Methodological aspects of the use of heart rate stratified RR interval histograms in the analysis of atrioventricular conduction during atrial fibrillation

Publisher Website
Google Scholar
Abstract
A new computer assisted method for analysing RR interval histograms was devised in 32 patients with sustained atrial fibrillation. Twenty four hour ECG recordings were divided consecutively into periods with 64 heart beats and mean heart rates calculated for each period. Using increments of 10 beats·min−1 serial heart rate stratified histograms with 5120 RR intervals were constructed according to the calculated heart rate. In all histograms the RR intervals were sorted into 20 ms wide subgroups between 180 and 2100 ms. The analysis showed a bimodal RR distribution in 26 patients. Several indices were chosen to characterise a given RR interval peak in the histogram analysis. The rate dependence of peak was calculated for each individual peak. In the presence of two or more peaks a peak dominance change value was defined. Furthermore, differences and ratios between peak values at peak dominance change value (peak gap and peak value ratio) were calculated. A contour index was used to describe the smoothness of the appearance of the histogram. Comparisons were made between the histograms with periods composed of different numbers of heart beats (8, 16, 32, 64) for the mean heart rate calculation and between the histograms constructed with different numbers of RR intervals (512, 1024, 2560, 5120, 10 240, pooled data) in six patients. A significant finding was that >5000 RR intervals should be used for the construction of a heart rate stratified histogram. Comparison was also made between the repeated histograms with a mean time interval of 137 days (7-413 days) in eight patients. The findings showed a high reproducibility of the heart rate stratified histogram.

This publication has 0 references indexed in Scilit: