LAUSR

BACKGROUND AND AIMS Current colon capsule cleansing grading scales rely on subjective parameters and lack proper interobserver agreement. We should strive for higher intra- and interobserver agreement for the evaluation of colon capsules' cleansing quality. Validation of a new grading scale for the evaluation of colon capsule cleansing. METHODS For the new grading scale (CC-CLEAR), the colon was divided in 3 segments: right, transverse, and left colon. Each segment was scored according to an estimation of the percentage of visualized mucosa (0: <50%; 1: 50%-75%; 2: >75%; 3: >90%). The overall cleansing classification was a sum of each segment scores, grading between excellent (8-9), good (6-7) and inadequate (0-5). Any segment scoring ≤1 resulted in inadequate overall classification. Videos were reviewed and scored using CC-CLEAR and the Leighton-Rex grading scale, by 2 experienced operators, blinded to each other. Kendall's coefficient evaluated inter- and intraobserver agreement. RESULTS We included 58 consecutive colon capsules, corresponding to 75.9% women, with mean age 65 years. Overall cleansing CC-CLEAR classifications were as follows: Reader A - 22.4% (n=13) excellent, 31% (n=18) good, 46.5% (n=27) inadequate; Reader B - 24.1% (n=14) excellent, 22.4% (n=13) good, 53.4% (n=31) inadequate. CC-CLEAR interobserver agreement was superior to Leighton-Rex scale, Kendall's W 0.911 vs 0.806, p<0.01. The intraobserver agreement for CC-CLEAR was excellent for both readers (p > 0.01). CONCLUSION CC-CLEAR is a new practical and reliable grading scale for the evaluation of bowel preparation quality in colon capsule, with excellent inter- and intraobserver agreement.