2000 Volume 45 Issue 4 Pages 584-595
Objective : To evaluate the reliability of spectral turbulence (ST) analysis of signalaveraged electrocardiograms (SAE), the specificity and the reproducibility of the diagnoses were studied and compared to time domain (TD) analysis. Materials : Two hundred healthy subjects consisting of 100 males and 100 females, with an average age of 29 years, were involved in this study. Methods : We recorded SAE in all subjects. In 65 of 200 subjects, two recordings were performed within 15 minutes, and in 24 of the 65 subjects, SAE was also recorded 1 week later in order to evaluate short-term and long-term reproducibility. The obtained signals were analyzed by ST and TD methods. The normal values for the parameters, the gender differences, and the correlation with physical characteristics were evaluated. The reproducibility of measurement values in each parameter and the diagnostic concordance were compared by applying previously proposed criteria and the criteria set by 95% confidence interval of measurements. Results : The measurement values of TD analysis were influenced by gender and physical characteristics, whereas those of ST analysis were not. Applying the previously proposed criteria, diagnostic specificity was 93% in TD analysis and it was 99% using the new criteria. The values were 98% and 99.5% in ST analysis, respectively. A significant correlation of the measurement value in each parameter was observed in the two recordings separated by a 15 minute interval. The diagnostic concordance with the short-term or long-term intervals was 89-100% in TD analysis and 96-100% in ST analysis by applying each criterion. Conclusions : ST analysis of SAE is not influenced by gender or physical characteristics and it is highly specific and reproducible. Therefore, the clinical usefulness including a screening test is suggested.
