Abstract 11: Early Magnetic Resonance Imaging to Quantify Ischemic Brain Insult Early Following Cardiac Arrest in a Swine Model

Abstract

Introduction: Hypoxic ischemic brain injury (HIBI) results from loss of oxygen or perfusion during cardiac arrest and is the most common cause of death/disability in patients surviving to admission. Diagnostic tests are crucial to identify irreversible injury thresholds for neuroprognostication. MRI DTI can detect cellular injury in response to HIBI, but imaging timing and severity of hypoxia vary widely. Hypothesis: We hypothesized that a model with fixed periods of ischemia followed by both standardized reperfusion and time to image acquisition would demonstrate a reliable gradient of injury. Such a model would allow neuroprotective strategies to be tested using early imaging biomarkers to quantify effects on cytoarchitecture. Methods: Eleven healthy swine (49 + 5 kg) under isoflurane anesthesia underwent control MRI with DTI at b2000/20 directions. Two animals each were then randomized to sham, 10, 15, 20, and one to 30-minutes of arrest. For reperfusion, extracorporeal membrane oxygenation (ECMO) catheters were placed in the distal aorta and right atrium, after which ventricular fibrillation was induced by a bipolar pacing catheter. Following the fixed period of ischemia, animals were reperfused at 2.8-3.5 L/min with unblended oxygenated blood. After 4 minutes they were defibrillated. Over two hours they were weaned from pump and decannulated with epinephrine and fluid boluses for hemodynamic support. Animals were re-imaged within two hours post-resuscitation using the same pre-arrest protocol. Whole brain ADC measurements were performed in the gray and white matter. Results: Baseline hemodynamic parameters: BP, Heart Rate, End-Tidal CO2, and SpO2 were similar between animals. When comparing pre- to post-arrest MRI, there was a strong linear correlation between percent gray matter ADC change (r=-0.87, p<0.001) and white matter ADC change (r=-0.79, p<0.001) with ischemia duration. Conclusion: MRI DTI changes observed following fixed periods of ischemia with ADC in gray and white matter correlated well with ischemia duration in the first few hours post arrest. These findings support the feasibility of MRI for measuring severity of insult in HIBI and an animal model that might serve as translational endpoint for neuroprotectant strategies.