The results of this study demonstrate that one brief (median 20-minute) training session using a high-fidelity three-dimensional simulation module, when used to supplement traditional lecture-based learning can significantly accelerate novice or beginner trainees’ attainment of the competency to correctly identify coronary angiographic projections. Although the effect was modest, the amount of time investment involved relative to the gain in knowledge should be noted as it sometimes takes weeks to months to achieve the same level via traditional learning. Manipulating the C-arm in simulations allows the trainee to explore virtual three-dimensional coronary anatomies actively in real-time, thereby facilitating internal mental anatomical model construction, developing hand-eye coordination skills, and improving confidence in troubleshooting technical challenges in a safe learning environment. The ability to continuously track the coronary arteries in these simulation training sessions is a distinct advantage in visual-spatial learning compared to traditional interrupted 2-dimensional representation of coronary anatomy between shots in real world angiography.
Prior studies have consistently shown the greatest impact of simulation in novice trainees, consistent with findings reported from other simulation-based studies 9 16 20 21. Correspondingly, the greatest improvement in our study was noted in ‘novice’ trainees, specifically residents, which are best representative of a new cardiology fellow with no prior cath lab experience. These findings further support the need for more studies to justify the adoption of a simulation curriculum early on in undergraduate and graduate medical education programs22.
The discrepancy between cardiology fellows and novice trainees is likely explained by the fellows’ previous attainment of the tested competency (basic anatomical identification on coronary angiograms) during their clinical training and experience. The study was conducted later in the academic year, and even our first-year cardiology fellows had already been exposed to coronary angiographic interpretation. Therefore, our described simulation training methodology should be integrated into a curriculum as part of introductory training. More broadly, our work demonstrates the importance of targeting a training protocol to the appropriate trainee. We speculate that cardiology fellows would best benefit from more advanced training protocols, such as teaching how to anticipate C-arm positioning to best visualize coronary anatomy.
To our knowledge, this is the first randomized controlled study to investigate the additive role of high-fidelity simulation training to traditional methods in teaching basic coronary angiography view interpretation to junior physicians. A recent study from France by Fischer et al23 randomized 118 medical school students into simulation and traditional power-point based teaching. They reported that the simulation group did better in identifying coronary anatomy and coronary angiographic projections after a single simulation session. Although our main findings were similar, our study design is different. Firstly, we recruited a spectrum of trainees at various levels of clinical training, beyond medical students, to examine if more experienced trainees would benefit. Next, in order to allow for different learning speeds and preferences, subjects in the simulation arm were provided one-on-one instruction at the start of the exercise and then allowed independent unobserved practice time with no restriction on the amount of time spent on the simulator. Finally, since our subjects had varying amounts of exposure to clinical cardiology and familiarity with coronary angiography, we decided to focus on improvement in test performance from baseline, pre-intervention to post-intervention (delta scores) as our major primary outcome rather than an isolated post-intervention score by itself as reported by Fischer et al23.
There are limitations to our study that are inherent with our sample size and study design. Our results for fellows are likely affected by their small sample size and varied amount of exposure to CA prior to the study. However, a study specific to cardiology fellows would require a multi-year and multi-center study, which would be limited by the general availability of coronary simulators. We did not perform quantitative assessment of baseline visuospatial skills of our study participants and so their influence if any on the study outcome is unknown. We were also unable to explore the effect of a single structured simulation session on long-term retention. Also, the additive effect of periodic booster training sessions on knowledge acquisition and retention was not studied. It would have been interesting to see if subjects in the control arm would have benefitted from crossing over to simulation training at the end of the study by administering a repeat assessment. This study was not blinded, but as the outcomes measurement was the performance on a multiple choice question test, the lack of blinding is unlikely to have caused bias.