Surg Today DOI 10.1007/s00595-014-0846-z
ORIGINAL ARTICLE
The prognostic factors for survival after curative resection of distal cholangiocarcinoma: perineural invasion and lymphovascular invasion Hee Joon Kim • Choong Young Kim • Young Hoe Hur • Yang Seok Koh • Jung Chul Kim Hyun Jong Kim • Chol Kyoon Cho
•
Received: 11 February 2013 / Accepted: 11 October 2013 Ó Springer Japan 2014
Abstract Purpose The purpose of this study was to elucidate the prognostic factors for distal cholangiocarcinoma after curative resection, and to assess the significance of perineural invasion (PNI) and lymphovascular invasion (LVI) as prognostic factors. Methods A retrospective analysis of 91 patients who underwent radical surgery for distal cholangiocarcinoma between March 2004 and October 2011 was performed. We analyzed the survival rate and prognostic factors affecting the survival. Results The overall 1-, 3- and 5-year survival rates were 84.1, 49.7 and 38.9 %, respectively. In the univariate analysis, the prognostic factors influencing the survival were the histological differentiation, lymph node (LN) involvement and TNM stage. In the multivariate analysis, LN metastasis was the only independent prognostic factor. Although patients with PNI tended to show poorer survival, it was not a statistically significant factor (3- and 5-year OS; 62.0 and 54.6 % vs. 42.8 and 30.9 %, P = 0.166). In the patients with a total lymph node count (TLNC) of 11 or less, PNI was a significant prognostic factor; however, it was not a significant factor in the patients with a TLNC
H. J. Kim C. Y. Kim Y. H. Hur Y. S. Koh J. C. Kim H. J. Kim C. K. Cho Department of Surgery, Chonnam National University Medical School, Gwangju, Korea e-mail: [email protected] C. K. Cho (&) Chonnam National University Hwasun Hospital, 322 Seoyang-ro, Hwasun-eup, Hwasun, Jeollanam 519-763, Korea e-mail: [email protected]
over 11. Overall, the LVI had no influence on the patient survival. Conclusions LN metastasis was the only significant prognostic factor after the curative resection of distal cholangiocarcinoma. In cases where adequate dissection was performed, it appeared that the PNI and LVI had no influence on the survival. Keywords Distal cholangiocarcinoma Prognostic factor Lymph node involvement Perineural invasion Lymphovascular invasion
Introduction Cholangiocarcinoma has commonly been classified into three broad categories: intrahepatic, perihilar and distal, based on the tumor location [1]. Distal cholangiocarcinomas are defined as tumors that involve the intrapancreatic portion of the bile duct, and these require pancreaticoduodenectomy for complete tumor resection. With the recent improvements in the techniques used for hepatobiliary surgery, a curative surgical resection of distal bile duct cancer can be accomplished with acceptable morbidity and mortality. However, the prognosis for such patients is poor. Numerous reports have suggested that the surgical margin status is the most useful prognostic factor [1–9]. However, even if a negative margin is obtained, the longterm survival is relatively poor because of frequent recurrence. The spread of the tumor through the perineurium and its clinical significance have been described in pancreatic cancer [10] and ampulla of Vater cancer [11]. The presence of lymphovascular invasion (LVI) has been reported as a significant prognostic factor in patients with pancreatic cancer and ampulla of Vater cancer [11, 12].
123
Surg Today
The purpose of this study was to identify useful prognostic factors and to evaluate the prognostic significance of perineural invasion (PNI) and lymphovascular invasion in patients undergoing radical pancreaticoduodenectomy with negative surgical margins for distal cholangiocarcinoma.
Patients and methods Pancreaticoduodenectomy was performed in 339 patients with pancreatobiliary diseases at the Department of Surgery, Chonnam University Hwa-sun Hospital between March 2004 and October 2011. The records for the 113 patients with intrapancreatic cholangiocarcinoma undergoing pancreaticoduodenectomy were reviewed retrospectively. Eight patients diagnosed with residual tumor cells at the resection margins based on the pathological findings, five patients underwent R2 resections, five patients with postoperative mortality, two patients with coexistent other malignancies and four patients without sufficient data were excluded from the study. Finally, the 91 eligible patients were selected for the present investigation.
perineural invasion, lymphovascular invasion and lymph node metastasis were all examined pathologically. The resection margins were considered to be positive if the postoperative pathological examination demonstrated the presence of infiltrating adenocarcinoma at the bile duct margin, pancreatic parenchymal margin, pancreatic retroperitoneal margin or posterior pancreatic margin. The total lymph node counts were recorded from the final pathological report.
Statistical analysis The Kaplan–Meier method was used to calculate the cumulative survival. Comparisons were made with the univariate log-rank test. Factors found to be significant in the univariate analysis were subjected to a multivariate analysis using a Cox proportional hazards model. The correlation of each factor was analyzed by the Chi-square test. A P value \0.05 was considered to be statistically significant. The statistical analyses were performed using the SPSS ver. 18.0 software program for windows (SPSS Inc., Chicago, IL, USA).
Surgical procedures Results A pylorus-preserving pancreaticoduodenectomy was performed in 77 patients, and a conventional pancreaticoduodenectomy (Whipple’s operation) was performed in 14 patients. All patients underwent dissection of the regional lymph nodes, including the nodes in the hepatoduodenal ligament, nodes along the common hepatic artery and the nodes around the superior mesenteric artery (SMA). Extended lymph node dissection involving the lymph nodes located left side to the SMA was not performed. The additional dissection of para-aortic lymph nodes was performed when necessary. The intraoperative pathological assessment of the bile duct resection margin, pancreatic parenchymal margin and retroperitoneal margin was performed using frozen sections. If the margin was positive for tumor cells, further resection was performed to the maximum extent possible. Partial resection of portal vein was performed when needed. The tumor stages were classified according to the AJCC 7th edition.
