R. Mantke a, *, U. Schmidt a,b, S. Wolff a, R. Kube a, H. Lippert a

Transcription

1 Available online at EJSO 38 (2012) 259e265 Incidence of synchronous liver metastases in atients with colorectal cancer in relationshi to clinico-athologic characteristics. Results of a German rosective multicentre observational study R. Mantke a, *, U. Schmidt a,b, S. Wolff a, R. Kube a, H. Liert a a Deartment of Surgery, Institute for Quality Control in Surgery, Otto-von-Guericke-University of Magdeburg, Germany b StatConsult Magdeburg, Germany Acceted 12 December 2011 Available online 29 December 2011 Abstract Background: The aim of this rosective observational multicentre study was to evaluate the incidence of synchronous liver metastases in colon and rectal cancer and to determine clinico-athologic factors of the colorectal cancer that influenced the develoment of synchronous liver metastases. Methods: Of 48,894 atients with colorectal cancer and who underwent surgery between January 2000 and December 2004, 7209 develoed heatic metastases and were analyzed. Results: Synchronous liver metastases occurred in 14.7% of the colorectal cancer cases. Colon cancer (15.4%) led significantly more frequently to haematogenous sread to the liver than rectal cancer (13.5%) in a univariate aroach. The N, V, and T stage, as well as the number of metastatic-involved local lymh nodes indeendently influenced the frequency of synchronous liver metastases in colon and rectal cancer in a multivariate analysis. Localization of the cancer in the colon led to a different number of synchronous liver metastases. Localization of the rectal cancer did not influence the rate of synchronous liver metastases. In the case of synchronous liver metastases, atients with colon cancer had significantly more eritoneal metastases (17.9 vs. 9.15%) but less lung (9.7 vs. 14%) and bone (0.7 vs. 1.6%) metastases. Simultaneous curative liver resections were done in 7% of colon cancer cases and in 8.8% of rectal cancer cases. Conclusion: In this national study the incidence of synchronous liver metastases in colon and rectal cancer were different. Indeendent factors leading to synchronous liver metastases could be identified. Venous infiltration seems to be imortant for the develoment of distant metastases. Ó 2011 Elsevier Ltd. All rights reserved. Keywords: Liver metastases; Colo-rectal cancer Introduction The liver is the most common site for colorectal distant cancer metastases. 1,2 Desite rogress in the quality of diagnosis, heatic metastases are discovered in many atients when they are diagnosed with colorectal cancer. 3e7 Substantial rogress has been made in the treatment strategies for these atients in recent years. 8e10 Most * Corresonding author. Klinik fuer Allgemein- und Viszeralchirurgie, Staedtisches Klinikum Brandenburg, Hochstrasse 29, D Brandenburg, Germany. Tel.: þ ; fax: þ address: mantke@klinikum-brandenburg.de (R. Mantke). studies of heatic metastases have only included atients from secialized academic centres. Their results are therefore limited by recruitment bias. Very few oulation-based studies have been conducted on heatic metastases. 1,11e13 Studies analyzing the clinico-athologic characteristics of atients with synchronous liver metastases for colon and rectal cancer searately using a large multicentre aroach were missing from the literature. The urose of the current study was to analyze the incidence of synchronous liver metastases in colon and rectal cancer in a large oulation-based rosective multicentre observational study and to determine clinico-athologic factors of the colorectal cancer that influenced the develoment of synchronous liver metastases /$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi: /j.ejso

2 260 R. Mantke et al. / EJSO 38 (2012) 259e265 Methods A total of 48,894 atients with curative or alliative surgically treated colorectal cancer were included in this rosective multicentre observational study from 1 January 2000 to 31 December Patient data were recorded rosectively into a standard form by the rimary surgeon. Overall, data were included from 346 hositals of all categories from all arts of Germany. This study was an otional quality assurance study triggered and managed by the Institute for Quality Control in Surgery of the Ottovon-Guericke-University of Magdeburg, Germany. Diagnoses of liver metastases were done by ultrasound, CT, MRI or during surgery. Statistical analysis was carried out with the statistical analysis ackage SPSS (version 16) in collaboration with statistical rofessionals (StatConsult, Magdeburg, Germany). Univariable analyses were erformed with the global chi-square test. A -value <0.05 was considered statistically significant. Initially, rognostic factors for synchronous liver metastases were investigated with a univariable aroach. Thereafter, a logistic regression model was used for a multile analysis of these otential rognostic factors. Relative risks were dislayed with 95 er cent confidence intervals. The study was conducted in accordance with the Declaration of Helsinki on Biomedical Research. As articiation was voluntary (written informed consent) and data recording done on an anonymous basis, and because the observational study had no influence on the theraeutic rocedure alied, no aroval by an ethics committee was required. Results Demograhics and atient characteristics Of 31,341 atients with colon cancer, 4831 had synchronous liver metastases at the time of admission. From 17,553 atients suffering from rectal cancer, 2378 had synchronous liver metastases. The frequency of all distant metastases and liver metastases of colorectal cancers did not differ over the course of four observation years. We found distant metastases including liver metastases in 19.7% (2000), 18.4% (2001), 18.6% (2002), 18.3% (2003) and 18.4% (2004). Synchronous liver metastases were detected in 15.5% (2000), 14.8% (2001), 14.4% (2002), 14.4% (2003) and 14.6% (2004). About two thirds of the distant metastases were liver metastases. Rectal cancers had significantly less distant metastases (16.5% vs. 19.9%) and less liver metastases (13.5% vs. 15.4%) than colon cancers in a univariate aroach. Liver metastases aeared more frequently in men with colon and rectal cancers than in women (Table 1). Patients with colon cancer who were younger than 40 years develoed liver metastases less often (11.9%) than atients in the 40e70 years age range (16.9%). Patients with rectal cancer who were younger than 40 years develoed liver metastases more often (15.7%) than the older atients (not significant, Table 1). Overall, 7% of atients with colon cancer and 8.8% of atients with rectal cancer and synchronous liver metastases were oerated curatively (simultaneous resection of the tumour and the liver metastases) (Table 1). Patients with liver metastases from colon cancer had more general morbidity than atients without liver metastases (Table 1). This difference could not be demonstrated for rectal cancer. In 17.9% of atients with colon cancer and synchronous liver metastases, eritoneal metastases were also resent (Table 1). In contrast, only 9.1% of atients with rectal cancer and synchronous liver metastases had eritoneal carcinomatosis. Distant lymh node metastases in addition to synchronous liver metastases occurred in 5.5% of colon and in 4.6% of rectal cancer cases (n.s.). For rectal cancer with synchronous liver metastases, lung and bone metastases were more frequent than for colon cancer (Table 1). Histoathological findings and liver metastases We analyzed the relationshi between synchronous liver metastases and tumour stage according to the UICC classification, 6th edition (T, N, M, L and V categories). With increasing T category of the colon or rectum carcinoma, the risk of heatic metastases continuously rises ( < 0.001) (Table 2). Colon carcinomas have more liver metastases than rectum carcinomas in category T2 (Table 2). In the T3 category, rectum carcinomas with 16.4% have more liver metastases than colon carcinomas (14.9%). The involvement of one to three lymh nodes (N1) and the involvement of four or more lymh nodes (N2) increases the risk of simultaneous liver metastases significantly (Table 2). N1 and N2 stages in colon cancer had more synchronous liver metastases than in rectal cancer (Table 3). Higher grading increased the risk of synchronous liver metastases significantly (Table 2). A G2 differentiated tumour was more frequent for colon cancer. In contrast, a G4 differentiated tumour was more frequent for rectal cancer (Table 2). Patients with a documented L0 category had the lowest frequency of synchronous liver metastases (Table 2). L1 staged atients had significantly more liver metastases. Patients in which no L stage was described in the athologic findings had significantly more synchronous liver metastases than L0 staged atients. This suggests that a non-documented L stage (LX) is not equivalent to an L0 stage. The atients with a documented V0 category had the lowest frequency of synchronous liver metastases (Table 2). V1 staged atients had significantly more liver metastases. Patients in which no V stage was described in the athologic findings had significantly more synchronous liver metastases than V0 staged atients. This suggests that a non-documented V stage (VX) is not equivalent to a V0 stage.

