Introduction: Because most patients with colorectal liver metastases (CRLM) present to general surgeons and oncologists without a specialist interest in their management, a computer program (OncoSurge) has been created that identifies individual patient resectability and recommends optimal treatment strategies. The aim of this study was to validate the Oncosurge strategy by comparing its recommendations with the decisions made by a multidisciplinary (MDT) meeting, in a supra-regional hepatobiliary referral centre, using real cases with known outcomes.
Methods: We reviewed the records and imaging of 98 consecutive patients with CRLM, who had been referred for decision making to our MDT meeting between January 1, 2004 and December 31, 2004. All patient and tumour characteristics were entered onto the Oncosurge decision model, which was accessed at www.evidis.com/oncosurge/
Results: There was concordance between Oncosurge and MDT decisions in 93/98 cases. The observed kappa (kappa) was 0.850 (95% CI: 0.728-0.972). In descriptive terms, a kappa score greater than 0.8 equates to "almost perfect agreement".
Conclusions: Our results demonstrate the validity of the Oncosurge system when compared to one year of decision making in an established hepatobiliary MDT meeting. This confirms the utility of the Oncosurge system for decision making, audit and educational purposes.