There is a large variety in quality of endoscopy, a well performed endoscopy reduces the chance that cancer are missed to a huge degree. The endoscopist, a human observer, is the most important variable factor, and the quality of the individual endoscopist is often expressed as average adenoma detection rate per procedure. Endoscopists with a high adenoma detection rate rarely miss large lesions an interventions leading to an increase in the adenoma detection rate have been shown to be effective in preventing colorectal cancer development from missed lesions. Unfortunately, there is substantial variation in adenoma detection rates across endoscopy services. Plausible explanations for the variation in adenoma detection rates are the difficulty of detecting subtle small and flat lesions, and incomplete inspection of the whole colorectal mucosa due to blind spots, especially at the flexures or on the proximal side of the folds.
In the past decades advanced endoscopy imaging techniques have been developed trying to reduce the chance that lesions are missed. These techniques either enhance the contrast between neoplastic and healthy mucosa, through approaches such as dye-spray or virtual chromoendoscopy, or used special devices to improve the inspection of colorectal mucosa such as special cuffs. Although there is some evidence that these techniques indeed increase adenoma detection rates, it has never led to widespread implementation of these techniques except for the niche of surveillance chromo-endoscopy in patients with inflammatory bowel disease in remission, but also there it is considered tedious and unpractical.
Computer-aided detection (CADe) techniques uses artificial intelligence algorithms to increase the detection of lesions. CADe may offer better prospects for broad adoption in clinical practice as it does not take more time to perform and does not add to patient discomfort.
In this systematic review and network meta-analysis, a comprehensive search of PubMed/Medline, Embase, and Scopus databases was performed for randomised controlled trials investigating the effectiveness of the following endoscopic techniques in detecting colorectal adenomas and cancers: CADe, high definition white-light endoscopy, chromoendoscopy, or add-on devices that increase mucosal visualisation.
A total of 50 randomised controlled trials were found investigating these new techniques, comprising 34 445 participants. High-definition white-light endoscopy was the control technique in all studies. Compared with the control technique, the adenoma detection rate was 7·4% higher with CADe, 4·4% higher with chromoendoscopy, and 4·1% higher with increased mucosal visualisation systems. CADe ranked as the superior technique for adenoma detection. For large adenomas (≥10 mm) there was a significant increase in the detection of large adenomas only with CADe when compared to white-light endoscopy. There were no significant differences in withdrawal time found for CADe versus the other techniques.
The authors conclude that detection rates of colorectal neoplasia are higher with CADe than with other techniques and that given the practical side of CADe, this technique is also more likely to be adopted into community endoscopy services.
Related papers
Comment to: Boxhoorn L, van Dijk SM, van Grinsven J, Verdonk RC, Boermeester MA, Bollen TL, Bouwense SAW, Bruno MJ, Cappendijk VC, Dejong CHC, van Duijvendijk P, van Eijck CHJ, Fockens P, Francken MFG, van Goor H, Hadithi M, Hallensleben NDL, Haveman JW, Jacobs MAJM, Jansen JM, Kop MPM, van Lienden KP, Manusama ER, Mieog [more info]
Gyawali CP, Tutuian R, Zerbib F, Rogers BD, Frazzoni M, Roman S, Savarino E, de Bortoli N, Vela MF, Sifrim D. Gastroenterology. 2021 Nov;161(5):1412-1422. doi: 10.1053/j.gastro.2021.07.004. When patients have persistent reflux symptoms while being treated with proton pump inhibitors (PPI), the key question is whether they have indeed PPI refractory reflux disease or another disorder [more info]
Comment to: Selvaraj EA, Mózes FE, Ajmer Jayaswal AN, Zafarmand MH, Vali Y, Lee JA, Levick CK, Joseph Young LA, Palaniyappan N, Liu CH, Aithal GP, Romero-Gómez M, Brosnan MJ, Tuthill TA, Anstee QM, Neubauer S, Harrison SA, Bossuyt PM, Pavlides M; LITMUS Investigators. Diagnostic accuracy of elastography and magnetic resonance imaging in patients with [more info]