Dynamic cerebral autoregulation measurement using rapid changes in head positioning: experiences in acute ischemic stroke and healthy control populations

Abstract

The ideal technique for dynamic cerebral autoregulation (dCA) assessment in critically ill patients should provide considerable variability in blood pressure (BP) but without the need for patient cooperation. We proposed using rapid head positioning (RHP) over spontaneous BP fluctuations for dCA assessment in patients with acute ischemic stroke (AIS). Cerebral blood velocity (transcranial Doppler), beat-to-beat BP (Finometer), and end-tidal CO2 (capnography) were recorded during 5-min baseline and RHP in 16 controls (8 women and 8 men, mean age: 57 ± 16 yr) and 15 patients with AIS (7 women and 8 men, mean age: 69 ± 8 yr) at two (12 ± 8 days) and three visits (13.3 ± 6.9 h, 4.8 ± 3.2 days, and 93.9 ± 11.5 days from the symptom onset), respectively. All participants were able to complete the RHP protocol without difficulty. Compared with controls, patients with AIS were hypocapnic (all visits, P < 0.0024) and hypertensive ( visit 1, P = 0.011), although BP gradually reduced after the acute phase. RHP demonstrated greater beat-to-beat BP variability (BPV) in controls ( visits 1 and 2, P < 0.001) but not in patients with AIS at any visit. Compared with controls, a reduced autoregulation index (ARI) was demonstrated in patients with AIS, at visit 2 for the baseline recording but not at other visits or during RHP. The area under the receiver-operating curve was 0.53 and 0.54 for baseline and RHP, respectively. The RHP paradigm required minimal patient cooperation and could be considered a feasible alternative for assessing dCA, mainly in conditions leading to increased BPV. The lack of BPV increase in AIS with RHP deserves further investigation. NEW & NOTEWORTHY This study used rapid head positioning (RHP) to enhance blood pressure (BP) variability (BPV) to improve BP signal-to-noise ratio and reliability of dynamic cerebral autoregulation (dCA). RHP was well accepted by controls and acute ischemic stroke (AIS); the increased BPV induced in controls was not observed in AIS, suggesting BPV at rest was already elevated. RHP did not improve detection of impaired CA in AIS; further work is needed to understand the different responses observed.