Effect of stage 1 sleep on auditory cortex during pure tone stimulation: evaluation by functional magnetic resonance imaging with simultaneous EEG monitoring

Abstract

Background and purpose: Responses of the auditory cortex to sound during sleep have been explored with somewhat discrepant results. The purpose of this study was to investigate the effect of stage 1 sleep on signal intensity changes in the auditory cortex in response to pure tone stimulus measured by functional MR imaging.

Methods: Six sleep-deprived subjects were exposed to a series of echo-planar images for 30-40 minutes. No medication was used to help the subjects go to sleep. A long repetition time of 12 seconds and a 1.9-second clustered multisection acquisition were used to minimize the effect of imager acoustic noise from the preceding acquisition and to make it possible to obtain electroencephalographs between image acquisitions. A pure tone stimulus (beep, 1,000-Hz sine waves, 30-millisecond duration, five beeps per second) was alternated with the baseline every 36 seconds.

Results: All subjects fell asleep. The effect of habituation evaluated by comparing the percentage of signal intensity change between the first and second half was not significant. The percentage of signal intensity changes in the right and left transverse temporal gyri were 0.49% and 0.43% during wakefulness and 0.05% and 0.07% during stage 1 sleep. The differences between wakefulness and stage 1 sleep were significant.

Conclusion: Transition to stage 1 sleep coincides with a decrease in functional MR imaging-determined signal intensity changes in the auditory cortex in response to pure tone stimulus. The limited response of the brain at this stage may protect the brain from sound and facilitate deepening of the sleep stage.

Fig 1.

Timing of auditory stimulation, functional MR imaging, and EEG measurements. Black vertical bars indicate echo-planar imaging measurements of 16 sections within 1.9 seconds followed by a noise-free and EEG measurement period of 10.1 seconds. Auditory stimulus was alternated with the baseline every 36 seconds. Acquisition time was 30–40 minutes, or 153–194 image acquisitions.

Fig 2.

Regions of interest in one subject. The region of interest were chosen so as to include the transverse temporal gyrus and its adjacent sulci in five contiguous sections.

Fig 3.

Evaluation of the effect of habituation. Percentage signal intensity changes during the first and second half of the measurements are shown. Panels A and B correspond to the right and left hemispheres, respectively. Two-factor factorial ANOVA demonstrated the effect of the timing of data acquisition was not significant, which implied that the effect of habituation measured with this method was insignificant. On the other hand, the effect of state of consciousness was found to be significant.

Fig 4.

Percentage signal intensity change in the transverse temporal gyri. A significant effect of state of consciousness was revealed by two-way repeated-measure ANOVA. The Tukey-Kramer post hoc test showed that the percentage signal intensity change during stage 1 sleep was significantly smaller than during wakefulness. The overall effect of the hemisphere was not significant.

Fig 5.

Positively or negatively correlated pixels (t >3.5; cluster > = 4 pixels; corrected P value < .05) during wakefulness and stage 1 sleep in one subject. Clusters of positively correlated pixels are seen in the bilateral transverse temporal gyri during wakefulness. No cluster is found during the other conditions.

Fig 6.

Number of activated pixels in the transverse temporal gyrus. A significant effect of state of consciousness was revealed by two-way repeated-measure ANOVA. The Tukey-Kramer post hoc test showed the number of positively correlated pixels during wakefulness was significantly larger than that under other conditions. The overall effect of the hemisphere was not significant.