Incident colorectal cancer in Lynch syndrome is usually not preceded by compromised quality of colonoscopy

Abstract

Background: Lifetime incidence of colorectal cancer (CRC) especially in carriers of MLH1 and MSH2 pathogenic germline variants in mismatch repair genes is high despite ongoing colonoscopy surveillance. Lynch syndrome (LS) registries have been criticized for not reporting colonoscopy quality adequately.

Methods: Prospective follow-up data from the national registry were combined with a retrospective assessment of the colonoscopy reports from Helsinki University Hospital electronic patients records in 2004–2019.

Results: Total of 366 MLH1, MSH2 and MSH6 carriers underwent 1564 colorectal endoscopies (mean 4.3 per patient, range 1–10) at a single unit. At least one subsequent examination was performed on 336 patients.

Bowel preparation was suboptimal (Boston Bowel Preparation Scale 0–2) on either right or left side of the colon in 12.9% of planned surveillance examinations. Caecal intubation rate for full-length colonoscopies was 98.9%. Adenoma detection rate (ADR) was 15.8% in 2004–2014 but substantially increased (21.9%) after introduction of high-definition (HD) technology in 2015–2019 (p = .004; 18.7% across all examinations).

CRCs were detected in 23 cases. Nineteen cancers were detected after 977 optimal quality colonoscopies and 4 after 151 compromised quality (BBPS <3 or non-complete examination; p = .16). Advanced neoplasias were not more frequently reported after compromised quality examinations.

Conclusion: The majority of LS-associated incident CRCs were detected after colonoscopies with proper bowel preparation and complete examination. There is a considerable time trend towards higher ADR after introducing HD technology of endoscopes. The effect of time trend in ADR to CRC incidence in LS needs to be studied in larger, prospective settings.

Keywords:

• Lynch Syndrome
• hereditary non-polyposis colorectal cancer
• surveillance colonoscopy
• colonoscopy
• colorectal cancer

Acknowledgements

We express our gratitude to Ms Kirsi Pylvänäinen and Ms Beatriz Alcala-Repo for their support in data acquisition.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

Work of the study group was supported by Emil Aaltosen Säätiö, Sigrid Juséliuksen Säätiö, and Suomen Lääketieteen Säätiö.