <14 h >/=14 h Blood loss <2,000 mL >/=2,000 mL Surgical complication POPF Absent or grade A Grade B or C Bile leakage Absent Present Surgical mortality Pathologic factors Differentiation pap, well mod, por, muc Lymphatic invasion Absent Present Venous invasion Absent Present Perineural invasion Absent Present Hepatic-side ductal margin Negative Positive
Gallbladder cancer (n = 5)
P value
7 7
4 1
.24
10 4
2 3
.21
5 9
3 2
.35
12 2
5 0
.37
10 4
1 4
.046*
7 7
3 2
.70
7 7
2 3
.70
2 12
0 5
.37
10 4 0
5 0 1 (20%)
.18
6 8
1 4
.36
3 11
0 5
.26
5 9
0 5
.12
2 12
0 5
.37
9 5
5 0
.12
Table I. (continued)
Radial margin Negative Positive Overall surgical margin Negative Positive T factor T1, T2 T3, T4 N factor N0 N1 M factor M0 M1 Stage I–III IV
Bile duct cancer (n = 14)
Gallbladder cancer (n = 5)
13 1
1 4
.001*
8 6
1 4
.15
11 3
0 5
.002*
5 9
0 5
.12
12 2
2 3
.046
11 3
0 5
.002*
P value
*P < .05. POPF, Postoperative pancreatic fistula; pap, papillary; mod, moderately; por, poorly; muc, mucinous.
.086
(continued)
Figure. Overall survival of patients undergoing extended hemihepatectomy plus pancreaticoduodenectomy for bile duct cancer and gallbladder cancer between 1989 and 2009. BD Ca, Bile duct cancer; GB Ca, gallbladder cancer.
series of patents from the past 20 years in a Japanese high-volume center, and no patients who experienced >2-year survival following HPD for gallbladder cancer were included in the present study. We have also reviewed the recent reports published by others after 2005 on outcomes of major HPD for bile duct cancer and
798 Sakamoto et al
Surgery June 2013
Table II. Reports on the short-term and long-term outcomes of hepatopancreaticoduodenectomy Year
Author
2006
Hirono
2007
Ebata
2007
Kaneoka
2007
Miwa
2008
Wakai
2010
Hemming
2012
Lim
2012 Ebata Present series
Disease BD GB BD GB BD GB BD GB BD GB BD GB BD GB BD BD GB
n 5* 1* 25 33 10 10 12 6 17 11 13** 9** 13 10 85** 14 5
Morbidity (%)
Mortality (%)
5-year survival rate (%)
P value
67*
33*
ND 64*
ND
ND
20
40 60 39
15
82
21
35
0
91
13
0
78 95
2.4 5.3
ND 64 0 52 33 12 9 24* 18* 32 10 54 53 0
<.001 ND .40 NS .09 ND <.001
*Excluding patients undergoing less than hemihepatectomy. **Including patients undergoing less than hemihepatectomy. BD, Bile duct cancer; GB, gallbladder cancer; ND, not determined; NS, not significant.
gallbladder cancer (Table II).5-10,20,21 Kaneoka reported results similar to those of our series, ie, a 20% 1-year survival rate without 2-year survivors among 10 patients undergoing HPD for gallbladder cancer. These series suggest that gallbladder cancer having biliary and peripancreatic region infiltration had already progressed to a systemic disease, and major HPD might be contraindicated in such patients. HPD carries high morbidity and mortality rates. The reported incidence of in-hospital mortality after HPD has been reported to be as high as 33%.1-4 The most troublesome complications following HPD are hepatic failure and pancreatic fistula. Preoperative PVE is effective in preventing postoperative hepatic failure. Ebata et al6 reported that the incidence of hepatic failure decreased from 56% to 14% after introduction of routine preoperative PVE. However, pancreatic or bile leakage and systemic infection can easily evoke secondary hepatic failure after HPD. The incidence of pancreatic fistula in the present series was very high (95%), and several reasons could be considered. As the multivariate analysis has clearly shown, small pancreatic duct size, soft gland texture, and increased operative blood loss greatly contributed to the increased risk for pancreatic fistulas.22 However, the present incidence appears higher than our previous report on pancreatic fistulas in patients with normal pancreas.23 It has also been reported that pancreatic fistulas occur more often
after HPD than after sole pancreaticoduodenectomy.21 Although the reason for an increase in pancreatic fistulas after HPD should be further evaluated, we suppose that the wide dissection area and dead space after removal of the hemiliver and the pancreatic head facilitated the extensive spread of pancreatic juice, which may have led to peritonitis and vast infection. In fact, in our series, 17 of the 19 patients (85%) developed grade B or C pancreatic fistulas, and 1 patient died of bleeding from the stump of the gastroduodenal artery. Although the incidence of pancreatic fistulas was very high, we wrapped the stump using the falciform ligament15 to prevent massive bleeding, and this technique was somewhat effective. Moreover, intermittent suction during the early postoperative period is very important to prevent the spread of pancreatic juice. A 2-stage pancreaticojejunostomy following complete drainage of pancreatic juice may be an alternative to prevent severe pancreatic fistulas, in which the pancreatic juice and jejunal contents will never be contaminated.4 In fact, Miwa et al8 achieved zero mortality in 52 patients undergoing HPD including major hepatectomy. We have previously reported the results in 35 patients with advanced gallbladder cancer undergoing extended right hemihepatectomy. Although the morbidity rate was 46%, there was no surgical mortality. The overall 5-year survival rate was 17%, with a median survival of 2.2 years, and 3 patients survived more than 5 years.24 Nishio et al25 also
Sakamoto et al 799
Surgery Volume 153, Number 6
