Aggressive surgical approaches to hilar cholangiocarcinoma_ hepatic or local resection_

S URGICAL TREATMENT FOR HILAR cholangiocarcino-ma has been reported as having a low resectability rate and poor outcome.1-3Recently, various inves-tigators4-8have reported that aggressive surgical approaches with extended hepatectomy brought about a better prognosis in some patients. However, most patients with hilar cholangiocarci-noma suffered from obstructive jaundice. In such patients, major hepatic resection could result in postoperative liver failure, which would usually have a fatal outcome.9,10Therefore, to avoid the occurrence of postoperative liver failure in patients with biliary tract malignancy involving hepatic hilus, new strategies have been developed. Preoperative portal embolization of the hepatic lobe to be resected is one such strategy.11Another approach used to avoid the occurrence of postop-erative liver failure is a minimal hepatic resection preserving as much hepatic mass as possible.6,8 For the selection of sites for minimal hepatic resection, precise diagnosis of cancer involvement is requi-site by intrahepatic cholangiogram through percu-taneous transhepatic biliary drainage catheter. However, most hilar cholangiocarcinomas have been demonstrated to have extensive extramucos-al extension.12Therefore major hepatic resection is often required for curative resection to obtain a negative surgical margin at the stump of the intra-Aggressive surgical approaches to hilarcholangiocarcinoma: Hepatic or localresection?Masaru Miyazaki, MD, Hiroshi Ito, MD, Koji Nakagawa, MD, Satoshi Ambiru, MD, Hiroaki Shimizu,MD, Yoshiaki Shimizu, MD, Atsushi Kato, MD, Shunta Nakamura, MD, Hideyuki Omoto, MD,Nobuyuki Nakajima, MD, Fumio Kimura, MD, and Toshikazu Suwa, MD, Chiba and Saitama, JapanBackground.It has been reported that surgical excision of hilar cholangiocarcinoma rather than pallia-tive surgical therapy, chemotherapy, or radiotherapy caused prolonged survival in some patients.However, excision is associated with high operative morbidity and mortality rates, particularly whenhepatic resection is also performed. The aim of this study was to evaluate the clinical implications ofhepatic resection in hilar cholangiocarcinoma.Methods.The study involved 76 patients with hilar cholangiocarcinoma who were undergoing surgicalresections. Twenty-one patients (28%) underwent a combined resection, with reconstruction of the portalvein in 20 patients and reconstruction of the hepatic artery in 7 patients. Sixty-five patients undergoingseven different types of hepatic resection with extrahepatic bile duct resection (BDR) and 11 patientsundergoing BDR only were retrospectively compared for background, operative morbidity and mortality,and survival.Results.Curative resection was obtained in 5 of 11 (45%) patients undergoing local resection and in49 of 65 (75%) patients undergoing hepatic resection (p < 0.05). The surgical morbidity rates were34% and 27% for hepatic and local resection, respectively. The 30-day mortality and hospital mortalityrates were 4.6% and 15% for hepatic resection and 0% and 0% for local resection, respectively. The 5-year survival rate was 26% for all resected patients (76 patients); it was 40% versus 0% for curativeversus noncurative resections (p < 0.05). No significant difference in surgical resection rates wasrevealed between hepatic and local resection among resected and curative resected patients.Conclusions.Aggressive surgical approaches to obtain curative resections could bring about a better prog-nosis in hilar cholangiocarcinoma independently of whether hepatic resection or local resection is per-formed. (Surgery 1998;123:131-6.)From the First Department of Surgery, School of Medicine, Chiba University, Chiba, Japan, and the Division ofSurgery, Ohmiya-Red Cross Hospital, Saitama, JapanSupported in part by The Inohana Foundation in ChibaUniversity.Accepted for publication May 20, 1997.Reprint requests: Masaru Miyazaki, MD, First Department ofSurgery, School of Medicine, Chiba University, 1-8-1 Inohana,Chuoh, Chiba 260, Japan.Copyright © 1998 by Mosby, Inc.0039-6060/98/$5.00 + 011/56/85440SURGERY 131hepatic bile duct.5,13Surgical morbidity and mor-tality rates have been high after combined hepatic resection