Post written by Mohamed Shiha, MRCP, from the Academic Unit of Gastroenterology, Sheffield Teaching Hospitals, Sheffield, UK.

Adenoma detection rate (ADR) is the main quality indicator for the effectiveness of colonoscopy in preventing colorectal cancer. However, it is estimated that 1 in 4 adenomas are missed during colonoscopy.

Recently, advances in artificial intelligence (AI) have enabled the development of computer-aided detection (CADe) systems that highlight colorectal polyps to endoscopists in real time with high accuracy. The focus of our systematic review and meta-analysis was to estimate the added benefit of using CADe on ADR compared with standard colonoscopy.

Several systematic reviews and meta-analyses of randomized controlled trials have been published on this topic. However, a common limitation among them is the inclusion of studies that used adjuvant AI systems to control colonoscopy quality or tandem colonoscopy methodology with adenoma miss rate as the primary outcome. Including these studies may have led to an overestimation of the effect of CADe on ADR.

Therefore, we aimed to provide more precise estimates of the true effect of CADe on ADR by including a homogeneous group of randomized controlled trials using a similar methodology and focusing on ADR as the primary outcome.

In our analysis, the pooled ADR was significantly higher in the CADe group than the standard colonoscopy group (41.4% vs 33%; relative risk [RR], 1.26; 95% confidence interval [CI], 1.18-1.35), which was mainly because of the increased detection of diminutive (≤5 mm) adenomas. This relative increase in ADR with CADe was lower than reported in previous meta-analyses.

Interestingly, there was evidence of significant publication bias, with studies reporting favorable outcomes with CADe more likely to be published. The difference in ADR between the CADe and standard colonoscopy groups remained significant after adjusting for publication bias (RR, 1.17; 95% CI, 1.09-1.27). Yet, the magnitude of the difference was more modest than initially expected (Supplementary Figure 3).

Long-term longitudinal studies are warranted to evaluate the effect of CADe on the risk of interval colorectal cancer and cancer-related deaths.

Funnel plot for adenoma detection rate, including the observed studies and the imputed studies using trim and fill analysis. CI, Confidence interval.

Visit iGIE’s Facebook, Twitter, and YouTube accounts for more content from the ASGE peer-reviewed journal that launched in December 2022.

Read the full article online.

The information presented in Endoscopedia reflects the opinions of the authors and does not represent the position of the American Society for Gastrointestinal Endoscopy (ASGE). ASGE expressly disclaims any warranties or guarantees, expressed or implied, and is not liable for damages of any kind in connection with the material, information, or procedures set forth.