Global and regional differences in cerebral blood flow after asphyxial versus ventricular fibrillation cardiac arrest in rats using ASL-MRI

Abstract

Both ventricular fibrillation cardiac arrest (VFCA) and asphyxial cardiac arrest (ACA) are frequent causes of CA. However, only isolated reports compared cerebral blood flow (CBF) reperfusion patterns after different types of CA, and even fewer reports used methods that allow serial and regional assessment of CBF. We hypothesized that the reperfusion patterns of CBF will differ between individual types of experimental CA. In a prospective block-randomized study, fentanyl-anesthetized adult rats were subjected to 8min VFCA or ACA. Rats were then resuscitated with epinephrine, bicarbonate, manual chest compressions and mechanical ventilation. After the return of spontaneous circulation, CBF was then serially assessed via arterial spin-labeling magnetic resonance imaging (ASL-MRI) in cortex, thalamus, hippocampus and amygdala/piriform complex over 1h resuscitation time (RT). Both ACA and VFCA produced significant temporal and regional differences in CBF. All regions in both models showed significant changes over time (p<0.01), with early hyperperfusion and delayed hypoperfusion. ACA resulted in early hyperperfusion in cortex and thalamus (both p<0.05 vs. amygdala/piriform complex). In contrast, VFCA induced early hyperperfusion only in cortex (p<0.05 vs. other regions). Hyperperfusion was prolonged after ACA, peaking at 7min RT (RT7; 199% vs. BL, Baseline, in cortex and 201% in thalamus, p<0.05), then returning close to BL at ∼RT15. In contrast, VFCA model induced mild hyperemia, peaking at RT7 (141% vs. BL in cortex). Both ACA and VFCA showed delayed hypoperfusion (ACA, ∼30% below BL in hippocampus and amygdala/piriform complex, p<0.05; VFCA, 34-41% below BL in hippocampus and amygdala/piriform complex, p<0.05). In conclusion, both ACA and VFCA in adult rats produced significant regional and temporal differences in CBF. In ACA, hyperperfusion was most pronounced in cortex and thalamus. In VFCA, the changes were more modest, with hyperperfusion seen only in cortex. Both insults resulted in delayed hypoperfusion in all regions. Both early hyperperfusion and delayed hypoperfusion may be important therapeutic targets. This study was approved by the University of Pittsburgh IACUC 1008816-1.

Keywords: Brain; Cardiac arrest; Cerebral blood flow; Magnetic resonance imaging; Resuscitation.

Figure 1

Representative CBF maps of rat brains, including baseline and post-resuscitation up to RT60 in both ACA and VFCA groups. CBF, cerebral blood flow; RT, resuscitation time; ACA, asphyxial cardiac arrest; VFCA, ventricular fibrillation cardiac arrest.

Figure 2

Analysis of the CBF map for the left hemisphere for ACA and VFCA groups, respectively. The bars represent SD. CBF, cerebral blood flow; ACA, asphyxial cardiac arrest; VFCA, ventricular fibrillation cardiac arrest.* p<0.05 ACA vs. respective baseline; # p<0.05 VFCA vs. respective baseline; a, p<0.05 ACA vs. VFCA; b, p<0.01 ACA vs. VFCA.

Figure 3

Regional analyses of the CBF maps for the individual regions of left hemisphere for ACA and VFCA groups, respectively. A, cortex; B, thalamus; C, hippocampus; D, amygdala/piriform complex. The bars represent SD. CBF, cerebral blood flow; ACA, asphyxial cardiac arrest; VFCA, ventricular fibrillation cardiac arrest.* p<0.05 ACA vs. respective baseline; # p<0.05 VFCA vs. respective baseline; a, p<0.05 ACA vs. VFCA; b, p<0.01 ACA vs. VFCA.

Figure 4

Regional analyses of the CBF maps in ACA group. There was an overall difference between groups (p=0.003). a, p<0.05 thalamus vs. amygdala/piriform complex; b, p<0.01 thalamus vs. amygdala/piriform complex; c, p<0.05 cortex vs. amygdala/piriform complex; d, p<0.01 cortex vs. amygdala/piriform complex; e, p<0.05 thalamus vs. hippocampus. SD bars are omitted to improve clarity of the figure.

Figure 5

Regional analyses of the CBF maps in VFCA group. There was an overall difference between groups (p=0.005). a, p<0.01 cortex vs. thalamus; b, p<0.05 cortex vs. amygdala/piriform complex; c, p<0.05 cortex vs. hippocampus; d, p<0.01 cortex vs. amygdala/piriform complex; e, p<0.01 cortex vs. hippocampus. SD bars are omitted to improve clarity of the figure.