Post-polypectomy surveillance colonoscopy:
In whom and when?

Abstract

Introduction
Post-polypectomy surveillance by colonoscopy is recommended in national and international guidelines. While colonoscopy is the gold standard colorectal investigation, it carries a risk of adverse events as well as being inconvenient and often uncomfortable for the patient.
It is established that population screening reduces mortality from colorectal cancer (CRC). The effect of post-polypectomy surveillance, however, is less clear. An increasing number of colonoscopies are being performed worldwide for both symptoms and screening. The adenoma detection rate at colonoscopy is also increasing with improved technology and training against the backdrop of an ageing population. As a result, an increasing number of individuals are entering post-polypectomy surveillance.

Aims & Objectives
The aim of the analysis was to evaluate the findings of post-polypectomy surveillance within the English Bowel Cancer Screening Programme (BCSP). This was done by assessing linked data from the BCSP database and the National Cancer Registration and Analysis Service (NCRAS). Objectives were:
1. To document surveillance pathways among the intermediate and high risk groups.
2. To determine the risk factors (adenoma and person-specific) at screening which predict the outcome of initial surveillance.
3. To determine the adenoma, advanced adenoma (AA) and CRC yield at initial surveillance of each colonoscopy surveillance cohort (and subcategories within each cohort) within the BCSP.

Methods
Data on individuals participating in the BCSP is entered prospectively onto the screening programme’s relational database, BCSS. BCSS was interrogated for individuals who had attended for post-polypectomy surveillance at any time from the start of the programme in 2006 until the end of 2016. In addition, linked data on CRCs diagnosed in this cohort were obtained from NCRAS.
Two separate analyses were performed. The first focussed on the detection of any AA (size ≥10mm or ≥25% villous or high-grade dysplasia) at the first surveillance attended by an individual. A separate analysis was performed with a diagnosis of CRC as the primary outcome.

Results
Of individuals with high risk findings at baseline colonoscopy, 12.3% of those attending first surveillance were found to have AA, 48.0% non-advanced adenoma, 39.1% no adenoma, and 0.5% CRC.
In the case of individuals with intermediate risk findings at baseline, of those attending first surveillance, 8.0% were found to have AA, 35.3% non-advanced adenoma, 56.1% no adenoma, and 0.4% CRC. In those categorised as intermediate risk based on the finding of a single adenoma (≥10mm) at baseline, 6.3% of those attending first surveillance were found to have AA and 0.3% CRC.
The most significant factor increasing the risk of AA at first surveillance was a higher total number of adenomas at baseline colonoscopy.

Conclusions
The rates of AA and CRC at first surveillance are relatively low and were found to be higher in the high risk group compared to intermediate risk. Those individuals categorised as intermediate risk based on a single adenoma (≥10mm) at baseline, had a particularly low rate of AA and CRC at first surveillance.
These findings support the hypothesis that the incidence of AA and CRC are low at post-polypectomy surveillance colonoscopy. The particularly low yield in the subgroup with a single adenoma at baseline suggests that surveillance is not be needed in this group and may not be necessary for the intermediate risk cohort as a whole.