recently reported a 5-year survival rate of 23% in 73 patients with pathologic biliary infiltration, and they concluded that gallbladder cancer involving the extrahepatic bile duct is worthy of resection. Miwa et al8 reported a 5-year survival rate of 33% in 6 patients undergoing HPD for gallbladder cancer. In contrast to these favorable results for advanced gallbladder cancer, the prognosis of the present 5 patients was dismal, because they already had stage IV cancer, ie, systemic disease. The differences among patients in the present case series and those undergoing extended right hemihepatectomy involved the necessity of performing pancreaticoduodenectomy. In the 5 patients with gallbladder cancer included in the present study, HPD was determined to be appropriate before surgery; however, 3 patients were found to have distant metastases. It is difficult to arrive at a definitive conclusion regarding the indication of HPD for advanced gallbladder cancer on the basis of the results to date. As such, optimal indication for HPD for gallbladder cancer should be further evaluated in a subgroup of patients with locally advanced gallbladder cancer. Previous reports have insisted that the benefit of performing HPD for bile duct cancer is based on the possibility of achieving clear ductal margins (R0). In 5 of the 14 patients (36%) with bile duct cancer, the hepatic-side ductal margin was positive for carcinoma in situ because of superficial spread of the tumor, but the prognoses for these patients were favorable. We previously reported that positive ductal margins in carcinoma in situ were not a poor prognostic factor, whereas positive radial margins excluding the ductal margins were an independent predictive factor for poor survival in patients with bile duct cancer.13,14 Igami et al26 also reported that in many patients with cholangiocarcinoma with carcinoma in situ, survival did not depend on complete resection of all of the superficial spread but depended on the stage of the main lesion. If a frozen section of the ductal margin is positive for carcinoma in situ, additional pancreaticoduodenectomy or major hemihepatectomy is not indicated, because carcinoma in situ has a very limited impact on the prognoses of patients.13,14 However, if the ductal margin is positive for invasive cancer after major hepatectomy, we first recommend additional resection of the lower bile duct in the pancreas. If the additional resection is still positive for invasive cancer, then pancreaticoduodenectomy can be performed. The present report is a retrospective analysis of only 19 patients undergoing major HPD in the past
20 years. The clinical stage of 5 patients with gallbladder cancer was very advanced. Therefore, the indication for major HPD in patients with gallbladder cancer should be further evaluated by additional studies. A multicenter large series would be necessary to comment conclusively on the indications of HPD for biliary cancer. In conclusion, HPD can be an acceptable treatment option for widespread bile duct cancer. However, the indication for HPD in advancedstage gallbladder cancer should be considered carefully, considering the high morbidity rate and advanced stage of the disease. Part of this paper was presented at the International Association of Surgeons, Gastroenterologists and Oncologists, Continuing Medical Education postgraduate course, in Tokyo, on April 15, 2012.
REFERENCES 1. Nimura Y, Hayakawa N, Kamiya J, Maeda S, Kondo S, Yasui A, et al. Hepatopancreatoduodenectomy for advanced carcinoma of the biliary tract. Hepatogastroenterology 1991; 38:170-5. 2. Tsukada K, Yoshida K, Aono T, Koyama S, Shirai Y, Uchida K, et al. Major hepatectomy and pancreatoduodenectomy for advanced carcinoma of the biliary tract. Br J Surg 1994;81:108-10. 3. Nakamura S, Nakamura S, Nishiyama R, Yokoi Y, Serizawa A, Nishiwaki Y, et al. Hepatopancreatoduodenectomy for advanced gallbladder carcinoma. Arch Surg 1994;129:625-9. 4. Miyagawa S, Makuuchi M, Kawasaki S, Hayashi K, Harada H, Kitamura H, et al. Outcome of major hepatectomy with pancreatoduodenectomy for advanced biliary malignancies. World J Surg 1996;20:77-80. 5. Hirono S, Tani M, Kawai M, Ina S, Uchiyama K, Yamue H. Indication of hepatopancreatoduodenectomy for biliary tract cancer. World J Surg 2006;30:567-73. 6. Ebata T, Nagino M, Nishio H, Arai T, Nimura Y. Right hepatopancreatoduodenectomy: improvements over 23 years to again acceptability. J Hepatobiliary Pancreat Surg 2007;14:131-5. 7. Keneoka Y, Yamaguchi A, Isogai M. Hepatopancreatoduodenectomy: its suitability for bile duct cancer versus gallbladder cancer. J Hepatobiliary Pancreat Surg 2007;14: 142-8. 8. Miwa S, Kobayashi A, Akahane Y, Nakata T, Mihara M, Kusama K, et al. Is major hepatectomy with pancreatoduodenectomy justified for advanced biliary malignancy? J Hepatobiliary Pancreat Surg 2007;14:136-41. 9. Wakai T, Shirai Y, Tsuchiya Y, Nomura T, Akazawa K, Hatakeyama K. Combined major hepatectomy and pancreaticoduodenectomy for locally advanced biliary carcinoma: long-term results. World J Surg 2008;32:1067-74. 10. Lim CS, Jang JY, Lee SE, Kang MJ, Kim SW. Reappraisal of hepatopancreatoduodenectomy as a treatment modality for bile duct and gallbladder cancer. J Gastrointest Surg 2012; 16:1012-8. 11. Sano T, Shimada K, Sakamoto Y, Yamamoto J, Yamasaki S, Kosuge T. One hundred two consecutive hepatobiliary resections for perihilar cholangiocarcinoma with zero mortality. Ann Surg 2006;244:240-7.
