Use of a Novel 3D‐Printed Anatomical Model for Ultrasound‐Guided Hip Arthrocentesis in Emergency Department Training Sessions

Abstract

The anatomy of the hip joint presents specific practical difficulties for arthrocentesis compared to other peripheral joints, which makes evaluation of suspected hip joint sepsis a challenge for emergency medicine physicians. Delaying treatment of sepsis in the hip joint can lead to serious sequelae, so it is valuable for emergency medicine (EM) personnel to be trained to perform this challenging, high‐risk intervention when suspected. The use of ultrasound in guiding hip arthrocentesis has been proven to be effective (Berona, et al. 2017), but a lack of adequate simulations for training EM physicians remains an obstacle. To address this deficit, we constructed a 3D printed model for use in an educational intervention module. CT imaging data was segmented to acquire a volume model of the left hip joint which was exported to the digital sculpting software Zbrush 2018. This program was used to correct small imperfections in the model's surface; to fix any non‐manifold geometry; and to export the model in a format suitable for 3D printing at true to life scale. The model was sliced in Simplify3D version 4.0.1 and printed in polylactic acid using a Lulzbot Taz 5 3D printer (Aleph Objects, Inc.) as two separate parts, a left pelvis and left femur. An Ambu bag was rigged to act as a proxy synovial joint capsule with a piece of tubing arranged to allow for filling of the “joint capsule” with colored liquid; this colored liquid would then act as a reporter for successful needle placement. The assembled model and “joint capsule” were embedded in a mixture of gelatin and psyllium husk (as per Kwon, et al. 2017). EM residents (N=26) from the UCLA system of hospitals were recruited to a training intervention consisting of an online video module viewed individually and in‐person training with the embedded 3D model. Prior to intervention, residents were assessed using a practice skills checklist and 10 question quiz to measure skill level of ultrasound methods and anatomical knowledge; the cohort was re‐assessed with this checklist and quiz post‐intervention. Pre‐intervention, fewer than 50% of residents were able to successfully complete 7 of the 9 items on the skills checklist, with performance on the remaining 2 items not exceeding 70%. Performance on the quiz ranged widely, with scores as low as 2 out of 10, and a mean of 5.13/10 points. After training with our 3D‐model, all but one of the residents was able to successfully complete each individual skill on the checklist for the ultrasound task, and the mean performance on the quiz increased to 7.9/10 points. These results provide support for the use of 3D printed anatomical models as a cost‐effective avenue for clinical departments to develop effective training tools designed to meet their individual educational needs.This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.