The use of a real-time, computer-aided quality improvement system (ENDOANGEL) increased the detection of adenomas in routine colonoscopies, according to a study published in The Lancet Gastroenterology and Hepatology.
The aim of this single-center, single-blind, randomized, parallel-group trial was to assess if the ENDOANGEL system could improve adenoma detection during a routine colonoscopy. The ENDOANGEL system was created by using neural networks and perceptual hash algorithms to aid endoscopists in monitoring withdrawal speeds, withdrawal timing, and blind spot identification. Patients (mean age, 50 years) were randomly assigned in a 1:1 ratio to either the test group, receiving an ENDOANGEL-assisted colonoscopy, or the control group, receiving an unassisted colonoscopy. Basic demographics and clinical evaluations were completed before the colonoscopy, and adenomas were diagnosed based on pathological results.
The primary outcome was the proportion of patients having at least 1 adenoma detected during the colonoscopy, and secondary outcomes focused on size, location, polyp detection, withdrawal time, and adverse events that occurred during the colonoscopy.
Of the 704 patients included in the intent-to-treat population, 355 were assigned to the ENDOANGEL-assisted colonoscopy group and 349 were assigned to the control group. The adenoma detection rate was significantly greater in the ENDOANGEL group than in the control group (16% vs 8%; odds ratio [OR], 2.30; 95% CI, 1.40-3.77; P =.001). A greater number of diminutive polyps (P =.0039) and diminutive adenomas (P =.015) were found in the ENDOANGEL-assisted colonoscopy group when compared with the control group.
Of the 642 patients included in the per-protocol population, 324 were assigned to have an ENDOANGEL-assisted colonoscopy and 318 were assigned to have an unassisted colonoscopy. Consistent with the intent-to-treat analysis, 17% of the patients in the ENDOANGEL-assisted colonoscopy group were diagnosed with adenomas and 8% of the patients in the unassisted colonoscopy group were diagnosed with adenomas (OR, 2.18; 95% CI, 1.31-3.62; P =.0026). Again, a greater number of diminutive polyps (P =.0059) and diminutive adenomas (P =.027) were found in the ENDOANGEL-assisted colonoscopy group when compared with the control group. Overall, these results were not affected by age or endoscope type, and no adverse events were reported.
Future studies are need to include more medical centers to increase generalizability and increase sample size to verify performance.
The researchers concluded that “the ENDOANGEL system significantly improved the adenoma yield during colonoscopy and seems to be effective and safe for use during routine colonoscopy.”
Several study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.
Gong D, Wu L, Zhang J, et al. Detection of colorectal adenomas with a real-time computer-aided system (ENDOANGEL): a randomised controlled study [published online January 22, 2020]. Lancet Gastroenterol Hepatol. doi:10.1016/S2468-1253(19)30413-3