Mean heart rate and short-term heart rate variability in early period of myocardial infarction: Value for mortality prediction

For evaluation of prognostic power of mean heart rate and shout-term restin g heart rate variability (HRV) 134 patients (79 men) with myocardial infarc tion (MI) aged 35-75 years (mean 59.8+/-9.3) were enrolled to a follow up s tudy. MI was Q-wave in 75.4%, anterior in 50,7% and recurrent in 21.6% of p atients. Thrombolytic therapy was performed in 32.8% of cases. Mean of norm al RR intervals (RRNN), standard deviation of normal RR intervals (SDNN), c oefficient of variance, total power (TP, 0.003-0.40 Hz), very low (VLF; 0.0 03-0.04 Hz), low (LF; 0.04-0.15 Hz) and high (HF; 0.15-0.40 Hz) frequency p owers together with LF/HF ratio were evaluated on short ECC strips (1024 ca rdiac cycles) at supine rest from 9:00 to 13:00 on days 2-4 of MI. 88.1% of patients received b-blockers. During follow up from 1 to 2 years (median 1 .8 years) 18 deaths occurred tall cardiovascular, 11 sudden). Cox proportio nal hazard regression was used for mortality modeling with 38 clinical, ech ocardiographic and HRV variables as potential predictors. Results. In age a nd sex adjusted analysis history of MI, MI localization, presence of angina , clinical signs of left ventricular failure, left ventricular volumes and election fraction as well as RRNN, SDNN, TP, VLF, LF and HF powers were sig nificantly associated with mortality. In multivariate analysis using step-u p approach clinical signs of left ventricular failure on the day of HRV ass essment was selected first i.e. was the variable with strongest independent association with mortality (relative risk 5.7; 95% CI 1.7-18.7; p=0.004). Addition of left ventricular end-systolic volume (relative risk 3.6; 95% CI 1.1-11.9; p=0.04), RRNN (relative risk 0,005; 95% CI 0-0,3; p=0,01) and VL F power (relative risk 1,7; 95% CI 0.9-3.0; p=0.08) improved the outcome pr ediction. Those 4 variables were used for construction of a prognostic inde x. Patients were divided into quartiles of this index. Mortality was 1.5% i n combined third and forth quartiles (low risk group), 8.8% in the second q uartile (medium risk group) and 42.4% in the first quartile thigh risk grou p). Relative risks of death for medium- and high risk groups compared with low risk one were 14.9 (95% CI 1.7-128,3; p=0.01) and 43.7 (95% CI 5,6-343. 1; p=0,0003) respectively. Conclusion: RRNN (reciprocal of mean heart rate) and only one of HRV characteristics (VLF power) obtained under standardize d conditions early in the course of the disease were the variables which in combination with some more conventional predictors improved assessment of medium term prognosis in patients with acute MI.