for hilar cholangiocarcinoma, and it is not yet clear whether better outcomes may be achieved. We have taken aggressive surgical approaches with appropriately limited hepatic resection and have combined vascular resection and reconstruction, if required, for the treatment of hilar cholangiocarcinoma.In this study we evaluated the clinical implica-tions of hepatic resection in surgical excision for hilar cholangiocarcinoma.PATIENTS AND METHODSPatients. Between January 1981 and December 1995, 120 patients with hilar cholangiocarciono-mas were admitted to our institution. Seventy-six of those patients underwent surgical excision and are the subject of this report. There were 46 men and 30 women, 40 to 84 years of age (median, 64 years).Preoperative clinical assessment and a laborato-ry and imaging workup, generally including ultra-sonography, cholangiography by the percutaneous transhepatic approach and by endoscopic retro-grade approach, selective celiac and hepatic angiography, and enhanced computed tomogra-phy, were performed to establish the nature and extent of the disease and to define ductal and vas-cular anatomic detail. Bile cytology and endoscop-ic biopsy by percutaneous transhepatic approach were performed in some patients for evaluation of tumor extent and histologic confirmation. The yield of these investigations and the criteria used to evaluate the potential resectability of the lesions are as follows.The operative procedures were selected so as to limit resection as much as possible to what was nec-essary for curative resection, according to the pre-operative and intraoperative evaluation of cancer.Tumor involvement of the portal vein and the hepatic artery of the liver remaining after hepatic resection could be excised by combined vascular resection and reconstruction in 21 patients (28%).All resected patients were free from peritoneal dis-semination, multiple hepatic metastases, and dis-tant metastases. All resected specimens were histo-logically confirmed to be hilar cholangiocarcinomas.No patients were given chemotherapy or radiother-apy before or after operation.Statistical analysis. Comparison between hepa-tectomy and local resection patient groups accord-ing to background, curability, surgical morbidity and mortality rates, and recurrent patterns were analyzed by the chi-squared test. Survival was com-puted and plotted according to the Kaplan-Meier method. Statistical analysis of survival rates was made by the generalized Wilcoxon test. A p value less than 0.05 was considered significant.RESULTSSurgical treatments. The operative procedures performed are shown in Table I. In 11 patients the tumor was excised by local resection encompassing132Miyazaki et al.Surgery February 1998Fig. 1.Survival after surgical resection in hilar cholangio-carcinoma. All resected patients undergoing curative and noncurative resections are included.Fig. 2. Survival after curative and noncurative resections in hilar cholangiocarcinoma. Survival rate of curative resec-tions was significantly higher than that of noncurative resections (p < 0.05).removal of the extrahepatic bile ducts and gall-bladder. The other 65 patients underwent some kind of hepatic resection for en bloc resection of the extrahepatic bile ducts together with the affect-ed hepatic segments or lobes. In two patients a pan-creaticoduodenectomy was needed to obtain a neg-ative surgical margin at the bile duct stump of the duodenal side. Seven kinds of hepatic resection were performed among the 65 patients of the hepatectomy group to match the appropriate type of resection to each patient. Portal vein resection and reconstruction were done in 20 patients and hepatic artery resection and reconstruction were also done in 7 patients. There were no significant differences as to age, gender, maximum tumor size, nodal involvement, depth of cancer invasion, oper-ation time, operative blood loss, or combined vas-cular resection between patients undergoing hepatic resection and those undergoing local resection (Table II). Histologic curative resections were done in 54 patients (71%) who underwent surgical resection, and noncurative resections of a positive surgical margin were done in 22 patients (29%). The histologic curability rate was signifi-cantly higher in