A new method to detect erratic sinus rhythm in RR-interval files generated from Holter recordings

Episodes of erratic sinus rhythm (ESR), assumed to be as a high-degree of non-respiratory sinus arrhythmia with normal p-waves, are seen on Holter recordings in the elderly and cardiac patients. ESR increases heart rate variability (HRV) and may confound its predictive value. We hypothesized that a triplet of normal-to-normal (NN) intervals represented by (N₁N₂N₃N₄) is ESR if N₁N₂/N₂N₃ and N₂N₃/N₃N₄ are both outliers in the distribution of NN-interval ratios. The time spent in ESR in each ten-minute segment was visually estimated using a heart-rate tachogram of NN intervals (HRTACH) and compared with the calculated values. A total of 858 ten-minute segments in six randomly selected 24-Hour recordings of older adults were examined. A qualitative test of the algorithm yielded a positive predictive value of 0.83 and a negative predictive value of 0.87 for 552 positive tests. Results suggest that the method can identify subjects with and without ESR in RR-interval files.