- Email: [email protected]
Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified
HPB
http://dx.doi.org/10.1016/j.hpb.2017.01.025
ORIGINAL ARTICLE
Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified Russell Hodgson, Harsheet Sethi, Andrew H. Ling & Peter Lodge St James’s University Hospital, Beckett Street, Leeds LS9 7TF, West Yorkshire, UK
Abstract Background: The aim of this study was to describe the outcome of patients with colorectal liver metastases (CRLM) and radiological or clinical evidence of metastatic hepatic lymph node involvement who underwent combined hepatectomy and hepatic pedicle lymphadenectomy. Methods: Retrospective analysis of a prospectively maintained audit of 2082 patients undergoing liver resection for CRLM between 1994 and 2014. Age, type of resection, CT/MRI/PET detection, location, disease recurrence and survival were analysed. Results: Combined hepatectomy and hepatic pedicle lymphadenopathy was performed on 76 patients who met the inclusion criteria. 46% of enlarged lymph nodes were located in the hepatic ligament, with 38% retroportal, 38% common hepatic and 33% coeliac nodes. 50% of lymph node resections were positive for metastatic tumour. Pre-operative CT, MRI and CT/PET failed to detect histologically proven lymph node disease in 25/38 patients. Patients with negative nodal histology had a significant overall (44 vs 20 months, p = 0.008) and disease free (20 vs 11 months, p < 0.001) survival advantage. Conclusion: Combined hepatectomy and lymph node resection for CRLM in the setting of enlarged or suspicious lymphadenopathy is justified as imaging and operative findings are poor guides in determining positive lymph node disease. Received 12 November 2016; accepted 24 January 2017
Correspondence Russell Hodgson, Department of Surgery, Austin Health, Studley Rd, Heidelberg 3084, Victoria, Australia. E-mail: [email protected]
Introduction Liver resection for isolated colorectal metastases is well recognised as the gold standard of care with excellent long term outcomes with five year survival rates >50% now commonplace.1,2 The frontier for the treatment of colorectal metastases is now expanding to treating patients with additional extrahepatic metastases, the most prevalent of these being lung metastases which are also amenable to resection.2 The role of liver resection with extrahepatic lymph node (LN) disease in the hepatic pedicle continues to be debated.3–5 It appears the therapeutic value of Previous presentation: This paper is based on a free paper presented to the EA-HPBA meeting in Manchester on 24/4/2015: R Hodgson, H Sethi, JPA Lodge, Combined Hepatectomy and Hepatic Pedicle Lymph Node Resection in Colorectal Liver Metastases is Justified. EA-HPBA, Manchester UK, 24/4/2015.
HPB 2017, -, 1–5
routine lymphadenectomy is low if not non-existent, but there could be a good argument for lymphadenectomy for staging purposes as many micrometastases are missed by imaging and intra-operative assessment.5,6 The pattern of LN involvement in the hepatic pedicle is not consistent, however, and skipped LN appear to be common, necessitating thorough LN clearance which has not yet become common practice.7,8 The other clinical question raised is what to do with enlarged or suspicious LN detected on radiological imaging or intraoperative assessment. Results over the last 20 years have consistently shown poor survival and have led to some suggesting that LN involvement should be a contraindication to liver resection. However not all enlarged or suspicious LN are positive on histology.9 Median survival for metastatic colorectal cancer without hepatectomy is now up to 24 months,10 and it remains to be answered if chemotherapy can add further
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Hodgson R, et al., Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.01.025
HPB
2
gains to patients undergoing hepatectomy and hepatic pedicle lymphadenectomy. The aim of this study was to assess the outcome of patients undergoing combined hepatectomy and LN resection in the setting of clinically enlarged or suspicious LN within the liver’s main lymphatic drainage.
Methods All patients undergoing liver resection for colorectal metastases at St James’s University Hospital, Leeds, UK between the 1st of January 1994 and the 31st of December 2014 were entered into a prospectively maintained database. Data recorded included gender, date of operation, age at operation, type of resection, lymph node group (hepatoduodenal ligament, retroportal, common hepatic, coeliac or para-aortic adjacent to coeliac), R status of resection (R1 < 1 mm clearance of hepatic disease), and number of tumours. Medical records were sourced for pre-operative reports of CT, MRI and PET scans. PET scans were used only in the last 10 years of the study and only when there was increased suspicion of extrahepatic disease. Lymph nodes were defined as enlarged if described as enlarged or suspicious by the reporting radiologists, typically if >1 cm in size, or increased uptake of FDG (PET). Suspicious lymph nodes were defined by individual surgeons at their discretion, typically if either enlarged, firm or tethered to surrounding structures. Lymph nodes with extranodal extension into main structures (such as hepatic arteries or bile ducts) precluded resection. Lymph node resection was not standardised and was performed according to individual surgeon’s preferences, either with en bloc resection of lymph node zones or resection of individual nodes. All specimens were reviewed by pathologists with experience in HPB pathology. Post-operatively, all patients were entered into an intense surveillance programme with LFT, CEA, CA19-9 and chest, abdominal and pelvis CT at 3, 6, 12, 18, 24 months then years 3, 4, 5, 7 and 10 and clinical exam and bloods at years 6, 8 and 9. Due to this intensive surveillance, all episodes of disease recurrence together with the outcome were recorded. Survival data up to 30/6/2016 including causes of death were obtained from hospital records and by cross referencing this data with that held by the Northern and Yorkshire Cancer Registry Information Service. Statistical analysis was performed using SPSS (IBM, New York, USA). Statistical significance was calculated using the log rank (Mantel–Cox) method and Fisher’s exact test where appropriate. A p-value of <0.05 was considered statistically significant.
