Modeling Brain Hypoxia to Predict Clinical Outcomes

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Impaired O2 delivery to the brain is thought to be a major etiologic factor in the neurocognitive dysfunction experienced by patients after certain surgical procedures and after exposure to carbon monoxide (CO). Knowledge of the extent to which these events impair oxygen (O2) delivery to specific regions of the brain is key to understanding the extent of neurologic damage and to assessing the efficacy of treatment strategies designed to prevent neurologic sequelae in these patients. For CO-poisoned patients, for example, there is neither an accepted clinical measure that predicts neurological outcome nor agreement regarding the efficacy of normobaric (NBO) vs. hyperbaric (HBO) hyperoxia, treatments that are presumed to increase partial pressure of oxygen (PO2) in brain tissue and minimize injury. MR imaging studies of CO-poisoned patients, however, are in general agreement regarding the specific regions of the brain in which gross changes occur (e. g., frontal cortex and white matter, basal ganglia, hippocampus). We propose to develop a mathematical model of tissue PO2 distributions in specific brain regions and test the hypothesis that model-based estimates of regional brain tissue hypoxia will correlate with neurological outcome in CO-poisoned patients. Our model will represent 6 specific regions of the brain: frontal cortex, frontal white matter, basal ganglia, hippocampus, generic gray matter, and generic white matter. In this R03 (Small Research Grant) proposal, we will first expand our existing model of O2 delivery to 1 brain region so that it represents the 6 specified brain regions. Next we will use our model to estimate, by retrospective analyses of a large (existing) clinical database, the duration and degree of regional brain hypoxia in individual CO-poisoned patients during both CO exposure and therapy. Finally, we will correlate the clinical outcomes (MRI findings, neurologic sequelae) from CO-poisoned patients with model-derived indices of the degree, duration, and spatial extent of local hypoxia in specific regions of the brain. We anticipate that this modeling approach will help to explain outcome and enhance the understanding of the efficacy of HBO vs. NBO in individual cases of CO poisoning. Furthermore, with appropriate modifications, this simulation modeling approach can be applied to other clinical and experimental situations involving impaired O2 delivery to the brain such as stroke and cardiac surgery.