Umeå universitets logga

Background: Neurofluid flow dynamics are frequently studied from asynchronous blood and CSF flow measurements from real-time imaging using separate phase contrast (PC) MRI scans. Asynchronous measures can be influenced by changes in heart rate, respiration, and other physiological processes, obfuscating neurofluids assessment. Here we present an approach for synchronous measures of neurofluids using simultaneous real-time 2D PC MRI and investigated the effects of different breathing patterns on synchronous and asynchronous blood and CSF flow in a group of healthy participants.

Methods: Interleaved dual-velocity encoding 2D PC MRI with retrospective real-time reconstruction was utilized for synchronous neurofluid measures during free breathing, paced breathing and breath holds. Data were collected on a clinical 3.0T scanner at the level of C1/C2 vertebrae in 10 participants. From real-time images, flow rates repeated measures, and cardiac and respiratory flow power were assessed using Bland-Altman, power spectral analyses, and breathing pattern group differences. Neurofluids coupling from cross-correlation between arterial and venous blood and CSF flow signals were quantified from synchronous and asynchronous measures. Real-time images were re-binned to the cardiac cycle and compared to high-temporal resolution cardiac-resolved images in terms of flow rate, pulsatility index, and stroke volume.

Results: Flow repeatability was highest in free breathing scans and in arteries and spinal canal compared to veins from Bland-Altman and repeatability coefficients. Significant differences were measured in cardiac and respiratory flow power across breathing patterns in various vessel segments and spinal canal (P ≤ 0.006). Synchronous blood and CSF cardiac coupling were significantly higher than asynchronous results in all vessels (P = 0.002). For example, free breathing synchronous cardiac couplings ranged from [0.81, 0.93], compared to asynchronous range [0.49, 0.53]. Synchronous cardiac coupling showed significant differences across breathing patterns in most vessels (P = 0.002). Comparison between real-time cardiac re-binned images and high-temporal resolution cardiac-resolved images showed high correlations for flow rate and spinal canal stroke volume (ρ ≥ 0.95) and lower for pulsatility index (ρ = [0.45, 0.88]).

Conclusions: Breathing patterns induced significant responses across neurofluids including in flow rates, flow power, and coupling parameters. Higher cross-correlation among synchronous measures support benefits over asynchronous measures for neurofluids coupling characterization.