800 Sakamoto et al
12. Shimada K, Sano T, Sakamoto Y, Kosuge T. Clinical implications of combined portal vein resection as a palliative procedure in patients undergoing pancreaticoduodenectomy for pancreatic head carcinoma. Ann Surg Oncol 2006;13:1569-78. 13. Sakamoto Y, Kosuge T, Shimada K, Sano T, Ojima H, Yamamoto J, et al. Prognostic factors of surgical resection in middle and distal bile duct cancer: an analysis of 55 patients concerning the significance of ductal and radial margins. Surgery 2005;137:396-402. 14. Sakamoto Y, Shimada K, Nara S, Esaki M, Ojima H, Sano T, et al. Surgical management of infrahilar/suprapancreatic cholangiocarcinoma: an analysis of the surgical procedures, surgical margins, and survivals of 77 patients. J Gastrointest Surg 2010;14:335-43. 15. Sakamoto Y, Shimada K, Esaki M, Kajiwara T, Sano T, Kosuge T. Wrapping the stump of the gastroduodenal artery using the falciform ligament during pancreatoduodenectomy. J Am Coll Surg 2007;204:334-6. 16. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J, et al. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 2005;138:8-13. 17. Koch M, Garden OJ, Padbury R, Rahbari NN, Adam R, Capussotti L, et al. Bile leakage after hepatobiliary and pancreatic surgery: a definition and grading of severity by the International Study Group of Liver Surgery. Surgery 2011; 149:680-8. 18. Rahbari NN, Garden OJ, Padbury R, Brooke-Smith M, Crawford M, Adam R, et al. Posthepatectomy liver failure: a definition and grading by the International Study Group of Liver Surgery (ISGLS). Surgery 2011;149:713-24.
Surgery June 2013
19. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:250-3. 20. Hemming AW, Magliocca JF, Fujita S, Kayler LK, Hochwakd S, Zendejas I, et al. Combined resection of the liver and pancreas for malignancy. J Am Coll Surg 2010;210: 808-14. 21. Ebata T, Yokoyama T, Igami T, Sugawara G, Takahashi Y, Nimra Y, et al. Hepatopancreatoduodenectomy for cholangiocarcinoma: a single-center review of 85 consecutive patients. Ann Surg 2012;256:297-305. 22. Pratt WB, Callery MP, Vollmer CM Jr. Risk prediction for development of pancreatic fistula using the ISGPF classification. World J Surg 2008;32:419-28. 23. Kajiwara T, Sakamoto Y, Morofuji N, Nara S, Esaki M, Shimada K, et al. An analysis of risk factors for pancreatic fistula after pancreaticoduodenectomy: clinical impact of bile juice infection on day 1. Langenbeck Arch Surg 2010; 395:707-12. 24. Shimada K, Nara S, Esaki M, Sakamoto Y, Kosuge T, Hiraoka N. Extended right hemihepatectomy for gallbladder carcinoma involving the hepatic hilum. Br J Surg 2011;98: 117-23. 25. Nishio H, Ebata T, Yokoyama Y, Igami T, Sugawara G, Nagino M. Gallbladder cancer involving the extrahepatic bile duct is worthy of resection. Ann Surg 2011;253:953-60. 26. Igami T, Nagino M, Oda K, Nishio H, Ebata T, Yokoyama Y, et al. Clinicopathologic study of cholangiocarcinoma with superficial spread. Ann Surg 2009;249:296-302.
