Assessment of cerebral hemodynamic parameters using pulsatile versus non-pulsatile cerebral blood outflow models

Prior methods evaluating the changes in cerebral arterial blood volume (∆C_aBV) assumed that brain blood transport distal to big cerebral arteries can be approximated with a non-pulsatile flow (CFF) model. In this study, a modified ∆C_aBV calculation that accounts for pulsatile blood flow forward (PFF) from large cerebral arteries to resistive arterioles was investigated. The aim was to assess cerebral hemodynamic indices estimated by both CFF and PFF models while changing arterial blood carbon dioxide concentration (EtCO₂) in healthy volunteers.

Continuous recordings of non-invasive arterial blood pressure (ABP), transcranial Doppler blood flow velocity (CBFV_a), and EtCO₂ were performed in 53 young volunteers at baseline and during both hypo- and hypercapnia. The time constant of the cerebral arterial bed (τ) and critical closing pressure (CrCP) were estimated using mathematical transformations of the pulse waveforms of ABP and CBFV_a, and with both pulsatile and non-pulsatile models of ∆C_aBV estimation. Results are presented as median values ± interquartile range.

Both CrCP and τ gave significantly lower values with the PFF model when compared with the CFF model (p ≪ 0.001 for both). In comparison to normocapnia, both CrCP and τ determined with the PFF model increased during hypocapnia [CrCP_PFF (mm Hg): 5.52 ± 8.78 vs. 14.36 ± 14.47, p = 0.00006; τ_PFF (ms): 47.4 ± 53.9 vs. 72.8 ± 45.7, p = 0.002] and decreased during hypercapnia [CrCP_PFF (mm Hg): 5.52 ± 8.78 vs. 2.36 ± 7.05, p = 0.0001; τ_PFF (ms): 47.4 ± 53.9 vs. 29.0 ± 31.3, p = 0.0003]. When the CFF model was applied, no changes were found for CrCP during hypercapnia or in τ during hypocapnia.

Our results suggest that the pulsatile flow forward model better reflects changes in CrCP and in τ induced by controlled alterations in EtCO₂.