3 R. Mantke et al. / EJSO 38 (2012) 259e Table 1 Characterization of atients with colon and rectal cancer. Relationshi between sex, age, and frequency of synchronous liver metastases of colon and rectal cancer. There are significant differences between the sexes. Significant differences are observed between the three age grous in colon or rectal cancer as well as between colon and rectal cancer within the selected age grou. Colon cancer: <40 vs. >40 ¼ n.s.; Rectal cancer: <40 vs. >40 ¼ n.s. Tye of surgery in atients with colon cancer or rectal cancer and synchronous liver metastases (all cases: resection of the rimary cancer). Relationshi between synchronous liver metastases of colon and rectal cancer, and at least one general morbidity. There are significant differences between atients with and without liver metastases in colon cancer. Frequency of metastases of colon and rectal cancer in addition to synchronous liver metastases. There are significant differences between colon and rectal cancer. Sex Colon cancer Rectal cancer n Liver metastases n Liver metastases Male % % Female % % Age Colon cancer Rectal cancer -Value colon vs. n Liver metastases n Liver metastases rectal cancer < % % e % % <0.001 > % % <0.001 Tye of surgery Colon cancer Rectal cancer n [%] n [%] Curative Palliative Colon cancer Rectal cancer Liver metastases n General morbidity n General morbidity Yes % ,7% No % ,3% < ¼ Colon cancer Rectal cancer -value colon vs. rectal cancer. Localization (other distant metastases) n (4831) Metastases (100%) n (2378) Metastases (100%) Peritoneal % % <0.001 Lung % % <0.001 Bone % % <0.001 Ovarian % % Skin % % Brain % 6 0.3% Distant lymh nodes % % Other % % Location of colon cancer and liver metastases Analyses of the localization of the colon cancer showed differences in the frequency of synchronous liver metastases (Table 3). Aendix cancer had the lowest rate of liver metastases (5.9%). Sigmoid cancer and the cancer of the slenic flexure had the highest rate of liver metastases (16.4 and 17.4%, resectively). Localization of the rectal cancer had no effect on the frequency of simultaneous liver metastases (Table 3). Influencing factors in statistical analysis We used multile logistic regression analysis to identify influencing factors for synchronous liver metastases for colon and rectal cancer searately (Tables 4 and 5). Lymh node involvement, the number of metastatic lymh nodes, L category, V category, and T category were indeendent factors for the develoment of synchronous liver metastases of colon cancer (Table 4). Lymh node involvement, the number of metastatic lymh nodes, V category, G category and T category were indeendent factors for the develoment of synchronous liver metastases of rectal cancer (Table 5). Discussion Colorectal cancer is one of the most common cancers worldwide. 14 Eidemiological data for cancer in general are very imortant for lanning and interreting scientific studies.