Background. Major hepatopancreaticoduodenectomy (HPD) is an extensive surgical procedure offering the highest curability for patients with advanced biliary cancer. However, surgical morbidity associated with major HPD is high, and optimal indications for this procedure remain unclear. Methods. Between 1989 and 2010, 14 patients with widespread bile duct cancer and 5 with gallbladder cancer having biliary infiltration underwent major HPD at our hospital. Preoperative portal vein embolization was performed in 17 patients undergoing right HPD. Clinicopathologic factors and survivals following HPD were compared between patients with bile duct cancer and those with gallbladder cancer. Results. One patient who underwent right HPD for gallbladder cancer died of hepatic failure (5.3%) and 18 of the 19 patients (95%) developed postoperative pancreatic fistulas. The median hospital stay was 47 days. Depth of invasion was T3 in 1 patient and T4 in 2 patients with bile duct cancer and was T4 in all 5 patients with gallbladder cancer (P = .002). The clinical stage was IV in 3 patients (21%) with bile duct cancer and in all 5 patients with gallbladder cancer (P = .002). The 5-year survival rates and median survival rates of patients with bile duct cancer and gallbladder cancer were 45% vs 0 and 3.3 years vs 8 months, respectively (P < .001). Conclusion. HPD can be an acceptable treatment option for widespread bile duct cancer. However, the indication for HPD in advanced-stage gallbladder cancer should be considered carefully, considering the high morbidity rate and the advanced stage of the disease. (Surgery 2013;153:794-800.) From the Hepato-Biliary-Pancreatic Surgery Division,a Department of Surgery, Graduate School of Medicine, The University of Tokyo, and the Hepatobiliary and Pancreatic Surgery Division,b National Cancer Center Hospital, Tokyo, Japan
SURGICAL RESECTION for advanced biliary cancer remains challenging. Patients with widespread bile duct cancer and advanced gallbladder cancer having biliary and peripancreatic infiltration are possible candidates for hepatopancreaticoduodenectomy (HPD). HPD, especially extended hemihepatectomy plus pancreaticoduodenectomy, offers the highest curability to obtain negative ductal or radial
Supported in part by a grant-in-aid for scientific research from the Ministry of Health and Welfare of Japan and the National Cancer Center Research and Development Fund. Accepted for publication November 28, 2012. Reprint requests: Yoshihiro Sakamoto, MD, Hepato-BiliaryPancreatic Surgery Division, Department of Surgery, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan. E-mail: [email protected]. 0039-6060/$ - see front matter Ó 2013 Mosby, Inc. All rights reserved. http://dx.doi.org/10.1016/j.surg.2012.11.024
794 SURGERY
margins for advanced biliary cancer.1-4 However, this procedure is associated with high morbidity and mortality rates. The mortality rate for HPD has been reported to be as high as 33%,5-10 and the major complications associated with this procedure include pancreatic fistula, hepatic failure, and other infectious complications. Therefore, surgical indications for major HPD should be evaluated carefully after considering the balance between the risks and the expected prognosis of the patient. We have performed 19 HPDs in the past 20 years for biliary malignancies. The objective of the present article is to discuss optimal indications for HPD in patients with advanced biliary cancer by reviewing our experiences and other reported case series. METHODS Between 1989 and 2010, 283 patients with bile duct cancer (excluding intrahepatic cholangiocarcinoma) and 204 patients with gallbladder cancer
Surgery Volume 153, Number 6
underwent surgical resection at the Hepatobiliary and Pancreatic Surgery Division, National Cancer Hospital, Tokyo. Of these, 14 patients with widespread bile duct cancer (4.9%) underwent right or left HPD, and 5 patients with gallbladder cancer having biliary infiltration (2.5%) underwent right HPD. One patient underwent right HPD on the preoperative diagnosis of diffuse-type bile duct cancer, but the postoperative pathologic examination revealed primary sclerosing cholangitis without evidence of malignancy. This patient was therefore excluded from the present series. Imaging studies for biliary cancer. A preoperative diagnosis of biliary cancer was made on the basis of multi-detector-row computed tomography, magnetic resonance imaging, and abdominal ultrasonography. In patients with obstructive jaundice, a percutaneous biliary drain was placed and cytologic examination of the bile juice was performed. Endoscopic biliary drainage, endoscopic ultrasonography, and intraductal ductal ultrasonography were not performed. On the basis of multiple preoperative imaging study results, 19 patients with the following conditions were considered as candidates for major HPD: (1) diffuse-type bile duct cancer ranging from the intrahepatic bile duct to the bifurcation of the hilar bile duct (n = 6); (2) hilar bile duct cancer with possible involvement of the intrapancreatic bile duct or with apparent nodal metastasis in the region of the pancreatic head (n = 7); (3) middle to lower bile duct cancer with involvement of the right hepatic artery, or a possible extension to the hilar bile duct (n = 2); and (4) gallbladder cancer with biliary involvement and pancreatic or duodenal infiltration (n = 4). Preoperative portal vein embolization. Preoperative portal vein embolization (PVE) was indicated if the estimated resectional volume exceeded 50% of the whole liver, taking into consideration the hepatic functional reserve or invasiveness of major HPD, as described elsewhere.11 In 17 patients scheduled to undergo extended right hemihepatectomy plus pancreaticoduodenectomy, right PVE was performed in 24 (range, 19 to 35) days before the surgery. Preoperative PVE was performed via the ileocolic vein (transileocolic portal vein embolization) under general anesthesia in 8 patients until 2002 and percutaneously (percutaneous transhepatic portal vein embolization) in 9 patients since 2003. Procedures of hepatopancreaticoduodenectomy. Details of the surgical procedures of extended hemihepatectomy plus HPD11 and pancreaticoduodenectomy12 in our institute have
