冠心病与心力衰竭患者短期QT间期变异性研究

135 浏览量更新于2024-08-28 收藏 1MB PDF 举报

"这篇研究论文探讨了短期QT间隔变异性在冠状动脉疾病(CAD)和充血性心力衰竭(CHF)患者与健康对照组之间的差异。研究涉及29名健康志愿者、29名年龄匹配的CAD患者和20名年龄匹配的CHF患者。通过5分钟的静息II导联心电图(ECG)获取QT时间序列，然后计算时域指数（如平均值、标准差和QT变异性指数(QTVI)）、频域指数（LF和HF）以及非线性指数（如样本熵(SampEn)、排列熵(PE)和动力学模式）。为了排除心率(HR)对QT变异性的影响，除了QTVI之外的所有计算都在使用三种不同的校正方法得到的心率校正QT时间序列(QTc)上进行重复。" 本文的重点在于评估QT间隔变异性作为心血管疾病指标的潜力。QT间隔是心电图上代表心脏复极化过程的时间，其变异性可以反映心脏电活动的不稳定性，可能与心脏病风险相关。在CAD和CHF患者中，QT间隔的异常变异性可能预示着心律失常的风险增加或心脏功能的恶化。首先，研究者比较了健康个体与CAD及CHF患者的QT间隔变异性。他们发现，与健康对照组相比，CAD和CHF患者的QT间隔变异性可能表现出显著的不同，这可能反映了这些疾病对心脏电信号稳定性的负面影响。时域分析中的平均值和标准差可以帮助识别QT间隔的一致性和变化范围，而QTVI则可以量化这种变异性。其次，频域分析（LF和HF）通常用于评估自主神经系统对心脏活动的影响。LF通常与交感神经活性有关，HF则与副交感神经活性有关。在CAD和CHF患者中，这些指数的变化可能揭示了疾病状态下的神经平衡失调。非线性分析如样本熵和排列熵是评估复杂系统（如心脏）动态行为的有效工具。在本研究中，这些非线性指标可能揭示了CAD和CHF患者心脏动态模式的异常，表明疾病导致的心脏活动规律性的丧失。最后，通过心率校正QT间隔，研究者试图消除心率对QT间隔的影响，以更准确地评估QT变异性。心率校正的方法包括Bazett、Fridericia和Janse-Mittag等，每种方法都有其理论依据和适用情况。这项研究提供了一个全面的分析框架，用以对比CAD和CHF患者与健康人的QT间隔变异性，这有助于理解这些疾病对心脏电生理特性的影响，并可能为临床诊断和治疗决策提供有价值的参考。未来的研究可能进一步探讨这些发现如何应用于个体化治疗和风险预测。

In all three equations, QT interval is in milliseconds and the

RR interval is in seconds.

2.5 QTV indices

2.5.1 Time domain indices

Time domain indices include mean and standard deviation

(SD)ofQT,QTc,aswellasQTatimeseries,meanofT

amplitudes and QTVI, a metric that incorporates HRV in

QTV using the following formula [10]:

QTVI ¼ log

=QT

=HR

: ð4Þ

Here, QT

and QT

are the variance and mean of QT inter-

vals, respectively. HR

and HR

are the variance and mean of

heart rate series, respectively. QTVI was calculated only for

the raw QT time series.

2.5.2 Frequency-domain indices

Before the power spectral analysis, QT and QTa time series

were evenly resampled to 4 Hz by spline interpolation.

Power spectral density (PSD) was estimated by

autoregressive (AR) model based parametric method. In

this study, the Burg’s method was used to estimate the

model coefficients with an order of 16 [17].Basedonthe

frequency-domain characteristics of HRV, two frequency

bands were defined: (1) a low-frequency band from 0.04

to 0.15 Hz and (2) a high-frequency band from 0.15 to

0.40 Hz. The frequency-domain i ndices for QT and QTa

time series include the power of low frequency bands (LF)

and high frequency bands (HF).

2.5.3 Sample entropy

For the time series {u

= u(i), 1 ≤ i ≤ N}, form N − m + 1 vectors

¼ uiðÞ; uiþ 1ðÞ; …; uiþ m−1ðÞ; 1≤i≤N−m



.Define

the distance between all possible pair of vectors X

and X

by d

i, jm

=max{|u(i + k) − u(j + k)|, 0 ≤ k ≤ m − 1}. Denote the

average number of j that meets d

i, jm

≤ r for all 1 ≤ j ≤ N − m,

j ≠ i by B

rðÞ. Then increase the dimension m to m +1 and

repeat the above-described procedures to calculate B

mþ1

rðÞ.

The SampEn for u

can be defined by [16]:

SampEn m; r; NðÞ¼−ln ∑

N−m

t¼1

mþ1

rðÞ= ∑

N−m

t¼1

rðÞ



: ð5Þ

The threshold r value has been recommended to choose

between 0.1 × SD and 0.25 × SD [27]. In t his study, the

threshold was set at r =0.2×SDin order to minimize the pos-

sibility of getting invalid SampEn. The dimension was set at

m = 2 in accordance with the previous study [40].

2.5.4 Permutation entropy and dynamical patterns

For a time series {u

= u(i), 1 ≤ i ≤ N}, form N − m + 1 vectors

¼ uiðÞ; uiþ 1ðÞ; …; uiþ m−1ðÞ; 1≤i≤N−m þ 1



symbolic counterpart is then obtained by coarse-graining the

time series using a given resolution Δ,i.e.,:

φ iðÞ¼floor

uiðÞ−min X





: ð6Þ

The N − m + 1 vectors can thus be represented by

¼ φ iðÞ; φ i þ 1ðÞ; …; φ i þ m−1ðÞðÞ. The ordinal pattern

corresponding to φ

can be defined by the index vector when

rearranging φ

in ascending order. For instance, if

¼ 10; 30; 20½, the index vector when rearranging it in

ascending order should be (132) which is defined as the ordi-

nal pattern for φ

Five dynamical patterns were defined by categorizing the

ordinal patterns into five different groups. The five dynamical

patterns for m = 3 were illustrated in Table 2.

The probability (frequency) of each dynamical pattern was

counted and denoted by {p(D

), 1 ≤ i ≤ 5}. PE is then calculat-

ed by:

PE m; ΔðÞ¼−∑



lnpD



: ð7Þ

Table 1 Demographical and

clinical characteristics of all

subjects

Variables Healthy group CAD group CHF group

N 29 (16/13) 29 (22/7)** 20 (9/11)

Age (year) 56 ± 8 58 ± 8 59 ± 9

Body mass index (BMI) (kg/m

)24±826±325±4

Systolic blood pressure (SBP) (mmHg) 114 ± 13 117 ± 14 118 ± 14

Diastolic blood pressure (DBP) (mmHg) 72 ± 9 74 ± 9 75 ± 9

Left ventricular ejection fraction (LVEF) (%) 65 ± 4 60 ± 6 39 ± 7*

Data are expressed as number (male/female) or mean ± SD.

*p < 0.05 vs. healthy group as revealed by t test

**p < 0.01 vs. healthy group as revealed by chi-squared test

Med Biol Eng Comput (2019) 57:389–400 391

剩余11页未读，继续阅读

weixin_38664469

粉丝: 5
资源: 896

冠心病与心力衰竭患者短期QT间期变异性研究

Distribution entropy for short-term QT interval variability analysis: a comparison between the heart failure and normal control groups

Measuring Electromechanical Coupling in Patients with Coronary Artery Disease and Healthy Subjects

js-leetcode题解之57-insert-interval.js

preserved.heart.beat.interval: 1000

最新资源