Background: Approximately half of all patients treated for colorectal carcinoma by bowel resection have neither lymph node metastases nor known residual tumor (clinicopathologic Stages A and B). The aim of this study was to compare the survival of these patients with that of the general population and to explain any significant difference.
Methods: Prospectively collected data recorded for 910 patients from one institution during a period of 21.5 years were used in the analysis. Patient follow-up ranged from 6 months to 21.5 years. The "Survival" procedure, developed by the Finnish Cancer Registry, was used to compare the observed survival of patients with their expected survival, based on age- and sex-matched data from the population of New South Wales. Survival analysis was performed by the Kaplan-Meier method. Multivariate models were examined using Cox proportional hazards regression.
Results: Males with tumor spread beyond the muscularis propria (Stage B) was the only group with significantly poorer survival than expected. The reduced survival in this group was due to the effects of four clinical variables (cardiovascular complication, permanent stoma, urgent operation, respiratory complication) and one pathologic variable (direct spread involving a free serosal surface) acting independently.
Conclusion: The survival of patients with clinicopathologic Stages A or B tumors closely matched their expected survival as predicted from the general population. Males with Stage B tumors were the only exception and their significantly reduced survival was largely due to clinical, as distinct from pathologic factors. These findings suggest that the risk of occult metastases is low for patients with Stages A and B tumors using this classification.