Hyperacute autonomic and cortical function recovery following cardiac arrest resuscitation in a rodent model

Figure 1. Schematic diagram of experimental design. ACA, asphyxial cardiac arrest; ECG, electrocardiography; EEG, electroencephalography; NDS, neurological deficit score; ROSC, return of spontaneous circulation.

Figure 2. HR dynamics following resuscitation in ACA rats and its correlation with neurological outcomes. (A) HR elevated remarkably from 15‐ to 30‐min post‐ROSC (p < 0.001 vs. 15‐min, respectively; n = 10). (B) The increase in HR at 25‐min from 15‐min post‐ROSC was significantly higher in animals with good NDS results than that in the poor outcome group (p = 0.04; n = 5 per group). (C) The AUC of increased HR (from 15‐min post‐ROSC) over 20–30 min post‐ROSC was significantly higher in the good outcome group as compared with the poor outcome group (p ≤ 0.05). (D) The changes in HR over 20–30 min from 15‐min post‐ROSC were positively correlated with NDS (rs = 0.559, p = 0.001). (E) The increased HR at 20‐, 25‐, and 30‐min from 15‐min post‐ROSC was strongly correlated with higher NDS, respectively (rs = 0.671, p = 0.04; rs = 0.814, p = 0.006; rs = 0.760, p = 0.001, respectively). ACA, asphyxial cardiac arrest; AUC, the area under the curve; BL, baseline; bpm, beats per minute; HR, heart rate; NDS, neurological deficit score; ROSC, return of spontaneous circulation. Data represent as mean ± SD. Comparisons among HR temporal changes were performed with repeated measures one‐way ANOVA with…

Figure 3. HRV analysis following resuscitation in ACA rats and its correlation with neurological outcomes. (A) Normalized LF power and LF/HF ratio over 15–30 min post‐ROSC were relatively elevated in the good outcome group as compared with the poor outcome group, especially over 15–20 min post‐ROSC (p > 0.05, respectively; n = 5 per group). (B) The increase of normalized LF power and LF/HF ratio during 15–20 min post‐ROSC was positively associated with higher NDS (rs = 0.600, p = 0.074; rs = 0.749, p = 0.046, respectively). ACA, asphyxial cardiac arrest; NDS, neurological deficit score; nHF, normalized high frequency; nLF, normalized low frequency; ROSC, return of spontaneous circulation. Data represent as mean ± SD. Comparisons between poor and good outcome groups in nLF, nHF, and LF/HF ratio were conducted by repeated measures two‐way ANOVA with Bonferroni correction; correlations between HRV parameters and NDS were performed with Spearman's correlation.

Figure 4. EEG changes following resuscitation in ACA rats and its correlation with neurological outcomes. (A) Total power over 15–30 min post‐ROSC was increased significantly in comparison with the first 15 min post‐resuscitation (vs. 0–5, 5–10, and 10–15 min, p < 0.01, respectively; n = 10). Among six EEG subbands, beta, gamma, and super‐gamma band power exhibited significant increases over 15–30 min post‐ROSC (p < 0.05 vs. 0–5, 5–10, and 10–15 min, respectively) and reached the peaks at 20–25 min during the first‐hour post‐ROSC (p ≤ 0.05 vs. 30–35, 35–40, 40–45, 45–50, 50–55, and 55–60 min, respectively). (B) Shannon entropy manifested an increase during 15–30 min post‐resuscitation (p < 0.001 vs. 0–5, 5–10, and 10–15 min, p < 0.01, respectively). Shannon entropy over 20–25 min post‐ROSC was correlated positively with NDS, though statistical significance was not found (rs = 0.608, p = 0.068). ACA, asphyxial cardiac arrest; EEG, electroencephalography; NDS, neurological deficit score; ROSC, return of spontaneous circulation. Data represent as mean ± SD. Comparisons among temporal changes of EEG parameters were performed with repeated measures one‐way ANOVA with Bonferroni correction;…

Figure 5. EEG subband power fraction over 15–30 min following resuscitation and its correlation with neurological outcomes. (A) The gamma band fractions over 15–30 min post‐ROSC were significantly higher than the other subbands, including delta, theta, alpha, and beta band fractions (p < 0.05, respectively). (B and C) Strong and positive correlations were identified between gamma band power over 15–30 min post‐ROSC and higher NDS (rs = 0.705, p = 0.027, rs = 0.693, p = 0.031, and rs = 0.614, p = 0.064, respectively) and between gamma band fraction during the same period and better neurological outcomes (rs = 0.796, p = 0.008, rs = 0.912, p < 0.001, and rs = 0.669, p = 0.04, respectively). EEG, electroencephalography; NDS, neurological deficit score; ROSC, return of spontaneous circulation. Data represent as mean ± SD. Comparisons among temporal changes of EEG parameters were performed with repeated measures one‐way ANOVA with Bonferroni correction; correlations between EEG parameters and NDS were performed with Spearman's correlation. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 6. Correlations between changes in HR/HRV and dynamics of EEG signals during the hyperacute stage after resuscitation. (A) Very strong correlations were revealed between HR changes and gamma band power at 15–20 min (rs = 0.827, p = 0.005), 20–25 min (rs = 0.912, p < 0.001), and 25–30 min (rs = 0.891, p = 0.001) post‐ROSC. (B) Gamma band fraction over 15–20 min post‐ROSC exhibited significant correlations with LF power (rs = 0.770, p = 0.013) and LF/HF ratio (rs = 0.845, p = 0.003). bpm, beats per minute; HR, heart rate; LF/HF, low frequency/high frequency; nLF, normalized low frequency; ROSC, return to spontaneous circulation. Correlations between HR/HRV parameters and EEG parameters were performed with Spearman's correlation.