Open AccessSignificance of the Accuracy of RR Interval Detection for the Analysis of New Dynamic Measures of Heart Rate Variability
Author(s) -
Tapanainen Jari M.,
Seppänen Tapio,
Laukkanen Raija,
Loimaala Antti,
Huikuri Heikki V.
Publication year - 1999
Publication title -
annals of noninvasive electrocardiology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.494
H-Index - 48
eISSN - 1542-474X
pISSN - 1082-720X
DOI - 10.1111/j.1542-474x.1999.tb00359.x
Subject(s) - medicine , rr interval , heart rate variability , cardiology , interval (graph theory) , heart rate , blood pressure , mathematics , combinatorics
Background: Interest in the analysis of RR interval variability by new dynamic methods is growing, because these methods give different, complementary information on heart rate behavior compared to the traditional measures of HRV. Several new methods have been recently introduced for studying complex heart rate dynamics, but the validity of these methods in analyzing RR interval behavior from Holter recordings is not well established. Methods: A real‐time microprocessor QRS detector system using a 1‐ms timing accuracy and a conventional 24‐hour ECG recording system with a sampling frequency of 128 Hz were compared in the measurement of new dynamic components of HRV in a sample of ten healthy subjects and ten patients with acute Ml. Results: Significant errors were observed in the analysis of approximate entropy (ApEn; mean difference 11 %) and in the short‐term scaling exponent value a 1 (mean difference 3%) in detrended fluctuation analysis (DFA). In the Poincaré analysis, notable differences were also observed in short‐term instantaneous RR‐interval variability (SD1; mean difference 5%). In spectral analysis, low‐to‐high frequency ratio was also influenced by the sampling frequency (mean difference 14%). The magnitude of all these differences grew proportionately when the overall heart HRV diminished. Conclusions: The data suggest that conventional Holter recordings should be used with caution in the analysis of new dynamic HRV parameters, especially approximate entropy and the ratio between low and high frequency spectral components, particularly in patient populations with low total HRV.