Comprehensive evaluation of the learning curve to achieve satisfactory adenoma detection rate

The number of colonoscopies required to reach satisfactory adenoma detection rate (ADR) is not well established. The aim of this study was to identify the appropriate number of procedures required to attain satisfactory ADR for those well-trained endoscopists who have a cecal intubation rate (CIR) ≥ 90% and start to perform colonoscopy independently.

All endoscopists with compelete independent colonoscopy data during career in our database were enrolled. The number of procedures required to achieve ADR ≥ 20% was identified by cumulative summation (Cusum), learning curve Cusum (LC-Cusum), and moving average method. Mixed effect logistic regression model was developed to determine the relationship between endoscopist as well as patient-related factors and adenoma detection.

A total of 24 943 procedures and 14 endoscopists were enrolled. By Cusum analysis, the interest point was at 207 procedures. By LC-Cusum analysis, 71% (10/14) and 86% (12/14) of endoscopists had attained satisfactory ADR after 200 and 300 procedures, respectively. By moving average method, endoscopists reached a mean ADR of 20% at 216 and 261 procedures over blocks of 50 and 100 procedures, respectively. The total number of procedures, number of daily procedures, patient age and gender, bowel preparation, sedation, and diverticulosis were significantly associated with adenoma detection.

This is the first study to investigate the learning curve of ADR for those well-trained endoscopists who have a CIR ≥ 90% and start to perform colonoscopy independently. Two hundred procedures might be an optimal number required to reach an ADR ≥ 20%.