metastases. From this cohort, 76 patients (4%) who had concomitant hepatic pedicle LN resection for enlarged or suspicious nodes were identified. There were no significant differences between those who underwent combined liver and nodal resection vs liver resection alone with regard to age, gender, number of tumours and R status (data not shown). In the LN excision group 38 (50%) patients were found to have positive histology. Demographics, tumour, operative characteristics, and detection on radiological imaging by nodal histology are shown in Table 1. Disease-free and overall survival The median (range) followup was 28 (0–214) months. At the end of the study period, 15 (20%) patients were alive with no disease, 7 (9%) were alive with disease and 54 (71%) had died. The LN excision group had a significantly worse disease free survival compared to the total population of patients undergoing hepatic resection for CRLM as shown in Fig. 1 with 1-year, 3-year
Table 1 Demographics of patients who underwent concomitant LN
resection with either positive or negative histology Variable
Histology negative
Histology positive
Significance
Number of patients
38
38
p = ns
Age, median (range)
60 (25–80)
63 (36–83)
p = ns
23
19
p = ns
2 (1–7)
3 (1–10)
p = ns
R0
25
21
R1
12
15
R2
1
2
Gender Male Number of tumours, median (range) R status
p = ns
Type of resection
p = ns
Minor liver resection (<3 segments)
15
15
Major liver resection (3 segments)
23
23
Radiologically enlarged/suspicious LN CT
2/37
7/38
p = 0.153
MRI
5/35
13/36
p = 0.055
PET
0/7
4/4
p < 0.003
Hepatoduodenal ligament
16
17
Retroportal
7
12
Results
Common hepatic artery
10
18
Patient demographics From the 1st of January 1994 until the 31st of December 2014 there were 2082 liver resections performed for colorectal liver
Coeliac
16
8
Para-aortic (at level of coeliac)
2
2
HPB 2017, -, 1–5
LN region
p = ns
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Hodgson R, et al., Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.01.025
HPB
3
The disease free survival of the negative histology group mirrored the total population and was significantly better than that of the positive histology group as shown in Fig. 3 with 1year, 3-year and 5-year survivals of 63%, 44% and 44%, compared with 46%, 16% and 12% respectively (p = 0.008). There was a similar pattern in overall survival as shown in Fig. 4 with a significant increased survival in the negative histology group with 1-year, 3-year and 5-year survivals of 92%, 70% and 55%, compared with 78%, 20% and 13% in the positive histology group (p < 0.001).
Discussion
Figure 1 Disease free survival of all patients undergoing hepatectomy
for CRLM compared with patients undergoing hepatectomy and lymphadenectomy
and 5-year survivals of 55%, 30% and 29%, and 78%, 60% and 54% respectively (p < 0.001). This pattern was similarly observed for overall survival as shown in Fig. 2 with 1-year, 3-year and 5year survivals of 85%, 44% and 31%, and 89%, 64% and 53% respectively (p < 0.001).
The incidence of hepatic pedicle LN involvement has been reported to be up to 28% in patients undergoing hepatectomy for colorectal liver metastases.11 In this study, only 76 patients from a total population of 2082 were identified using radiologically or through intra-operative assessment, and only 36 (2%) had positive histology. Thus if this reported incidence of up to 28% was assumed for the 2082 patients undergoing hepatectomy for CRLM in this study, the vast majority of LN metastases were not detected. Additionally, of those suspected in this study, only 50% were positive. This poses two questions: what should be done when enlarged or suspicious nodes are encountered, and should hepatic pedicle lymphadenectomy during hepatectomy for colorectal liver metastases be performed routinely? Most of the patients in this study were investigated preoperatively with CT (75/76) and MRI (73/76), with PET scanning increasing in prevalence in the latter years of the study.
