Heart rate variability can be used to estimate sleepiness-related decrements in psychomotor vigilance during total sleep deprivation

Abstract

Study objectives: To assess whether changes in psychomotor vigilance during sleep deprivation can be estimated using heart rate variability (HRV).

Design: HRV, ocular, and electroencephalogram (EEG) measures were compared for their ability to predict lapses on the Psychomotor Vigilance Task (PVT).

Setting: Chronobiology and Sleep Laboratory, Duke-NUS Graduate Medical School Singapore.

Participants: Twenty-four healthy Chinese men (mean age ± SD = 25.9 ± 2.8 years).

Interventions: Subjects were kept awake continuously for 40 hours under constant environmental conditions. Every 2 hours, subjects completed a 10-minute PVT to assess their ability to sustain visual attention.

Measurements and results: During each PVT, we examined the electrocardiogram (ECG), EEG, and percentage of time that the eyes were closed (PERCLOS). Similar to EEG power density and PERCLOS measures, the time course of ECG RR-interval power density in the 0.02-0.08-Hz range correlated with the 40-hour profile of PVT lapses. Based on receiver operating characteristic curves, RR-interval power density performed as well as EEG power density at identifying a sleepiness-related increase in PVT lapses above threshold. RR-interval power density (0.02-0.08 Hz) also classified subject performance with sensitivity and specificity similar to that of PERCLOS.

Conclusions: The ECG carries information about a person's vigilance state. Hence, HRV measures could potentially be used to predict when an individual is at increased risk of attentional failure. Our results suggest that HRV monitoring, either alone or in combination with other physiologic measures, could be incorporated into safety devices to warn drowsy operators when their performance is impaired.

Keywords: ECG; EEG; HRV; PERCLOS; PVT; Sleepiness; heart rate.

Figure 1

Protocol for measuring sleepiness-related changes in physiology and performance. (A) Subjects participated in a 4-day laboratory protocol that included a 40-h constant routine procedure in dim light (< 5 lux). The protocol shown is for a representative subject with habitual bedtime at midnight. Black bars show scheduled sleep episodes in darkness. Every 2 hours, subjects completed the Karolinska Sleepiness Scale, a visual analogue scale for sleepiness, and a 10-minute Psychomotor Vigilance Task (PVT; asterisks). (B) Lapses on the PVT increased during the usual hours of sleep and reached their highest levels at about 24 h after wake. Thereafter, performance showed partial recovery toward the baseline rested state. (C) Subjective sleepiness measured with the visual analogue scale (VAS) showed a similar profile to PVT lapses, with highest levels of sleepiness reported between 20 and 26 h after wake. Vertical reference lines in B and C indicate habitual bedtime and wake time. Error bars show SEM.

Figure 2

Time course of heart rate variability, electroencephalographic (EEG) spectral power, and ocular measures during sleep deprivation. (A) The RR-interval time series during each Psychomotor Vigilance Task (PVT) was determined from the peak of consecutive QRS complexes (open circles) in the electrocardiogram (ECG). (B) As shown in a representative subject, variability in the RR-interval time series was low during the habitual hours of wakefulness, and increased during sleep deprivation. (C) RR-interval power spectral density (PSD) in the 0.02- to 0.08-Hz frequency band correlated most strongly with PVT lapses. (D) The time course of RR-interval PSD in this frequency range increased during the usual hours of sleep, and then decreased after 24 h of wake, similar to the profile of PVT lapses shown in Figure 1B. (E) For EEG PSD-derived measures, delta power (1.0-4.5 Hz) measured in the frontal derivation correlated most strongly with PVT performance. (F) The time course of delta power tracked the profile of PVT lapses and subjective sleepiness. (G) Eye blinks showed an inverted profile relative to PVT lapses, whereas (H) the time course of PERCLOS matched the profile of PVT performance during 40 h of sustained wakefulness. Vertical reference lines in D, F, G, and H indicate habitual bedtime and wake time. Error bars show SEM.

Figure 3

RR-interval power spectral density (PSD) can be used to identify an increase in Psychomotor Vigilance Task (PVT) lapses above threshold. (A) Receiver operating characteristic (ROC) curves show the relative performance of different physiologic and subjective measures at identifying a > 25% increase in PVT lapses, measured relative to each individual's range of PVT performance. (B) Using leave-1-out cross-validation to normalize data for classification, ROC curves are shown for RR-interval PSD (0.02-0.08 Hz) versus percentage of eyelid closure over the pupil over time (PERCLOS) for classifying subject performance at the 25% PVT lapse threshold. (C) Based on area under the curve (AUC) for ROC curves at different PVT lapse thresholds, RR-interval PSD (red bars) and PERCLOS (black bars) were similar in their ability to predict a relative increase in PVT lapses. Error bars show the standard error of AUC values.