Chaos analysis of blood pressure and heart rate variability in hypertension

为了研究高血压患者血压和心率变异性(HRV)的非线性和线性成分，我们构建了时间序列数据的混沌和谱分析系统，并将其应用于临床。将无创连续血压监护仪或动态心电图的R-R间期数据输入计算机后，进行混沌分析和频谱分析。(1)对高血压病患者(EHT)和血压正常者(NT)的心率变异性(HRV)进行混沌和频谱分析，研究直立倾斜时交感神经活动的动态变化。EHT组静息时%HF降低，头高位倾斜使两组的LF/HF升高，%HF降低。另一方面，作为复杂性指标的关联维，EHT组在静息状态下低于NT组，且EHT组对仰卧位的反应没有改变，而NT组则降低。两组患者的Lyapunov指数，即对初始条件敏感依赖的指标，在仰卧位倾斜过程中均未发生改变。(2)研究了门诊HT和NT患者的混沌和频谱成分的昼夜变化。HT组的%HF无昼夜节律性变化，而NT组白天%HF呈下降趋势。低频/高频的变化与%HF的变化相反。另一方面，NT和成人HT的关联维度在白天或晚上都有所下降，但老年HT的关联维度没有变化。老年高血压患者夜间的Lyapunov指数明显小于老年NT患者。综上所述，HRV中包含的混沌成分似乎反映了心率变异与频谱成分不同的方面。

To investigate nonlinear and linear components of arterial pressure and heart rate variability (HRV) in essential hypertension, we constructed the chaos and spectral analysis system of the time series data, and applied it on clinical settings. After taking the R-R interval data of non-invasive continuous blood pressure monitor or Holter ECG in the computer, both the chaos and spectral analyses were performed. The following results were obtained by using this system.(1) Chaos and spectral analyses of HRV were performed in essential hypertensive subjects (EHT) and normotensive controls (NT) to investigate the serial changes of sympathetic nerve activity during head-up tilting. The %HF was decreased in EHT at rest, and head-up tilting increased the LF/HF and reduced the %HF in both groups. On the other hand, the correlation dimension, an index of complexity, was lower at rest in EHT than in NT, and did not change in EHT in response to head-up tilting, but decreased in NT. Lyapunov exponents, an index of sensitive dependence on initial condition, did not change during head-up tilting in both groups.(2) Circadian changes of chaos and spectral components were investigated in outpatient HT and NT. The %HF did not show circadian changes in HT, but it decreased during the daytime in NT. The LF/HF showed an opposite change to the %HF. On the other hand, the correlation dimension decreased in the daytime or in the evening in both NT and adult HT, but did not change in elderly HT. The Lyapunov exponents were significantly smaller in the elderly HT than in elderly NT during nighttime.In conclusion, the chaos components included in the HRV seems to reflect the different aspect of HRV from spectral components.