Pathological assessments All specimens were examined pathologically, and each tumor was classified as either well-differentiated, moderately differentiated or poorly differentiated adenocarcinoma according to the predominant pathological grading or differentiation. Pancreatic invasion, duodenal invasion,
123
Of the 91 patients, 59 (64.8 %) were male and 32 (35.2 %) were female, and their ages ranged from 48 to 83 years, averaging 66.6 years. The patients older than 65 years constituted 59.3 % of the cases. The mean follow-up of all patients was 28.7 months (range 1–79 months). Grossly, polypoid tumors were found in 27 patients. Most of the cases were adenocarcinoma, two cases were adenosquamous carcinoma and one case was a mixed tumor (adenocarcinoma with neuroendocrine carcinoma). Pathologically, perineural invasion, lymphovascular invasion, pancreatic invasion and duodenal invasion were identified in 57, 23, 53 and 17 patients, respectively. The tumor was identified as a welldifferentiated adenocarcinoma in 26 patients, moderately differentiated in 40 and poorly differentiated adenocarcinoma in 25 patients. Lymph node metastases were found in 32 patients. According to the AJCC 7th edition, the tumors were classified as stage IA in nine patients, IB in 12, IIA in 38 and IIB in 32 patients. The total number of pathologically assessed lymph nodes was recorded from the final pathology report for each surgical specimen. The mean number of assessed lymph nodes for all specimens was 11 (range 4–38). A total of 11 or fewer lymph nodes were found in 56 % of all cases (n = 51). Adjuvant chemotherapy was performed in 26 patients using a gemcitabine-based regimen for advanced disease. None of the patients received preoperative chemotherapy.
Surg Today Table 1 Results of the univariate analysis of prognostic factors for the overall survival Factor
No.
3 YSR
5 YSR
MST (months)
P value
B65
37
48.2
34.5
33.0
0.988
[65
54
50.3
42.5
40.0
Age (years)
Sex Male
59
39.6
29.3
24.0
Female
32
70.2
57.9
64.0
CA 19-9 (U/mL) \39 31
0.022
55.7
49.5
47.0
60
46.7
33.4
24.0
Polypoid
27
50.0
42.9
33.0
Nodular
9
33.3
33.3
15.0
Infiltrative
55
52.0
37.1
37.0
Well
26
74.6
61.1
77.0
Moderate/poor
65
38.7
28.6
24.0
T1
9
66.7
66.7
77.0
T2
19
50.3
26.8
40.0
T3
63
46.6
36.8
32.0
59 32
70.2 12.3
58.2 6.1
64.0 16.0
\0.001
0.166
C39
0.338
Gross appearance 0.991
Differentiation
Fig. 1 Overall survival rate of 91 patients with distal cholangiocarcinoma following surgical resection. Overall survival rates were 84.1 % at 1 year, 49.7 % at 3 years, and 38.9 % at 5 years, respectively. At end point of this study, of 45 patients who survived, 35 patients were observed less than 5 years after resection in the observation period
0.013
T stage 0.420
N stage
Overall survival
N0 N1
Perineural invasion
The 1-, 3- and 5-year overall survival rates were 84.1, 49.7 and 38.9 %, respectively (median survival time, 33.0 months) (Fig. 1). At the end of this study, 45 patients were still alive. Of these patients, ten patients had survived for more than 5 years, and 35 patients had been observed for less than 5 years after the resection. In the univariate analysis, the prognosis was significantly related to the histological differentiation, lymph node involvement and TNM stage. The following factors were not significantly associated with the prognosis: gross tumor appearance, T stage, perineural invasion, lymphovascular invasion, pancreatic invasion and duodenal invasion (Table 1). Adjuvant chemotherapy did not influence the overall survival in the T3 group, and also did not enhance the survival in the N0 and N1 groups. In the multivariate analysis using the Cox proportional hazards model involving the three significant factors identified by the univariate analysis, only lymph node metastasis exhibited an independent association with the long-term survival (P \ 0.001) (Table 2). The 1-, 3- and 5-year survival rates were 94.7, 70.2 and 58.2 %, respectively, for patients without lymph node involvement, whereas they were 59.9, 12.3 and 6.1 %, respectively, for patients with lymph node involvement (Fig. 2).
Negative
34
62.0
54.6
64.0
Positive
57
39.5
30.9
24.0
Lymphovascular invasion Negative
68
50.4
44.9
40.0
Positive
23
47.7
19.9
26.0
0.376
Perineural or lymphovascular invasion Negative
25
66.4
66.4
74.0
Positive
66
44.1
30.9
26.0
0.068
Pancreatic invasion Negative
38
51.0
39.1
40.0
Positive
53
48.8
39.5
33.0
Negative
74
67.8
41.8
40.0
Positive
17
41.2
29.4
21.0
0.915
Duodenal invasion 0.306
YSR year survival rate, MST median survival time
Perineural invasion, lymphovascular invasion and survival The overall survival (OS) rate was higher in the PNInegative group than in the PNI-positive group (3- and 5-year OS; 62.0 and 54.6 % vs. 39.5 and 30.9 %), but the differences were not statistically significant (P = 0.166)
123
Surg Today Table 2 Results of the multivariate analysis of prognostic factors for the overall survival Factor
P value
Hazard ratio
95 % CI
0.345
1.469
0.661–3.265
\0.001
3.768
1.834–7.742
Differentiation Well Moderate/poor N stage N0 N1
than that of the 13 LVI-positive patients (3- and 5-year OS; 61.9 and 57.5 % vs. 51.3 and 17.1 %, P = 0.084) (Fig. 4b). When the patients were grouped into a PNI and LVI double-negative group and a PNI- or LVI-positive group, the 25 patients in the double-negative group showed a higher survival rate than the 66 patients in the PNI- or LVI-positive group (3- and 5-year OS; 66.4 and 66.4 % vs. 44.1 and 30.9 %, P = 0.068) (Fig. 5a). In the group of patients with a total lymph node count of 11 or less, the survival rate of the 16 patients in the double-negative group was significantly higher than the survival rate of the 35 patients in the PNI- or LVI-positive group (3- and 5-year OS; 87.5 and 87.5 % vs. 46.4 and 28.8 %, P = 0.06) (Fig. 5b). However, in the group with a total lymph node count over 11, there were no survival differences between the double-negative patients and the PNI- or LVI-positive patients (P = 0.383). When the patients were grouped according to the lymph node status, the PNI and LVI did not affect the survival rate in either group.