the group of hepatectomy patients (75%) than in the group of nonhepatectomy patients (45%) (p< 0.05; Table III).Surgical morbidity and mortality. Surgical mor-bidity rates were 34% and 27% in the hepatectomy and local resection groups, respectively (Table IV).In the hepatectomy group the breakdown of hepaticojejunostomy occurred in 14 patients, hyperbilirubinemia in 12, rupture of a pseudo-aneurysm in 5, sepsis in 2, pleural effusion in 2, pneumonia in 2, and hepatic abscess in 1. In the local resection group pneumonia occurred in two patients and pleural effusion in one. Surgical mor-tality rates were 15% and 0% in the hepatectomy and local resection groups, respectively. There were no significant differences in the surgical mor-tality and morbidity rates between the two groups.Prognosis. Survival rates for all 76 resected patients including curative and noncurative resec-tions were 57%, 39%, 31%, 31%, and 26% at 1, 2, 3, 4, and 5 years, respectively (Fig. 1).Survival rates for the 54 curative resections were remarkably better than for the 22 noncurative resections: 62%, 48%, 40%, 40%, and 40% versus 44%, 8%, 8%, 8%, and 0% at 1, 2, 3, 4, and 5 years, respectively (p< 0.05; Fig. 2). In all 76 resected patients, survival rates for the hepatectomy group (n= 65) and the local resection group (n= 11) were 56%, 42%, 33%, 33%, and 27% versus 63%, 16%, 16%, —, and — at 1, 2, 3, 4, and 5 years, respectively. No significant difference was found between the two groups (Fig. 3). In 54 curative resections, survival rates for the hepatectomy and local resection groups were 60%, 49%, 37%, 37%, and 37% versus 100%, 100%, 100%, —, and — at 1, 2, 3, 4, and 5 years, respectively. No significantSurgery Miyazaki et al.133 Volume 123, Number 2difference was found between the two groups (Fig.4). On the contrary, none of the unresected patients survived more than 2 years after diagnosis (15% at 1 year).Recurrence rates were 39% and 20%, respec-tively, in the hepatectomy and local resection groups after curative resection (Table V). Local lymph node metastasis was the most frequent pat-tern of recurrence (12 patients, 24%). Recurrent patterns were peritonitis carcinomatosa in five,stump recurrence of the bile duct in three, hepatic metastasis in one, and distant metastasis in one.DISCUSSIONHilar cholangiocarcinoma is usually not diag-nosed until the bile duct is occluded and the patient is jaundiced.1,14Even though the lesion may remain small, its location and close proximity to important structures in the hepatic hilum make curative resection technically difficult and often impossible.9,15However, surgical resection for hilar cholangiocarcinoma is now generally regarded as the best possible treatment.4-8For purposes of improving the outlook for survival, we have aggres-sively performed surgical resection for hilar cholangiocarcinoma without any chemotherapy or radiotherapy, using variable types of hepatic resec-tion minimally required for curative resection.These were selected according to the cancer extent in each patient. In this study we evaluated the clin-ical implications of hepatic resection in surgical resection for hilar cholangiocarcinoma.Baer et al.7and Pinson and Rossi 16reported that no significant differences of survival were demon-strated between patients treated by hepatic resec-tion and those treated by local resection. On the contrary, Tsuzuki et al.5advocated extensive resec-tion of the bile ducts combined with resection of the liver to obtain long-term survival. Furthermore,Nimura et al.6and Childs and Hart 8advocated that curative resection should be designed according to the preoperative findings of the extent of cancer in each segmental duct, and that hepatic resection should be performed in the smallest hepatic seg-mentectomy possible.Our study showed that the surgical curability134Miyazaki et al.Surgery February 1998Fig. 4.Survival after surgical resection with hepatic resec-tion or local resection in patients with curative resection in hilar cholangiocarcinoma. No significant difference wasfound between the two groups.Fig. 3.Survival after surgical resection with hepatic resec-tion or local resection in hilar cholangiocarcinoma. All resected patients undergoing curative and noncurative resections are included. No