Sakamoto et al 795
been described elsewhere. Extended right hemihepatectomy involves resection of the right hemiliver, the caudate lobe, and the inferior portion of segment 4. Extended left hemihepatectomy involves resection of the left hemiliver, the caudate lob, and the inferior portion of segment 5. In brief, pancreaticoduodenectomy is performed before hepatectomy during conventional major HPD. Lymph nodes around the common hepatic artery or superior mesenteric artery are dissected. In the extended hemihepatectomy that follows, the hepatic artery and portal vein are ligated and divided prior to hepatic transection. Portal vein resection and reconstruction are performed at the end of hepatic transection if necessary. The caudate lobe is removed in patients with bile duct cancer but is preserved in patients with gallbladder cancer. The bile duct is transected at a point as far from the tumor as possible at the end of hepatic transection. Frozen sections of the cut end of the bile duct are examined in all patients.13,14 If the hepatic-side ductal margin is positive for cancer, additional resection is performed if possible. When extended hemihepatectomy precedes pancreaticoduodenectomy, if the duodenal-side ductal margin is positive even after additional resection, then pancreaticoduodenectomy is also performed. Pancreaticojejunostomy is performed with 2-layer anastomosis. The stump of the gastroduodenal artery is wrapped using the falciform ligament to prevent massive bleeding from the stump in case of pancreatic fistula.15 Postoperatively, drains placed beside the pancreaticojejunostomy are suctioned intermittently during the first 3 to 5 days after surgery. Definition of surgical complications. Postoperative pancreatic fistula and bile leakage were defined according to the definition of international study group of liver surgery.16,17 Liver failure was defined also according to the international study group of liver failure.18 Other complications were classified according to the grading system of Dindo et al,19 and grade III or further complications were considered to be significant. Comparison of clinicopathological factors and survival in patients with bile duct cancer and gallbladder cancer. Clinicopathologic factors and survival were compared in patients with bile duct cancer and gallbladder cancer using the following categorized variables: operative period (1989–1999 vs 2000–2010); age (>65 years vs <65 years); gender; operative time (>14h vs <14h); blood loss (>2,000 mL vs <2,000 mL); disease (bile duct cancer vs gallbladder cancer); surgical margins; and disease stage (I–III vs IV). Each threshold
796 Sakamoto et al
value was determined on the basis of the median value of each category. Statistic analysis. Results are reported as the median and range unless otherwise specified. A parametric statistical analysis was performed using the chi-square test analysis or the Fisher exact test. The cumulative survival rates were generated using a Kaplan-Meier method, and the differences among the rates of the groups were assessed using the log-rank test. Statistical significance was defined as a P value of < .05. All statistical analyses were performed using software package SPSSII 19.0 (IBM, Chicago, IL). RESULTS Surgical outcomes. Right HPD was performed in 17 patients, and left HPD was performed in 2 patients. Of these, HPD was scheduled at the beginning of surgery in 12 patients (7 with bile duct cancer and 5 with gallbladder cancer), according to imaging or intraoperative findings, including palpation of the hepatoduodenal ligament in 4 patients with bile duct cancer. In the remaining 3 patients, HPD proceeded based on the findings in the ductal margins; in 2 patients, extended right hemihepatectomy was performed when the duodenal-side ductal margin was positive for carcinoma in situ in 1 patient and the ductal wall was positive in the other. The hepatic-side ductal margin was negative in both. Pancreaticoduodenectomy was performed in these 2 patients, and the final surgical margins were determined to be negative on the basis of pathologic examination. In 1 patient, pancreaticoduodenectomy was performed, and the hepatic-side ductal margin was positive for carcinoma in situ. In this patient, PVE had been performed before surgery, and an extended right hemihepatectomy was subsequently completed. The final ductal margins on the left hepatic duct were positive for carcinoma in situ in this patient. Portal vein resection was performed in 4 of the 14 patients (29%) with bile duct cancer and in 4 of 5 patients (80%) with gallbladder cancer (P = .046). The median operation time was 13.5 h (range, 8.2 to 22.7 h). The median blood loss was 2,300 mL (range, 900 to 7,760 mL). A postoperative pancreatic fistula was found in 18 patients--grade A in 1, grade B in 14, and grade C in 3. Grade B bile leakage was found in 4 patients (21%). Other complications included grade III infectious abscess in 5 patients, intestinal bleeding in 2, and wound infection in 1. According to the international definition, posthepatectomy liver
