Overview
During colonoscopy, the endoscopist will document colonoscopy indication; BBPS score; withdrawal time; adenoma and polyp detection rate at index and follow-up colonoscopy; completeness of polypectomy; polyp location, size, surface, morphology (Paris classification), histopathology; complications.
Description
Colorectal cancer (CRC) ranks second among worldwide cancer related deaths and third in terms of cancer incidence. Colonoscopy-based screening programs have been established to reduce CRC morbidity and mortality. Multiple guidelines have established surveillance recommendations for repeat colonoscopies based on findings at index colonoscopy. Serrated lesions (SLs), including sessile serrated polyps/adenomas (SSP) and traditional serrated adenomas (TSA) have become of increased interest for their role as precursors of CRC. The optimal timing of follow-up colonoscopies after detection of SLs has been controversial as studies looking into optimal surveillance timing are lacking. The US Multi Society Task Force (USMSTF) 2020 guidelines recommend 5-10y surveillance intervals for detection of 1-2 SSPs, 3-5y for 3-4 SSPs, 3y for >4 SSPs or TSA. In contrast, the 2020 European Society of Gastrointestinal Endoscopy (ESGE) Guidelines state that 1-10mm SLs do not require follow-up. It is unclear what the appropriate surveillance intervals is for patients with SLs which is evidenced by diverging recommendations from USMSTF/ESGE. High-grade dysplasia (HGD) is an exceedingly rare finding in colorectal polyps. The current literature on the yield of colonoscopy after index HGD is sparse, with conflicting data on risk of metachronous HRA due to low numbers of included HGD leading to high variability in reported outcomes.
The primary aim of this study is to determine the rate of metachronous advanced neoplasia (TMAN) detection after index detection of serrated lesions [sessile serrated polyps (SSPs), traditional serrated adenomas (TSAs)], and metachronous high-risk adenoma (HRA) after index detection of high-grade dysplasia (HGD).
Patient with SL or HGD diagnosed from 2010-2022 with lack of follow-up will be contacted by phone, then invited to undergo follow-up colonoscopy. Data collected will include patient age; sex; ASA class; past medical history; family history of CRC; procedure date; name of endoscopist; colonoscopy indication; BBPS score; withdrawal time; adenoma and polyp detection rate at index and follow-up colonoscopy; completeness of polypectomy; polyp location, size, surface, morphology (Paris classification), histopathology; complications, immediate and late (14 days).
We expect a higher percentage of high-risk lesions in our study group compared to our retrospective findings due to follow-up delays. As there is no published literature on the subject, our assumptions will be relative to our retrospective cohort. Assuming 60% participation in our study (patients deceased, unable to be contacted, received follow-up elsewhere, do not fit inclusion criteria), we expect 362 out of 603 patients to be enrolled in the SL group and 75 out of 124 in the HGD group. Assuming a 35% high-risk lesion (T-MAN) detection in the SL group compared to 22.1% detected retrospectively, we expect to include 127 cases of T-MAN. Assuming a 40% high-risk lesion (metachronous HRA) detection in the HGD group compared to 23.8% detected retrospectively [given the longer surveillance delays for these patients (7y) with 1% already with CRC at 1.8y median follow-up] we expect to include 30 cases of metachronous HRA. To perform univariate regression, we require inclusion of 10 patients with T-MAN in the SL group (minimum sample size: 29) and 10 patients with metachronous HRA in the HGD group (minimum sample size: 25). To perform our multivariate analyses, we require inclusion of 30 patients with T-MAN in the SL group (minimum sample size: 86) and 30 patients with metachronous HRA in the HGD group (minimum sample size: 75). Descriptive analysis with presentation of crude numbers, proportions, or medians with interquartile range will be used to present patient, procedure, and polyp outcomes. The rate of T-MAN in the SL group and metachronous HRA in the HGD group will be reported as proportions with exact 95% confidence intervals (CI). Primary outcomes will be illustrated using Kaplan Meier survival analyses for each group. We will perform COX proportional hazards regression to determine the effects of polyp size, location, and synchronous polyp status (no adenoma, LRA, or HRA) on risk of T-MAN, and effects of polyp size and location on risk of metachronous HRA for the SL and HGD groups respectively. Comparisons will be presented as Hazard ratios (HR) with 95% CI. We will perform multivariate COX regression to determine the effects of confounders on development of T-MAN or metachronous HRA. Our model will be adjusted family history of CRC and smoking status. Further confounders such as age, sex, and aspirin use will be evaluated in sensitivity analyses for possible inclusion in the multivariate analysis. The effect of duration of surveillance delays on risk of T-MAN or metachronous HRA will be evaluated using logistic regression. A two-tailed p<0.05 will be considered statistically significant for all analyses.
Eligibility
Inclusion Criteria:
- Patients 45-80 who underwent colonoscopy from 2009 to 2022 at the Montreal University Hospital Center (CHUM) with 1+ SL or HGD detected at index colonoscopy and lacking follow-up within or beyond the surveillance interval recommended by 2020 USMSTF guidelines.
Exclusion Criteria:
- 1) Patients with a diagnosis of inflammatory bowel disease;
- 2) Hereditary CRC syndromes;
- 3) CRC at index colonoscopy;
- 4) Serrated polyposis syndrome;
- 5) Life expectancy too short to benefit from colonoscopy;
- 6) Follow-up colonoscopy not yet due according to USMSTF guidelines. Patients with concomitant HRA and SL at index will be invited to participate if the index (or last) colonoscopy was performed more than 1 year ago. This is based on the high rates of HRA we identified in our retrospective study posing increased risks for these patients.