Development of an in-house endoanal ultrasound teaching phantom.

OBJECTIVES

The aim of this study was to construct an anatomically correct phantom for visualization of anal sphincter complex structures.

METHODS

To ake an endoanal ultrasound phantom, we experimented with a variety of commercially available materials to simulate hyperechoic, hypoechoic, and isoechoic tissue consistency. We created external anal sphincter muscle and the levator plate using stand-alone density gel and microbubbles to simulate the echogenicity of the muscles. A mold was created in a container, the muscles were inserted, and the mold was filled with high-density clear gel and allowed to fill. Once finished, the container was sealed tightly and stored for later use. To evaluate the accuracy of the model, 6 ultrasound imaging raters from different disciplines were included. The raters were administered a 6-item Internet-based interactive test using 2-dimensional images obtained from scanning the phantom. Rater agreement was determined. The Fleiss κstatistic was calculated to determine inter-rater reliability.

RESULTS

The raters identified the structures in 32 (89%) of 36 test questions. There was good to excellent agreement among the readers. Agreement rates for visualization of the external anal sphincter, perineal body, and internal anal sphincter were 67%, 83%, and 100%, respectively.

CONCLUSIONS

An endoanal phantom can be constructed as a simulator for endoanal sonography.