Is abnormal heart rate variability a specific feature of congestive heart failure?

Abstract

H rate variability (HRV), analyzed in the time and frequency domain, is reduced in patients with congestive heart failure (CHF).1–4 Fractal characteristics of heart rate (HR) behavior are also altered in patients with CHF,5,6 and traditional and fractal analysis methods of HRV provide prognostic information on patients with CHF.2,5,6 Despite these clinical observations, the pathophysiologic background for altered HRV is not well established. Two mechanisms have been proposed: (1) altered HRV in CHF results from abnormal afferent neural input from the failing heart to central cardiovascular autonomic regulatory systems,7,8 or (2) changes in neurohumoral activity caused by CHF result in altered efferent cardiac autonomic outflow.9 In this study, we tested these hypotheses by comparing traditional with nonlinear indexes of HRV in a variety of cases of appropriate sinus tachycardia and CHF to evaluate the specificity of the changes in HR dynamics among patients with CHF. • • • We compared conventional measures of HRV and fractal scaling properties in 3 ageand sex-matched patient groups: 20 patients with CHF, 20 patients with appropriate sinus tachycardia, and 20 healthy subjects. The CHF group consisted of patients who were hospitalized with worsening CHF with an ejection fraction ,35%. The appropriate sinus tachycardia group comprised patients with a wide variety of etiologies, including: sepsis (n 5 3), pericarditis (n 5 1), endocarditis (n 5 1), pneumonia (n 5 1), cellulitis (n 5 1), complicated pregnancy (n 5 3), hyperthyroidism (n 5 1), hepatic failure (n 5 1), psychotic crisis (n 5 1), anemia (n 5 1), uncontrolled diabetes (n 5 2), emphysema (n 5 1), meningitis (n 5 1), second-degree burns (n 5 1), and skull fracture (n 5 1). The control group comprised subjects with no history of cardiovascular disease and normal blood pressure, electrocardiography, echocardiography, and stress testing. Patients with CHF, appropriate sinus tachycardia, and healthy controls were matched with respect to age and gender. CHF and appropriate sinus tachycardia patients were also matched to mean HR. The study protocol was approved by the ethics committees of the participating institutes. All subjects were monitored with a 24-hour Holter. Before the RR interval tachograms were analyzed, the recordings were edited to eliminate premature ectopic beats and artifacts. Recordings with qualified beats for at least a 16-hour period and with .80% of qualified sinus beats were included in our analysis. Indexes of HR dynamics were obtained from custom-made analysis programs described in detail previously.10 Traditional measures of HRV were analyzed by the methods recommended by the Task Force of the European Society of Cardiology.11 The SD of all normalto-normal RR intervals and mean HR were computed as time domain measures of HRV. The low frequency, high frequency, and very low frequency were computed as frequency domain measures of HRV. For analysis of long-term scaling properties of the HR dynamics, the power-law relation of RR interval variability was calculated from the frequency range 10 to 10 Hz. Details of this method have been described previously.12 The detrended fluctuation analysis technique was used to quantify the short-term fractal scaling properties of RR interval time series. In this method, the root-mean-square fluctuation of integrated and detrended time series is measured at each observation window and plotted against the size of the observation window on a log-log scale. In the present study, the HR correlations were defined for short-term (,11 beats, a) RR interval data on the basis of the previous finding of a crossover point on the log-log plot. For a, mean values of the entire recording period were computed. Details of this method have been described previously.6,13 Results are expressed as mean 6 SD. Analysis of variance followed by Bonferroni’s post hoc multiple range tests were used to compare the differences between the groups. SPSS for Windows version 9.0 (SPSS Inc., Chicago, Illinois) was used in the analysis. A p value ,0.05 was considered statistically significant. Characteristics of the patients are listed in Table 1. There were no significant differences in sex and age between the 3 groups. Also, mean 24-hour HR did not differ among the appropriate sinus tachycardia and CHF patients. From the time and frequency domain measures of HRV, the SD of all normal-to-normal RR intervals and spectral measures were significantly lower among patients with appropriate sinus tachycardia than among patients with CHF. Fractal HR variability indexes did not differ between appropriate sinus tachycardia and CHF patients (Tables 2 and 3). Traditional HRV measures were significantly lower among appropriate sinus tachycardia and CHF patients tham among From the University of Miami School of Medicine, Miami, Florida; and University of Oulu, Oulu, Finland. Dr. Lopera’s address is: Division of Cardiology (D62), University of Miami School of Medicine. P.O Box 016960, Miami, Florida 33101. E-mail: Loperag@aol.com. Manuscript received September 22, 2000; revised manuscript received and accepted December 18, 2000.