Conventional QT Variability Measurement vs. Template Matching Techniques: Comparison of Performance Using Simulated and Real ECG

Natasha M Maurits,Mathias Baumert,Vito Starc,Alberto Porta

doi:10.1371/journal.pone.0041920

Natasha M Maurits, Mathias Baumert + Show 2 more

Open Access

https://doi.org/10.1371/journal.pone.0041920

Copy DOI

Abstract

Increased beat-to-beat variability in the QT interval (QTV) of ECG has been associated with increased risk for sudden cardiac death, but its measurement is technically challenging and currently not standardized. The aim of this study was to investigate the performance of commonly used beat-to-beat QT interval measurement algorithms. Three different methods (conventional, template stretching and template time shifting) were subjected to simulated data featuring typical ECG recording issues (broadband noise, baseline wander, amplitude modulation) and real short-term ECG of patients before and after infusion of sotalol, a QT interval prolonging drug. Among the three algorithms, the conventional algorithm was most susceptible to noise whereas the template time shifting algorithm showed superior overall performance on simulated and real ECG. None of the algorithms was able to detect increased beat-to-beat QT interval variability after sotalol infusion despite marked prolongation of the average QT interval. The QTV estimates of all three algorithms were inversely correlated with the amplitude of the T wave. In conclusion, template matching algorithms, in particular the time shifting algorithm, are recommended for beat-to-beat variability measurement of QT interval in body surface ECG. Recording noise, T wave amplitude and the beat-rejection strategy are important factors of QTV measurement and require further investigation.

Highlights

The QT interval of body surface ECG reflects ventricular depolarization and repolarization
Measuring beat-to-beat variability in the QT interval (QTV) has received increased attention over the last 15 years, since several clinical studies provided evidence regarding the predictive value of elevated QTV causing sudden cardiac death in a variety of cardiac conditions [3,4,5]
The presence of white Gaussian noise introduced a notable amount of artificial QT variability, ranging between 1 ms at an T wave acquisition range (TWAR) of 6.4% up to 9 ms at the lowest

Summary

Introduction

The QT interval of body surface ECG reflects ventricular depolarization and repolarization. Prolongation of the QT interval is a clinically accepted risk factor for malignant cardiac arrhythmia and used for guiding ICD implantation and drug development [1,2]. Measuring beat-to-beat variability in the QT interval (QTV) has received increased attention over the last 15 years, since several clinical studies provided evidence regarding the predictive value of elevated QTV causing sudden cardiac death in a variety of cardiac conditions [3,4,5]. Animal studies demonstrated increased QTV before the onset of drug-induced Torsades de Pointes (TdP) with a predictive value higher than that of standard QT interval assessment [6,7,8]. The mechanisms contributing to beat-to-beat QTV are incompletely understood, autonomous nervous system activity and repolarisation reserve have both been implicated [9,10,11,12], in addition to the well-known action potential duration adaptation to heart rate changes [13].

Objectives

Methods

Results

Conclusion