Aging and Time-Domain and Spectral Turbulence Parameters of Signal-Averaged Electrocardiograms

The effect of age on the time-domain and spectral parameters of the signal-averaged ECG (SAECG) is not known. In this study we recorded SAECGs from 138 normal healthy subjects (84 females and 54 males) aged between 7 and 102 years. The recordings were processed in time-domain and by spectral turbulence analysis. Spearman correlation coefficients were used to assess the relationship among each of the parameters obtained and age in the total population and in the subpopulations of males and females. Linear regression models of individual parameters with age were calculated and compared in males and females. Using conventional diagnostic criteria, 13 recordings (9.4%) were abnormal in time-domain analysis and 13 (9.4%) were abnormal in spectral turbulence analysis. However, no recording was abnormal in both types of analysis, and the incidence of abnormal SAECG was not age dependent. There were significant correlations between age and 4 of the 5 parameters used to evaluate spectral turbulence analysis (low segment correlation ratio [LSCR], intersegment correlation mean, intersegment standard deviation, and spectral entropy [SE]). However, there was no systematic significant correlation between age and time-domain parameters. The values of tQRS, mean peaks per slice, LSCR, and SE indices were significantly higher in males than in females, irrespective of age. The study concludes that with increasing age, there is a tendency for the parameters of spectral turbulence analysis to become abnormal, possibly reflecting an increase in conduction abnormalities with age.