Correlation among perineural invasion, lymphovascular invasion and other pathological factors
Fig. 2 Comparison of overall survival in patients, who underwent curative resection for distal cholangiocarcinoma according to lymph node metastasis (P \ 0.001). The 1-, 3-, and 5-year survival rates of both groups were 94.7, 70.2, 58.2 and 59.9, 12.3, 6.1 %, respectively. 29 patients in LN (-) group and 7 patients in LN (?) group had the observation period less than 5 years after resection at end point of this study
(Fig. 3a). When the extent of dissection was stratified according to the total lymph nodes harvested, in the group with a total lymph node count of 11 or less, the 22 PNInegative patients showed a significantly higher survival rate than the 29 PNI-positive patients (3- and 5-year OS; 76.4 and 68.7 % vs. 45.5 and 32.5 %, P = 0.037) (Fig. 3b). However, in the group with a total lymph node count higher than 11, there were no differences in the survival between the PNI-negative and PNI-positive patients (P = 0.795). Similarly, although the LVI had no effect on the overall survival in the overall patient population (Fig. 4a), in the group with a total lymph node count of 11 or less, the survival rate of the 38 LVI-negative patients was higher
123
PNI was more frequently observed in tumors with invasion to the pancreas. Among the patients with pancreatic invasion, 76.5 % showed PNI, and 57.5 % of pancreatic invasion-negative patients had no PNI (P = 0.001). Furthermore, PNI was also observed to be associated with lymph node metastasis (P = 0.014). However, no significant associations between the PNI and duodenal invasion (P = 0.267), lymphovascular invasion (P = 0.877) or the histological differentiation (P = 0.154) were observed (Table 3). LVI was significantly associated with lymph node metastasis. A total of 74.2 % of LVI-negative patients had no lymph node metastasis, and 60.9 % of LVI-positive patients showed lymph node metastasis (P = 0.003). LVI was found to be correlated with duodenal invasion (P = 0.031). However, there was no correlation between LVI and pancreatic invasion, perineural invasion or the histological differentiation (Table 4).
Discussion Distal cholangiocarcinomas are rare compared to hilar cholangiocarcinoma. The optimal treatment for distal cholangiocarcinomas is complete surgical resection. Distal cholangiocarcinoma requires pancreatoduodenectomy for complete surgical resection. In the current study, the 1-, 3and 5-year survival rate and median survival time were observed to be 84.1, 49.7 and 38.9 % and 33.0 months, respectively, consistent with values reported in previous
Surg Today
Fig. 3 a Comparison of overall survival of all patients according to perineural invasion. There was no statistically significant survival difference (P = 0.166). 16 patients in PNI (-) group and 19 patients in PNI (?) group had the observation period less than 5 years after resection at end point of this study. b In the patients group of total
lymph node counts with 11 or less, the survival was significantly different according to perineural invasion status (P = 0.037). 11 patients in PNI (-) group and 13 patients in PNI (?) group had the observation period less than 5 years after resection at end point of this study
Fig. 4 a Comparison of overall survival of all patients according to lymphovascular invasion. There was no statistically significant difference in overall survival (P = 0.390). 28 patients in LVI (-) group and 8 patients in LVI (?) group had the observation period less than 5 years after resection at end point of this study. b In the patients
group of total lymph node counts with 11 or less, LVI-negative patients had higher survival rate than LVI-positive patients (P = 0.084). 19 patients in LVI (-) group and 4 patients in LVI (?) group had the observation period less than 5 years after resection at end point of this study
123
Surg Today
Fig. 5 a Comparison of overall survival of all patients among PNI, LVI both-negative group and PNI or LVI-positive group. 25 patients of both-negative group showed higher survival rate than 66 patients of PNI or LVI-positive group (P = 0.068). 13 patients in both (-) group and 22 patients in PNI or LVI (?) group had the observation period less than 5 years after resection at end point of this study. b In the
patients group of total lymph node counts with 11 or less, bothnegative patients had significantly higher survival rate than PNI or LVI-positive patients (P = 0.006). 9 patients in both (-) group and 14 patients in PNI of LVI (?) group had the observation period less than 5 years after resection at end point of this study
Table 3 The correlation between perineural invasion and other factors
tumor differentiation [1, 3] have been reported as independent prognostic factors for distal cholangiocarcinoma, as indicated by a multivariate survival analysis. In the present study, the multivariate analysis showed that lymph node metastasis was the only significant prognostic factor for the overall survival after R0 resection. Ito et al. [15] have reported that patients who underwent R0 resection and were classified as N0, based on a total lymph node count \11, had a disease-specific survival that was significantly worse than that of patients classified as N0 based on a total lymph node count C11. Moreover, Oshiro et al. [16] have reported that the lymph node ratio (LNR) was an independent prognostic factor for extrahepatic cholangiocarcinoma. Yoshida et al. [8] have similarly reported that the number of metastatic lymph nodes was a significant prognostic factor. Nevertheless, in the present study, the total lymph node count, the lymph node ratio and the number of metastatic lymph nodes had no influence on the survival. The presence of metastatic lymph nodes was the only significant prognostic factor, irrespective of the total lymph node count, lymph node ratio and number of metastatic lymph nodes. Several authors have reported that PNI was the significant prognostic factor for cholangiocarcinoma [17–21]. However, in these reports, all types of cholangiocarcinoma were enrolled, regardless of the location. In most of the
Cancer involvement
Perineural invasion
P value
Positive (%)
Negative (%)
Pancreas positive
42/57 (73.7)
11/34 (32.3)
0.001
Duodenum positive
13/57 (22.8)
4/34 (11.7)
0.267
Lymph node positive
25/57 (43.8)
7/34 (20.5)
0.014
Lymphovascular positive
14/57 (24.5)
9/34 (26.4)
0.877
Table 4 The correlation between lymphovascular invasion and other factors Cancer involvement
Pancreas positive Duodenum positive
Lymphovascular invasion
P value
Positive (%)
Negative (%)
16/23 (69.5)
37/68 (54.4)
0.185
8/23 (34.7)
9/68 (13.2)
0.033
Lymph node positive
14/23 (60.8)
18/68 (26.4)
0.003
Perineural positive
14/23 (60.8)
43/68 (63.2)
0.877
studies [2, 3, 5, 8]. In the past, there have been several studies concerning the prognostic factors for distal cholangiocarcinoma. The surgical margin status [1–8], lymph node invasion [1–8, 13, 14] and pathological grading of
123
Surg Today
reports that have analyzed the prognostic factors for distal cholangiocarcinoma, PNI was not considered as a significant prognostic factor [3, 5–8, 14]. In the present study, we analyzed only the patients treated by pancreatoduodenectomy for distal cholangiocarcinoma without any microscopic residual tumor. Although patients with PNI tended to show a poorer survival than those without, the result was not statistically significant. The 3- and 5-year survival rates of patients with and without PNI were 39.5 and 30.9 % and 62.0 and 54.6 %, respectively (P = 0.166). When the patients were divided into two groups according to the total lymph node counts, in the group with a total lymph node count of 11 or less, the patients without PNI had a significantly better survival than those with PNI (P = 0.037). However, in the group with a total lymph node count over 11, there was no survival difference between PNI-positive and PNI-negative patients. Similar results were observed for the survival analysis according to LVI status. Therefore, PNI and LVI reflect the local extent of disease, but did not significantly influence the survival if adequate dissection was performed. Perineural invasion was more frequently observed in tumors with invasion to the pancreas. Among the PNIpositive patients, 70.9 % showed pancreatic invasion, and 64.7 % of PNI-negative patients had no pancreatic invasion. This result suggests that there is a significant relationship between PNI and pancreatic invasion (P = 0.001). Moreover, LVI was more frequently observed in tumors with invasion to the duodenum (P = 0.033). However, the pancreatic invasion and duodenal invasion were not associated with the survival. This suggested that pathological factors that define the extent of disease, such as the pancreatic invasion, duodenal invasion, PNI and LVI, had no significant influence on the survival if a radical resection was performed.