4 262 R. Mantke et al. / EJSO 38 (2012) 259e265 Table 2 Histoathological characterization: Relationshi between tumour category, nodal stage, grading, L category, V category and frequency of synchronous liver metastases of colon and rectal cancer. Significant differences between the categories in colon or rectal cancer. Colon cancer: L0 vs. LX und L1 vs. LX ¼ < 0.001; Rectal cancer: L0 vs. LX und L1 vs. LX ¼ < 0.001; Colon cancer: V0 vs. VX und V1 vs. VX ¼ < 0.001; Rectal cancer: V0 vs. VX und V1 vs. VX ¼ < Colon cancer Rectal cancer -value colon vs. n Liver metastases n Liver metastases rectal cancer T stage T % % T % % <0.001 T % % T % % N stage N % % <0.001 N % % <0.001 N % % <0.001 Grading G % % G % % <0.001 G % % G % % L stage L % % <0.001 L % % <0.001 LX % % V stage V % % <0.001 V % % <0.001 VX % % The data from multicentre trials that include several tyes of hositals gives a much more realistic view of the treatment situation than single-centre studies, which are limited by recruitment bias. Most ublished data focused Table 3 Relationshi between localization of the cancer in the colon or rectum and frequency of synchronous liver metastases. n Liver metastases [%] -value Localization Aendix <0.001 Coecum n.s. Colon ascendens Flexura heatica n.s. Colon transversum n.s. Flexura lienalis Colon descendens n.s. Colon sigmoideum <0.001 Rectal cancer localization <4 cm e7.9 cm e11.9 cm e16 cm n.s. Table 4 Histoathological findings of colon cancer associated with synchronous liver metastases. Univariable and multivariable logistic regression analysis (27,871 comlete cases, including 3949 with synchronous liver metastases, 2000e2004). Variables n (%) (100.0) univariate multivariate N category < < N0 15,438 (55.4) 1 N1 12,433 (44.6) 3.4 N (number of metastatic nodes) < < to3 22,095 (79.3) (20.7) 2.2 L category < < L0 11,509 (41.3) 1 L (27.8) 1.24 LX 8617 (30.9) 1.03 L0 vs. L1 < L0 vs. LX V category < < V0 15,032 (53.9) 1 V (9.9) 2.8 VX 10,072 (36.1) 1.4 V0 vs. V1 < V0 vs. VX G category < G1 þ G2 21,511 (77.2) 1 G3 þ G (22.8) 0.96 T category < < T (7.9) 1 T (14.6) 2.8 T3 16,987 (60.9) 6.2 T (16.6) 8.6 T1 vs. T2 < < T1 vs. T3 < < T1 vs. T4 < < Odds ratio on liver metastases are single-centre studies that reort the results of heatic resections for colorectal cancer metastases with a limited number of cases and sometimes with a very long recruitment time. 15,16 Very few oulation-based studies have been conducted on heatic metastases. 1,11e13 To our knowledge, multicentre studies analyzing the clinico-athologic characteristics of atients with synchronous liver metastases for colorectal cancer were missing from the literature. We resent a large oulation-based analysis of atients with synchronous liver metastases (48,894 atients with colorectal cancer). Manfredi et al. ublished data over the eriod from 1976 to 2000 from 13,463 atients with colorectal cancer in France. 17 They found that 14.5% had synchronous liver metastases. We found that 14.7% had liver metastases at the time of surgery for colorectal cancer. Interestingly, the rogress in diagnosis since this French study did not increase the frequency of diagnosed metastases during colorectal cancer surgery. Additionally we did not find any change in the frequency of synchronous metastases from

5 R. Mantke et al. / EJSO 38 (2012) 259e Table 5 Histoathological findings of rectal cancer associated with synchronous liver metastases. Univariable and multivariable logistic regression analysis (14,392 comlete cases, including 1721 with synchronous liver metastases, 2000e2004). Variables n (%) 14,392 (100.0) univariate multivariate N category < < N (55.9) 1 N (44.1) 3.2 N (number of metastatic nodes) < < to3 11,320 (78.6) (21.4) 2.2 L category < L (42.2) 1 L (27.7) 1.14 LX 4327 (30.1) 1.04 L0 vs. L L0 vs. LX V category < < V (54.0) 1 V (10.8) 3.0 VX 5058 (35.1) 1.4 V0 vs. V1 < V0 vs. VX G category < G1 þ G2 11,672 (81.1) 1 G3 þ G (18.9) 1.2 T category < < T (8.7) 1 T (27.6) 1.8 T (56.2) 6.2 T (7.5) 8.8 T1 vs. T2 < < T1 vs. T3 < < T1 vs. T4 < < Odds ratio 2000 to Data from Australia showed a frequency of simultaneous liver metastases of 18e19.4%. 