Real-Time Brain Monitoring by Near-Infrared Spectroscopy Predicts Neurological Outcome after Cardiac Arrest and Resuscitation in Rats: A Proof of Concept Study of a Novel Prognostic Measure after Cardiac Arrest

Figures

• 

  Figure 1. Flow diagram of the experimental groups. Data for two samples (in 6 and 9 min CA groups) in the PCR analysis were missing, due to technical failures. Brain samples from sham-operated rats were used for brain immunohistochemistry (n = 2) and PCR analysis (n = 7). EEG; electroencephalography, O2; Oxygen, CO2; carbon dioxide, PCR; polymerase chain reaction, CA; cardiac arrest, rSO2; regional cerebral oxygen saturation, ROSC: return of spontaneous circulation, NFS; neurological function score, ROC analyses; receiver operating characteristic analyses.

• 

  Figure 2. Brain electrical recovery after cardiac arrest/cardiopulmonary resuscitation. The duration of both the appearance of the first EEG amplitude activity and continuous EEG activity were significantly longer in the 12 min CA group than in the 6 min CA group. Data are presented as mean ± SD. n = 5 each group. * p < 0.05, **** p < 0.0001. CA: cardiac arrest.

• 

  Figure 3. Time-dependent changes in cerebral oxygen metabolisms between 6 and 12 min cardiac arrest and cardiopulmonary resuscitation. (Left) Cerebral O2 consumption and (right) cerebral CO2 production at baseline, 20 min, and 60 min after the return of spontaneous circulation in each group; n = 5 each group. Data are presented as mean ± SD. # p < 0.05, ## p < 0.01, #### p < 0.0001 vs. at baseline in the same group, † p < 0.05, ††† p < 0.001 vs. at 20 min in the same group, *** p < 0.001 between each group.

• 

  Figure 4. Daily neurological functional scores (NFS) in the good and poor outcome groups after the return of spontaneous circulation (ROSC). The good neurological outcome group exhibited markedly higher NFS at 24, 48, and 72 h post-ROSC. Lines indicate mean; *** p < 0.001, **** p < 0.0001.

• 

  Figure 5. Serial change in regional cerebral oxygen saturation (rSO2) during first 2 h after the return of spontaneous circulation (ROSC) in the good (a red line) and poor outcome (a blue line) groups. The 72-h survived rats were classified into binary groups using the NFS cut-off for good (NFS ≥ 60) and poor (NFS < 60) outcomes at 72 h post-ROSC. The rSO2 value increased rapidly with ROSC and decreased to a “nadir” level after reaching the peak level. After reaching the nadir, the rSO2 value increased gradually, showing a biphasic pattern. Fraction of inspired oxygen (FIO2) was maintained at 1.0 after initiation of CPR. FIO2 was gradually reduced to 0.6, 0.3, and 0.25 at 110, 115, and 120 min, respectively, followed by disconnection from mechanical ventilator and extubation. There were no significant differences in rSO2 over time between the good and poor neurological outcome groups; p = 0.92 for the two-way analysis of variance. The red and blue arrows indicate the “nadir” of rSO2 in the good and poor outcome groups, respectively. Data are presented as mean ± SD.

• 

  Figure 6. The association of Tnadir with neurological function score (NFS) at 72 h post-return of spontaneous circulation. (A) Animals in the good outcome group exhibited a shorter Tnadir than those in the poor outcome group (22.2 ± 3.2 min vs. 31.5 ± 7.7 min, *** p < 0.001). Closed circles indicate Tnadir in each individual. The tentacles indicate mean ± SD in each group. (B). There was a significant negative correlation between Tnadir and NFS at 72 h after return of spontaneous circulation (r = −0.48; p = 0.007). Solid line indicates a correlation coefficient (r). Dashed lines show 95% confidence intervals.

• 

  Figure 7. ROC analyses. (A) The ROC-AUC revealed that a Tnadir cut-off of 24 min provided optimal sensitivity and specificity for predicting good neurological outcomes (AUC, 0.88; 95% CI, 0.75–1.0; sensitivity, 88.9%; specificity, 85.7%; p = 0.001). (B) The AUC of plasma IL-6 levels at 1 h after return of spontaneous circulation, for predicting good neurological outcomes, was 0.59 (95% CI, 0.35–0.83; p = 0.47). The AUC of Tnadir was significantly superior to that of plasma IL-6 (p = 0.027). AUC: area under the curve, CI: confidence interval, IL-6: interleukin 6.

• 

  Figure 8. The associations between Tnadir and biomarkers in the post-cardiac arrest brain. The levels of (A) interleukin (IL)-6, (B) IL-1β, (C) tumor necrosis factor (TNF) α, (D) high-mobility group box 1 (HMGB-1), (E) dynamin 1-like protein (Dnm1L), (F) fission 1 protein (Fis 1), and (G) mitochondrial elongation factor 1 (Mief 1) in the brains of sham-operated rats and rats 2 h post return of spontaneous circulation. Data are presented as mean ± SD. n = 7, 10, and 21 in sham, Tnadir ≤ 24 min, and Tnadir > 24 min groups, respectively; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

• 

  Figure 9. Fluoro-Jade B (FJB) immunostaining in the cortex, caudoputamen, and hippocampus. (Upper) Representative photomicrographs of brain sections (cortex, caudoputamen (CPu), and hippocampus (Hipp) CA1 region), showing FJB-positive cells in sham-operated rats and post-cardiac arrest rats having Tnadir ≤ 24 min and Tnadir > 24 min. (Lower) Number of FJB-positive cells in the cortex, caudoputamen, and hippocampus regions in survived rats having Tnadir ≤ 24 min (n = 9) and Tnadir > 24 min (n = 6). Data are presented as mean ± SD. * p < 0.05.