Surgery June 2013
failure was found in 14 patients (74%), but the bilirubin value decreased without any specific treatment in 1 patient (7%, grade A) or after conservative treatments, including administering fresh-frozen plasma in 12 of the 14 patients (86%, grade B). The remaining patient, who underwent right HPD after percutaneous transhepatic portal vein embolization, died of bleeding caused by a pancreatic fistula and subsequent hepatic failure on day 8 (7%, grade C). Thus, the mortality rate in the 19 patients included in the study was 5%, and the median hospital stay was 47 days (range, 8 to 390days). Comparison of clinicopathologic factors and survival between patients with bile duct cancer and gallbladder cancer. Table I shows a comparison of clinicopathologic factors in patients with bile duct cancer and gallbladder cancer. The radial margins were positive in 1 patient (7%) with bile duct cancer but in 4 (80%) with gallbladder cancer (P = .001). The final hepatic-side ductal margin was positive for carcinoma in situ in 4 patients with bile duct cancer. The T stage in TNM classification was T3 in 1 patient and T4 in 2 with bile duct cancer, but it was T4 in all 5 patients (100%) with gallbladder cancer (P = .002). The clinical stage was IV in 3 patients (21%) with bile duct cancer and IV in all 5 patients (100%) with gallbladder cancer (P = .002). All 5 patients with gallbladder cancer had microscopically biliary infiltration; hepatic metastases were found in the resected livers of 2 patients, and the para-aortic nodes were positive in 1 patient. The overall survival of patients undergoing HPD are shown in the Figure. The 5-year survival rates and median survival rates of patients with bile duct cancer and gallbladder cancer were 45% vs 0 and 3.3 years vs 8 months, respectively (P < .001). DISCUSSION The present results showed a remarkable contrast with regard to the prognoses of patients undergoing HPD for bile duct cancer and gallbladder cancer. In 14 patients with bile duct cancer, 11 (79%) had T1 or T2 tumors, and the final stage was I or II in 11 other patients. The radial margin was negative in 93% of these patients, and the overall 5-year survival rate was 45%. In contrast, in 5 patients with gallbladder cancer, all had T4 tumors, and the final stage was IV in all. Biliary infiltration was also positive in all 5 patients, and the radial margin was positive in 4 (80%). The 5-year survival rate was 0. This report details a
Sakamoto et al 797
Surgery Volume 153, Number 6
Table I. Comparison of clinicopathologic factors and survivals in 14 patients with bile duct cancer and 5 patients with gallbladder cancer undergoing extended hemihepatectomy plus pancreaticoduodenectomy between 1989 and 2009 Bile duct cancer (n = 14) Patient characteristics Age =/<65 >65 Gender Male Female Operative period 1989–1999 2000–2010 Operative parameters Resection side Right Left Portal vein resection Not performed Performed Operative time <14 h >/=14 h Blood loss <2,000 mL >/=2,000 mL Surgical complication POPF Absent or grade A Grade B or C Bile leakage Absent Present Surgical mortality Pathologic factors Differentiation pap, well mod, por, muc Lymphatic invasion Absent Present Venous invasion Absent Present Perineural invasion Absent Present Hepatic-side ductal margin Negative Positive
Gallbladder cancer (n = 5)
P value
7 7
4 1
.24
10 4
2 3
.21
5 9
3 2
.35
12 2
5 0
.37
10 4
1 4
.046*
7 7
3 2
.70
7 7
2 3
.70
2 12
0 5
.37
10 4 0
5 0 1 (20%)
.18
6 8
1 4
.36
3 11
0 5
.26
5 9
0 5
.12
2 12
0 5
.37
9 5
5 0
.12
Table I. (continued)
Radial margin Negative Positive Overall surgical margin Negative Positive T factor T1, T2 T3, T4 N factor N0 N1 M factor M0 M1 Stage I–III IV
Bile duct cancer (n = 14)
Gallbladder cancer (n = 5)
13 1
1 4
.001*
8 6
1 4
.15
11 3
0 5
.002*
5 9
0 5
.12
12 2
2 3
.046
11 3
0 5
.002*
P value
*P < .05. POPF, Postoperative pancreatic fistula; pap, papillary; mod, moderately; por, poorly; muc, mucinous.
.086
(continued)
Figure. Overall survival of patients undergoing extended hemihepatectomy plus pancreaticoduodenectomy for bile duct cancer and gallbladder cancer between 1989 and 2009. BD Ca, Bile duct cancer; GB Ca, gallbladder cancer.
series of patents from the past 20 years in a Japanese high-volume center, and no patients who experienced >2-year survival following HPD for gallbladder cancer were included in the present study. We have also reviewed the recent reports published by others after 2005 on outcomes of major HPD for bile duct cancer and
798 Sakamoto et al
Surgery June 2013
Table II. Reports on the short-term and long-term outcomes of hepatopancreaticoduodenectomy Year
Author
2006
Hirono
2007
Ebata
2007
Kaneoka
2007
Miwa
2008
Wakai
2010
Hemming
2012
Lim
2012 Ebata Present series
Disease BD GB BD GB BD GB BD GB BD GB BD GB BD GB BD BD GB
n 5* 1* 25 33 10 10 12 6 17 11 13** 9** 13 10 85** 14 5
Morbidity (%)
Mortality (%)
5-year survival rate (%)
P value
67*
33*
ND 64*
ND
ND
20
40 60 39
15
82
21
35
0
91
13
0
78 95
2.4 5.3
ND 64 0 52 33 12 9 24* 18* 32 10 54 53 0
<.001 ND .40 NS .09 ND <.001
*Excluding patients undergoing less than hemihepatectomy. **Including patients undergoing less than hemihepatectomy. BD, Bile duct cancer; GB, gallbladder cancer; ND, not determined; NS, not significant.