Figure 2 Overall survival of all patients undergoing hepatectomy for
Figure 3 Disease free survival comparison of positive and negative LN
CRLM compared with patients undergoing hepatectomy and
histology in those undergoing hepatectomy and lymphadenectomy for
lymphadenectomy
CRLM
HPB 2017, -, 1–5
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Hodgson R, et al., Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.01.025
HPB
4
Figure 4 Overall survival comparison of positive and negative LN
histology in those undergoing hepatectomy and lymphadenectomy for CRLM
While the positive predictive values of CT and MRI were individually above 50%, combined they failed to detect 25/38 (66%) patients with histologically positive nodes. This correlates with the study performed by Rau et al., which found that CT had only a positive predictive value of 56%.6 PET scanning however, while performed less frequently, appeared to accurately determine positive nodes with a 100% positive predictive value, albeit with only 11 scans performed in selected patients in this series. If there was convincing evidence to not offer surgery to those with hepatic pedicle LN disease, there may be an argument for routinely performing PET scans on all potential hepatectomy patients, however more data are still required. When analysing outcomes for those with positive LN histology, disease free (12% at 5 years) and overall survival (13% at 5 years) is dismal, which is consistent with published findings.4,11–13 The outcome has slightly improved with only isolated survivors noted in the pre-2000 literature.14,15 Within the current series 23/38 patients with positive histology, however, were not identified on pre-operative imaging and were only detected at intraoperative assessment, suggesting that the surgeon is better able to recognise extrahepatic disease. This study only had a positive histology rate of 50% after resection of suspicious LN on either radiological imaging or intra-operative assessment. Correlating with other studies, it appears that there is no objective method to detect lymph node metastases prior to thorough histological analysis,5,6,8 however there is little evidence for frozen section with previous studies recommending continuing resection along with completion zonal lymphadenectomy despite poor survival data.15,16 EUS was not available for the first 15 years of this study, but as use and expertise HPB 2017, -, 1–5
increase it may have a role in the future. Thus for patients with histologically positive LN, hepatic resection as planned regardless of suspicion of hepatic pedicle LN involvement is still recommended, however it may be that lymph node sampling techniques that detect LN involvement pre-resection, combined with modern chemotherapy, may offer patients similar outcomes. The hepatic pedicle lymph node chain was characterised into 5 groups, being hepatoduodenal ligament, retroportal, common hepatic, coeliac and para-aortic (at the level of the coeliac). Others have described the lymph drainage from the liver with the main lymph drainage along the right side of the hepatoduodenal ligament towards the retroportal/retropancreatic nodes and some drainage along the left side of the hepatoduodenal ligament and straight into the common hepatic nodes.8,17,18 A small amount of drainage can be identified as draining into the falciform ligament and to the anterior mediastinum, and some may cross from the retroportal nodes to the nodes draining the superior mesenteric artery. It is thus clear that there will be variation in the lymph drainage patterns from patient to patient which may explain why there is variation in which LN are found to be positive for metastatic disease.7,8,15,18 In this study, there were roughly even numbers of patients with positive nodes in the hepatoduodenal ligament, retroportal and common hepatic groups, with many patients in the common hepatic group not having known positive nodes in the other locations. Others have also noted “skip” LN involvement, thus disease free and survival outcome may be difficult to risk assess according to the location of enlarged or suspicious nodes.8,16,19,20 Jaeck’s group has suggested grouping LN into 2 areas and originally suggested that survival outcome was better if LN involvement was limited to distal nodes in area 1, however a later study by the same group found no difference,16,19 thus suggesting that the specific location of positive LN within the hepatic pedicle is irrelevant. Routine lymphadenectomy in at least high risk patients has been proposed by several authors with up to 28% of patients undergoing hepatectomy having occult LN metastases.3,6,8,11,12,16,19,21 Routine lymphadenectomy will be continued to be debated as there is yet no clear benefit in terms of survival, whether through a direct effect of tumour clearance or through adjusting management if used as a prognostic marker. There are few adverse effects of performing a lymphadenectomy, however vascular and biliary injuries have been described, as well as vastly increased difficulty of re-operation in the porta hepatis for recurrence.3 Median survival for metastatic colorectal cancer without hepatectomy is now approximately 24 months,10 and the patients in this study with positive hepatic pedicle LN had a similar survival thus justifying the common practice of not performing routine lymphadenectomy. LN disease was only detected if large enough to be seen on imaging or through intraoperative assessment and it is interesting to note that some studies that detect subclinical metastases through routine lymphadenectomy, particularly when using
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Hodgson R, et al., Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.01.025
HPB
5
immunohistochemistry, find that these patients can have similar survival to those without extrahepatic disease.3,12 Future chemotherapy regimens may be more effective against extrahepatic disease and thus detecting these patients earlier may be of benefit. It is possible that routine lymphadenectomy in at least high risk patients will become standard practice as the balance of risk shifts. The time of this, the extent of the lymphadenectomy and what constitutes a high risk patient are all yet to be determined. In conclusion, combined hepatectomy and hepatic pedicle lymphadenopathy for CRLM in the setting of radiologically or clinically suspicious or enlarged LN is justified on the basis that half of all suspected hepatic pedicle disease is benign on histology. Those with hepatic pedicle LN metastases do poorly when compared to those without extrahepatic disease, however detecting these patients pre-operatively is difficult. CT and MRI imaging detect less than half of histologically proven metastases, however PET possibly has a high positive predictive value although the numbers in this study are too small to make any further conclusions. There may be future roles for the use of frozen section and EUS, but these have not yet been defined. There is no current role for routine lymphadenectomy, however this is likely to change in the future where identifying patients with hepatic pedicle metastases can direct them towards aggressive therapy at an earlier stage.