Conclusion The presence of lymph node metastasis was the only significant prognostic factor for distal cholangiocarcinoma after curative resection. Perineural invasion and lymphovascular invasion were not significant prognostic factors if adequate dissection was performed. Therefore, efforts should be made to ensure adequate lymphatic and neural dissection, and if lymph node involvement is found during the pathological examination; adjuvant chemotherapy should be considered, even if a negative resection margin is obtained. Conflict of interest Hee Joon Kim and co-authors have no conflicts of interest to declare.
References 1. Nakeeb A, Pitt HA, Sohn TA, Coleman J, Abrams RA, Piantadosi S, et al. Cholangiocarcinoma. A spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg. 1996;224(4):463–73 (discussion 473–5). 2. Bahra M, Jacob D, Langrehr JM, Neumann UP, Neuhaus P. Carcinoma of the distal and middle bile duct: surgical results, prognostic factors, and long-term follow-up. J Hepatobiliary Pancreat Surg. 2008;15(5):501–7. 3. Choi SB, Park SW, Kim KS, Choi JS, Lee WJ. The survival outcome and prognostic factors for middle and distal bile duct cancer following surgical resection. J Surg Oncol. 2009;99(6):335–42. 4. Fong Y, Blumgart LH, Lin E, Fortner JG, Brennan MF. Outcome of treatment for distal bile duct cancer. Br J Surg. 1996;83(12):1712–5. 5. Kayahara M, Nagakawa T, Ohta T, Kitagawa H, Tajima H, Miwa K. Role of nodal involvement and the periductal soft-tissue margin in middle and distal bile duct cancer. Ann Surg. 1999;229(1):76–83. 6. Murakami Y, Uemura K, Hayashidani Y, Sudo T, Hashimoto Y, Ohge H, et al. Prognostic significance of lymph node metastasis and surgical margin status for distal cholangiocarcinoma. J Surg Oncol. 2007;95(3):207–12. 7. Nomura T, Tsuchiya Y, Nashimoto A, Yabusaki H, Takii Y, Nakagawa S, et al. Prognostic factors for radical resection of middle and distal bile duct cancer. Hepatogastroenterology. 2009;56(90):294–8. 8. Yoshida T, Matsumoto T, Sasaki A, Morii Y, Aramaki M, Kitano S. Prognostic factors after pancreatoduodenectomy with extended lymphadenectomy for distal bile duct cancer. Arch Surg. 2002;137(1):69–73. 9. Uno M, Shimada K, Yamamoto Y, Nara S, Esaki M, Sakamoto Y, et al. Periductal infiltrating type of intrahepatic cholangiocarcinoma: a rare macroscopic type without any apparent mass. Surg Today. 2012;42(12):1189–94. 10. Chatterjee D, Katz MH, Rashid A, Wang H, Iuga AC, Varadhachary GR, et al. Perineural and intraneural invasion in posttherapy pancreaticoduodenectomy specimens predicts poor prognosis in patients with pancreatic ductal adenocarcinoma. Am J Surg Pathol. 2012;36(3):409–17. 11. Carter JT, Grenert JP, Rubenstein L, Stewart L, Way LW. Tumors of the ampulla of vater: histopathologic classification and predictors of survival. J Am Coll Surg. 2008;207(2):210–8. 12. Chen JW, Bhandari M, Astill DS, Wilson TG, Kow L, BrookeSmith M, et al. Predicting patient survival after pancreaticoduodenectomy for malignancy: histopathological criteria based on perineural infiltration and lymphovascular invasion. HPB (Oxford). 2010;12(2):101–8. 13. Murakami Y, Uemura K, Hayashidani Y, Sudo T, Ohge H, Sueda T. Pancreatoduodenectomy for distal cholangiocarcinoma: prognostic impact of lymph node metastasis. World J Surg. 2007;31(2):337–42 (discussion 343–4). 14. Woo SM, Ryu JK, Lee SH, Yoo JW, Park JK, Kim YT, et al. Recurrence and prognostic factors of ampullary carcinoma after radical resection: comparison with distal extrahepatic cholangiocarcinoma. Ann Surg Oncol. 2007;14(11):3195–201. 15. Ito K, Ito H, Allen PJ, Gonen M, Klimstra D, D’Angelica MI, et al. Adequate lymph node assessment for extrahepatic bile duct adenocarcinoma. Ann Surg. 2010;251(4):675–81. 16. Oshiro Y, Sasaki R, Kobayashi A, Murata S, Fukunaga K, Kondo T, et al. Prognostic relevance of the lymph node ratio in surgical patients with extrahepatic cholangiocarcinoma. Eur J Surg Oncol. 2011;37(1):60–4.
123
Surg Today 17. Bhuiya MR, Nimura Y, Kamiya J, Kondo S, Fukata S, Hayakawa N, et al. Clinicopathologic studies on perineural invasion of bile duct carcinoma. Ann Surg. 1992;215(4):344–9. 18. Bhuiya MR, Nimura Y, Kamiya J, Kondo S, Nagino M, Hayakawa N. Clinicopathologic factors influencing survival of patients with bile duct carcinoma: multivariate statistical analysis. World J Surg. 1993;17(5):653–7. 19. He P, Shi JS, Chen WK, Wang ZR, Ren H, Li H. Multivariate statistical analysis of clinicopathologic factors influencing
123
survival of patients with bile duct carcinoma. World J Gastroenterol. 2002;8(5):943–6. 20. Kaneoka Y, Yamaguchi A, Isogai M, Kumada T. Survival benefit of hepatopancreatoduodenectomy for cholangiocarcinoma in comparison to hepatectomy or pancreatoduodenectomy. World J Surg. 2010;34(11):2662–70. 21. Nakagohri T, Asano T, Kinoshita H, Kenmochi T, Urashima T, Miura F, et al. Aggressive surgical resection for hilar-invasive and peripheral intrahepatic cholangiocarcinoma. World J Surg. 2003;27(3):289–93.