11,13 Leorrier et al. described a rate of 18.8% synchronous liver metastases. 14 In our study, colon cancer led significantly more frequently to haematogenous sread into the liver than rectal cancer (15.4 vs. 13.5%). This fact has never before been significantly described in oulation-based studies. 1,11e14,17,18 Of course, this fact was shown only in a univariate aroach. The reason could be the difference in T categories between rectal and colon cancer. We found a T3/T4 category in 77.2% in colon cancer and in 61.7% in rectal cancer. The higher number of T3/T4 categories in colon cancer robably led to a higher frequency of liver metastases. We demonstrated in a multivariable aroach that besides the T and N category of the colorectal cancer, the V category is most relevant for the develoment of synchronous liver metastases. The V category alone does not usually lead to an adjuvant theray. About 20e30% of the atients will develo liver metastases and could rofit from an adjuvant theray based on the V category. In a relevant number of atients, the athologist did not note the V category (VX, 62% colon, 65% rectal cancer). These atients clearly had more synchronous liver metastases than V0 staged cases. The relationshi between V category and metastatic colorectal cancer has already been reorted in several athologic studies. 19e21 The number of analyzed atients was always small. Using H þ E stain alone for identifying a venous invasion is ineffective. 19 The additional use of an elastic fibre stain could increase the registration of venous invasion from 38.5 to 70.4% in stage IV colorectal cancer. 19 For that reason, athologists should ay more attention to sufficient venous staging of colorectal cancer atients. Further investigations should analyze how many atients with colorectal cancer UICC stage I or II had venous infiltration and missed adjuvant theray. Because of the high number of colon cancer cases included, our study shows more synchronous liver metastases from cancers at the flexura lienalis and at the colon sigmoideum. Because of the small differences in frequency of simultaneous liver metastases in relation to the localization of the colon cancer, we see no imact of this finding at this time from a clinical ersective. Cancer of the aendix vermiformis clearly had the smallest number of liver metastases (5.9%). Cancer of the aendix is rare and eritoneal carcinomatosis, esecially in mucinous aendix cancer, is more common than liver metastases. 22e25 Data on synchronous liver metastases in aendix cancer have never been ublished. On the other hand, localization for rectal cancer did not influence the rate of synchronous liver metastases. In the case of synchronous liver metastases, atients with colon cancer had significantly more eritoneal metastases (17.9 vs. 9.15%) but less lung (9.7 vs. 14%) and less bone (0.7 vs. 1.6%) metastases than rectal cancer. This is likely because higher stages of colon cancer have easier access to the eritoneal cavity than rectal cancers. Higher stages of rectal cancer can based on the venous drainage directly metastasized into distant organs. The ostoerative in-hosital morbidity for colon cancer alone was higher in atients with synchronous liver metastases (27.9%) comared to atients without liver metastases (22.9%). The morbidity rate in rectal cancer is usually higher than in colon cancer. On the other hand, the number of acute oerations in colon surgery is usually higher than in rectal surgery. The morbidity range of surgery for rectal cancer varies from 24 to 35% in several studies and age seems to be a relevant cause for high morbidity. 26e29 Some authors have reorted that simultaneous resection of liver metastases is as safe as metachronous resection. 30,31 In secialized heatic surgery centres 18e36% receive simultaneous resections. 32,33 Because colorectal surgery is done in a variety of tyes of hositals the rates of simultaneous liver resection with an intent of curing colorectal heatic metastases were quite low in our study (colon 7% and rectal cancer 8.8%)

