Diagnostic yield and miss rate of EndoRings in an organized colorectal cancer screening program: the SMART (Study Methodology for ADR-Related Technology) trial.

ABSTRACT

BACKGROUND AND AIMS: The EndoRings add-on has been claimed to improve adenoma detection at colonoscopy, but available data are inconsistent. When testing a new technology, parallel and crossover methodologies measure different outcomes, leaving uncertainty about their correspondence. The aims of this study were to compare the diagnostic yield and miss rate of the EndoRings for colorectal neoplasia. METHODS: Consecutive patients undergoing colonoscopy after a positive fecal immunochemical test (FIT) within an organized screening program in 7 Italian centers were randomized between a parallel (EndoRings or standard) or a crossover (EndoRings/standard or standard/EndoRings) methodology. Outcomes measures were the adenoma detection rate (ADR) and advanced adenoma detection rate (AADR) in the parallel arms and the miss rate of adenomas in the crossover arms. RESULTS: Of 958 eligible patients, 927 (317 EndoRings; 317 standard; 142 EndoRings/standard; 151 standard/Endo-Rings) were included in the final analysis. In the parallel arms (mean ADR, 51.3%; mean AADR, 25.4%), no difference between standard and EndoRings was found for both ADR (relative risk [RR], 1.10; 95% confidence interval [CI], 0.95-1.28) and AADR (RR, 1.16; 95% CI, 0.88-1.51), as well as for the mean number of adenomas and advanced adenomas per patient (EndoRings, 1.9 ± 1.3 and 1.0 ± 1.2; standard, 2.1 ± 1.5 and 1.0 ± 1.2; P = not significant for both comparisons). In the crossover arms, no difference in the miss rate for adenomas between EndoRings and standard was found at per polyp (RR, 1.43; 95% CI, 0.97-2.10) or per-patient analysis (24% vs 26%; P = .76). CONCLUSIONS: No statistically significant difference in diagnostic yield and miss rate between EndoRings and standard colonoscopy was detected in patients with a positive FIT result. A clinically relevant correspondence between miss and detection rates was shown, supporting a cause-effect relationship. (ISRCTN registry ISRCTN10357435.).