ORIGINAL ARTICLE
The prognostic factors for survival after curative resection of distal cholangiocarcinoma: perineural invasion and lymphovascular invasion Hee Joon Kim • Choong Young Kim • Young Hoe Hur • Yang Seok Koh • Jung Chul Kim Hyun Jong Kim • Chol Kyoon Cho
•
Received: 11 February 2013 / Accepted: 11 October 2013 Ó Springer Japan 2014
Abstract Purpose The purpose of this study was to elucidate the prognostic factors for distal cholangiocarcinoma after curative resection, and to assess the significance of perineural invasion (PNI) and lymphovascular invasion (LVI) as prognostic factors. Methods A retrospective analysis of 91 patients who underwent radical surgery for distal cholangiocarcinoma between March 2004 and October 2011 was performed. We analyzed the survival rate and prognostic factors affecting the survival. Results The overall 1-, 3- and 5-year survival rates were 84.1, 49.7 and 38.9 %, respectively. In the univariate analysis, the prognostic factors influencing the survival were the histological differentiation, lymph node (LN) involvement and TNM stage. In the multivariate analysis, LN metastasis was the only independent prognostic factor. Although patients with PNI tended to show poorer survival, it was not a statistically significant factor (3- and 5-year OS; 62.0 and 54.6 % vs. 42.8 and 30.9 %, P = 0.166). In the patients with a total lymph node count (TLNC) of 11 or less, PNI was a significant prognostic factor; however, it was not a significant factor in the patients with a TLNC
H. J. Kim C. Y. Kim Y. H. Hur Y. S. Koh J. C. Kim H. J. Kim C. K. Cho Department of Surgery, Chonnam National University Medical School, Gwangju, Korea e-mail: [email protected] C. K. Cho (&) Chonnam National University Hwasun Hospital, 322 Seoyang-ro, Hwasun-eup, Hwasun, Jeollanam 519-763, Korea e-mail: [email protected]
over 11. Overall, the LVI had no influence on the patient survival. Conclusions LN metastasis was the only significant prognostic factor after the curative resection of distal cholangiocarcinoma. In cases where adequate dissection was performed, it appeared that the PNI and LVI had no influence on the survival. Keywords Distal cholangiocarcinoma Prognostic factor Lymph node involvement Perineural invasion Lymphovascular invasion
Introduction Cholangiocarcinoma has commonly been classified into three broad categories: intrahepatic, perihilar and distal, based on the tumor location [1]. Distal cholangiocarcinomas are defined as tumors that involve the intrapancreatic portion of the bile duct, and these require pancreaticoduodenectomy for complete tumor resection. With the recent improvements in the techniques used for hepatobiliary surgery, a curative surgical resection of distal bile duct cancer can be accomplished with acceptable morbidity and mortality. However, the prognosis for such patients is poor. Numerous reports have suggested that the surgical margin status is the most useful prognostic factor [1–9]. However, even if a negative margin is obtained, the longterm survival is relatively poor because of frequent recurrence. The spread of the tumor through the perineurium and its clinical significance have been described in pancreatic cancer [10] and ampulla of Vater cancer [11]. The presence of lymphovascular invasion (LVI) has been reported as a significant prognostic factor in patients with pancreatic cancer and ampulla of Vater cancer [11, 12].
123
Surg Today
The purpose of this study was to identify useful prognostic factors and to evaluate the prognostic significance of perineural invasion (PNI) and lymphovascular invasion in patients undergoing radical pancreaticoduodenectomy with negative surgical margins for distal cholangiocarcinoma.
Patients and methods Pancreaticoduodenectomy was performed in 339 patients with pancreatobiliary diseases at the Department of Surgery, Chonnam University Hwa-sun Hospital between March 2004 and October 2011. The records for the 113 patients with intrapancreatic cholangiocarcinoma undergoing pancreaticoduodenectomy were reviewed retrospectively. Eight patients diagnosed with residual tumor cells at the resection margins based on the pathological findings, five patients underwent R2 resections, five patients with postoperative mortality, two patients with coexistent other malignancies and four patients without sufficient data were excluded from the study. Finally, the 91 eligible patients were selected for the present investigation.
perineural invasion, lymphovascular invasion and lymph node metastasis were all examined pathologically. The resection margins were considered to be positive if the postoperative pathological examination demonstrated the presence of infiltrating adenocarcinoma at the bile duct margin, pancreatic parenchymal margin, pancreatic retroperitoneal margin or posterior pancreatic margin. The total lymph node counts were recorded from the final pathological report.
Statistical analysis The Kaplan–Meier method was used to calculate the cumulative survival. Comparisons were made with the univariate log-rank test. Factors found to be significant in the univariate analysis were subjected to a multivariate analysis using a Cox proportional hazards model. The correlation of each factor was analyzed by the Chi-square test. A P value \0.05 was considered to be statistically significant. The statistical analyses were performed using the SPSS ver. 18.0 software program for windows (SPSS Inc., Chicago, IL, USA).
Surgical procedures Results A pylorus-preserving pancreaticoduodenectomy was performed in 77 patients, and a conventional pancreaticoduodenectomy (Whipple’s operation) was performed in 14 patients. All patients underwent dissection of the regional lymph nodes, including the nodes in the hepatoduodenal ligament, nodes along the common hepatic artery and the nodes around the superior mesenteric artery (SMA). Extended lymph node dissection involving the lymph nodes located left side to the SMA was not performed. The additional dissection of para-aortic lymph nodes was performed when necessary. The intraoperative pathological assessment of the bile duct resection margin, pancreatic parenchymal margin and retroperitoneal margin was performed using frozen sections. If the margin was positive for tumor cells, further resection was performed to the maximum extent possible. Partial resection of portal vein was performed when needed. The tumor stages were classified according to the AJCC 7th edition.
