Artificial Intelligence Increased Fraction of Colonoscopy Patients Needing 3-Year Surveillance: Relative Increase of 35% and Absolute Increase of 2.9%
Douglas K. Rex, MD, MASGE, reviewing Mori Y, et al. Clin Gastroenterol Hepatol 2022 Aug 26.
This study examined the fraction of patients undergoing colonoscopy with versus without artificial intelligence (AI) in 9 randomized controlled trials (6 parallel and 3 tandem) to determine the number of patients whose recommended surveillance interval was 3 years because of increased detection with AI. Adhering to U.S. and Japanese guidelines, which are similar, the fraction of patients with recommended 3-year surveillance intervals increased from 8.4% without AI to 11.3% with AI, for an absolute difference of 2.9% and a relative increase of 35%. Using the European guidelines, the fraction increased from 6.1% to 7.4%, for an absolute difference of 1.3% and a relative increase of 22%.
COMMENTIncreased numbers of surveillance colonoscopies and their associated cost and risk is always a consequence of an increased adenoma detection rate (ADR). Of course, the entire goal of high-quality colonoscopy is to maximize ADR and cancer prevention; in previous modeling, reduction of cancer risk associated with higher ADR offsets the cost of more colonoscopies.
Note to readers: At the time we reviewed this paper, its publisher noted that it was not in final form and that subsequent changes might be made.
CITATION(S)
Mori Y, Wang P, Løberg M, et al. Impact of artificial intelligence on colonoscopy surveillance after polyp removal: a pooled analysis of randomized trials. Clin Gastroenterol Hepatol 2022 Aug 26. (Epub ahead of print) (https://doi.org/10.1016/j.cgh.2022.08.022)
