Low frequency power in cerebral blood flow is a biomarker of neurologic injury in the acute period after cardiac arrest

Abstract

Aim: Cardiac arrest often results in severe neurologic injury. Improving care for these patients is difficult as few noninvasive biomarkers exist that allow physicians to monitor neurologic health. The amount of low-frequency power (LFP, 0.01-0.1 Hz) in cerebral haemodynamics has been used in functional magnetic resonance imaging as a marker of neuronal activity. Our hypothesis was that increased LFP in cerebral blood flow (CBF) would be correlated with improvements in invasive measures of neurologic health.

Methods: We adapted the use of LFP for to monitoring of CBF with diffuse correlation spectroscopy. We asked whether LFP (or other optical biomarkers) correlated with invasive microdialysis biomarkers (lactate-pyruvate ratio - LPR - and glycerol concentration) of neuronal injury in the 4 h after return of spontaneous circulation in a swine model of paediatric cardiac arrest (Sus scrofa domestica, 8-11 kg, 51% female). Associations were tested using a mixed linear effects model.

Results: We found that higher LFP was associated with higher LPR and higher glycerol concentration. No other biomarkers were associated with LPR; cerebral haemoglobin concentration, oxygen extraction fraction, and one EEG metric were associated with glycerol concentration.

Conclusion: Contrary to expectations, higher LFP in CBF was correlated with worse invasive biomarkers. Higher LFP may represent higher neurologic activity, or disruptions in neurovascular coupling. Either effect may be harmful in the acute period after cardiac arrest. Thus, these results suggest our methodology holds promise for development of new, clinically relevant biomarkers than can guide resuscitation and post-resuscitation care. Institutional protocol number: 19-001327.

Keywords: Cardiac arrest; Diffuse correlation spectroscopy; Low frequency power; Microdialysis; Optical neuromonitoring.

Figure 1.

Experimental design of the post-arrest monitoring study. (A) After a baseline period, asphyxial cardiac arrest was induced by endotracheal tube clamping. Cardiopulmonary resuscitation was performed for up to 20 minutes. Return of spontaneous circulation (ROSC) was induced by defibrillation, if necessary. Animals that achieved ROSC were monitored for 4 hours with continuous optical methods and EEG, and microdialysis was sampled every 30 minutes. (B) Diagram of the placement of neurologic monitoring on the head of the animals (viewed from above; the right side of the image is the left side of the animal).

Figure 2.

Example data (from a single piglet) after the return of spontaneous circulation (ROSC) demonstrating relative blood flow index (rBFI) and relative low frequency power (rLFP) and their relationship to lactate-pyruvate ratio (LPR) and glycerol concentration. The optical tracing is shown in blue with the median value of each 30-minute interval shown in black dashed lines (left-sided axis). The red stars show the microdialysis data (right-sided axis). (A-B) Comparison of rBFI with microdialysis data. There is substantial variability in BFI in the first hour after ROSC. But overall, notice that rBFI is low after ROSC before recovering and then slightly falling, which does not match the microdialysis time courses. (C-D) Comparison of rLFP with microdiaylsis. rLFP is initially volatile, but with a high median that decreases over time before stabilizing below baseline levels. This time course follows a similar time course to that from the microdialysis data.