Background and Methods. Surveillance by repeated colonoscopy is currently recommended for patients with colorectal adenomas. We assessed the long-term risk of colorectal cancer after rigid-instrument sigmoidoscopy and polypectomy in 1618 patients with rectosigmoid adenomas (tumors of the rectum or distal sigmoid colon) who did not undergo surveillance. A total of 22,462 person-years of observation were accrued (mean, 14 years per patient). Results. The incidence of subsequent rectal cancer in these patients was similar to that in the general population (standardized incidence ratio, 1.2; 95 percent confidence interval, 0.7 to 2.1). Most rectal cancers developed in patients whose adenomas had been inadequately removed; the risk was very low after complete removal. The risk of subsequent colon cancer depended on the histologic type, size, and number of adenomas in the rectosigmoid. Among 842 patients with a rectosigmoid adenoma that was tubulovillous, villous, or large (greater-than-or-equal-to 1 cm), colon cancer developed in 31 patients. The standardized incidence ratio was 3.6 overall (95 percent confidence interval, 2.4 to 5.0) and 6.6 (95 percent confidence interval, 3.3 to 11.8) if there were multiple rectosigmoid adenomas. Among the remaining 776 patients with only small, tubular adenomas (whether single or multiple), colon cancer developed in only 4 patients. The standardized incidence ratio in this group was 0.5 (95 percent confidence interval, 0.1 to 1.3). Conclusions. Follow-up colonoscopic examinations may be warranted in patients with tubulovillous, villous, or large adenomas in the rectosigmoid, particularly if the adenomas are also multiple. In patients with only a single, small tubular adenoma that is only mildly or moderately dysplastic (43 percent of our series), however, surveillance may not be of value because the risk of cancer is so low.
