Heart rate variability in dilated cardiomyopathy

Abstract

Chronic heart failure is associated with excessive neurohormonal activation. Analysis of heart rate variability is considered a valid technique for assessment of the autonomic balance of the heart. Twenty symptomatic patients of dilated cardiomyopathy in NYHA class II-IV symptomatic status and as many normal controls were subjected to 24 hours Holter monitoring to assess the heart rate variability with both time domain and frequency domain analysis. Age of the patients ranged from 12 to 67 years (mean +/- SD 38.6 +/- 7 years), the male-female ratio was 4:1. The left ventricular ejection fraction of the patients was between 18-42 percent (mean +/- SD 30.2 +/- 9%) and all received diuretics, digoxin and angiotensin-converting enzyme inhibitors. Heart rate variability parameters measured included mean heart rate with standard deviation, hourly heart rate with SD and the mean of all normal RR intervals from the 24-hour recording. Time domain measures calculated were SD of all normal RR intervals, SD of 5 minute mean RR intervals and root mean square of difference of successive RR intervals. Using spectral plots, frequency domain subsets of low frequency and high frequency were analysed and expressed in normalised units. Total power was also measured. In the dilated cardiomyopathy patients, mean 24-hour heart rate in beats per minute was significantly higher in comparison to controls (82 +/- 13 vs 72 +/- 8; p < 0.001) whereas mean hourly heart rate with standard deviation (msec) was significantly lower (97 +/- 41 vs 232 +/- 25; p < 0.001), SD of all normal RR intervals (msec) was 85.5 +/- 26.3 vs 139.4 +/- 16.9 in controls (p < 0.001), SD of 5 minute mean RR intervals (msec) was also significantly less in patients in comparison to controls (75.8 +/- 39.6 vs 130.8 +/- 20.3; p < 0.001). However, although root mean square of difference of successive RR intervals (msec) was reduced in patients (30.1 +/- 9.3 vs 37.3 +/- 11.7; p < 0.05), the difference was non-significant. Low frequency power (0.05-0.15 Hz) (normalised units) was reduced in the dilated cardiomyopathy group (0.0721 +/- 0.003 vs 0.136 +/- 0.047 in the control group; p < 0.001). High frequency power (0.35-0.50 Hz) (normalised units) (0.08 +/- 0.05 in patients vs 0.09 +/- 0.02 in controls; p > 0.1) and total power frequency (0.02-0.50 Hz) (normalised units) (0.34 +/- 0.05 in patients vs 0.35 +/- 0.12 in controls; p > 0.1) was non-significantly different in the two groups. Regression analysis showed a significant decrease in SD of all normal RR intervals, SD of 5 minute mean RR intervals, low frequency, high frequency, total power and a non-significant decrease in root mean square of difference of successive RR intervals with a decrease in ejection fraction percent whereas there was a significant decrease in SD of all normal RR intervals, SD of 5 minute mean RR intervals, low frequency and total power and a less significant decrease in root mean square of difference of successive RR intervals and high frequency power with an increase in NYHA class. At 6 months duration, 6 patients were lost to follow-up, 3 patients were readmitted (2 for congestive cardiac failure, one of paroxysmal supraventricular tachycardia). One patient who was NYHA class IV at baseline was readmitted for congestive cardiac failure and showed much lower heart rate variability parameters compared to the average of the patients. We conclude that in symptomatic dilated cardiomyopathy patients, heart rate variability parameters are significantly reduced in comparison to control subjects.