A comparison of the time constant of the cerebral arterial bed using invasive and non-invasive arterial blood pressure measurements

Abstract

Abstract Objective: The time constant of the cerebral arterial bed (τ), which is an index of brain haemodynamics, can be estimated in patients using continuous monitoring of arterial blood pressure (ABP), transcranial Doppler cerebral blood flow velocity (CBFV) and intracranial pressure (ICP) if these measures are available. But, in some clinical scenarios invasive measurement of ABP is not feasible. Therefore, in this study we aimed to investigate whether invasive ABP can be replaced with non-invasive ABP, monitored using the Finapres photoplethysmograph (fABP). Approach: Forty-six recordings of ICP, ABP, fABP, and CBFV in the right and left middle cerebral arteries were performed daily for approximately 30 min in 10 head injury patients. Two modelling approaches (constant flow forward [CFF, pulsatile blood inflow and steady blood outflow] and pulsatile flow forward [PFF, where both blood inflow and outflow are pulsatile]) were applied to estimate τ using either invasive ABP (τCFF, τPFF) or non-invasive ABP (fτCFF, fτPFF). Main results: Bland–Altman analysis showed quite poor agreement between the fτ and τ methods of estimation. The fτ method produced significantly higher values than the τ method when calculated using both the CFF and PFF models (p < .001 for both). The correlation between fτCFF and τCFF was moderately high (r s = 0.63; p < .001), whereas that between fτPFF and τPFF was weaker (r s = 0.40; p = .009). Significance: Our results suggest that using non-invasive ABP for estimation of τ is inaccurate in head injury patients.