Introduction
Sessile serrated lesions (SSLs), alongside adenomas, play a significant role in the development of colorectal cancer (CRC). Recent studies suggest that the adenoma detection rate (ADR) alone may not be sufficient for predicting post-colonoscopy CRC. Consequently, the ASGE/ACG Quality Task Force has recently recommended the sessile serrated lesion detection rate (SSLDR) as a new priority indicator for colonoscopy quality. Reviewing the variability of SSLDR in the control groups of previous randomized controlled trials (RCTs) may help identify factors that influence study outcomes.
Aims & Methods
A systematic review was conducted using PubMed, Embase, and the Cochrane Library databases to collect data from control groups of RCTs that assessed methods for increasing polyp detection with high-definition white-light colonoscopy (up to March 14, 2025). Random-effects meta-analysis pooled SSLDRs and their 95% confidence intervals (CIs). Heterogeneity was assessed with the I² statistic. Univariable and multivariable meta-regressions were performed to examine associations between SSLDR and study settings (publication year, sample size, number of centers, study design, and geographic origin), demographics (mean age, male proportion), and technical parameters (type of intervention, proportion of screening colonoscopy, ADR, and withdrawal time). This study was registered in PROSPERO (ID: CRD420251012475).
Results
A total of 23,461 patients from 48 studies were included in the analysis. SSLDR in the control arms ranged from 0.3% to 21.5%, with a pooled value of 4.4% (95% CI: 3.3%-5.8%; I² = 93%). Univariable meta-regression analysis identified patient age (OR 1.112; 95% CI: 1.043-1.186) and gender (OR 1.038; 95% CI: 1.003 to 1.074), geographic origin (OR 2.848; 95% CI: 1.585-5.116 for Europe and OR 4.044; 95% CI: 2.205-7.416 for North America), and ADR in the control arms (OR 1.054; 95% CI: 1.031-1.077) as significant predictors of SSLDR, with sample size (OR 0.999; 95% CI: 0.998-1.000) showing a negative correlation. Other factors, such as publication year, number of study centers (single or multiple), study design (parallel or tandem), type of intervention (imaging or mechanical), proportion of screening colonoscopy, and withdrawal time, had no significant effect. In multivariable meta-regression, RCTs conducted in Europe (OR 2.179; 95% CI: 1.242-3.820) and North America (OR 2.664; 95% CI: 1.550-4.580), as well as ADR in control arms (OR 1.035; 95% CI: 1.006-1.065), remained significant predictors of SSLDR, while sample size (OR 0.998; 95% CI: 0.997-0.999) and proportion of screening colonoscopy (OR 0.991; 95% CI: 0.983-0.998) were negatively correlated with SSLDR.
Conclusion
A high level of variability in SSLDR was found in the control groups of RCTs. Standardization of methodological and technical parameters is essential to improve the generalizability and reproducibility of findings from RCTs and enhance SSL detection in colonoscopy.