Dexmedetomidine promotes biomimetic non-rapid eye movement stage 3 sleep in humans: A pilot study

Abstract

Objectives: Sleep, which comprises of rapid eye movement (REM) and non-REM stages 1-3 (N1-N3), is a natural occurring state of decreased arousal that is crucial for normal cardiovascular, immune and cognitive function. The principal sedative drugs produce electroencephalogram beta oscillations, which have been associated with neurocognitive dysfunction. Pharmacological induction of altered arousal states that neurophysiologically approximate natural sleep, termed biomimetic sleep, may eliminate drug-induced neurocognitive dysfunction.

Methods: We performed a prospective, single-site, three-arm, randomized-controlled, crossover polysomnography pilot study (n = 10) comparing natural, intravenous dexmedetomidine- (1-μg/kg over 10 min [n = 7] or 0.5-μg/kg over 10 min [n = 3]), and zolpidem-induced sleep in healthy volunteers. Sleep quality and psychomotor performance were assessed with polysomnography and the psychomotor vigilance test, respectively. Sleep quality questionnaires were also administered.

Results: We found that dexmedetomidine promoted N3 sleep in a dose dependent manner, and did not impair performance on the psychomotor vigilance test. In contrast, zolpidem extended release was associated with decreased theta (∼5-8 Hz; N2 and N3) and increased beta oscillations (∼13-25 Hz; N2 and REM). Zolpidem extended release was also associated with increased lapses on the psychomotor vigilance test. No serious adverse events occurred.

Conclusions: Pharmacological induction of biomimetic N3 sleep with psychomotor sparing benefits is feasible.

Significance: These results suggest that α2a adrenergic agonists may be developed as a new class of sleep enhancing medications with neurocognitive sparing benefits.

Keywords: Biomimetic sleep; Dexmedetomidine; N3 sleep; Sedation; Zolpidem.

Fig. 1. Dexmedetomidine increases N3 sleep in a dose dependent fashion

A) Total sleep time was not significantly different between the sleep groups. However, dexmedetomidine significantly increased the total time spent in non-REM sleep compared to natural sleep (p = 0.021). This increase was compensated for by a significantly decreased time spent in REM sleep compared to natural sleep (p =0.006). B) N1 and N2 sleep were not significantly different between the groups. However, dexmedetomidine significantly increased the total time spent in N3 sleep compared to natural sleep (p = 0.032). C) The effect of dexmedetomidine on N3 sleep increase was most prominent during the first half of the night. Dexmedetomidine significantly increased the total time spent in N3 sleep compared to natural sleep during the first half of the night (p = 0.0001). Zolpidem also significantly increased the total time spent in N3 sleep compared to natural sleep during the first half of the night (p = 0.031). D) Dexmedetomidine significantly increased the total time spent in N2 sleep compared to natural sleep during the second half of the night (p = 0.042). E) There was a dose dependent association between dexmedetomidine and total N3 sleep with higher doses of dexmedetomidine resulting in increased total N3 sleep (Spearman’s ρ = 0.58, p = 0.08). F) There was also a dose dependent association between dexmedetomidine and total REM sleep with higher doses of dexmedetomidine resulting in decreased total REM sleep (Spearman’s ρ = −0.44, p = 0.2). REM, rapid eye movement; *, p <0.05; **, p <0.05. Error bars represent standard deviation.

Fig. 2. Spectral comparison of EEG power during non-rapid eye movement stage N2 sleep

