Symbolic analysis of heart rate variability and its interaction with p-value of normality tests applied on RR intervals

Abstract

OBJECTIVE: A previous study suggested that the p-value of normality test applied to RR intervals is an index able to quantify Heart Rate Variability (HRV) through correlation of traditional time and frequency domain indices. We investigate the association between the p-value of normality test applied in RR intervals and symbolic analysis of HRV. METHOD: We evaluated 32 healthy women between 18 and 30 years old. RR intervals were used for HRV analysis and we performed symbolic analysis, where RR intervals are joined by symbols. Sets of three consecutive symbols (RR intervals) were grouped into four types of clusters, as follows: 1. Three equal symbols (zero variation); 2: two equal, one divergent symbol (one variation); 3. Three different symbols monotonically ascending or descending (two like variations); 4. Three different symbols, forming a peak or a trough (two unlike variations). Frequency of occurrence of each type of cluster was calculated. Normality tests were applied to all RR intervals and the p-value was calculated. We computed the correlations between the p-value of normality test and symbolic analysis of HRV. RESULTS: Correlation coefficients between the p-value of normality test from Kolmorogov-Smirnov test and the four types of clusters showed no correlation for any of them. Likewise, the correlation coefficient index between the p-value of normality test calculated from the Shapiro-Wilk test and symbolic analysis produced no significant results for any of the four types of clusters. CONCLUSION: There was no significant correlation between the p-value of normality test and HRV symbolic analysis. The physiological significance of this result is that the PNT is not related to chaotic behavior of HRV.