Abstract 181: MR Oxygen Metabolic Index (OMI) Thresholds for Delineating Core, Penumbra, and Oligemia in Acute Ischemic Stroke Patients
Background: PET cerebral oxygen metabolic rate (CMRO2) provides a measure of tissue viability during acute ischemic stroke. Oxygen metabolic index (OMI), an MR-derived parameter closely related to CMRO2, was developed as a more accessible imaging approach for acute stroke patients.
Methods: 38 acute ischemic stroke patients were imaged at 3.0 hours (tp1), 6.2 hours (tp2), and 1 month (tp3) after onset. Dynamic susceptibility contrast method measured CBF. An asymmetry spin echo sequence was used to calculate OEF; OMI=CBFxOEF. Co-registered voxels were normalized to the non-ischemic hemisphere. Two OMI thresholds to distinguish core from penumbra and penumbra from oligemia were derived by separating brain tissue into 4 groups: (1) tissue that died regardless of reperfusion (core); (2) tissue that died without reperfusion (nonreperfused_penumbra); (3) tissue that survived with reperfusion (reperfused_penumbra); and (4) tissue that survived regardless of reperfusion (oligemia). An exhaustive search method was used to determine an optimal pair of core/penumbra and penumbra/oligemia thresholds for each patient by minimizing the “average prediction error (APE)”, a metric averaging the differences for each tissue group’s actual infarct probability (IP) from the ideal IP (Table). The predictive abilities of the thresholds were tested in the same cohort using “leave-one-out” cross-validation, and the APE averaged across the population.
Results: The core/penumbra OMI threshold and the penumbra/oligemia OMI threshold ranged from 0.21-0.22 and 0.41-0.43, respectively. The median IP’s [IQR] for the 4 tissue groups are shown (Table). The population averaged APE was 15% [5%, 24%].
Conclusions: OMI thresholds delineated the ischemic penumbra with high predictive ability: reperfused penumbra demonstrated low IP (6.79%), while non-reperfused penumbra demonstrated high IP (73.57%). These thresholds will require further testing in independent patient cohorts.
- © 2012 by American Heart Association, Inc.