Correlation between time-domain measures of heart rate variability and scatterplots in postinfarction patients.

Heart rate variability (HRV) is usually measured in time or frequency domains. Beat-to-beat variability, which cannot be assessed by frequency-domain analysis, and can only be assessed globally by time-domain analysis, provides information concerning the nonlinear behavior of heart rate. This beat-to-beat variability can be displayed on scatterplots, where each RR interval is plotted against the preceding RR interval. However, the relationship between scatterplots and other measures of HRV is unknown. We studied the correlations between time-domain measures and scatterplot length, width, and area in 50 postinfarction patients. Scatterplot length and width were measured after printing. Scatterplot area was calculated from length and width, assimilating the plot to an ellipse. Long-term variability indexes (SDNN and SDANN) were strongly correlated with scatterplot length (r > 0.9, P < 0.0001), and short-term variability parameters (pNN50 and variability index) with scatterplot width (r > 0.9; P < 0.0001). Scatterplots are, therefore, a simple way of providing information concerning long- and short-term HRV. Furthermore, measurement of scatterplot width at different given RR intervals could be an approach to the evaluation of short-term HRV for different heart rates. This could provide a simple way of assessing cardiac parasympathetic modulation at different heart rates.