Hepatic Resection for Colorectal Metastases: Value for Risk Scoring Systems?

摘要: Introduction: Predictors of outcome in patients with metastatic colorectal cancer remain inconsistent. We aimed to identify predictors these patients, develop a prognostic scoring system, and assess the general applicability current major risk systems. Materials Methods: Following IRB approval, medical records 662 consecutive undergoing resection metastases liver during 1960 1995 were reviewed. Clinicopathologic data assessed from mailed questionnaire. variables tested using univariate multivariate analyses; best-fit models then generated study effect each independent factor on outcome. To validate existing models, our set was applied those scores. The relative concordance probability estimates calculated for compared that proposed Mayo model. Results: overall disease-specific 5-year survival rates 37% 42%, respectively. recurrence at any site 65% 5 years. Perioperative blood transfusion positive hepatoduodenal nodes determinants recurrence. systems, we imported patient population into 3 other Neither nor among stratified discretely by Based estimates, all only marginally better than chance alone predicting Conclusion: Broad application systems has limited clinical value refinement external validation should be undertaken before utilization.