Pathological assessments All specimens were examined pathologically, and each tumor was classified as either well-differentiated, moderately differentiated or poorly differentiated adenocarcinoma according to the predominant pathological grading or differentiation. Pancreatic invasion, duodenal invasion,
123
Of the 91 patients, 59 (64.8 %) were male and 32 (35.2 %) were female, and their ages ranged from 48 to 83 years, averaging 66.6 years. The patients older than 65 years constituted 59.3 % of the cases. The mean follow-up of all patients was 28.7 months (range 1–79 months). Grossly, polypoid tumors were found in 27 patients. Most of the cases were adenocarcinoma, two cases were adenosquamous carcinoma and one case was a mixed tumor (adenocarcinoma with neuroendocrine carcinoma). Pathologically, perineural invasion, lymphovascular invasion, pancreatic invasion and duodenal invasion were identified in 57, 23, 53 and 17 patients, respectively. The tumor was identified as a welldifferentiated adenocarcinoma in 26 patients, moderately differentiated in 40 and poorly differentiated adenocarcinoma in 25 patients. Lymph node metastases were found in 32 patients. According to the AJCC 7th edition, the tumors were classified as stage IA in nine patients, IB in 12, IIA in 38 and IIB in 32 patients. The total number of pathologically assessed lymph nodes was recorded from the final pathology report for each surgical specimen. The mean number of assessed lymph nodes for all specimens was 11 (range 4–38). A total of 11 or fewer lymph nodes were found in 56 % of all cases (n = 51). Adjuvant chemotherapy was performed in 26 patients using a gemcitabine-based regimen for advanced disease. None of the patients received preoperative chemotherapy.
Surg Today Table 1 Results of the univariate analysis of prognostic factors for the overall survival Factor
No.
3 YSR
5 YSR
MST (months)
P value
B65
37
48.2
34.5
33.0
0.988
[65
54
50.3
42.5
40.0
Age (years)
Sex Male
59
39.6
29.3
24.0
Female
32
70.2
57.9
64.0
CA 19-9 (U/mL) \39 31
0.022
55.7
49.5
47.0
60
46.7
33.4
24.0
Polypoid
27
50.0
42.9
33.0
Nodular
9
33.3
33.3
15.0
Infiltrative
55
52.0
37.1
37.0
Well
26
74.6
61.1
77.0
Moderate/poor
65
38.7
28.6
24.0
T1
9
66.7
66.7
77.0
T2
19
50.3
26.8
40.0
T3
63
46.6
36.8
32.0
59 32
70.2 12.3
58.2 6.1
64.0 16.0
\0.001
0.166
C39
0.338
Gross appearance 0.991
Differentiation
Fig. 1 Overall survival rate of 91 patients with distal cholangiocarcinoma following surgical resection. Overall survival rates were 84.1 % at 1 year, 49.7 % at 3 years, and 38.9 % at 5 years, respectively. At end point of this study, of 45 patients who survived, 35 patients were observed less than 5 years after resection in the observation period
0.013
T stage 0.420
N stage
Overall survival
N0 N1
Perineural invasion
The 1-, 3- and 5-year overall survival rates were 84.1, 49.7 and 38.9 %, respectively (median survival time, 33.0 months) (Fig. 1). At the end of this study, 45 patients were still alive. Of these patients, ten patients had survived for more than 5 years, and 35 patients had been observed for less than 5 years after the resection. In the univariate analysis, the prognosis was significantly related to the histological differentiation, lymph node involvement and TNM stage. The following factors were not significantly associated with the prognosis: gross tumor appearance, T stage, perineural invasion, lymphovascular invasion, pancreatic invasion and duodenal invasion (Table 1). Adjuvant chemotherapy did not influence the overall survival in the T3 group, and also did not enhance the survival in the N0 and N1 groups. In the multivariate analysis using the Cox proportional hazards model involving the three significant factors identified by the univariate analysis, only lymph node metastasis exhibited an independent association with the long-term survival (P \ 0.001) (Table 2). The 1-, 3- and 5-year survival rates were 94.7, 70.2 and 58.2 %, respectively, for patients without lymph node involvement, whereas they were 59.9, 12.3 and 6.1 %, respectively, for patients with lymph node involvement (Fig. 2).
Negative
34
62.0
54.6
64.0
Positive
57
39.5
30.9
24.0
Lymphovascular invasion Negative
68
50.4
44.9
40.0
Positive
23
47.7
19.9
26.0
0.376
Perineural or lymphovascular invasion Negative
25
66.4
66.4
74.0
Positive
66
44.1
30.9
26.0
0.068
Pancreatic invasion Negative
38
51.0
39.1
40.0
Positive
53
48.8
39.5
33.0
Negative
74
67.8
41.8
40.0
Positive
17
41.2
29.4
21.0
0.915
Duodenal invasion 0.306
YSR year survival rate, MST median survival time
Perineural invasion, lymphovascular invasion and survival The overall survival (OS) rate was higher in the PNInegative group than in the PNI-positive group (3- and 5-year OS; 62.0 and 54.6 % vs. 39.5 and 30.9 %), but the differences were not statistically significant (P = 0.166)
123
Surg Today Table 2 Results of the multivariate analysis of prognostic factors for the overall survival Factor
P value
Hazard ratio
95 % CI
0.345
1.469
0.661–3.265
\0.001
3.768
1.834–7.742
Differentiation Well Moderate/poor N stage N0 N1
than that of the 13 LVI-positive patients (3- and 5-year OS; 61.9 and 57.5 % vs. 51.3 and 17.1 %, P = 0.084) (Fig. 4b). When the patients were grouped into a PNI and LVI double-negative group and a PNI- or LVI-positive group, the 25 patients in the double-negative group showed a higher survival rate than the 66 patients in the PNI- or LVI-positive group (3- and 5-year OS; 66.4 and 66.4 % vs. 44.1 and 30.9 %, P = 0.068) (Fig. 5a). In the group of patients with a total lymph node count of 11 or less, the survival rate of the 16 patients in the double-negative group was significantly higher than the survival rate of the 35 patients in the PNI- or LVI-positive group (3- and 5-year OS; 87.5 and 87.5 % vs. 46.4 and 28.8 %, P = 0.06) (Fig. 5b). However, in the group with a total lymph node count over 11, there were no survival differences between the double-negative patients and the PNI- or LVI-positive patients (P = 0.383). When the patients were grouped according to the lymph node status, the PNI and LVI did not affect the survival rate in either group.
