Postural Response of Low-Frequency Component of Heart Rate Variability: Survival Analysis

Postural Response of Low-Frequency Component of Heart Rate Variability: Survival AnalysisUnivariate Cox regression analysis of baseline clinical features revealed that age, previous myocardial infarction, reduced ejection fraction, and the number of diseased coronary arteries were associated with increased risks for cardiac death and that age was associated with noncardiac death (Table 3). Although no significant association with a risk for death was detected for heart rate, BP, or their postural change during the HUTT, several measures of HRV showed univariate association with increased risks for cardiac and noncardiac death (Table 4). After adjustment for other cardiovascular risks, including age, sex, hypertension, smoking, cholesterol levels, diabetes, body mass index, previous infarction, ejection fraction, and the number of diseased coronary arteries, only an increase in the postural LF response was associated with an increased risk for cardiac death. No significant association with a risk for noncardiac death was detected in any HRV measure after the adjustment.
Postural LF Response and Mortality
Interestingly, the postural LF response distributed normally around a slightly negative mean value, and the distribution was similar between the patients with CAD and the control subjects (Fig 2). The cardiac death events occurred more frequently on the left side of the distribution that reflected patients who showed a postural LF rise.

The prognostic association of postural LF response was further analyzed by stratifying the patients by the response. The patients were divided into the following three groups with almost equivalent sizes at values of —0.6 and 0.0 ln(ms2): large LF drop (group D2; n = 82); small LF drop (group D1; n = 83); and LF rise (group R; n = 85). The three groups did not differ in any baseline clinical feature or in the therapies performed during the follow-up except for receiving regular p-blocker medication, which was more frequent in group R (Table 5). Kaplan-Meier curves for the three groups showed significant group differences in cardiac mortality, but not in noncardiac mortality (Fig 3). The 8-year cardiac mortality rates in groups D2, D1, and R were 0%, 6%, and 12%, respectively (p = 0.008 [log rank test]). Furthermore, when the survival analysis was performed after excluding 64 patients who were regularly taking p-blockers during the follow-up, the group difference in the cardiac death rate was enhanced. The 8-year cardiac mortality rates in groups D2, D1, and R were 0%, 7%, and 15%, respectively (p = 0.006 [log-rank test]). Among the 64 patients receiving p-blockers, cardiac death occurred in two patients in group R.
Fig2
Figure 2. Bar graphs showing the relative frequency of patients with CAD (top) and age-matched and gender-matched healthy control subjects (bottom) classified by the postural LF response. Vertical dashed lines in the top panel indicate the cutoff points by which the LF responses were classified into large drop (D2), small drop (D1), and rise (R) among the patients with CAD. The letters at the top of the bars indicate the causes of death that were observed in each group during the follow-up period. A = acute myocardial infarction; F = fatal stroke; N = noncardiac causes; and S = sudden cardiac death.
Fig3
Figure 3. Kaplan-Meier curves for cardiac and noncardiac death in patients stratified by the postural LF response. The 8-year cardiac mortality rates for the patients who showed LF responses of large drop (D2), small drop (D1), and rise (R) were 0%, 6%, and 12%, respectively.

Table 3—RR of Clinical Variables for Different Forms of Death in 250 Patients Undergoing Coronary Angiography

Cardiac Death (n = 13) Noncardiac Death (n = 12)
Variables RR (95% CI) RR (95% CI)
Age 1.15 (1.07-1.23)T 1.11 (1.03-1.19)|
Male sex 1.73 (0.38-7.83) 1.13(0.23-5.42)
Hypertension 1.53 (0.42-5.58) 0.63 (0.08-5.07)
Smoking 0.51 (0.17-1.56) 0.68 (0.18-2.53)
LDL cholesterol 1.00(0.99-1.01) 1.01 (1.00-1.03)
HDL cholesterol 0.99(0.95-1.04) 0.94 (0.87-1.02)
Diabetes mellitus 1.19 (0.33-4.34) 0.37 (0.05-2.88)
Body mass index 0.75 (0.59-1.00) 1.10(0.87-1.39)
Previous myocardial 3.55 (1.09-11.5)| 1.31 (0.35-4.88)
infarction
Ejection fraction 0.94 (0.90-0.98)| 0.99 (0.93-1.05)
LVEDP 1.03(0.92-1.14) 0.91 (0.78-1.06)
Diseased coronary arteries 2.84 (1.41-5.70)t 1.54 (0.81-2.91)

