Implementation and Evaluation of a Perceptual Learning Module on Visual Estimation of LVEF

Medical Education

This study is designed to: 1. Implement a Web platform intended to host perceptual learning modules (PLMs) 2. Implement and assess a PLM designed to improve the capacity of first-year residents and fourth-year medical students to visually estimate the left ventricular ejection fraction (LVEF) with transesophageal echocardiography (TEE) images. The hypothesis of the study is that the PLM will improve the visual assessment of LVEF by TEE in junior residents and medical students.

TEE is increasingly used for perioperative diagnosis and monitoring. However, learning TEE may be long and difficult: guidelines suggest that learners perform and interpret hundreds of TEE exams during their training. Even if this learning process has been used successfully for many years, it has important limitations, especially in the context of an ever-increasing number of trainees: access to patients and TEE experts is limited, and the extensive period of time required to train is also a problem. Simulators have gained popularity as tools for teaching echocardiography, showing benefits mostly regarding probe handling and image acquisition. Some simulators may also be useful to replicate some diseases, but most simulators can only replicate a limited variety of diseases, and mostly typical presentations. This is an important limitation when it comes to teaching image interpretation. In order to develop their expertise, trainees have to be exposed to a wide range of normal variations and to more subtle abnormalities. PLMs represent an alternative modality to teach image interpretation, and are already used in other fields. PLMs are online tools created to enhance the development of pattern recognition, a skill required to efficiently interpret TEE images. With PLMs, learners observe a series of images or videoclips and must answer a question about each image in a short period of time. Each answer is followed by an immediate feedback (the correct answer). By observing a large number of images online and by receiving systematic and expert feedback, the trainee quickly learns to extract efficiently the required information from the image. This study will aim to evaluate the effects of a PLM on the ability of junior residents and medical students to correctly estimate visually the left ventricular ejection faction using TEE images. Methods: A team of experienced programmers will develop a Web platform, which will allow the identification of both participants and TEE experts, manage the consent process, host the tests used to assess the participants' expertise and the PLM itself, as well as manage the TEE images and the study data. The pilot study designed to assess the impact of the PLM on the visual estimation of LVEF will be composed of the following steps: 1. selection and validation of TEE cases, 2. development of two versions of a test and of the PLM, 3. analysis of the impact of the PLM on participants' performance vs a control group. Each TEE case used in this study will comprise 3 short video loops containing the following standardised echocardiographic views: * midesophageal four chamber view * midesophageal two chamber view * midesophageal long axis view Prior to inclusion in the study, each case will be anonymized and then validated by two cardiologists expert in TEE interpretation. Only cases where the visual estimation of the LVEF provided independently by the two TEE experts will differ by no more than 10% will be used. Two different sets of validated images will then be created. The first set will be comprised of two groups of 20 validated cases (total of 40 cases). These cases will be used to create two versions of a test used to assess participants' expertise in the visual estimation of LVEF at different stages in the research protocol. These cases will not be part of the PLM. The second set will be comprised of 96 validated cases destined to the PLM. The PLM is designed not to assess the participants' expertise, but to develop their ability to estimate LVEF. When taking a test on the platform, participants will see the three short video loops of each case simultaneously on the screen for a maximum of 20 seconds, and will then have to provide an estimation of the LVEF. Each answer will be followed immediately by another case, and no feedback will be provided. When doing the PLM, participants will see the three short video loops of each case simultaneously on the screen for a maximum of 20 seconds, and will then have to provide an estimation of the LVEF. However, each answer will be followed by feedback in the form of the average value of the experts' visual estimates of LVEF. This sequence of case visualisation - participant estimation - expert feedback will be repeated 96 times during this PLM. Residents and medical students will be randomised by the platform to a Control Group and a PLM group. Both groups will participate in two on-line sessions. During the first session, demographic and clinical experience data will be collected. All participants will also perform a pretest to measure their baseline ability to estimate LVEF. Then only the participants randomised to the PLM group will do the PLM. All participants will do a posttest at the end of the first session. Six months later, all participants will return for a second session to perform the delayed test. At the end of the study, participants of the Control group will be allowed to complete the PLM if they wish to.

Canada