significant difference was found between hepatic and local resections.rate, meaning the free surgical margin, was remark-ably high in patients undergoing hepatic resection (75%) as compared with that in patients undergo-ing local resection (45%) (p< 0.05). However, patients were not randomized but were selected in this study. Furthermore, no significant improve-ment of survival was obtained by hepatic resection among all resected patients, including both cura-tive and noncurative resections. In the case of cura-tive resections, survival in patients undergoing hepatic resection (37% at 5 years) was also not sig-nificantly different from survival in patients under-going local resection (100% at 3.9 years). In this series the 54 patients undergoing curative resec-tion had a remarkably better survival rate of 40% at 5 years than the 22 patients undergoing noncura-tive resection (8% at 4 years, 0% at 5 years) (p< 0.05). It has also been revealed by other investiga-tors13,16that improved survival rates can be achieved if histologically free surgical margins are obtained. It has been recognized that in hilar cholangiocarcinoma blood-borne metastases are distinctly unusual and, although nodal metastases may occur in as much as one third of the instances, the tumors spread more frequently by local exten-sion.2Therefore from this evidence, including our results, achievement of curative resection with min-imal morbidity and mortality rates is the most effec-tive therapeutic strategy.Although a high surgical curability rate (75%) was obtained in patients undergoing hepatic resec-tion in this study, it brought about high surgical morbidity and mortality rates (34% and 15%) as compared with local resection (27%, 0%), but there were no significant differences between the two groups. Blumgart et al.9similarly reported that hepatic resection in patients with jaundice is more hazardous than resection for primary or metastatic parenchymal tumors and in some patients a period of biliary drainage should precede definitive surgery, especially if a major operation is to be car-ried out in elderly high-risk patients. All patients in this study received preoperative biliary drainage until the serum total bilirubin level was reduced to less than 5 mg/dl. To reduce the surgical mortality rate after hepatic resection in patients with obstruc-tive jaundice, other prophylactic modalities such as partial portal embolization11and more limited hepatectomy6,8might be required. At present, hepatic resection for hilar cholangiocarcinoma seems to improve the outlook for survival at the cost of a high surgical mortality rate.In this study we achieved a high curability rate of 71% (54 of 76 patients) with a relatively justified surgical mortality rate of 13% (10 of 76 patients). This high operative curability might be dependent on the selection of appropriate operative proce-dures. In any kind of hepatic resection we under-went a combined resection of the caudate lobe (S1). The importance of caudate lobe resection in curative resection has been reported previously.5,6 The intrahepatic branch of the bile duct into the caudate lobe is usually made up of branches initial-ly originating bilaterally from the first intrahepatic bile duct.6Therefore caudate segmentectomy might often be required in combination with local resection at first, depending on the extent of the cancer into the intrahepatic bile duct. However, this study did not clarify whether caudate lobe resection contributed to the improvement of the surgical curability rate in patients undergoing hepatic resection.For unresectable hilar cholangiocarcinoma, internal biliary drainage, in conjunction with radio-therapy, was reported to be safe and effective palli-ation.17Percutaneous stenting or operative shunt-ing for internal biliary drainage has been reported as a beneficial palliative treatment.18However, sat-isfactory prognoses were not obtained by these pal-liative treatments. Furthermore, O’Grady et al.19 showed that liver transplantation for hilar cholan-giocarcinoma could not obtain a favorable progno-sis, similar to the result achieved by Iwatsuki et al.20 In conclusion, aggressive surgical resection to obtain histologically curative resection could bring about a better prognosis in hilar