examination can predict metastatic lymph node in the hepatic pedicle in patients with colorectal liver metastasis. Ann Surg Oncol 19: 163 – 168. 7. Elias D, Saric J, Jaeck D, Arnaud JP, Gayet B, Rivoire M et al. (1996) Prospective study of microscopic lymph node involvement of the hepatic pedicle during curative hepatectomy for colorectal metastases. Br J Surg 83:942–945. 8. Moszkowicz D, Cauchy F, Dokmak S, Belghiti J. (2012) Routine pedicular lymphadenectomy for colorectal liver metastases. J Am Coll Surg 214:e39–e45. 9. Liu W, Yan X-L, Wang K, Bao Q. (2014) The outcome of liver resection and lymphadenectomy for hilar lymph node involvement in colorectal liver metastases. Int J Colorectal Dis 29:737–745. 10. Kopetz S, Chang GJ, Overman MJ, Eng C, Sargent DJ, Larson DW et al. (2009) Improved survival in metastatic colorectal cancer is associated with adoption of hepatic resection and improved chemotherapy. J Clin Oncol 27:3677–3683. 11. Beckurts K, Hölscher A, Thorban S, Bollschweiler E, Siewert J. (1997) Significance of lymph node involvement at the hepatic hilum in the resection of colorectal liver metastases. Br J Surg 84:1081–1084. 12. Bennett J, Schmidt C, Klimstra D, Grobmyer S, Ishill N, D’Angelica M et al. (2008) Perihepatic lymph node micrometastases impact outcome after partial hepatectomy for colorectal metastases. Ann Surg Oncol 15: 1130–1136. 13. Laurent C, Sa Cunha A, Rullier E, Smith D, Rullier A, Saric J. (2004) Impact of microscopic hepatic lymph node involvement on survival after resection of colorectal liver metastasis. J Am Coll Surg 198: 884 –891. 14. Rodgers M, McCall J. (2000) Surgery for colorectal liver metastases with
Funding
hepatic lymph node involvement: a systematic review. Br J Surg 87: 1142–1155.
No funding source for this paper.
15. Elias D, Ouellet JF. (2003) Incidence, distribution and significance of hilar lymph node metastases in hepatic colorectal metastases. Surg
Conflicts of interest
Oncol Clin N Am 12:221–229. 16. Jaeck D, Nakano H, Bachellier P, Inoue K, Weber J-C, Oussoultzoglou E
None declared.
et al. (2002) Significance of hepatic pedicle lymph node involvement in References
patients with colorectal liver metastases: a prospective study. Ann Surg
1. Abdalla EK, Vauthey JN, Ellis LM, Ellis V, Pollock R, Broglio KR et al. (2004) Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver
Oncol 9:430–438. 17. Ito M, Mishima Y, Sato T. (1991) An anatomical study of the lymphatic drainage of the gallbladder. Surg Radiol Anat 13:89–104. 18. Kokudo N, Sato T, Seki M, Ohta H, Azekura K, Ueno M et al. (1999)
metastases. Ann Surg 239:818–825. 2. Carpizo D, D’Angelica M. (2009) Liver resection for metastatic colorectal cancer in the presence of extrahepatic disease. Ann Surg Oncol 16:
Hepatic lymph node involvement in resected cases of liver metastases from colorectal cancer. Dis Colon Rectum 42:1285–1291. 19. Oussoultzoglou E, Romain B, Panaro F, Rosso E, Pessaux P,
2411–2421. 3. Jaeck D, Oussoultzoglou E, Rosso E. (2007) Hepatectomy for colorectal
Bachellier P et al. (2009) Long-term survival after liver resection for
metastases in the presence of extrahepatic disease. Surg Oncol Clin N
colorectal liver metastases in patients with hepatic pedicle lymph nodes
Am 16:507–523.
involvement in the era of new chemotherapy regimens. Ann Surg 249:
4. Elias D, Ouellet JF, Bellon N, Pignon JP, Pocard M, Lasser P. (2003) Extrahepatic disease does not contraindicate hepatectomy for colorectal liver metastases. Br J Surg 90:567–574. 5. Grobmyer S, Wang L, Gonen M, Fong Y, Klimstra D, D’Angelica M et al. (2006) Perihepatic lymph node assessment in patients undergoing partial hepatectomy for malignancy. Ann Surg 244:260–264. 6. Rau C, Blanc B, Ronot M, Dokmak S, Aussilhou B, Faivre S et al.
879–886. 20. Adam R, de Haas RJ, Wicherts DA, Aloia TA, Delvart V, Azoulay D et al. (2008) Is hepatic resection justified after chemotherapy in patients with colorectal liver metastases and lymph node involvement? J Clin Oncol 26:3672–3680. 21. Agrawal S, Belghiti J. (2011) Oncologic resection for malignant tumors of the liver. Ann Surg 253:656–665.