Correlation among perineural invasion, lymphovascular invasion and other pathological factors
Fig. 2 Comparison of overall survival in patients, who underwent curative resection for distal cholangiocarcinoma according to lymph node metastasis (P \ 0.001). The 1-, 3-, and 5-year survival rates of both groups were 94.7, 70.2, 58.2 and 59.9, 12.3, 6.1 %, respectively. 29 patients in LN (-) group and 7 patients in LN (?) group had the observation period less than 5 years after resection at end point of this study
(Fig. 3a). When the extent of dissection was stratified according to the total lymph nodes harvested, in the group with a total lymph node count of 11 or less, the 22 PNInegative patients showed a significantly higher survival rate than the 29 PNI-positive patients (3- and 5-year OS; 76.4 and 68.7 % vs. 45.5 and 32.5 %, P = 0.037) (Fig. 3b). However, in the group with a total lymph node count higher than 11, there were no differences in the survival between the PNI-negative and PNI-positive patients (P = 0.795). Similarly, although the LVI had no effect on the overall survival in the overall patient population (Fig. 4a), in the group with a total lymph node count of 11 or less, the survival rate of the 38 LVI-negative patients was higher
123
PNI was more frequently observed in tumors with invasion to the pancreas. Among the patients with pancreatic invasion, 76.5 % showed PNI, and 57.5 % of pancreatic invasion-negative patients had no PNI (P = 0.001). Furthermore, PNI was also observed to be associated with lymph node metastasis (P = 0.014). However, no significant associations between the PNI and duodenal invasion (P = 0.267), lymphovascular invasion (P = 0.877) or the histological differentiation (P = 0.154) were observed (Table 3). LVI was significantly associated with lymph node metastasis. A total of 74.2 % of LVI-negative patients had no lymph node metastasis, and 60.9 % of LVI-positive patients showed lymph node metastasis (P = 0.003). LVI was found to be correlated with duodenal invasion (P = 0.031). However, there was no correlation between LVI and pancreatic invasion, perineural invasion or the histological differentiation (Table 4).
Discussion Distal cholangiocarcinomas are rare compared to hilar cholangiocarcinoma. The optimal treatment for distal cholangiocarcinomas is complete surgical resection. Distal cholangiocarcinoma requires pancreatoduodenectomy for complete surgical resection. In the current study, the 1-, 3and 5-year survival rate and median survival time were observed to be 84.1, 49.7 and 38.9 % and 33.0 months, respectively, consistent with values reported in previous
Surg Today
Fig. 3 a Comparison of overall survival of all patients according to perineural invasion. There was no statistically significant survival difference (P = 0.166). 16 patients in PNI (-) group and 19 patients in PNI (?) group had the observation period less than 5 years after resection at end point of this study. b In the patients group of total
lymph node counts with 11 or less, the survival was significantly different according to perineural invasion status (P = 0.037). 11 patients in PNI (-) group and 13 patients in PNI (?) group had the observation period less than 5 years after resection at end point of this study
Fig. 4 a Comparison of overall survival of all patients according to lymphovascular invasion. There was no statistically significant difference in overall survival (P = 0.390). 28 patients in LVI (-) group and 8 patients in LVI (?) group had the observation period less than 5 years after resection at end point of this study. b In the patients
group of total lymph node counts with 11 or less, LVI-negative patients had higher survival rate than LVI-positive patients (P = 0.084). 19 patients in LVI (-) group and 4 patients in LVI (?) group had the observation period less than 5 years after resection at end point of this study
123
Surg Today
Fig. 5 a Comparison of overall survival of all patients among PNI, LVI both-negative group and PNI or LVI-positive group. 25 patients of both-negative group showed higher survival rate than 66 patients of PNI or LVI-positive group (P = 0.068). 13 patients in both (-) group and 22 patients in PNI or LVI (?) group had the observation period less than 5 years after resection at end point of this study. b In the
patients group of total lymph node counts with 11 or less, bothnegative patients had significantly higher survival rate than PNI or LVI-positive patients (P = 0.006). 9 patients in both (-) group and 14 patients in PNI of LVI (?) group had the observation period less than 5 years after resection at end point of this study
Table 3 The correlation between perineural invasion and other factors
tumor differentiation [1, 3] have been reported as independent prognostic factors for distal cholangiocarcinoma, as indicated by a multivariate survival analysis. In the present study, the multivariate analysis showed that lymph node metastasis was the only significant prognostic factor for the overall survival after R0 resection. Ito et al. [15] have reported that patients who underwent R0 resection and were classified as N0, based on a total lymph node count \11, had a disease-specific survival that was significantly worse than that of patients classified as N0 based on a total lymph node count C11. Moreover, Oshiro et al. [16] have reported that the lymph node ratio (LNR) was an independent prognostic factor for extrahepatic cholangiocarcinoma. Yoshida et al. [8] have similarly reported that the number of metastatic lymph nodes was a significant prognostic factor. Nevertheless, in the present study, the total lymph node count, the lymph node ratio and the number of metastatic lymph nodes had no influence on the survival. The presence of metastatic lymph nodes was the only significant prognostic factor, irrespective of the total lymph node count, lymph node ratio and number of metastatic lymph nodes. Several authors have reported that PNI was the significant prognostic factor for cholangiocarcinoma [17–21]. However, in these reports, all types of cholangiocarcinoma were enrolled, regardless of the location. In most of the
Cancer involvement
Perineural invasion
P value
Positive (%)
Negative (%)
Pancreas positive
42/57 (73.7)
11/34 (32.3)
0.001
Duodenum positive
13/57 (22.8)
4/34 (11.7)
0.267
Lymph node positive
25/57 (43.8)
7/34 (20.5)
0.014
Lymphovascular positive
14/57 (24.5)
9/34 (26.4)
0.877
Table 4 The correlation between lymphovascular invasion and other factors Cancer involvement
Pancreas positive Duodenum positive
Lymphovascular invasion
P value
Positive (%)
Negative (%)
16/23 (69.5)
37/68 (54.4)
0.185
8/23 (34.7)
9/68 (13.2)
0.033
Lymph node positive
14/23 (60.8)
18/68 (26.4)
0.003
Perineural positive
14/23 (60.8)
43/68 (63.2)
0.877
studies [2, 3, 5, 8]. In the past, there have been several studies concerning the prognostic factors for distal cholangiocarcinoma. The surgical margin status [1–8], lymph node invasion [1–8, 13, 14] and pathological grading of
123
Surg Today
reports that have analyzed the prognostic factors for distal cholangiocarcinoma, PNI was not considered as a significant prognostic factor [3, 5–8, 14]. In the present study, we analyzed only the patients treated by pancreatoduodenectomy for distal cholangiocarcinoma without any microscopic residual tumor. Although patients with PNI tended to show a poorer survival than those without, the result was not statistically significant. The 3- and 5-year survival rates of patients with and without PNI were 39.5 and 30.9 % and 62.0 and 54.6 %, respectively (P = 0.166). When the patients were divided into two groups according to the total lymph node counts, in the group with a total lymph node count of 11 or less, the patients without PNI had a significantly better survival than those with PNI (P = 0.037). However, in the group with a total lymph node count over 11, there was no survival difference between PNI-positive and PNI-negative patients. Similar results were observed for the survival analysis according to LVI status. Therefore, PNI and LVI reflect the local extent of disease, but did not significantly influence the survival if adequate dissection was performed. Perineural invasion was more frequently observed in tumors with invasion to the pancreas. Among the PNIpositive patients, 70.9 % showed pancreatic invasion, and 64.7 % of PNI-negative patients had no pancreatic invasion. This result suggests that there is a significant relationship between PNI and pancreatic invasion (P = 0.001). Moreover, LVI was more frequently observed in tumors with invasion to the duodenum (P = 0.033). However, the pancreatic invasion and duodenal invasion were not associated with the survival. This suggested that pathological factors that define the extent of disease, such as the pancreatic invasion, duodenal invasion, PNI and LVI, had no significant influence on the survival if a radical resection was performed.
