Nocturnal short-term heart rate variability reflects impaired daytime vigilance better than overnight heart rate variability in suspected obstructive sleep apnea patients

Figures

• 

  Figure 1.. Inclusion and exclusion criteria for patients and 5-min segment pairs of R peak time series and nasal pressure selection. Abbreviations: CRC, cardiorespiratory coupling; ECG, electrocardiogram; ESS, Epworth sleepiness scale, HRV, heart rate variability; Kubios, Kubios Premium 3.4.1 software (Kubios Oy, Kuopio, Finland); OSA, obstructive sleep apnea; PSG, polysomnography; PVT, psychomotor vigilance task.

• 

  Figure 2.. An illustration of heart rate variability (HRV) and cardiorespiratory coupling (CRC) of a patient (a) without sleep apnea (apnea-hypopnea index (AHI) = 0.4 events/h) accompanied by low nocturnal hypoxic load (Desaturation severity (DesSev) = 0.008%) and (b) with severe sleep apnea (AHI = 81.2 events/h) accompanied by severe hypoxic load (DesSev = 5.022%). Intermittent hypoxia over-activates the sympathetitc nervous system and induces arousals from sleep. Together they further lead to fragmented sleep, lower normalized high-frequency (HF) power (decreased HRV), and weakened spectral coherence in the HF band (consistently impaired CRC). The level of significant CRC was based on surrogate data analysis [34]: it is determined only if the coupling level exceeded the threshold (ρ = 0.6241) for significant coupling. Otherwise, the coupling was marked as zero.

• 

  Figure 3.. Variation of overnight and short-term heart rate variability (HRV) and cardiorespiratory coupling (CRC) parameters in men based on their performance in psychomotor vigilance task (PVT). In general, the variation within parameters is higher in the short-term data: by averaging the parameter values overnight, we lose important physiological information. Abbreviations: DesSev, desaturation severity; HFc, cross-spectral coherence in the HF band; HFnu, normalized power in the HF band (HF/(LF+HF)); LFc, cross-spectral coherence in the low-frequency band; LF/HF, ratio of LF to HF band power; LFnu, normalized power in the low-frequency band (LF/(LF+HF)); NN, interval between two successive normal R peaks; RMSSD, root mean square of successive differences; SDNN, standard deviation of NN intervals. *Statistically significant difference compared with the vigilant group Q1 (p < 0.05), **statistically significant difference compared with vigilant group Q1 (p < 0.001).

• 

  Figure 4.. Variation of overnight and short-term heart rate variability (HRV) and cardiorespiratory coupling (CRC) parameters in women based on their performance in psychomotor vigilance task (PVT). In general, the variation within parameters is higher in the short-term data: by averaging the parameter values overnight, we lose important physiological information. Abbreviations: DesSev, desaturation severity; HFc, cross-spectral coherence in the HF band; HFnu, normalized power in the HF band (HF/(LF+HF)); LFc, cross-spectral coherence in the low-frequency band; LF/HF, ratio of LF to HF band power; LFnu, normalized power in the low-frequency band (LF/(LF+HF)); NN, interval between two successive normal R peaks; RMSSD, root mean square of successive differences; SDNN, standard deviation of NN intervals. **Statistically significant difference compared with the vigilant group Q1 (p < 0.001).