The management alternatives of segmental colonic resection versus follow-up alone for colonic adenomas containing early invasive carcinoma (i.e., polypectomy resection margin free of tumor) were evaluated by decision tree analysis. Using data from the literature, the major variables influencing the decision were as follows: probability of residual disease after polypectomy, operative efficacy (defined as the chance of curing a tumor that would otherwise recur without surgery), and operative mortality. For a hypothetical patient with low operative risk (operative mortality of 0.2%), resection would yield the best outcome in terms of life expectancy as long as the probability of residual disease was greater than 0.5%. Extensive analysis of the impact of changes in assumptions about the parameters characterizing the problem showed this conclusion to be unchanging over wide ranges of operative efficacy and probability of residual disease. However, the decision to do a secondary resection in this situation was still close because surgery was only marginally justifiable on economic grounds. Observation would be preferred in patients with higher operative risk (operative mortality greater than 2%).