6 264 R. Mantke et al. / EJSO 38 (2012) 259e265 comared to data from secialized centres. In contrast, other authors recently showed that simultaneous resection of liver metastases led to a higher 90-day mortality rate based on administrative Medicare data. 34 The rate of synchronous resections of liver metastases was about 32% of all atients with colorectal cancer and liver metastases in this study. To our knowledge the administrative Medicare data could not differentiate between curative resections and resections done only to diagnose the metastases during colorectal surgery (diagnostic, alliative intent). This is a limitation of this study. Patients with resectable liver metastases were transferred to secialized liver centres after successful colorectal surgery for secondary liver resection in our study too. This two-ste surgery for atients with colorectal cancer and resectable liver metastases is wellestablished in Germany. The imlementation of secialized colorectal cancer centres in the last decade will lead to a higher rate of synchronous and secondary liver resections and should be the focus of further investigations. Our study has limitations too, given that the data are not controlled. To reach the highest documentation quality, all data were roofed for lausibility during the documentation rocess into the database. Since the study was rosective, the quality of documented data was excellent. Also, because this study was a rosective multicentre observational study including such a large number of atients the data are very reresentative. In this national study the incidence of synchronous liver metastases in colon and rectal cancer were different. Indeendent factors leading to synchronous liver metastases could be identified. Venous infiltration seems to be imortant for develoment of distant metastases. The attention given to diagnosing venous infiltration should be increased. Further studies are necessary to investigate the imact of venous infiltration on long-term survival. Acknowledgements The authors thank all enlisted hositals and their staff for excellent cooeration. Conflict of interest statement The author s declare that there is no conflict of interest. References 1. Cummings LC, Payes JD, Cooer GS. Survival after heatic resection in metastatic colorectal cancer: a oulation-based study. Cancer 2007;109(4): Grundmann RT, Hermanek P, Merkel S, et al. Diagnosis and treatment of colorectal liver metastases e workflow. Zentralbl Chir 2008;133(3): Bloed W, van Leeuwen MS, Borel Rinkes IH. Role of intraoerative ultrasound of the liver with imroved reoerative heatic imaging. Eur J Surg 2000;166(9): Wildi SM, Gubler C, Hany T, et al. Intraoerative sonograhy in atients with colorectal cancer and resectable liver metastases on reoerative FDG-PET-CT. J Clin Ultrasound 2008;36(1): Wiering B, Vogel WV, Ruers TJ, Oyen WJ. Controversies in the management of colorectal liver metastases: role of PET and PET/CT. Dig Surg 2008;25(6): Rafaelsen SR, Jakobsen A. Contrast enhanced ultrasonograhy versus multidetector-comuted tomograhy in detection of liver metastases from colorectal cancer A rosective, blinded, atient by atient analysis. Colorectal Dis Kulemann V, Schima W, Tamandl D, et al. Preoerative detection of colorectal liver metastases in fatty liver: MDCTor MRI? Eur J Radiol Alberts SR. Udated otions for liver-limited metastatic colorectal cancer. Clin Colorectal Cancer 2008;7(Sul. 2):S Van den Eynde M, Hendlisz A. Treatment of colorectal liver metastases: a review. Rev Recent Clin Trials 2009;4(1): Hebbar M, Pruvot FR, Romano O, Triboulet JP, de GA. Integration of neoadjuvant and adjuvant chemotheray in atients with resectable liver metastases from colorectal