【摘要】 目的:基于自回归模型5min心率变异性频域分析的临床价值,检验糖尿病患者5min短时心率变异性自回归模型分析频域指标的变化。方法:实验对象为50岁以上糖尿病患者9例和正常人7例,糖尿病史>6年7例,糖尿病史<3年2例。被试者采用坐姿,静息状态,测试30min的ECG信号,采样率为1000Hz。对每个样本取第10～15分钟的时长5min的RR间期作为心率变异性信号,进行基于自回归模型的频域分析。依据功率谱估计计算16例受试者的总能量,低频段能量,高频段能量,低频段高频段能量比值等指标参数。结果:通过对5min心率变异性自回归模型计算结果得出,糖尿病史>6年患者总能量、低频段能量、高频段能量低于正常人,差异有显著性意义(t=2.4952,2.7027,1.7299,P<0.05)。糖尿病史<3年患者总能量、低频段能量、高频段能量与正常人差异不明显。两类糖尿病患者与正常人低频段高频段能量比值无明显差异。结论:5min短时心率变异性自回归模型分析能够有效反映糖尿病患者自主神经的病变,可能成为检测自主神经病变的方法。更多还原

【Abstract】 AIM: To test the clinical value of 5-minute heart rate variability (HRV) frequency analysis based on autoregressive (AR) model, and observe the diabetic 5-minute HRV frequency domain indices. METHODS: The experimental subjects were 9 diabetics and 7 normal subjects who were over 50 years old, including 7 severe diabetics whose history of disease exceeded 6 years and 2 slight diabetics whose history disease was less than 3 years. All the subjects were in seated resting states and their ECG signals were recorded for 30 minutes, with sample rate 1 000 Hz. Five-minute RR intervals between 10 and 15 minutes as HRV signals were analyzed in frequency domain based on AR model. Indices of frequency domain such as total power (TP), low frequency (LF) band power, high frequency (HF) band power and LF/HF power ratio, etc were calculated for each subject. RESULTS: Results of calculating 5-minute HRV AR model showed that there were significant differences in TP, LF band power and HF band power between severe diabetes and normal group (t =2.495 2, 2.702 7, 1.729 9, P < 0.05). The differences were not significant between slight diabetes and normal group. LF/HF power ratio was insignificant among severe diabetes, slight diabetes and normal group. CONCLUSION: Five-minute HRV AR model analysis is valid to reflect the autonomic neuropathy in the diabetic patients, and it seems to become the method of detecting autonomic neuropathy.更多还原