Establishment of an MR-Conditional Porcine Model for Real-Time Assessment of Cerebral Blood Flow During Extracorporeal Circulation

Background and Purpose: Neurological injury remains a major complication of pediatric cardiac surgery and is closely related to alterations in cerebral blood flow during extracorporeal circulation (ECC). However, the real-time assessment of cerebral perfusion under these conditions has been limited by the lack of magnetic resonance (MR)-compatible perfusion systems. The aim of this pilot feasibility study was to establish a porcine model enabling simultaneous cardiopulmonary bypass (CPB) and real-time MR-based assessment of cerebral blood flow during simulated pediatric cardiac surgery. Methods: We conducted a pilot study on 11 Duroc pigs (14.6 ± 1.4 kg BW), designed in iterative cycles. The experimental setup included an MR-conditional heart-lung machine and a surgical protocol closely mimicking pediatric cardiac surgery. After the initiation of CPB and hemodynamic stabilization, animals were cooled to target temperatures (20 °C or 28 °C) depending on the perfusion strategy. Structural and functional MRI, including phase-contrast imaging, arterial spin labeling, diffusion-weighted imaging, and MR spectroscopy, were performed during cooling and rewarming. Procedural feasibility, technical challenges, and optimization strategies were systematically documented. Results: The study successfully established a reproducible porcine model enabling MR imaging during extracorporeal circulation. Key technical challenges, including vascular access, cannulation of the ascending aorta, and blood volume management, were identified and addressed through the iterative refinement of the surgical and perfusion protocols. The use of the Seldinger technique significantly improved cannulation safety and reduced blood loss. Stable CPB conditions and target hypothermic temperatures were achieved in successfully cannulated animals. MRI acquisition during CPB was feasible, providing simultaneous structural and functional assessment of cerebral perfusion. Representative imaging data demonstrate the capability of the model to capture cerebral hemodynamics in real time. Conclusions: This pilot study establishes a novel MR-compatible porcine model for the real-time assessment of cerebral blood flow during extracorporeal circulation. The platform provides a robust foundation for future quantitative investigations of cerebral perfusion, mechanisms of brain injury, and neuroprotective strategies in pediatric cardiac surgery.