Signal evolution and infarction risk for apparent diffusion coefficient lesions in acute ischemic stroke are both time-and perfusion-dependent

BACKGROUND AND PURPOSE-This study aimed to examine the temporal relationship between tissue perfusion and apparent diffusion coefficient (ADC) changes within 6 hours of ischemic stroke onset and how different reperfusion patterns may affect tissue outcome in ADC lesions. METHODS-Thirty-one participants were sequentially imaged at 3 hours, 6 hours, and 1 month post-stroke. Three regions of interest (ROIs) were defined within initial ADC lesions: ROI (1)_reperf-3hour hyperacute reperfusion (within 3 hours), ROI (2)_reperf-6hour acute reperfusion (3 to 6 hours), and ROI (3)_nonreperf no reperfusion (by 6 hours). For each ROI, changes in ADC (ΔADC) from 3 to 6 hours and risks of infarction were examined. RESULTS-The magnitude of initial ADC reduction was similar in all 3 ROIs (P=0.51). ΔADC was strongly associated with reperfusion (P<0.0001) but not with initial ADC reduction (P=0.83). ΔADC in ROI (1) _reperf-3hour and ROI (2)_reperf-6hour was significantly larger than that of ROI (3)_nonreperf (P<0.05). Positive ΔADC was obtained from 3 to 6 hours in ROI (1)_reperf-3hour that had restored perfusion before 3 hours, demonstrating a temporal delay between reperfusion and ADC changes. Risks of infarction were significantly higher in ROI (3)_nonreperf than those in ROI (1)_reperf-3hour and ROI (2)_reperf-6hour. CONCLUSIONS-Improvement in ADC did not occur coincidently with reperfusion but showed a temporal delay. Regions with similar initial ADC reductions at 3 hours had different evolution of ADC and infarction risks depending on when or if tissue reperfused. These findings provide a physiological basis for the observation that a single ADC measurement at a fixed time after stroke onset may not accurately predict tissue outcome.