Lower- vs Higher-Than-Expected Long-Term Colorectal Cancer Mortality After Removal of Lower- vs Higher-Risk Adenomas

Get Permission

Magnus Løberg, MD

Adenoma Removal and Colorectal Cancer Mortality

Our finding that the removal of low-risk adenomas reduces the risk of death from colorectal cancer over a period of 8 years to a level below the risk in the general population is consistent with the hypothesis that surveillance every 5 years after removal of low-risk adenomas may confer little benefit over less intensive surveillance strategies.

—Magnus Løberg, MD, and colleagues

Few data are available on long-term risk of colorectal cancer mortality after adenoma removal. In a Norwegian study reported in The New England Journal of Medicine, Magnus Løberg, MD, of the Department of Health Management and Health Economics, University of Oslo, and colleagues found that patients who had low-risk adenomas removed had lower colorectal cancer mortality risk and those who had high-risk adenomas removed had higher colorectal cancer mortality risk compared with the general population over 7.7 years of follow-up.1 The findings suggest that frequent colonoscopic surveillance may be unnecessary after removal of low-risk adenomas.

Study Details

In the study, the linked Cancer Registry and Cause of Death Registry of Norway were used to estimate colorectal cancer mortality among patients who underwent removal of colorectal adenomas between 1993 and 2007 and who were followed through 2011. Incidence-based standardized mortality ratios (SMRs) were generated using colorectal cancer mortality rates for the Norwegian general population. Polyp size and exact number were not available from the cancer registry. High-risk adenomas were defined as multiple adenomas or adenomas with a villous component or high-grade dysplasia.

Current Norwegian guidelines recommended colonoscopy after 10 years for patients with high-risk adenomas (eg, those with high-grade dysplasia, a villous component, or size ≥ 10 mm) and after 5 years for patients with at least three adenomas, with no surveillance recommended for those with low-risk adenomas.

Patients and Risk Characterization

A total of 40,826 patients who had undergone removal of colorectal adenomas were identified. In the cohort: 50.8% of patients were male; age at first adenoma removal was 40 to 49 years for 9.3%, 50 to 59 years for 22.9%, 60 to 69 years for 28.0%, 70 to 79 years for 26.7%, and ≥ 80 years for 13.2%.

The period of first adenoma removal was 1993 to 1999 for 34.7% and 2000 to 2007 for 65.3%. Duration of follow-up was 0 to 4 years for 16.9%, 5 to 9 years for 44.0%, 10 to 14 years for 27.6%, and 15 to 19 years for 11.5%. The number of adenomas per occurrence was one for 88.9% and two or more for 11.1%, and adenoma location was distal in 47.4%, proximal in 13.7%, and multiple or unspecified in 39.0%.

A total of 23,449 (51.3%) had removal of low-risk adenomas and 22,306 (48.8%) had removal of high-risk adenomas, including at least two adenomas (22.8%), villous or tubulovillous growth pattern adenomas (28.8%), and adenomas with high-grade dysplasia (15.3%). Review of detailed pathology reports for 442 patients showed that 8.2% of 220 patients with adenomas classified as high-risk had low-risk adenomas and that 30.2% of those with adenomas characterized as low-risk had high-risk adenomas.

Colorectal Cancer Mortality Rates

Median follow-up was 7.7 years (maximum, 19.0 years). Overall, 1,273 patients (387 per 100,000 person-years) were diagnosed with colorectal cancer; 383 (115/100,000 person-years) died of colorectal cancer compared with 398 expected colorectal cancer deaths in the general population (SMR = 0.96, 95% confidence interval [CI] = 0.87–1.06).

Compared with colorectal cancer mortality in the general population, colorectal cancer mortality was significantly increased among patients with high-risk adenomas removed (242 observed deaths vs 209 expected deaths, SMR = 1.16, 95% CI = 1.02–1.31) and significantly decreased among patients with low-risk adenomas removed (141 observed deaths vs 189 expected deaths, SMR = 0.75, 95% CI = 0.63–0.88). Colorectal cancer mortality was also increased in the adenoma cohort vs the general population among patients with first adenoma removal in 1993 to 1999 (SMR = 1.17, 95% CI = 1.03–1.33), those with at least two adenomas per occurrence (SMR = 1.19, 95% CI = 1.00–1.43), those with villous/tubulovillous growth pattern (SMR = 1.26, 95% CI = 1.08–1.47), and those with high-grade dysplasia (SMR = 1.40, 95% CI = 1.15–1.72).

Colorectal cancer mortality was decreased among men (SMR = 0.86, 95% CI = 0.74–1.00; SMR = 1.06, 95% CI = 0.93–1.22, for women), those with first adenoma removal in 2000 to 2007 (SMR = 0.76, 95% CI = 0.65–0.89), those with one adenoma per occurrence (SMR = 0.89, 95% CI = 0.79–1.00), those with tubulous growth pattern (SMR = 0.83, 95% CI = 0.73-0.94), and those with low-grade dysplasia (SMR = 0.87, 95% CI = 0.78–0.98).

Multivariate Analysis

The investigators performed a multivariate analysis of colorectal cancer mortality including age, sex, number of adenoma occurrences, period of adenoma removal, adenoma location, number of adenomas, grade of dysplasia, and growth pattern. In this analysis, first adenoma removal in the 2000s (hazard ratio [HR] = 0.63, P < .001, vs 1990s) was associated with significantly reduced mortality.

Factors associated with increased mortality included two or more adenomas per occurrence (HR = 1.31, P = .02, vs 1 per occurrence), villous/tubulovillous growth pattern (HR = 1.40, P = .002, vs tubulous growth pattern), high-grade dysplasia (HR = 1.45, P = .002, vs low-grade dysplasia), and increasing age at first adenoma removal (HRs = 2.43 for 50–59 years, 3.85 for 60–69 years, 7.38 for 70–79 years, and 17.74 for ≥ 80 year age groups, all P ≤ .01, vs 40–49 year group).

The investigators concluded:

After a median of 7.7 years of follow-up, colorectal-cancer mortality was lower among patients who had had low-risk adenomas removed and moderately higher among those who had had high-risk adenomas removed, as compared with the general population…. Our finding that the removal of low-risk adenomas reduces the risk of death from colorectal cancer over a period of 8 years to a level below the risk in the general population is consistent with the hypothesis that surveillance every 5 years after removal of low-risk adenomas may confer little benefit over less intensive surveillance strategies. Furthermore, complications associated with colonoscopy are not trivial and might offset the benefit of surveillance….

Dr. Løberg and Mette Kalager, MD, PhD, also of the University of Oslo, contributed equally to The New England Journal of Medicine article.   ■

Disclosure: The study was funded by the Norwegian Cancer Society and others. For full disclosures of the study authors, visit


1. Løberg M, Kalager M, Holme Ø, et al: Long-term colorectal-cancer mortality after adenoma removal. N Engl J Med 371:799-807, 2014.


Related Articles

Colonoscopic Polypectomy and Predicting Cancer Risk: A Work in Progress

Colon cancer screening using colonoscopy has significantly decreased the incidence and mortality of colorectal cancer in the United States. In the National Polyp Study (NPS), colorectal cancer was prevented by removal of adenomatous polyps.1 A more recent study looking at long-term follow-up from...