TY - JOUR
T1 - Survival of patients operated for colorectal liver metastases and concomitant extra-hepatic disease
T2 - External validation of a prognostic model
AU - Mavros, Michael N.
AU - Hyder, Omar
AU - Pulitano, Carlo
AU - Aldrighetti, Luca
AU - Pawlik, Timothy M.
PY - 2013/4
Y1 - 2013/4
N2 - Background The presence of extra-hepatic disease (EHD) is no longer an absolute contraindication to surgery in patients with colorectal liver metastases (CRLM). Recently, a novel prognostic model predicting overall survival in such patients was proposed using five risk factors (EHD other than isolated lung metastases, CEA ≥10 ng/ml, >5 liver metastases, right colon as the primary CRC location, and diagnosis of EHD concomitant to CRLM recurrence). A bi-institutional database was used to perform an external validation of this model. Methods Ninety-seven patients operated for CRLM and EHD between 1982 and 2011 in two institutions was analyzed. The proposed prognostic model was validated in this cohort using Cox proportional hazards models and the concordance index (c). Results Of the five proposed risk factors, only EHD other than isolated lung metastases was found to independently predict overall survival [Hazards Ratio (HR) = 2.10 (95% CI: 1.01-4.40)]. Although, the number of risk factors was marginally associated with overall survival in univariate analysis (P = 0.049), the performance of the proposed prognostic model was poor when applied to our cohort (c = 0.64). Conclusion The examined prognostic model of survival in patients with CRLM and EHD had poor performance. Further research is warranted to delineate the subset of patients who will benefit from surgery.
AB - Background The presence of extra-hepatic disease (EHD) is no longer an absolute contraindication to surgery in patients with colorectal liver metastases (CRLM). Recently, a novel prognostic model predicting overall survival in such patients was proposed using five risk factors (EHD other than isolated lung metastases, CEA ≥10 ng/ml, >5 liver metastases, right colon as the primary CRC location, and diagnosis of EHD concomitant to CRLM recurrence). A bi-institutional database was used to perform an external validation of this model. Methods Ninety-seven patients operated for CRLM and EHD between 1982 and 2011 in two institutions was analyzed. The proposed prognostic model was validated in this cohort using Cox proportional hazards models and the concordance index (c). Results Of the five proposed risk factors, only EHD other than isolated lung metastases was found to independently predict overall survival [Hazards Ratio (HR) = 2.10 (95% CI: 1.01-4.40)]. Although, the number of risk factors was marginally associated with overall survival in univariate analysis (P = 0.049), the performance of the proposed prognostic model was poor when applied to our cohort (c = 0.64). Conclusion The examined prognostic model of survival in patients with CRLM and EHD had poor performance. Further research is warranted to delineate the subset of patients who will benefit from surgery.
KW - colorectal cancer
KW - extra-hepatic disease
KW - metastases
KW - secondary liver malignancies
UR - http://www.scopus.com/inward/record.url?scp=84875253599&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84875253599&partnerID=8YFLogxK
U2 - 10.1002/jso.23260
DO - 10.1002/jso.23260
M3 - Article
C2 - 22952157
AN - SCOPUS:84875253599
VL - 107
SP - 481
EP - 485
JO - Journal of Surgical Oncology
JF - Journal of Surgical Oncology
SN - 0022-4790
IS - 5
ER -