cholangiocarcino-ma without any chemotherapy or radiotherapy, independent of whether hepatic resection or local resection is performed. The operative procedures, limited as much as possible to what is necessary forSurgery Miyazaki et al.135 Volume 123, Number 2curative resection, should be performed on hepat-ic segments chosen according to preoperative and intraoperative evaluations of cancer extent in each patient to avoid the occurrence of postoperative liver failure.REFERENCES1.Bismuth H, Castaing D, Traynor O. Resection or palliation:priority of surgery in the treatment of hilar carcinoma.World J Surg 1988;12:39-47.2.Stain SC, Baer HU, Dennison AR, Blumgart LH. Currentmanagement of hilar cholangiocarcinoma. Surg Gynecol Obstet 1992;175:579-88.3.Yeo CJ, Pitt HA, Cameron JL. Cholangiocarcinoma. SurgClin North Am 1990;70:1429-47.nger JC, Langer B, Taylor BR, Zeldin R, Cummings B.Carcinoma of the extrahepatic bile ducts: results of an aggressive surgical approach. Surgery 1985;98:752-9.5.Tsuzuki T, Ueda M, Kuramachi S, Iida S, Takahashi S, Iri H.Carcinoma of the main hepatic duct junction: indications, operative morbidity and mortality, and long-term survival.Surgery 1990;108:495-501.6.Nimura Y, Hayakawa N, Kamiya J, Kondo S, Shionoya S.Hepatic segmentectomy with caudate lobe resection for bile duct carcinoma of the hepatic hilus. World J Surg 1990;14:535-44.7.Baer HU, Stain SC, Dennison AR, Eggers B, Blumgart LH.Improvements in survival by aggressive resection of hilar cholangiocarcinoma. Ann Surg 1993;217:20-7.8.Childs T, Hart M. Aggressive surgical therapy for Klaskintumors. Am J Surg 1993;165:554-7.9.Blumgart LH, Hadjis NS, Benjamin IS, Beazley R. Surgicalapproaches to cholangiocarcinoma at confluence of hepat-ic ducts. Lancet 1984;i:66-9.10.Miyazaki M, Itoh H, Ambiru S, Shimizu H, Togawa A,Gohchi E, et al. Radical surgery for advanced gallbladder carcinoma. Br J Surg 1996;83:478-81.11.Makuuchi M, Thai BL, Takayasu K, Takayama T, Kosuge T,Gunven P, et al. Preoperative portal embolization for hilar bile duct carcinoma: a preliminary report. Surgery 1990;107:521-7.12.Hayashi S, Miyazaki M, Kondo Y, Nakajima N. Invasivegrowth patterns of hepatic hilar ductal carcinoma: a histo-logic analysis of 18 surgical cases. Cancer 1994;73:2922-9.13.Bismuth H, Nakache R, Diamond T. Management strate-gies in resection for hilar cholangiocarcinoma. Ann Surg 1992;215:31-8.14.Suzuki M, Takahashi T, Ouchi K, Matsuno S. The develop-ment and extension of hepatohilar bile duct carcinoma.Cancer 1989;64:658-63.15.Boerma EJ, Bronkhorst FB, van Halest UJGM, de BoerHHM. An anatomical investigation of radical resection of tumor in the hepatic duct confluence. Surg Gynecol Obstet 1985;161:223-8.16.Pinson CW, Rossi RL. Extended right hepatic lobectomy,left hepatic lobectomy, and skeletonization resection for proximal bile duct cancer. World J Surg 1988;12:52-9. 17.Kuvshinoff BW, Armstrong JG, Fong Y, Schupak K,Getradjrman BW, Heffernan V, et al. Palliation of irre-sectable hilar cholangiocarcinoma with biliary drainage and radiotherapy. Br J Surg 1995;82:1522-5.18.Nordback IH, Pitt HA, Coleman J, Venbrux AC, DooleyWC, Yeu NN, et al. Unresectable hilar cholangiocarciono-ma: percutaneous versus operative palliation. Surgery 1994;115:597-603.19.O’Grady JG, Polson RJ, Rolles K, Calne RY, Williams R.Liver transplantation for malignant diseases: results in 93 consective patients. Ann Surg 1988;207:373-9.20.Iwatsuki S, Gordon RD, Shaw BW, Starzl TE. Role of livertransplantation in cancer therapy. Ann Surg 1985;202:401-7.136Miyazaki et al.SurgeryFebruary 1998Availability of Journal Back IssuesAs a service to our subscribers, copies of back issues of Surgery for the preceding 5 years aremaintained and are available for purchase from Mosby until inventory is depleted at a cost of$12.00 per issue. The following quantity discounts are available: 25% off on quantities of 12 to23, and one third off on quantities of 24 or more. Please write to Mosby, Inc., SubscriptionServices, 11830 Westline Industrial Drive, St. Louis, MO 63146-3318, or call (800)453-4351 or(314)453-4351 for information on availability of particular issues. If unavailable from the pub-lisher, photocopies of complete issues may be purchased from UMI, 300 N. Zeeb Rd., AnnArbor, MI 48106 (313)761-4700.。