Introduction: Resuscitation simulations immerse learners into the complexity of emergency patient management. Head-Mounted Display Virtual Reality (VR) has been used for stress inoculation therapy for phobias and posttraumatic stress disorder. However, VR for stress inoculation in resuscitation leadership training has not been studied. We sought to develop VR simulation for stress inoculation, as exposure therapy training, for resuscitations. Methods: We explain the conceptual design, development, production, and initial evaluation process for 2 VR simulations in infant status epilepticus and pediatric anaphylactic shock. We further describe deliberate game mechanic choices to maximize psychological fidelity. In-virtual reality performance data for time-to-critical actions and stress physiology markers (heart rate, salivary cortisol) were collected from expert pediatric emergency physicians and novice pediatric residents. Data were analyzed to examine differences between the 2 groups for both outcome types to determine the extent of stress response or performance deficit the VR induced. Results: Multiple difficulties and distractions were designed for the 2 scenarios; we evaluated the highest difficulty and environmental distraction versions. Between 19 expert physicians and 15 novice physicians, no performance differences were found in typical airway, breathing, and circulation actions. Residents preferred more lorazepam first-line antiepileptics than attendings (P = 0.003) and performed a cricothyrotomy later than attendings (P = 0.02). Residents, however, manifested higher salivary cortisol levels than attendings (+0.07 μg/dL, 95% confidence interval = 0.03-0.12, P = 0.001). Conclusions: A VR resuscitation simulation manifested expected stress physiology changes in physicians. Further evaluation is needed to determine the effect of VR simulation as longitudinal stress inoculation for healthcare providers.