gallbladder cancer (Table II).5-10,20,21 Kaneoka reported results similar to those of our series, ie, a 20% 1-year survival rate without 2-year survivors among 10 patients undergoing HPD for gallbladder cancer. These series suggest that gallbladder cancer having biliary and peripancreatic region infiltration had already progressed to a systemic disease, and major HPD might be contraindicated in such patients. HPD carries high morbidity and mortality rates. The reported incidence of in-hospital mortality after HPD has been reported to be as high as 33%.1-4 The most troublesome complications following HPD are hepatic failure and pancreatic fistula. Preoperative PVE is effective in preventing postoperative hepatic failure. Ebata et al6 reported that the incidence of hepatic failure decreased from 56% to 14% after introduction of routine preoperative PVE. However, pancreatic or bile leakage and systemic infection can easily evoke secondary hepatic failure after HPD. The incidence of pancreatic fistula in the present series was very high (95%), and several reasons could be considered. As the multivariate analysis has clearly shown, small pancreatic duct size, soft gland texture, and increased operative blood loss greatly contributed to the increased risk for pancreatic fistulas.22 However, the present incidence appears higher than our previous report on pancreatic fistulas in patients with normal pancreas.23 It has also been reported that pancreatic fistulas occur more often
after HPD than after sole pancreaticoduodenectomy.21 Although the reason for an increase in pancreatic fistulas after HPD should be further evaluated, we suppose that the wide dissection area and dead space after removal of the hemiliver and the pancreatic head facilitated the extensive spread of pancreatic juice, which may have led to peritonitis and vast infection. In fact, in our series, 17 of the 19 patients (85%) developed grade B or C pancreatic fistulas, and 1 patient died of bleeding from the stump of the gastroduodenal artery. Although the incidence of pancreatic fistulas was very high, we wrapped the stump using the falciform ligament15 to prevent massive bleeding, and this technique was somewhat effective. Moreover, intermittent suction during the early postoperative period is very important to prevent the spread of pancreatic juice. A 2-stage pancreaticojejunostomy following complete drainage of pancreatic juice may be an alternative to prevent severe pancreatic fistulas, in which the pancreatic juice and jejunal contents will never be contaminated.4 In fact, Miwa et al8 achieved zero mortality in 52 patients undergoing HPD including major hepatectomy. We have previously reported the results in 35 patients with advanced gallbladder cancer undergoing extended right hemihepatectomy. Although the morbidity rate was 46%, there was no surgical mortality. The overall 5-year survival rate was 17%, with a median survival of 2.2 years, and 3 patients survived more than 5 years.24 Nishio et al25 also
Sakamoto et al 799
Surgery Volume 153, Number 6
recently reported a 5-year survival rate of 23% in 73 patients with pathologic biliary infiltration, and they concluded that gallbladder cancer involving the extrahepatic bile duct is worthy of resection. Miwa et al8 reported a 5-year survival rate of 33% in 6 patients undergoing HPD for gallbladder cancer. In contrast to these favorable results for advanced gallbladder cancer, the prognosis of the present 5 patients was dismal, because they already had stage IV cancer, ie, systemic disease. The differences among patients in the present case series and those undergoing extended right hemihepatectomy involved the necessity of performing pancreaticoduodenectomy. In the 5 patients with gallbladder cancer included in the present study, HPD was determined to be appropriate before surgery; however, 3 patients were found to have distant metastases. It is difficult to arrive at a definitive conclusion regarding the indication of HPD for advanced gallbladder cancer on the basis of the results to date. As such, optimal indication for HPD for gallbladder cancer should be further evaluated in a subgroup of patients with locally advanced gallbladder cancer. Previous reports have insisted that the benefit of performing HPD for bile duct cancer is based on the possibility of achieving clear ductal margins (R0). In 5 of the 14 patients (36%) with bile duct cancer, the hepatic-side ductal margin was positive for carcinoma in situ because of superficial spread of the tumor, but the prognoses for these patients were favorable. We previously reported that positive ductal margins in carcinoma in situ were not a poor prognostic factor, whereas positive radial margins excluding the ductal margins were an independent predictive factor for poor survival in patients with bile duct cancer.13,14 Igami et al26 also reported that in many patients with cholangiocarcinoma with carcinoma in situ, survival did not depend on complete resection of all of the superficial spread but depended on the stage of the main lesion. If a frozen section of the ductal margin is positive for carcinoma in situ, additional pancreaticoduodenectomy or major hemihepatectomy is not indicated, because carcinoma in situ has a very limited impact on the prognoses of patients.13,14 However, if the ductal margin is positive for invasive cancer after major hepatectomy, we first recommend additional resection of the lower bile duct in the pancreas. If the additional resection is still positive for invasive cancer, then pancreaticoduodenectomy can be performed. The present report is a retrospective analysis of only 19 patients undergoing major HPD in the past
20 years. The clinical stage of 5 patients with gallbladder cancer was very advanced. Therefore, the indication for major HPD in patients with gallbladder cancer should be further evaluated by additional studies. A multicenter large series would be necessary to comment conclusively on the indications of HPD for biliary cancer. In conclusion, HPD can be an acceptable treatment option for widespread bile duct cancer. However, the indication for HPD in advancedstage gallbladder cancer should be considered carefully, considering the high morbidity rate and advanced stage of the disease. Part of this paper was presented at the International Association of Surgeons, Gastroenterologists and Oncologists, Continuing Medical Education postgraduate course, in Tokyo, on April 15, 2012.
