Table 5. HRV characteristics in morning wake according to OSA severity and MACCE.
| Controls n = 15 |
AHI <20 n = 30 |
AHI ≥20 n = 30 |
No MACCE n = 43 |
MACCE n = 17 |
|
|---|---|---|---|---|---|
| HR, 1/min | 64 (54–87) | 68 (59–83) | 68 (57–80) | 68 (59–83) | 67 (56–79) |
| NNI Min, ms | 688 (129) | 727 (121) | 747 (102) | 727 (118) | 764 (92) |
| NNI Max, ms | 1121 (119) | 1024 (212) | 1055 (205) | 1025 (190) | 1076 (248) |
| NNI Mean, ms | 933 (102) | 878 (158) | 887 (120) | 878 (145) | 898 (126) |
| NNI Dev, ms | 52 [46] | 46 [36] | 37 [27] * | 43 [34] | 34 [31] * |
| NNI RMSSD, ms | 20 [16] | 20 [17] | 20 [21] | 20 [18] | 18 [16] |
| NNI pNN50, % | 2.1 [4.4] | 1.6 [5.4] | 2.2 [5.8] | 1.8 [6.1] | 2.2 [4.9] |
| NNI SampEn | 1.0 (0.3) | 1.1 (0.3) | 1.1 (0.3) | 1.1 (0.3) | 1.1 (0.3) |
| NNIS Total, ms2 | 2174 [4952] | 1430 [3487] | 996 [2043] * | 1193 [2926] | 943 [1903] |
| NNIS ULF, ms2 | 223 [881] | 140 [488] | 69 [225] * | 118 [455] | 103 [322] * |
| NNIS VLF, ms2 | 1482 [2674] | 762 [2309] | 580 [1096] * | 633 [1692] | 490 [1096] * |
| NNIS LF, ms2 | 395 [1038] | 254 [283] | 218 [371] | 254 [323] | 173 [367] |
| NNIS HF, ms2 | 75 [120] | 76 [99] | 105 [110] | 91 [110] | 61 [116] |
| NNIS LF/HF | 5.4 [5.0] | 3.9 [3.3] | 2.5 [3.5] * ‡ | 3.7 [3.7] | 3.0 [3.7] |
| NNIS nLF, nu | 80 (14) | 73 (19) | 64 (20) * | 69 (20) | 67 (20) |
| NNIS nHF, nu | 20 (12) | 24 (18) | 32 (16) * | 28 (17) | 29 (18) |
| NNI Alpha 1 | 1.5 [0.18] | 1.34 (0.35) | 1.12 (0.34)†‡ | 1.25 (0.36) | 1.20 (0.35) |
Data are mean (standard deviation) or median [interquartile range] except for mean (range) for heart rate (HR). AHI = Apnea-hypopnea index, Alpha 1 = Fractal scaling exponent alpha 1, Dev = Standard deviation of NNI, HF = Power in the high frequency range (0.15–0.4 Hz), LF = Power in low frequency range 0.04–0.15 Hz), MACCE = Major adverse cardiovascular and cerebrovascular event, ms = millisecond, nHF = normalized HF ratio, nLF = normalized LF ratio, NNI = normal-to-normal interval (i.e. time between normal beats in the electrocardiogram), NNIS = NNI spectrum, pNN50 = Proportion of NNI >50 ms, RMSSD = Root mean square of the sum of successive differences (between adjacent normal-to-normal intervals, i.e. beat-to-beat variability), SampEn = Sample entropy, Total = Total power of the NNI spectrum (i.e. overall autonomic activity), ULF = Power in the ultra-low frequency range (≤0.003 Hz), VLF = Power in the very low frequency range (0.003–0.04 Hz).
*: p <0.05 vs. controls
†: p <0.01 vs. controls
‡: p<0.05 vs. AHI <20/hour (prior to adjustment for age, body mass index, coronary artery disease and PAD history). One-way analysis of variance with Dunnett’s post hoc was used to test for differences between each patient group (with and without OSA/MACCE) vs. controls. Analysis of covariance was used to test for independent differences between patient groups (with vs. without OSA/MACCE). Non-normally distributed variables were log-transformed (natural logarithm) before the statistical analyses (non-transformed values with IQRs in square brackets shown in tables). Kruskal-Wallis’s test and Mann-Whitney’s U-test with Bonferroni’s adjustment for multiple comparisons were used to test for differences in variables that were skewed after log-transformation.