Reduced harmonic complexity of brain parenchymal cardiovascular pulse waveforms in Alzheimer’s disease

• MREG (TR=100ms) enables analysis of brain cardiovascular waveforms • With relative harmonic power it is possible to quantify cardiovascular pulse shape • Cardiovascular pulse waveforms are more monotonous in Alzheimer's disease (AD) • Analysis method provides a non-invasive way to quantify AD pathology Alzheimer’s disease (AD) is characterized by specific neuropathologies, and is associated with arterial wall β-amyloid accumulations, which lead to radiologically detectable amplitude increases and variable propagation speed of cardiovascular impulses in brain. In this study, we developed a fast frequency domain imaging method know as relative harmonic power of magnetic resonance encephalography (MREG RHP ), aiming to investigate the configuration of the cardiovascular impulses independently of the mean magnetic resonance signal intensity and physiological impulse amplitude. In the initial analyses in healthy controls, we found that a wide 0.8 - 5Hz bandpass produced the most physiologically realistic cardiovascular waveforms. Whereas the data recorded in cerebrospinal fluid (CSF) with flip angle (FA) of 25° yielded up to 7-fold higher cardiac signal intensity as compared to FA of 5°, within the brain tissue recordings with FA of 5° were markedly more sensitive to cardiac waveform. We detected arterial impulses originating from major arteries and extending into the surrounding brain parenchyma, with simultaneous dampening of amplitude as a function of distance from source. Finally, we compared MREG RHP results in 34 AD patients (mean age: 60.7±4.7 years; 53% female) against 29 controls (mean age: 56.9±7.9 years; 66% female). We show that the harmonic power of cardiovascular brain pulses is significantly reduced in cortical frontoparietal areas of AD patients, indicating monotonous impulse patterns colocalizing with the previously reported areas of increased impulse propagation speed. In conclusion, the MREG RHP offers a fast Fourier transform (FFT)-based method to non-invasively quantify and locate human arterial blood vessel wall pathology.