REFERENCES 1. Nimura Y, Hayakawa N, Kamiya J, Maeda S, Kondo S, Yasui A, et al. Hepatopancreatoduodenectomy for advanced carcinoma of the biliary tract. Hepatogastroenterology 1991; 38:170-5. 2. Tsukada K, Yoshida K, Aono T, Koyama S, Shirai Y, Uchida K, et al. Major hepatectomy and pancreatoduodenectomy for advanced carcinoma of the biliary tract. Br J Surg 1994;81:108-10. 3. Nakamura S, Nakamura S, Nishiyama R, Yokoi Y, Serizawa A, Nishiwaki Y, et al. Hepatopancreatoduodenectomy for advanced gallbladder carcinoma. Arch Surg 1994;129:625-9. 4. Miyagawa S, Makuuchi M, Kawasaki S, Hayashi K, Harada H, Kitamura H, et al. Outcome of major hepatectomy with pancreatoduodenectomy for advanced biliary malignancies. World J Surg 1996;20:77-80. 5. Hirono S, Tani M, Kawai M, Ina S, Uchiyama K, Yamue H. Indication of hepatopancreatoduodenectomy for biliary tract cancer. World J Surg 2006;30:567-73. 6. Ebata T, Nagino M, Nishio H, Arai T, Nimura Y. Right hepatopancreatoduodenectomy: improvements over 23 years to again acceptability. J Hepatobiliary Pancreat Surg 2007;14:131-5. 7. Keneoka Y, Yamaguchi A, Isogai M. Hepatopancreatoduodenectomy: its suitability for bile duct cancer versus gallbladder cancer. J Hepatobiliary Pancreat Surg 2007;14: 142-8. 8. Miwa S, Kobayashi A, Akahane Y, Nakata T, Mihara M, Kusama K, et al. Is major hepatectomy with pancreatoduodenectomy justified for advanced biliary malignancy? J Hepatobiliary Pancreat Surg 2007;14:136-41. 9. Wakai T, Shirai Y, Tsuchiya Y, Nomura T, Akazawa K, Hatakeyama K. Combined major hepatectomy and pancreaticoduodenectomy for locally advanced biliary carcinoma: long-term results. World J Surg 2008;32:1067-74. 10. Lim CS, Jang JY, Lee SE, Kang MJ, Kim SW. Reappraisal of hepatopancreatoduodenectomy as a treatment modality for bile duct and gallbladder cancer. J Gastrointest Surg 2012; 16:1012-8. 11. Sano T, Shimada K, Sakamoto Y, Yamamoto J, Yamasaki S, Kosuge T. One hundred two consecutive hepatobiliary resections for perihilar cholangiocarcinoma with zero mortality. Ann Surg 2006;244:240-7.
800 Sakamoto et al
12. Shimada K, Sano T, Sakamoto Y, Kosuge T. Clinical implications of combined portal vein resection as a palliative procedure in patients undergoing pancreaticoduodenectomy for pancreatic head carcinoma. Ann Surg Oncol 2006;13:1569-78. 13. Sakamoto Y, Kosuge T, Shimada K, Sano T, Ojima H, Yamamoto J, et al. Prognostic factors of surgical resection in middle and distal bile duct cancer: an analysis of 55 patients concerning the significance of ductal and radial margins. Surgery 2005;137:396-402. 14. Sakamoto Y, Shimada K, Nara S, Esaki M, Ojima H, Sano T, et al. Surgical management of infrahilar/suprapancreatic cholangiocarcinoma: an analysis of the surgical procedures, surgical margins, and survivals of 77 patients. J Gastrointest Surg 2010;14:335-43. 15. Sakamoto Y, Shimada K, Esaki M, Kajiwara T, Sano T, Kosuge T. Wrapping the stump of the gastroduodenal artery using the falciform ligament during pancreatoduodenectomy. J Am Coll Surg 2007;204:334-6. 16. Bassi C, Dervenis C, Butturini G, Fingerhut A, Yeo C, Izbicki J, et al. Postoperative pancreatic fistula: an international study group (ISGPF) definition. Surgery 2005;138:8-13. 17. Koch M, Garden OJ, Padbury R, Rahbari NN, Adam R, Capussotti L, et al. Bile leakage after hepatobiliary and pancreatic surgery: a definition and grading of severity by the International Study Group of Liver Surgery. Surgery 2011; 149:680-8. 18. Rahbari NN, Garden OJ, Padbury R, Brooke-Smith M, Crawford M, Adam R, et al. Posthepatectomy liver failure: a definition and grading by the International Study Group of Liver Surgery (ISGLS). Surgery 2011;149:713-24.
Surgery June 2013
19. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240:250-3. 20. Hemming AW, Magliocca JF, Fujita S, Kayler LK, Hochwakd S, Zendejas I, et al. Combined resection of the liver and pancreas for malignancy. J Am Coll Surg 2010;210: 808-14. 21. Ebata T, Yokoyama T, Igami T, Sugawara G, Takahashi Y, Nimra Y, et al. Hepatopancreatoduodenectomy for cholangiocarcinoma: a single-center review of 85 consecutive patients. Ann Surg 2012;256:297-305. 22. Pratt WB, Callery MP, Vollmer CM Jr. Risk prediction for development of pancreatic fistula using the ISGPF classification. World J Surg 2008;32:419-28. 23. Kajiwara T, Sakamoto Y, Morofuji N, Nara S, Esaki M, Shimada K, et al. An analysis of risk factors for pancreatic fistula after pancreaticoduodenectomy: clinical impact of bile juice infection on day 1. Langenbeck Arch Surg 2010; 395:707-12. 24. Shimada K, Nara S, Esaki M, Sakamoto Y, Kosuge T, Hiraoka N. Extended right hemihepatectomy for gallbladder carcinoma involving the hepatic hilum. Br J Surg 2011;98: 117-23. 25. Nishio H, Ebata T, Yokoyama Y, Igami T, Sugawara G, Nagino M. Gallbladder cancer involving the extrahepatic bile duct is worthy of resection. Ann Surg 2011;253:953-60. 26. Igami T, Nagino M, Oda K, Nishio H, Ebata T, Yokoyama Y, et al. Clinicopathologic study of cholangiocarcinoma with superficial spread. Ann Surg 2009;249:296-302.