The Addition of Computer‐Aided Detection (CADe) to Texture and Color Enhancement Imaging (TXI) Does Not Improve Adenoma Detection Rate During Colonoscopy: A Prospective Study
Edward Young, Arvind Rajagopalan, Derrick Tee, Dharshan Sathananthan, Hamish Philpott, Dauda Bawa, Rajvinder SinghABSTRACT
Objectives
Early detection and resection of colorectal polyps prevent their progression toward advanced adenocarcinomas. The use of Texture and Color Enhancement Imaging (TXI) has been demonstrated to enhance adenoma detection compared to white light imaging. It remains unknown whether there is an additive benefit when using computer‐aided detection (CADe) in addition to enhanced imaging technologies.
Methods
This prospective study involved the combined use of TXI and CADe during colonoscope withdrawal in the intervention group, compared against historical controls from a prior randomized controlled trial using TXI alone. The presence and number of colonic polyps identified, resected, and analyzed via histopathology were documented and compared between groups.
Results
A total of 263 patients were included in the analysis, 100 in the TXI and CADe group and 163 in the TXI‐only group. There was no difference in the number of adenomas per colonoscopy (mean 1.71 in both groups, p > 0.99) or the adenoma detection rate (ADR) (62% with TXI and CADe vs. 55.8% with TXI alone, p = 0.33). In standard‐detector endoscopists, there was an improvement in ADR using TXI and CADe compared with TXI alone (63% vs. 44%, p = 0.046). There was an almost 1‐min increase in mean withdrawal time with TXI and CADe ( p < 0.001) and an increase in the resection of non‐neoplastic polyps (27% vs. 16.6%, p = 0.042).
Conclusions
The addition of CADe did not impact overall adenoma detection, although it did improve adenoma detection in those with an ADR < 55%. This comes at the cost of increased withdrawal time and more frequent resection of non‐neoplastic polyps.
Trial Registration
Australian New Zealand Clinical Trials Registry: ACTRN12625000996460