(2011) Neither preoperative computed tomography nor intra-operative
HPB 2017, -, 1–5
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Hodgson R, et al., Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.01.025
http://dx.doi.org/10.1016/j.hpb.2017.01.025
ORIGINAL ARTICLE
Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified Russell Hodgson, Harsheet Sethi, Andrew H. Ling & Peter Lodge St James’s University Hospital, Beckett Street, Leeds LS9 7TF, West Yorkshire, UK
Abstract Background: The aim of this study was to describe the outcome of patients with colorectal liver metastases (CRLM) and radiological or clinical evidence of metastatic hepatic lymph node involvement who underwent combined hepatectomy and hepatic pedicle lymphadenectomy. Methods: Retrospective analysis of a prospectively maintained audit of 2082 patients undergoing liver resection for CRLM between 1994 and 2014. Age, type of resection, CT/MRI/PET detection, location, disease recurrence and survival were analysed. Results: Combined hepatectomy and hepatic pedicle lymphadenopathy was performed on 76 patients who met the inclusion criteria. 46% of enlarged lymph nodes were located in the hepatic ligament, with 38% retroportal, 38% common hepatic and 33% coeliac nodes. 50% of lymph node resections were positive for metastatic tumour. Pre-operative CT, MRI and CT/PET failed to detect histologically proven lymph node disease in 25/38 patients. Patients with negative nodal histology had a significant overall (44 vs 20 months, p = 0.008) and disease free (20 vs 11 months, p < 0.001) survival advantage. Conclusion: Combined hepatectomy and lymph node resection for CRLM in the setting of enlarged or suspicious lymphadenopathy is justified as imaging and operative findings are poor guides in determining positive lymph node disease. Received 12 November 2016; accepted 24 January 2017
Correspondence Russell Hodgson, Department of Surgery, Austin Health, Studley Rd, Heidelberg 3084, Victoria, Australia. E-mail: [email protected]
Introduction Liver resection for isolated colorectal metastases is well recognised as the gold standard of care with excellent long term outcomes with five year survival rates >50% now commonplace.1,2 The frontier for the treatment of colorectal metastases is now expanding to treating patients with additional extrahepatic metastases, the most prevalent of these being lung metastases which are also amenable to resection.2 The role of liver resection with extrahepatic lymph node (LN) disease in the hepatic pedicle continues to be debated.3–5 It appears the therapeutic value of Previous presentation: This paper is based on a free paper presented to the EA-HPBA meeting in Manchester on 24/4/2015: R Hodgson, H Sethi, JPA Lodge, Combined Hepatectomy and Hepatic Pedicle Lymph Node Resection in Colorectal Liver Metastases is Justified. EA-HPBA, Manchester UK, 24/4/2015.
HPB 2017, -, 1–5
routine lymphadenectomy is low if not non-existent, but there could be a good argument for lymphadenectomy for staging purposes as many micrometastases are missed by imaging and intra-operative assessment.5,6 The pattern of LN involvement in the hepatic pedicle is not consistent, however, and skipped LN appear to be common, necessitating thorough LN clearance which has not yet become common practice.7,8 The other clinical question raised is what to do with enlarged or suspicious LN detected on radiological imaging or intraoperative assessment. Results over the last 20 years have consistently shown poor survival and have led to some suggesting that LN involvement should be a contraindication to liver resection. However not all enlarged or suspicious LN are positive on histology.9 Median survival for metastatic colorectal cancer without hepatectomy is now up to 24 months,10 and it remains to be answered if chemotherapy can add further
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Hodgson R, et al., Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.01.025
HPB
2
gains to patients undergoing hepatectomy and hepatic pedicle lymphadenectomy. The aim of this study was to assess the outcome of patients undergoing combined hepatectomy and LN resection in the setting of clinically enlarged or suspicious LN within the liver’s main lymphatic drainage.
Methods All patients undergoing liver resection for colorectal metastases at St James’s University Hospital, Leeds, UK between the 1st of January 1994 and the 31st of December 2014 were entered into a prospectively maintained database. Data recorded included gender, date of operation, age at operation, type of resection, lymph node group (hepatoduodenal ligament, retroportal, common hepatic, coeliac or para-aortic adjacent to coeliac), R status of resection (R1 < 1 mm clearance of hepatic disease), and number of tumours. Medical records were sourced for pre-operative reports of CT, MRI and PET scans. PET scans were used only in the last 10 years of the study and only when there was increased suspicion of extrahepatic disease. Lymph nodes were defined as enlarged if described as enlarged or suspicious by the reporting radiologists, typically if >1 cm in size, or increased uptake of FDG (PET). Suspicious lymph nodes were defined by individual surgeons at their discretion, typically if either enlarged, firm or tethered to surrounding structures. Lymph nodes with extranodal extension into main structures (such as hepatic arteries or bile ducts) precluded resection. Lymph node resection was not standardised and was performed according to individual surgeon’s preferences, either with en bloc resection of lymph node zones or resection of individual nodes. All specimens were reviewed by pathologists with experience in HPB pathology. Post-operatively, all patients were entered into an intense surveillance programme with LFT, CEA, CA19-9 and chest, abdominal and pelvis CT at 3, 6, 12, 18, 24 months then years 3, 4, 5, 7 and 10 and clinical exam and bloods at years 6, 8 and 9. Due to this intensive surveillance, all episodes of disease recurrence together with the outcome were recorded. Survival data up to 30/6/2016 including causes of death were obtained from hospital records and by cross referencing this data with that held by the Northern and Yorkshire Cancer Registry Information Service. Statistical analysis was performed using SPSS (IBM, New York, USA). Statistical significance was calculated using the log rank (Mantel–Cox) method and Fisher’s exact test where appropriate. A p-value of <0.05 was considered statistically significant.
