Interobserver Reliability and the Sigmoid Takeoff—An Interobserver Study

Abstract

Background: Colorectal cancer is the second most common cancer worldwide. The sigmoid takeoff is the landmark where the colon sigmoid curves toward the sacrum viewed from sagittal magnetic resonance imaging (MRI). The purpose of this study was to assess interobserver variability in the assessment of the anal verge and anorectal junction in patients diagnosed with rectal cancer on magnetic resonance imaging (MRI). Materials and Methods: The rectal MRI examinations were performed using a 1.5- or 3.0-tesla unit using an anterior coil and a standard scan protocol. Two senior radiologists assessed MRI scans from patients under investigation for rectal cancer. The two observers assessed the anal verge and takeoff in cm independently. Difference in agreement between the observers were evaluated using intraclass correlation (ICC) and graphically by Bland–Altman plots. Results: The study population (n = 122) included 68 (55.7%) female and 54 (44.3%) male subjects. The overall median age was 69.5 years (range 39–95 years). There was perfect agreement between the two observers when defining rectal tumor above or below the takeoff landmark. The reliability of measuring the distance from the anal verge to the sigmoid takeoff was 0.712. Conclusion: Overall, the study found a moderate reliability in assessing the location of the sigmoid takeoff, with a low difference in the distance measuring, as well as a good consensus concerning the determination of tumors in relation to the sigmoid takeoff. Routine implementation of this information within the report seems reasonable.