cancer. Cancer Treat Rev 2009; 35(8): Alley PG, McNee RK. Colorectal cancer in Auckland 1981e1982: atients with liver metastases. N Z Med J 1985;98(785): Faivre J, Manfredi S, Bouvier AM. Eidemiology of colorectal cancer liver metastases. Bull Acad Natl Med 2003;187(5): Kune GA, Kune S, Field B, et al. Survival in atients with large-bowel cancer. A oulation-based investigation from the Melbourne Colorectal Cancer Study. Dis Colon Rectum 1990;33(11): Leorrier J, Maurel J, Chiche L, Bara S, Segol P, Launoy G. A oulationbased study of the incidence, management and rognosis of heatic metastases from colorectal cancer. Br J Surg 2006;93(4): Assumcao L, Choti MA, Gleisner AL, et al. Patterns of recurrence following liver resection for colorectal metastases: effect of rimary rectal tumor site. Arch Surg 2008;143(8): Abdalla EK, Vauthey JN, Ellis LM, et al. Recurrence and outcomes following heatic resection, radiofrequency ablation, and combined resection/ablation for colorectal liver metastases. Ann Surg 2004; 239(6): Manfredi S, Leage C, Hatem C, Coatmeur O, Faivre J, Bouvier AM. Eidemiology and management of liver metastases from colorectal cancer. Ann Surg 2006;244(2): Gayowski TJ, Iwatsuki S, Madariaga JR, et al. Exerience in heatic resection for metastatic colorectal cancer: analysis of clinical and athologic risk factors. Surgery 1994;116(4): Sternberg A, Amar M, Alfici R, Groisman G. Conclusions from a study of venous invasion in stage IV colorectal adenocarcinoma. J Clin Pathol 2002;55(1): Dirschmid K, Lang A, Mathis G, Haid A, Hansen M. Incidence of extramural venous invasion in colorectal carcinoma: findings with a new technique. Hum Pathol 1996;27(11): Ouchi K, Sugawara T, Ono H, et al. Histologic features and clinical significance of venous invasion in colorectal carcinoma with heatic metastasis. Cancer 1996;78(11): Kraus T, Vogel CU, Stolte M. Adenocarcinoma of the vermiform aendix. Leber Magen Darm 1993;23(1): Misdraji J, Yantiss RK, Graeme-Cook FM, Balis UJ, Young RH. Aendiceal mucinous neolasms: a clinicoathologic analysis of 107 cases. Am J Surg Pathol 2003;27(8): Pai RK, Beck AH, Norton JA, Longacre TA. Aendiceal mucinous neolasms: clinicoathologic study of 116 cases with analysis of factors redicting recurrence. Am J Surg Pathol 2009;33(10): Bak M, Jorgensen LJ. Adenocarcinoid of the aendix resenting with metastases to the liver. Dis Colon Rectum 1987;30(2): Ptok H, Marusch F, Meyer F, Schubert D, Gastinger I, Liert H. Imact of anastomotic leakage on oncological outcome after rectal cancer resection. Br J Surg 2007;94(12): Marusch F, Koch A, Schmidt U, et al. The imact of the risk factor "age" on the early ostoerative results of surgery for colorectal

7 R. Mantke et al. / EJSO 38 (2012) 259e carcinoma and its significance for erioerative management. World J Surg 2005;29(8): Law WL, Chu KW. Anterior resection for rectal cancer with mesorectal excision: a rosective evaluation of 622 atients. Ann Surg 2004; 240(2): Pedrazzani C, Cerullo G, De MG, et al. Imact of age-related comorbidity on results of colorectal cancer surgery. World J Gastroenterol 2009;15(45): Sluski M, Wlodarczyk Z, Jasinski M, Masztalerz M, Tujakowski J. Outcomes of simultaneous and delayed resections of synchronous colorectal liver metastases. Can J Surg 2009; 52(6):E Hillingso JG, Wille-Jorgensen P. Staged or simultaneous resection of synchronous liver metastases from colorectal cancer e a systematic review. Colorectal Dis 2009;11(1): Weber JC, Bachellier P, Oussoultzoglou E, Jaeck D. Simultaneous resection of colorectal rimary tumour and synchronous liver metastases. Br J Surg 2003;90(8): Thelen A, Jonas S, Benckert C, et al. Simultaneous versus staged liver resection of synchronous liver metastases from colorectal cancer. Int J Colorectal Dis 2007;22(10): Robertson DJ, Stukel TA, Gottlieb DJ, Sutherland JM, Fisher ES. Survival after heatic resection of colorectal cancer metastases: a national exerience. Cancer 2009;115(4):752 9.