An fMRI approach to assess intracranial arterial-to-venous cardiac pulse delay in aging.

Each heartbeat generates a cardiac pressure wave that propagates through the brain and travels from large arteries through cerebrospinal fluid and brain tissue, compressing the venous sinuses and producing venous blood pulsatility. The delay between arterial and venous pulsation (A-V delay) is an insightful marker of intracranial compliance and the intracranial mechanical environment. We developed a novel approach to extract A-V delay from conventional resting-state functional MRI (fMRI) scans, leveraging fMRI's sensitivity to vessel pulsations in large cerebral arteries and the superior sagittal sinus (SSS). This fully automated method was applied to the Human Connectome Project - Aging dataset to analyze 578 participants aged 35 to 90 years. The mean A-V delay was 78 ± 32 msec; it shortened by 4 msec for every decade of aging and was 12 msec faster in men than women, highlighting age-related and sex-specific differences. We also identified a within-SSS pattern of pulsations, characterized by an earlier posterior pulsation and a later anterior pulsation. This pattern opposes the direction of blood flow, supporting that the SSS is passively compressed and tied to a distinct intracranial pulse transmission. Overall, this work demonstrates the feasibility of extracting an fMRI-based A-V delay, uncovering a previously unexplored capability of fMRI. This approach broadens the potential applications of fMRI by adding a biomechanical dimension to fMRI's established roles in evaluating neuronal and hemodynamic function. Given the widespread availability of fMRI, this approach can be applied in future studies to investigate biomechanical changes in various disease conditions.