State-of-the-art analysis

Science Behind HeartTrends' unique HRV Analysis

 

Heart rate variability (HRV) is a well-established marker of mortality and sudden death shown to be attenuated in patients with coronary artery disease (CAD) even at rest. Based on this clinical evidence, HeartTrends was developed to provide an innovative modality with a high sensitivity for detection of myocardial ischemia at rest.

 

  • Deterministic diagnosis of ischemia

  • Heart rate attenuation associated with CAD likely for early detection of myocardial ischemia

  • Measures Autonomic Nervous System functionality

  • Uses non-linear physics’ Chaos Theory multipole frequency domain parametric analysis

Why is HeartTrends more accurate than other heart rate variability (HRV) analyses?

HeartTrends is unique in that it has been clinically tested and peer-reviewed showing earlier prognostic therapeutic benefit than previous HRV analyses. HeartTrends uses the Non-linear Multipole Analysis method for deriving information from three domains: time, frequency, and RR randomness as opposed to the traditional one-dimensional HRV standard deviation. Notably, prior HRV algorithms were used mostly for risk stratification, while HeartTrends is the first to show that heart rate variability can be used to detect the presence of significant myocardial ischemia associated with significant coronary artery disease in individuals without known CAD.

 

WARNING: Highly technical in-depth technical explanation:

In clinical medicine, the dynamics of the beat-to-beat (RR) time series is commonly represented by a phase-space (or Poincaré) plot, where each RR interval is plotted against the previous one. The classification of the phase-space plot is traditionally performed by visual inspection and semi-quantitative analysis describing the features of the plot, as length or width, but that approach ignores the varying density of points leading to similar plots due to hearts with very different dynamics.

 

The Multipole HRV analysis is a relatively new way of investigating the Poincaré plot from complex time series. We interpret the Poincaré plot as a two-dimensional body, where each data point in the plot is assigned a unit mass, in order to describe the total mass distribution within the plot. The measures obtained from this kind of analysis bear intrinsic time dependence due to the very construction of the plot. As a result the Multipole method derives information from both the time and the frequency domains as well as reflecting increased randomness in the RR interval time series. The traditional HRV-measures derive only information from one of these domains, which seems to be the reason that The Multipole Method have shown more prognostic power than previous suggested risk markers.

 

From the time series one may calculate the leading multipoles: the quadrupoles, octoupoles, and the hexadecapoles, and from which the new HRV parameter Dyx is derived. The Quadrupole (Qyy), for example, describes the overall distribution of data points in the Poincaré Plot (i.e., the shape of the plot). It was found to be a strong predictor of mortality in a population of post-Myocardial Infarction (MI) patients with both a depressed and preserved Left Ventricular Ejection Fraction (LVEF) [1]. When used in combination with Dyx as a weighted multipole parameter, it has been shown to be a stronger predictor of mortality after MI than SDNN and the short-term scaling exponent Alpha-1 [2].

 

The varying density of data points implies that some other measures based on analysis of the plot incorrectly add the same significance to low populated areas of the plot as to higher populated areas. This is, for example, the case for SD12 which is the ratio between the length (SD2) and the width (SD1) of an imaginary ellipse fitted to the Poincaré Plot with the center in the average RR interval. In contrast to SD12, Dyx is a relative density measure obtained from the plot with prevalence of the densest populated area.

 

Joergensen et al. [1] compared the Multipole method with the traditional HRV measures in the Nordic ICD study. Patients with AMI were screened with 2D Echocardiography and 24h Holter-recordings 2-14 days post-MI. Reduced Dyx predicted both all-cause, cardiovascular mortality and sudden cardiovascular death in univariate Cox proportional hazard analysis.

 

In multivariate analysis with correction for known risk factors, Dyx continued to show independent predictive value with a hazard value of 2.1 (C.L. 1.1-4.2), whereas none of the traditional HRV measures reached statistical significance. The assumption is that, due to Dyx obtaining information from time as well as frequency intervals combined with focusing on only dense populated areas of the recurrence plot, prognostic power for VT/VF arrhythmias is enhanced relative to traditional HRV measures which do not receive information from the time domain.

 

[1] Joergensen RM, et al. Prediction of ventricular tachycardia in ICD patients based on the multipole method. European Heart Journal 2007, 28:414 (Abstract Supplement).

[2] Olesen RM et al. Statistical Analysis of Diamond MI Study by the Multipole Method. Physiol. Meas. 2005, 26:591-598.

Product of Lev-El Diagnostics Ltd. ©2020