(A) Top panel; Overlay of median first natural sleep N2 spectrum (black), and median first dexmedetomidine N2 spectrum (red). Bootstrapped median spectra are presented, and the shaded regions represent the 99% confidence interval for the uncertainty around each median spectrum. Bottom panel: Median bootstrap difference in spectra with shaded regions representing the 99% confidence of the difference. We did not find differences between the two spectra. (B) Top panel; Overlay of median first natural sleep N2 spectrum (black), and median first zolpidem N2 spectrum (red). Bootstrapped median spectra are presented, and the shaded regions represent the 99% confidence interval for the uncertainty around each median spectrum. Bottom panel: Median bootstrap difference in spectra with shaded regions representing the 99% confidence of the difference. We found differences in power between the first natural N2 spectrum and the first zolpidem N2 spectrum (zolpidem < natural sleep: 5.3 – 9.8 Hz). (C) Top panel; Overlay of median last natural sleep N2 spectrum (black), and median last dexmedetomidine N2 spectrum (red). Bootstrapped median spectra are presented, and the shaded regions represent the 99% confidence interval for the uncertainty around each median spectrum. Bottom panel: Median bootstrap difference in spectra with shaded regions representing the 99% confidence of the difference. We did not find differences between the two spectra. (D) Top panel; Overlay of median last natural sleep N2 spectrum (black), and median last zolpidem N2 spectrum (red). Bootstrapped median spectra are presented, and the shaded regions represent the 99% confidence interval for the uncertainty around each median spectrum. Bottom panel: Median bootstrap difference in spectra with shaded regions representing the 99% confidence of the difference. We found differences in power between the last natural N2 spectrum and the last zolpidem N2 spectrum (zolpidem > natural sleep: 15.5 – 24.9 Hz). Black lines in bottom panel represent regions that me our threshold for significance.

Fig. 3. Spectral comparison of EEG power during non-rapid eye movement stage N3 sleep

(A) Top panel; Overlay of median first natural N3 spectrum (black), and median first dexmedetomidine N3 spectrum (red). Bootstrapped median spectra are presented, and the shaded regions represent the 99% confidence interval for the uncertainty around each median spectrum. Bottom panel: Median bootstrap difference in spectra with shaded regions representing the 99% confidence of the difference. We did not find differences between the two spectra. (B) Top panel; Overlay of median first natural sleep N3 spectrum (black), and median first zolpidem N3 spectrum (red). Bootstrapped median spectra are presented, and the shaded regions represent the 99% confidence interval for the uncertainty around each median spectrum. Bottom panel: Median bootstrap difference in spectra with shaded regions representing the 99% confidence of the difference. We found differences in power between the first natural sleep N3 spectrum and the first zolpidem N3 spectrum (zolpidem < natural sleep: 6.8 – 9.8 Hz). (C) Top panel; Overlay of median last natural sleep N3 spectrum (black), and median last dexmedetomidine N3 spectrum (red). Bootstrapped median spectra are presented, and the shaded regions represent the 99% confidence interval for the uncertainty around each median spectrum. Bottom panel: Median bootstrap difference in spectra with shaded regions representing the 99% confidence of the difference. We did not find differences between the two spectra. (D) Top panel; Overlay of median last natural sleep N3 spectrum (black), and median last zolpidem N3 spectrum (red). Bootstrapped median spectra are presented, and the shaded regions represent the 99% confidence interval for the uncertainty around each median spectrum. Bottom panel: Median bootstrap difference in spectra with shaded regions representing the 99% confidence of the difference. We found differences in power between the last natural sleep N3 spectrum and the last zolpidem N3 spectrum (zolpidem > natural sleep: 2.5 – 8.9 Hz). Black lines in bottom panel represent regions that me our threshold for significance.

Fig. 4. Spectral comparison of EEG power during non-rapid eye movement stage N3 sleep

(A) Top panel; Overlay of median natural REM sleep spectrum (black), and median dexmedetomidine REM sleep spectrum (red). Bootstrapped median spectra are presented, and the shaded regions represent the 99% confidence interval for the uncertainty around each median spectrum. Bottom panel: Median bootstrap difference in spectra with shaded regions representing the 99% confidence of the difference. We did not find differences between the two spectra. (B) Top panel; Overlay of median natural REM sleep spectrum (black), and median last zolpidem REM sleep spectrum (red). Bootstrapped median spectra are presented, and the shaded regions represent the 99% confidence interval for the uncertainty around each median spectrum. Bottom panel: Median bootstrap difference in spectra with shaded regions representing the 99% confidence of the difference. We found differences in power between the last natural sleep N3 spectrum and the first zolpidem (zolpidem < natural sleep: 12.1 – 22.3 Hz, 23.1 – 24.9 Hz).