Comparison of four methods of assessing heart rate variability in patients with heart failure

BACKGROUND: Heart rate variability reflects autonomic tone and is used to assess progression and prognosis in a variety of illnesses. However, multiple heart rate variability methods exist and are not necessarily equivalent. OBJECTIVES: To compare four methods of heart rate variability in heart failure patients and healthy subjects. METHODS: Twenty-four-hour Holter recordings were obtained in 50 heart failure patients and 50 age- and gender-matched control patients. From these recordings, heart rate variability was assessed by histograms, standard deviation, Poincare plots, and spectral analysis. RESULTS: For R-R interval histograms, standard deviation, and Poincare plots, diminished heart rate variability was identified in 65% to 100% of heart failure patients versus 0% to 8% of controls. Agreement among these tests ranged from 69% to 96%. Spectral values varied greatly over the recording period, even in the same subject, possibly because of variations in activity. Only 16% of heart failure patients had spectral values that were identified as abnormal. Agreement between spectral analysis and the other methods ranged between 58% and 67%. CONCLUSIONS: Heart rate variability assessed over a 24-hour period with different techniques yields similar but not identical results. Heart rate variability assessed from spectral analysis of short periods of data varied markedly in a 24-hour period and should not be compared with measures obtained from 24-hour methods. Standardization of subject activity and recording time is necessary for comparison of spectral analysis of brief periods. Further research is required to determine if differences among methods assessing 24-hour heart rate variability yield complementary information.