metastases. From this cohort, 76 patients (4%) who had concomitant hepatic pedicle LN resection for enlarged or suspicious nodes were identified. There were no significant differences between those who underwent combined liver and nodal resection vs liver resection alone with regard to age, gender, number of tumours and R status (data not shown). In the LN excision group 38 (50%) patients were found to have positive histology. Demographics, tumour, operative characteristics, and detection on radiological imaging by nodal histology are shown in Table 1. Disease-free and overall survival The median (range) followup was 28 (0–214) months. At the end of the study period, 15 (20%) patients were alive with no disease, 7 (9%) were alive with disease and 54 (71%) had died. The LN excision group had a significantly worse disease free survival compared to the total population of patients undergoing hepatic resection for CRLM as shown in Fig. 1 with 1-year, 3-year
Table 1 Demographics of patients who underwent concomitant LN
resection with either positive or negative histology Variable
Histology negative
Histology positive
Significance
Number of patients
38
38
p = ns
Age, median (range)
60 (25–80)
63 (36–83)
p = ns
23
19
p = ns
2 (1–7)
3 (1–10)
p = ns
R0
25
21
R1
12
15
R2
1
2
Gender Male Number of tumours, median (range) R status
p = ns
Type of resection
p = ns
Minor liver resection (<3 segments)
15
15
Major liver resection (3 segments)
23
23
Radiologically enlarged/suspicious LN CT
2/37
7/38
p = 0.153
MRI
5/35
13/36
p = 0.055
PET
0/7
4/4
p < 0.003
Hepatoduodenal ligament
16
17
Retroportal
7
12
Results
Common hepatic artery
10
18
Patient demographics From the 1st of January 1994 until the 31st of December 2014 there were 2082 liver resections performed for colorectal liver
Coeliac
16
8
Para-aortic (at level of coeliac)
2
2
HPB 2017, -, 1–5
LN region
p = ns
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Hodgson R, et al., Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.01.025
HPB
3
The disease free survival of the negative histology group mirrored the total population and was significantly better than that of the positive histology group as shown in Fig. 3 with 1year, 3-year and 5-year survivals of 63%, 44% and 44%, compared with 46%, 16% and 12% respectively (p = 0.008). There was a similar pattern in overall survival as shown in Fig. 4 with a significant increased survival in the negative histology group with 1-year, 3-year and 5-year survivals of 92%, 70% and 55%, compared with 78%, 20% and 13% in the positive histology group (p < 0.001).
Discussion
Figure 1 Disease free survival of all patients undergoing hepatectomy
for CRLM compared with patients undergoing hepatectomy and lymphadenectomy
and 5-year survivals of 55%, 30% and 29%, and 78%, 60% and 54% respectively (p < 0.001). This pattern was similarly observed for overall survival as shown in Fig. 2 with 1-year, 3-year and 5year survivals of 85%, 44% and 31%, and 89%, 64% and 53% respectively (p < 0.001).
The incidence of hepatic pedicle LN involvement has been reported to be up to 28% in patients undergoing hepatectomy for colorectal liver metastases.11 In this study, only 76 patients from a total population of 2082 were identified using radiologically or through intra-operative assessment, and only 36 (2%) had positive histology. Thus if this reported incidence of up to 28% was assumed for the 2082 patients undergoing hepatectomy for CRLM in this study, the vast majority of LN metastases were not detected. Additionally, of those suspected in this study, only 50% were positive. This poses two questions: what should be done when enlarged or suspicious nodes are encountered, and should hepatic pedicle lymphadenectomy during hepatectomy for colorectal liver metastases be performed routinely? Most of the patients in this study were investigated preoperatively with CT (75/76) and MRI (73/76), with PET scanning increasing in prevalence in the latter years of the study.
Figure 2 Overall survival of all patients undergoing hepatectomy for
Figure 3 Disease free survival comparison of positive and negative LN
CRLM compared with patients undergoing hepatectomy and
histology in those undergoing hepatectomy and lymphadenectomy for
lymphadenectomy
CRLM
HPB 2017, -, 1–5
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Hodgson R, et al., Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.01.025
HPB
4
Figure 4 Overall survival comparison of positive and negative LN
histology in those undergoing hepatectomy and lymphadenectomy for CRLM
While the positive predictive values of CT and MRI were individually above 50%, combined they failed to detect 25/38 (66%) patients with histologically positive nodes. This correlates with the study performed by Rau et al., which found that CT had only a positive predictive value of 56%.6 PET scanning however, while performed less frequently, appeared to accurately determine positive nodes with a 100% positive predictive value, albeit with only 11 scans performed in selected patients in this series. If there was convincing evidence to not offer surgery to those with hepatic pedicle LN disease, there may be an argument for routinely performing PET scans on all potential hepatectomy patients, however more data are still required. When analysing outcomes for those with positive LN histology, disease free (12% at 5 years) and overall survival (13% at 5 years) is dismal, which is consistent with published findings.4,11–13 The outcome has slightly improved with only isolated survivors noted in the pre-2000 literature.14,15 Within the current series 23/38 patients with positive histology, however, were not identified on pre-operative imaging and were only detected at intraoperative assessment, suggesting that the surgeon is better able to recognise extrahepatic disease. This study only had a positive histology rate of 50% after resection of suspicious LN on either radiological imaging or intra-operative assessment. Correlating with other studies, it appears that there is no objective method to detect lymph node metastases prior to thorough histological analysis,5,6,8 however there is little evidence for frozen section with previous studies recommending continuing resection along with completion zonal lymphadenectomy despite poor survival data.15,16 EUS was not available for the first 15 years of this study, but as use and expertise HPB 2017, -, 1–5
increase it may have a role in the future. Thus for patients with histologically positive LN, hepatic resection as planned regardless of suspicion of hepatic pedicle LN involvement is still recommended, however it may be that lymph node sampling techniques that detect LN involvement pre-resection, combined with modern chemotherapy, may offer patients similar outcomes. The hepatic pedicle lymph node chain was characterised into 5 groups, being hepatoduodenal ligament, retroportal, common hepatic, coeliac and para-aortic (at the level of the coeliac). Others have described the lymph drainage from the liver with the main lymph drainage along the right side of the hepatoduodenal ligament towards the retroportal/retropancreatic nodes and some drainage along the left side of the hepatoduodenal ligament and straight into the common hepatic nodes.8,17,18 A small amount of drainage can be identified as draining into the falciform ligament and to the anterior mediastinum, and some may cross from the retroportal nodes to the nodes draining the superior mesenteric artery. It is thus clear that there will be variation in the lymph drainage patterns from patient to patient which may explain why there is variation in which LN are found to be positive for metastatic disease.7,8,15,18 In this study, there were roughly even numbers of patients with positive nodes in the hepatoduodenal ligament, retroportal and common hepatic groups, with many patients in the common hepatic group not having known positive nodes in the other locations. Others have also noted “skip” LN involvement, thus disease free and survival outcome may be difficult to risk assess according to the location of enlarged or suspicious nodes.8,16,19,20 Jaeck’s group has suggested grouping LN into 2 areas and originally suggested that survival outcome was better if LN involvement was limited to distal nodes in area 1, however a later study by the same group found no difference,16,19 thus suggesting that the specific location of positive LN within the hepatic pedicle is irrelevant. Routine lymphadenectomy in at least high risk patients has been proposed by several authors with up to 28% of patients undergoing hepatectomy having occult LN metastases.3,6,8,11,12,16,19,21 Routine lymphadenectomy will be continued to be debated as there is yet no clear benefit in terms of survival, whether through a direct effect of tumour clearance or through adjusting management if used as a prognostic marker. There are few adverse effects of performing a lymphadenectomy, however vascular and biliary injuries have been described, as well as vastly increased difficulty of re-operation in the porta hepatis for recurrence.3 Median survival for metastatic colorectal cancer without hepatectomy is now approximately 24 months,10 and the patients in this study with positive hepatic pedicle LN had a similar survival thus justifying the common practice of not performing routine lymphadenectomy. LN disease was only detected if large enough to be seen on imaging or through intraoperative assessment and it is interesting to note that some studies that detect subclinical metastases through routine lymphadenectomy, particularly when using
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Hodgson R, et al., Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.01.025
HPB
5
immunohistochemistry, find that these patients can have similar survival to those without extrahepatic disease.3,12 Future chemotherapy regimens may be more effective against extrahepatic disease and thus detecting these patients earlier may be of benefit. It is possible that routine lymphadenectomy in at least high risk patients will become standard practice as the balance of risk shifts. The time of this, the extent of the lymphadenectomy and what constitutes a high risk patient are all yet to be determined. In conclusion, combined hepatectomy and hepatic pedicle lymphadenopathy for CRLM in the setting of radiologically or clinically suspicious or enlarged LN is justified on the basis that half of all suspected hepatic pedicle disease is benign on histology. Those with hepatic pedicle LN metastases do poorly when compared to those without extrahepatic disease, however detecting these patients pre-operatively is difficult. CT and MRI imaging detect less than half of histologically proven metastases, however PET possibly has a high positive predictive value although the numbers in this study are too small to make any further conclusions. There may be future roles for the use of frozen section and EUS, but these have not yet been defined. There is no current role for routine lymphadenectomy, however this is likely to change in the future where identifying patients with hepatic pedicle metastases can direct them towards aggressive therapy at an earlier stage.