Conclusion The presence of lymph node metastasis was the only significant prognostic factor for distal cholangiocarcinoma after curative resection. Perineural invasion and lymphovascular invasion were not significant prognostic factors if adequate dissection was performed. Therefore, efforts should be made to ensure adequate lymphatic and neural dissection, and if lymph node involvement is found during the pathological examination; adjuvant chemotherapy should be considered, even if a negative resection margin is obtained. Conflict of interest Hee Joon Kim and co-authors have no conflicts of interest to declare.
References 1. Nakeeb A, Pitt HA, Sohn TA, Coleman J, Abrams RA, Piantadosi S, et al. Cholangiocarcinoma. A spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg. 1996;224(4):463–73 (discussion 473–5). 2. Bahra M, Jacob D, Langrehr JM, Neumann UP, Neuhaus P. Carcinoma of the distal and middle bile duct: surgical results, prognostic factors, and long-term follow-up. J Hepatobiliary Pancreat Surg. 2008;15(5):501–7. 3. Choi SB, Park SW, Kim KS, Choi JS, Lee WJ. The survival outcome and prognostic factors for middle and distal bile duct cancer following surgical resection. J Surg Oncol. 2009;99(6):335–42. 4. Fong Y, Blumgart LH, Lin E, Fortner JG, Brennan MF. Outcome of treatment for distal bile duct cancer. Br J Surg. 1996;83(12):1712–5. 5. Kayahara M, Nagakawa T, Ohta T, Kitagawa H, Tajima H, Miwa K. Role of nodal involvement and the periductal soft-tissue margin in middle and distal bile duct cancer. Ann Surg. 1999;229(1):76–83. 6. Murakami Y, Uemura K, Hayashidani Y, Sudo T, Hashimoto Y, Ohge H, et al. Prognostic significance of lymph node metastasis and surgical margin status for distal cholangiocarcinoma. J Surg Oncol. 2007;95(3):207–12. 7. Nomura T, Tsuchiya Y, Nashimoto A, Yabusaki H, Takii Y, Nakagawa S, et al. Prognostic factors for radical resection of middle and distal bile duct cancer. Hepatogastroenterology. 2009;56(90):294–8. 8. Yoshida T, Matsumoto T, Sasaki A, Morii Y, Aramaki M, Kitano S. Prognostic factors after pancreatoduodenectomy with extended lymphadenectomy for distal bile duct cancer. Arch Surg. 2002;137(1):69–73. 9. Uno M, Shimada K, Yamamoto Y, Nara S, Esaki M, Sakamoto Y, et al. Periductal infiltrating type of intrahepatic cholangiocarcinoma: a rare macroscopic type without any apparent mass. Surg Today. 2012;42(12):1189–94. 10. Chatterjee D, Katz MH, Rashid A, Wang H, Iuga AC, Varadhachary GR, et al. Perineural and intraneural invasion in posttherapy pancreaticoduodenectomy specimens predicts poor prognosis in patients with pancreatic ductal adenocarcinoma. Am J Surg Pathol. 2012;36(3):409–17. 11. Carter JT, Grenert JP, Rubenstein L, Stewart L, Way LW. Tumors of the ampulla of vater: histopathologic classification and predictors of survival. J Am Coll Surg. 2008;207(2):210–8. 12. Chen JW, Bhandari M, Astill DS, Wilson TG, Kow L, BrookeSmith M, et al. Predicting patient survival after pancreaticoduodenectomy for malignancy: histopathological criteria based on perineural infiltration and lymphovascular invasion. HPB (Oxford). 2010;12(2):101–8. 13. Murakami Y, Uemura K, Hayashidani Y, Sudo T, Ohge H, Sueda T. Pancreatoduodenectomy for distal cholangiocarcinoma: prognostic impact of lymph node metastasis. World J Surg. 2007;31(2):337–42 (discussion 343–4). 14. Woo SM, Ryu JK, Lee SH, Yoo JW, Park JK, Kim YT, et al. Recurrence and prognostic factors of ampullary carcinoma after radical resection: comparison with distal extrahepatic cholangiocarcinoma. Ann Surg Oncol. 2007;14(11):3195–201. 15. Ito K, Ito H, Allen PJ, Gonen M, Klimstra D, D’Angelica MI, et al. Adequate lymph node assessment for extrahepatic bile duct adenocarcinoma. Ann Surg. 2010;251(4):675–81. 16. Oshiro Y, Sasaki R, Kobayashi A, Murata S, Fukunaga K, Kondo T, et al. Prognostic relevance of the lymph node ratio in surgical patients with extrahepatic cholangiocarcinoma. Eur J Surg Oncol. 2011;37(1):60–4.
123
Surg Today 17. Bhuiya MR, Nimura Y, Kamiya J, Kondo S, Fukata S, Hayakawa N, et al. Clinicopathologic studies on perineural invasion of bile duct carcinoma. Ann Surg. 1992;215(4):344–9. 18. Bhuiya MR, Nimura Y, Kamiya J, Kondo S, Nagino M, Hayakawa N. Clinicopathologic factors influencing survival of patients with bile duct carcinoma: multivariate statistical analysis. World J Surg. 1993;17(5):653–7. 19. He P, Shi JS, Chen WK, Wang ZR, Ren H, Li H. Multivariate statistical analysis of clinicopathologic factors influencing
123
survival of patients with bile duct carcinoma. World J Gastroenterol. 2002;8(5):943–6. 20. Kaneoka Y, Yamaguchi A, Isogai M, Kumada T. Survival benefit of hepatopancreatoduodenectomy for cholangiocarcinoma in comparison to hepatectomy or pancreatoduodenectomy. World J Surg. 2010;34(11):2662–70. 21. Nakagohri T, Asano T, Kinoshita H, Kenmochi T, Urashima T, Miura F, et al. Aggressive surgical resection for hilar-invasive and peripheral intrahepatic cholangiocarcinoma. World J Surg. 2003;27(3):289–93.
