Table 3.
Methods of HRV measurement and obstructive sleep apnea findings.
| Study | Condition of recording and data lengh for analysis | Derived HRV measures | HRV and OSA findings |
|---|---|---|---|
| Noda et al., 199814 | 24-h Holter recordings | LF (0.04-0.15 Hz), HF (0.15-0.40 Hz), LF/HF | LF/HF ratio was elevated during sleep and daytime in patients with severe OSAS compared with patients with mild OSA and controls |
| Wiklund et al., 200017 | 10 min ECG recording at supine position a day after the sleep recording | PMF (0.04-0.15 Hz), PHF(0.15-0.40 Hz), PTOT | Decreased high-frequency component in supine position in OSAS |
| Aydin et al., 200418 | 24-h Holter recordings | Total power (0-0.4 Hz), ULF (0-0.0033 Hz), VLF (0.0033-0.04 Hz), LF (0.04-0.15 Hz), HF (0.15-0.40 Hz), LF/HF, SDNN, SDANN, RMSSD | SDNN and SDANN were lower in both mild and severe OSAS while RMSSD values were lower only in severe OSAS in comparison to controls. Total power, ULF, VLF, LF and LF/HF values of both groups of OSAS were higher than controls, but HF values were lower |
| Wakai et al., 200416 | 24-h Holter recordings | ULF (0.0001-0.003 Hz), VLF (0.003-0.04 Hz), LF (0.04-0.15 Hz), HF (0.15-0.40 Hz) | VLF and LF during sleep was higher in severe OSA patients than mild OSAS and controls |
| Chrysostomakis et al., 200619 | 24-h Holter recordings | NN, SDNN, SD, PNN50, RMSSD, SDANN | PNN50 and RMSSD were higher at night in patients with severe and moderate OSA |
| Coruzzi et al., 200620 | 20 min ECG recording at supine position | RRI, LF (0.04-0.14 Hz), HF (0.15-0.50 Hz) | RRI, RRI variation and HF were higher in controls than OSA patients but LF and LF/HF ration were lower |
| Aytemir et al., 200721 | 24-h Holter recordings | Total power, VLF (0-0.04 Hz), LF (0.04-0.15 Hz), HF (0.15-0.40 Hz), LF/HF, SDNN, RMSSD, pNN50 | At day time, SDNN was higher in control than OSA patients. During night time, RMSSD and 24h HFnu were higher, while 24h LF and LF/HF ratio were lower in controls than OSA patients |
| Zhu et al., 201222 | 24-h Holter recordings | VLF (0.003-0.04 Hz), LF (0.04-0.15 Hz), LFnu, HF (0.15-0.40 Hz), HFnu, LF/HF, SDNN, RMSSD, pNN50 | Patients with severe OSA exhibited a shorter mean NN overnight |
| Chang et al., 201324 | 15 min ECG recording in sitting position | LF (0.04-0.15 Hz), HF (0.15-0.40 Hz), LF/HF, RMSSD, SampEnRR | HF and SampEnRR were higher among controls, while LF and LF/HF were higher in OSA patients |
| Kim et al., 201523 | PSG recordings | TP, VLF (<0.04 Hz), LF (0.04-0.15 Hz), LFnu, HF (0.15-0.40 Hz), HFnu, LF/HF, SDNN, SDNNi, RMSSD, pNN50, TINN, HRVtri | All frequency domain parameters, except HF which was decreased, were increased in OSA patients. Among time domain parameters, all parameters were also increased in OSA group |
| Palma et al., 201525 | PSG recordings | LF (0.04-0.15 Hz), HF (0.15-0.40 Hz), LF/HF | OSA group during sleep exhibit higher LF and lower HF modulations |
| Xie et al., 201715 | PSG recordings | LF (0.04-0.15 Hz), HF (0.15-0.40 Hz), SDNN, RMSSD, pNN50 | SDNN and HF were higher in controls while LF/HF was lower in comparison to OSA group |
ECG=electrocardiography; HRV=heart rate variability; HRVtri=heart rate variability triangular index; HF=high frequency; HFnu=normalized units of high frequency component; LF=low frequency; LFnu=normalized units of low frequency component; LF/HF=ratio of low frequency and high frequency; OSA=obstructive sleep apnea; PHF=spectral power of the high-frequency component; PMF=spectral power of the mid-frequency component; pNN50=percentage of adjacent NN intervals; PTOT=variance of total spectral power; RMSSD=root mean squared successive difference; RRI=mean R-R interval; SampEnRR=sample entropy of R-R intervals; SDANN=stardard deviation of 5-min average NN intervals; SDNN=standard deviation of NN interval; TINN=triangular interpolation of NN intervals; T p = total power; ULF=ultralow frequency; VLF=very low frequency.