examination can predict metastatic lymph node in the hepatic pedicle in patients with colorectal liver metastasis. Ann Surg Oncol 19: 163 – 168. 7. Elias D, Saric J, Jaeck D, Arnaud JP, Gayet B, Rivoire M et al. (1996) Prospective study of microscopic lymph node involvement of the hepatic pedicle during curative hepatectomy for colorectal metastases. Br J Surg 83:942–945. 8. Moszkowicz D, Cauchy F, Dokmak S, Belghiti J. (2012) Routine pedicular lymphadenectomy for colorectal liver metastases. J Am Coll Surg 214:e39–e45. 9. Liu W, Yan X-L, Wang K, Bao Q. (2014) The outcome of liver resection and lymphadenectomy for hilar lymph node involvement in colorectal liver metastases. Int J Colorectal Dis 29:737–745. 10. Kopetz S, Chang GJ, Overman MJ, Eng C, Sargent DJ, Larson DW et al. (2009) Improved survival in metastatic colorectal cancer is associated with adoption of hepatic resection and improved chemotherapy. J Clin Oncol 27:3677–3683. 11. Beckurts K, Hölscher A, Thorban S, Bollschweiler E, Siewert J. (1997) Significance of lymph node involvement at the hepatic hilum in the resection of colorectal liver metastases. Br J Surg 84:1081–1084. 12. Bennett J, Schmidt C, Klimstra D, Grobmyer S, Ishill N, D’Angelica M et al. (2008) Perihepatic lymph node micrometastases impact outcome after partial hepatectomy for colorectal metastases. Ann Surg Oncol 15: 1130–1136. 13. Laurent C, Sa Cunha A, Rullier E, Smith D, Rullier A, Saric J. (2004) Impact of microscopic hepatic lymph node involvement on survival after resection of colorectal liver metastasis. J Am Coll Surg 198: 884 –891. 14. Rodgers M, McCall J. (2000) Surgery for colorectal liver metastases with
Funding
hepatic lymph node involvement: a systematic review. Br J Surg 87: 1142–1155.
No funding source for this paper.
15. Elias D, Ouellet JF. (2003) Incidence, distribution and significance of hilar lymph node metastases in hepatic colorectal metastases. Surg
Conflicts of interest
Oncol Clin N Am 12:221–229. 16. Jaeck D, Nakano H, Bachellier P, Inoue K, Weber J-C, Oussoultzoglou E
None declared.
et al. (2002) Significance of hepatic pedicle lymph node involvement in References
patients with colorectal liver metastases: a prospective study. Ann Surg
1. Abdalla EK, Vauthey JN, Ellis LM, Ellis V, Pollock R, Broglio KR et al. (2004) Recurrence and outcomes following hepatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver
Oncol 9:430–438. 17. Ito M, Mishima Y, Sato T. (1991) An anatomical study of the lymphatic drainage of the gallbladder. Surg Radiol Anat 13:89–104. 18. Kokudo N, Sato T, Seki M, Ohta H, Azekura K, Ueno M et al. (1999)
metastases. Ann Surg 239:818–825. 2. Carpizo D, D’Angelica M. (2009) Liver resection for metastatic colorectal cancer in the presence of extrahepatic disease. Ann Surg Oncol 16:
Hepatic lymph node involvement in resected cases of liver metastases from colorectal cancer. Dis Colon Rectum 42:1285–1291. 19. Oussoultzoglou E, Romain B, Panaro F, Rosso E, Pessaux P,
2411–2421. 3. Jaeck D, Oussoultzoglou E, Rosso E. (2007) Hepatectomy for colorectal
Bachellier P et al. (2009) Long-term survival after liver resection for
metastases in the presence of extrahepatic disease. Surg Oncol Clin N
colorectal liver metastases in patients with hepatic pedicle lymph nodes
Am 16:507–523.
involvement in the era of new chemotherapy regimens. Ann Surg 249:
4. Elias D, Ouellet JF, Bellon N, Pignon JP, Pocard M, Lasser P. (2003) Extrahepatic disease does not contraindicate hepatectomy for colorectal liver metastases. Br J Surg 90:567–574. 5. Grobmyer S, Wang L, Gonen M, Fong Y, Klimstra D, D’Angelica M et al. (2006) Perihepatic lymph node assessment in patients undergoing partial hepatectomy for malignancy. Ann Surg 244:260–264. 6. Rau C, Blanc B, Ronot M, Dokmak S, Aussilhou B, Faivre S et al.
879–886. 20. Adam R, de Haas RJ, Wicherts DA, Aloia TA, Delvart V, Azoulay D et al. (2008) Is hepatic resection justified after chemotherapy in patients with colorectal liver metastases and lymph node involvement? J Clin Oncol 26:3672–3680. 21. Agrawal S, Belghiti J. (2011) Oncologic resection for malignant tumors of the liver. Ann Surg 253:656–665.
(2011) Neither preoperative computed tomography nor intra-operative
HPB 2017, -, 1–5
© 2017 International Hepato-Pancreato-Biliary Association Inc. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Hodgson R, et al., Combined hepatectomy and hepatic pedicle lymphadenectomy in colorectal liver metastases is justified, HPB (2017), http://dx.doi.org/10.1016/j.hpb.2017.01.025