Abstract Body (Do not enter title and authors here): Introduction:
Cerebral blood flow (CBF) signals contain physiological information about the dynamics of the heart, brain, and their interaction during ischemic stroke. Traditional frequency-based analyses may miss critical dynamic changes due to inter-subject variability. We hypothesize that a time-frequency approach can extract more detailed information from CBF’s physiological frequency bands to better detect stroke-induced cerebral hemodynamic changes. This study applies empirical mode decomposition (EMD) and the Hilbert transform to extract dynamic-specific features from CBF signals in a rat stroke model.
Methods:
Sixteen rats underwent transient right middle cerebral artery (MCA) occlusion, with CBF recorded continuously from the right cortex. 5-minute CBF recordings were analyzed at three timepoints: baseline, 1 hour post-MCA occlusion, and 3 hours post-reperfusion. EMD decomposed each signal into 13 intrinsic mode functions (IMFs), with instantaneous frequencies extracted via the Hilbert transform. IMFs were then recombined into five physiological bands: cardiac (~2–5 Hz), respiratory (~0.4-2 Hz), myogenic (~0.15-0.4 Hz), sympathetic (~0.04-0.15 Hz), and endothelial (~0.0095-0.04 Hz). The Hilbert transform was then applied to each band to compute a 95% area metric (CBF 95%-Area Index) from the analytic signal’s complex-plane trajectory. Statistical tests compared physiological states across timepoints. P-value < 0.05 was considered significant.
Results:
The myogenic band (~0.15-0.4 Hz), associated with vascular smooth muscle activity, showed a significant reduction in the 95% area metric from baseline to occlusion (p<0.0001), and from baseline to reperfusion (p<0.05). No significant change between occlusion and reperfusion, suggesting persistent suppression of myogenic vasomotion during early reperfusion. Other bands showed minor or nonsignificant changes.
Conclusion:
Persistent myogenic oscillatory suppression during early reperfusion suggests a form of cerebral vascular stunning, analogous to stunned myocardium, where contractile function remains impaired despite restored perfusion. This may represent post-ischemic vascular dysfunction in the brain. Alternatively, it may reflect a microvascular no-reflow phenomenon within the cerebral circulation. Frequency-resolved CBF analysis offers a noninvasive approach to detect reperfusion-related vascular dysfunction, with the potential to guide acute stroke therapies and improve cerebrovascular outcomes.