Table 4—Unadjusted and Adjusted RR of HRV Measures and Their Postural Responses for Different Forms of Death

Variables Cardiac De1Univariate RR (95% CI) ath (n = 13)Adjusted RR(95% CI) Noncardiac D1Univariate RR(95% CI) eath (n = 12)Adjusted RR(95% CI)
Supine
Total power 0.46 (0.25-0.85)* 0.67 (0.33-1.39) 1.05 (0.49-2.24) 1.57 (0.67-3.69)
HF 0.74 (0.49-1.11) 1.01 (0.62-1.64) 0.52 (0.33-0.83)* 0.59 (0.33-1.03)
LF 0.67 (0.42-1.07) 0.79 (0.43-1.43) 0.74 (0.42-1.30) 0.86 (0.44-1.68)
LFnu 1.04 (0.81-1.34) 0.91 (0.67-1.23) 1.45 (1.01-2.08) 1.43 (0.97-2.11)
Tilt
Total power 0.82 (0.44-1.55) 1.20(0.66-2.16) 0.66 (0.30-1.42) 0.71 (0.31-1.64)
HF 0.97 (0.63-1.47) 1.29(0.78-2.12) 0.55 (0.36-0.86)* 0.60 (0.36-1.01)
LF 1.04 (0.62-1.74) 1.76(0.99-3.14) 0.62 (0.36-1.98) 0.72 (0.37-1.42)
LFnu 1.06 (0.82-1.36) 1.12 (0.83-1.51) 1.26 (0.89-1.78) 1.27 (0.89-1.79)
Change (tilt — supine)
Total power 2.22 (1.10-4.49)* 1.89 (0.92-3.87) 0.54 (0.22-1.32) 0.40 (0.15-1.09)
HF 1.47 (0.87-2.50) 1.41 (0.77-2.59) 1.01 (0.56-1.82) 0.92 (0.50-1.72)
LF 2.05 (1.11-3.79)* 4.36 (1.64-11.6)т 0.65 (0.40-1.05) 0.91 (0.53-1.58)
LFnu 1.02 (0.81-1.28) 1.18(0.91-1.52) 0.91 (0.69-1.21) 0.97 (0.72-1.29)

Table 5—Clinical Characteristics of Patients Grouped by Postural LF Response

Characteristics Group D2(n = 82) Group D1(n = 83) Group R(n = 85) p Value
Baseline
Age, yr 56 ± 10 58 ± 9 57 ± 9 NS
Male 62 (76) 60 (72) 59 (69) NS
Hypertension 14(17) 17 (20) 13(15) NS
Smoking 42 (51) 49 (59) 45 (53) NS
LDL cholesterol, mg/dL 119 ± 40 121 ± 43 120 ± 43 NS
HDL cholesterol, mg/dL 42 ± 13 42 ± 13 42 ± 13 NS
Diabetes mellitus 19 (23) 20 (24) 13(15) NS
Body mass index 23 ± 3 23 ± 3 23 ± 3 NS
Previous myocardial 30 (37) 37 (45) 31 (37) NS
infarction
Ejection fraction, % 64 ± 10 63 ± 11 65 ± 11 NS
LVEDP, mm Hg 13 ± 6 13 ± 5 13 ± 5 NS
Diseased coronary arteries
0 vessels 23 (28) 21 (25) 31 (36)
1 vessel 41 (50) 40 (48) 37 (44)
2 vessels 12 (15) 12 (15) 12 (14)
3 vessels 6(7) 10(12) 5 (6)
Supine
Mean R-R, ms 939 ± 158 897 ± 151 944 ± 134 NS
Systolic BP, mm Hg 123 ± 15 123 ± 15 119 ± 16 NS
Diastolic BP, mm Hg 76 ± 10 75 ± 10 73 ± 10 NS
Tilt
Mean R-R, ms 778 ± 150 774 ± 121 806 ± 134 NS
Systolic BP, mm Hg 118 ± 16 117 ± 15 116 ± 17 NS
Diastolic BP, mm Hg 76 ± 12 76 ± 11 75 ± 12 NS
Therapies during follow-up
PTCA 9(15) 7 (8) 9 (16) NS
CABG 2 (3) 3 (5) 2 (3) NS
p -blocker 12 (15) 19 (23) 33 (39) < 0.001
ACE inhibitor 9(11) 7 (8) 10(12) NS
anosiahuman.com