Our study of the MACH-3 HLHS VR curriculum demonstrated effectiveness of this medical education intervention for pediatric cardiology fellows. The intervention group, exposed to the VR curriculum, outperformed the control group on the assessment tool with a medium effect size. Additionally, participants in the intervention reported high satisfaction and perceived strong educational value in the MACH-3 curriculum.

Our findings further support the use of VR as a tool in the education of complex CHD. In prior studies, the Stanford Virtual Heart was well received and demonstrated to be efficacious in teaching medical students about simple CHD [7]. Among pediatric residents, the Stanford Virtual Heart improved participants’ abilities to answer questions about visuospatial concepts in common congenital heart defects [8], which was confirmed by validated assessments [9]. Additionally, VR models have been reported to improve understanding of morphology in pediatric cardiology fellows beyond two-dimensional and didactic learning [10] and to enhance knowledge beyond traditional two-dimensional echocardiography education [11]. Our study suggests VR may also benefit fellows’ understanding of complex physiology and surgical palliation in CHD.

The application of VR in pediatric cardiology medical education is not limited to categorical fellows. Within the specialty of pediatric cardiology, VR has been used in simulation for intensive care physicians [14] and in training for advanced electrophysiology trainees in both adult and pediatric cardiology [15]. Likewise, VR may be an important tool in interprofessional education of advanced practitioners and across specialties [16]. In addition, VR may also have an important role in patient and family education; in previous studies, VR has been well received by young adults with CHD in learning about their own heart conditions [17] and in preprocedural preparedness for children with CHD [18]. Beyond education, VR has been used in both adult cardiovascular disease [19] and pediatric cardiology [20, 21] for interventional and surgical planning. The MACH-3 curriculum and similar validated curricula have a multitude of applications across the CHD community.

It is interesting to note that the largest differences between intervention and control groups were demonstrated on questions about side effects of prostaglandin E1 and sources of systemic oxygen desaturation in patients who have undergone palliation for single ventricle physiology. It would be logical to hypothesize that VR is superior to teach visuospatial concepts when compared to traditional teaching methods, yet the assessment items with the largest differences between the intervention and control groups test factual knowledge and application of physiological concepts. While our study is not designed to assess the domains of pediatric cardiology best taught by VR, it is possible that the explanation for this finding is due to the immersive nature of VR which has been suggested by previous studies to improve knowledge retention in addition to confidence and skill in graduate medical trainees [22].

The results of our study should be interpreted with several limitations in mind. The sample size was modest, although the multi-institutional design supports greater generalizability. Despite the multi-institutional design across 13 programs, the sample size may underpower the analysis. Because of variation in fellowship program sizes and the availability of VR technology at participating centers, true randomization was not possible, which may have introduced selection bias. While the participants were all within their first year of fellowship and completed the study during a similar time of the academic year, there is inherently variability in what the participants may have been exposed to outside of the study that could affect their performance on the assessment unrelated to the VR curriculum. The assessment measured only short-term effectiveness of the curriculum, as the study was not designed to test long-term knowledge retention. Qualitative feedback from the intervention group was self-reported and may be subject to response bias. Finally, the current MACH-3 curriculum and assessment are designed only for HLHS, limiting the applicability of our findings to other CHD lesions or broader learner populations.

Future directions for the MACH-3 curriculum include evaluating long-term retention, assessing clinical application, and expanding similar VR curricula to other CHD lesions and broader learner populations. Further, additional multi-center studies may best delineate how the MACH-3 curriculum and similar curricula can be integrated alongside more traditional teaching methods. It is presently uncertain how the recent changes in the pediatric residency program requirements from the Accreditation Council for Graduate Medical Education [23] will affect preparedness of the incoming pediatric cardiology fellow. Educational tools like MACH-3 may provide an enhanced educational experience for fellows with larger pre-existing knowledge gaps in CHD.

Virtual reality, exemplified in this study by the MACH-3 curriculum, represents an effective and well-received education strategy for teaching complex CHD concepts to pediatric cardiology fellows. Continued research and curricular innovations are warranted to further maximize the impact of VR in CHD education.