Suppression of autonomic drive determined by nonlinear HRV analysis in therapeutic hypothermia after cardiac arrest

Therapeutic hypothermia is recommended as a cardiac arrest treatment for neuroprotection. In this study, we aimed to explore nonlinear heart rate variability (HRV) with other domains to clearify and setting up the significant index for survival. All patients were divided into two groups by 7 days survival outcome (n = 9 vs. 9). The limb lead II ECG signals are collected (10 minutes, sampling rate: 1000 Hz) for all phases; 6, 18 hours after induction (A and B), during rewarming (C) and recover (D). Fluctuation of heart rate (HR) was presented in both groups especially in C and D phases. Standard deviation of RR intervals (SDNN) was significantly decreased from phase A to C in survivors (p<;0.05). Abrupt increase in SDNN was observed in non survivors and caused a significant difference compared to survivors in C phase. During A-C, low-frequency to high-frequency power ratio (LF/HF) and SD2 of Poincare's plot were suppressed in survivors and they slightly increased in D phases. Less suppression of those LF/HF ratio and SD2 was evident in non survivors. Gradual increase in sample entropy (SampEn) was consistent with prolong suppression of LF/HF ratio and SD2 in survivors. Significant difference of SampEN in both groups was presented at C phase. Prominent SampEn at rewarm of survivors was consistent with large and more scatter signals in Poincare's plot that may explain the rebound response after autonomic suppression. Contrast to survivors, less and ineffective autonomic suppression from hypothermia lead to a crisis in rewarming phase. Also, we notice significant difference of SDNN, SD2 and SampEN in non survivors. In conclusion, nonlinear HRV with SDNN and LF/HF ratio are able to determine successful, effective and rebound responses of hypothermia through autonomic suppression during A-C periods even though HR of both groups is displayed in C and D periods.