癫痫患者心率变异性复杂度的多尺度熵分析

目的:研究癫痫疾病对心率变异性复杂度的影响,探索癫痫疾病的病理生理机制。方法:基于多尺度熵方法回顾性分析212例癫痫患者和212例健康对照清醒状态下相同时间段内4 h的RR间期序列,并量化SampEn(样本熵)、Slope_1-5(1~5连续5个尺度下SampEn值线性拟合的斜率)、Area_1-5(5个SampEn值的总和)、Area_6-15(6~15连续10个尺度下SampEn值的总和)、Area_6-20(6~20连续15个尺度下SampEn值的总和)以及SampEn₅(尺度5下的熵值)等特征参数以表征心率变异性复杂度。基于ROC曲线及其AUC值评估心率变异性复杂度指标区分癫痫患者和健康对照的能力。结果:癫痫患者的心率变异性复杂度指标Slope_1-5、SampEn₅、Area_1-5、Area_6-15、Area_6-20均显著低于健康对照(P<0.001),且与癫痫病程及发作频率无显著相关性。Slope_1-5区分癫痫患者和健康对照的性能最佳,AUC值为0.764,灵敏度和特异度分别为75.0%和68.9%。结论:基于多尺度熵方法量化的心率变异性复杂度指标为癫痫病理生理机制的探索提供了新的视角,在癫痫疾病的辅助诊断、治疗、预后及风险预测领域具有潜在的临床应用价值。

Multiscale entropy analysis of heart rate variability complexity in patients with epilepsy

Objective To investigate the effects of epilepsy on heart rate variability(HRV)complexity and to explore the pathophysiological mechanism of epilepsy.Methods 4-hour heartbeat interval time series from 212 epileptic patients and 212 healthy control subjects under awake state were retrospectively analyzed using multiscale entropy methods.The HRV complexity indices including SampEn(sample entropy),Slope_1-5(slope of the linear fit of SampEn values at 5 consecutive scales from 1 to 5),Area_1-5(sum of 5 SampEn values),Area_6-15(sum of SampEn values at 10 consecutive scales from 6 to 15),Area_6-20(sum of SampEn values at 15 consecutive scales from 6 to 20)and SampEn₅(entropy value at scale 5),were to characterize the complexity of heart rate variability.The ability of the heart rate variability complexity to differentiate between epileptic patients and healthy controls was assessed based on ROC curves and their AUC values.Results The heart rate variability complexity indexes Slope_1-5,SampEn₅,Area_1-5,Area_6-15and Area_6-20were significantly lower in epileptic patients than in healthy controls(P<0.001)and did not correlate significantly with the duration of epilepsy or seizure frequency.Slope_1-5had the best performances in differentiating epileptic patients from healthy controls,with an AUC value of 0.764 and the sensitivity and specificity of 75.0%and 68.9%,respectively.Conclusion The heart rate variability complexity indies quantified based on multiscale entropy method provides a new perspective for the exploration of the pathophysiological mechanism of epilepsy and has potential clinical application in the field of adjunctive diagnosis,treatment,prognosis and risk prediction of epilepsy.