Surveillance Colonoscopy After Screening Polypectomy Reduces Colorectal Cancer Incidence in Intermediate-Risk Patients

August 2017, Vol. 24, No. 8

Atkin W, Wooldrage K, Brenner A, et al. Adenoma surveillance and colorectal cancer incidence: a retrospective, multicenter, cohort study. Lancet Oncol 2017;18:823–34.


To Download a PDF of the Full Article:

Click Here


Study Overview

Objective. To examine the heterogeneity in colorectal cancer (CRC) incidence in intermediate-risk patients and the effect of surveillance on CRC incidence.

Design. Retrospective, multicenter cohort study.

Setting and participants. Study patients underwent colonoscopy between 1 January 1990 and 21 December 2010 at 17 hospitals in the United Kingdom. Patients  were eligible for the study if they had a baseline colonoscopy with a newly diagnosed intermediate-risk adenoma. Intermediate-risk adenomas (as defined by the UK guidelines) included 1 to 2 large adenomas ≥ 10 mm or 3 to 4 small adenomas < 10 mm in size. Patients with a history of prior resections, colorectal cancer, inflammatory bowel disease or a family history of CRC were excluded from the study. Patient, procedural, and polyp characteristics were assessed at baseline.

Main outcome measures. The primary outcome was inci-dence of CRC. Additional factors assessed included age at first polyp detection, sex, completeness of colonoscopy, preparation quality, number of adenomas, size of largest adenoma, histology, and location. Proximal polyps were defined as those proximal to the descending colon. Information regarding social history (eg, smoking status) was not available.

Results. The authors identified 253,798 patients who underwent colonoscopy between 1 January 1990 and 21 December 2010. Of those, 223,539 were excluded based on not meeting the pre-specified inclusion criteria, resulting in 30,259 eligible patients for analysis. Review of histological data confirmed intermediate-risk adenomas in 11,995 (40%) of the patients. The median age in this study was 66 years and 55% were men. Fifty-eight percent attended 1 or more follow-up surveillance visits while 42% had no follow-up surveillance colonoscopy. Those who attended more than 1 follow-up surveillance visits were younger, had a greater proportion of large adenomas (> 20 mm), or had an adenoma with high-grade dysplasia. Both groups had similar rates of villous histology (9% vs. 10%).

After a median follow-up of 7.9 years, 210 CRCs were diagnosed and 32% of patients died. In the group with no follow-up surveillance, 46% died and 2% were diagnosed with cancer. In the group who had 1 or more follow-up colonoscopies, 21% died and 1% were diagnosed with cancer. One or 2 surveillance visits were associated with a significant reduction in CRC incidence (HR 0.57 [95% confidence interval {CI} 0.4–0.8) and 0.51 [95% CI 0.31–0.84], respectively). Three or more surveillance exams were also associated with a similar reduction in CRC incidence (HR 0.54; CI 0.29–0.99). Characteristics associated with increased CRC incidence were older age, adenomas > 20 mm, high-grade dysplasia, proximal polyps, and colonoscopies that were either incomplete or with poor preparation. The number of adenomas was not independently associated with CRC incidence.

The authors divided the cohort into higher-risk (74%) and lower-risk (26%)  subgroups based on polyp and procedural characteristics. The higher-risk group included patients with baseline adenomas ≥ 20 mm, high-grade dysplasia, proximal polyps, or suboptimal evaluation. The lower-risk group included all others. CRC incidence was higher in the “higher-risk” subgroup (247 CRC per 100,000 vs. 93 CRC per 100,000). In the higher-risk group, risk of CRC decreased with more surveillance visits, a finding that was not observed in the lower-risk group. The 10-year incidence of CRC in the cohort overall was 2.7%, in the higher-risk group was 3.3% and in the lower risk group was 1.1%. CRC incidence was significantly higher in the higher-risk subgroup compared with the general population.

Conclusion. Colonoscopy surveillance significantly reducedthe incidence of CRC in intermediate-risk patients (1 to 3 large adenomas ≥ 10 mm or 3 to 4 small adenomas < 10 mm in size) who were offered surveillance at 3-year intervals. Moreover, the benefit of surveillance was particularly noted in a sub-group of patients who had large adenomas (≥ 20 mm), high-grade dysplasia, proximal polyps or poor endoscopic evaluation at the time of initial screening.



Screening colonoscopy with removal of adenomatous polyps prevents many CRCs and has been shown to reduce mortality [1]. The results of this retrospective study suggest that patients with intermediate-risk adenomas who underwent at least 1 surveillance colonoscopy at 3-year intervals had a significant reduction in the incidence of CRC. The authors have identified a subgroup of patients at higher risk for CRC, which included those who had a suboptimal initial colonoscopy including poor bowel preparation, adenomas ≥ 20 mm, adenomas with high-grade dysplasia, or proximal adenomas. In particular, ongoing surveillance in this high-risk cohort was associated with significant reductions in CRC incidence. Conversely, those in the lower-risk group had a CRC incidence lower than that of the general population,  raising some questions as to whether this group benefits from ongoing surveillance. However, definitive conclusions are difficult to make given the relatively low incidence of CRC in this group.

The risk of neoplasia in patients with colorectal ade-nomas has been evaluated in multiple studies. A pooled analysis by Martinez and colleagues examined over 9000 patients and noted advanced adenomas were found during follow up in 11.2% of the population, with 0.6% of the population developing invasive CRC [2]. Compared with adenomas < 5 mm, those with baseline adenomas 10–19 mm had a higher risk of advanced neoplasia (15.9% vs 7.7%; OR 2.2). Moreover, those with a baseline polyp ≥ 20 mm had a risk of advanced neoplasia at follow-up of 19.3% (OR 2.99). The results of the current investigation also suggest an increased risk of neoplasia with increased polyp size. Interestingly, the polyp size that conferred a higher risk in this study was ≥ 20 mm. The authors of this study suggest that polyps ≥ 20 mm along with the previously mentioned high-risk features may identify a subgroup within the intermediate-risk population who may benefit from close surveillance. One particularly interesting finding in this study was the identification of proximal colon polyps as a risk factor. While less well defined, previous investigations have noted a similar finding suggesting a risk of advanced neoplasia of up to 80% in patients with proximal polyps [3]. Given such, intensive surveillance may not be appropriate for all intermediate-risk patients and a more refined risk-adapted approach may be preferred.

There are some important limitations of the current study that warrant discussion. First, it should be emphasized that this study is observational in nature and therefore, definitive conclusions cannot be made despite the significant effect of surveillance colonoscopy in patients with high-risk features. In addition, the median follow-up in this study was 7.9 years and one could argue that longer-follow up is needed in order to validate the findings of this study, particularly in patients in the lower-risk cohort. Nevertheless, this study does suggest that there may be a population of patients that harbor higher-risk features and close surveillance limited to this group may be more appropriate. Furthermore, the duration of surveillance remains an important clinical question that requires further research.


Applications for Clinical Practice

In 2012, the United States Multi-Society Task Force (MSTF) on CRC issued updated guidelines defining adenoma risk and postpolypectomy surveillance. Low-risk adenomas (1 to 2 tubular adenomas ≤ 10 mm at baseline) should have repeat surveillance colonoscopy in 5 to 10 years. Advanced adenomas (≥ 10 mm, villous histology, or high-grade dysplasia) or those with 3 to 10 adenomas at baseline should undergo first surveillance in 3 years [4]. The authors of the current study suggest that surveillance colonoscopy at 3-year intervals for patients with particularly high-risk features including those with poor bowel preparation, adenomas ≥ 20 mm, adenomas with high-grade dysplasia or proximal adenomas benefit the greatest from at least 1 surveillance colonoscopy. Those with lower- risk features may not require such rigorous follow-up; however, further work to define which high-risk cohorts should undergo close surveillance is warranted. It is vital that the primary care provider understand such guidelines in order to facilitate the appropriate follow-up.


—Daniel Isaac, DO, MS, Michigan State University, East Lansing, MI



1.   Zauber AG, Winawer SJ, O’Brien MJ, et al. Colonoscopic polypectomy and long-term prevention of colorectal cancer deaths. N Engl J Med 2012;366:687–96.

2. Martinez ME, Baron JA, Lieberman DA, et al. A pooled analysis of advanced colorectal neoplasia diagnosis after colonoscopic polypectomy. Gastroenterology 2009;136:832–41.

3. Pinsky PF, Schoen RE, Weissfeld JL, et al. The yield of surveillance colonoscopy by adenoma history and time to examination. Clin Gastroenterol Hepatol 2009;7:86–92.

4. Lieberman DA, Rex DK, Winawer SJ, et al. Guidelines for colonoscopy surveillance after screening and polypectomy: A consensus update by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology 2012;143:844–57.

Comments are closed.

« »