Control of activity of the diaphragm in rapid-eye-movement sleep.

Respiration in rapid-eye-movement sleep (REMS) is known to be highly variable. The purpose of this study was to investigate the source of this variability and to determine which ordering principles remained operative in REM sleep. In unrestrained, naturally sleeping cats we recorded the electroencephalogram, electrooculogram, neck electromyogram, and diaphragmatic electromyogram (EMG) and computed its moving average (MAdi). As a reference, we first examined MAdi during "tonic" REMS, since breathing is fairly regular in this state. "Control" ranges for peak amplitude (PEMG), inspiratory time (TI), duration of postinspiratory inspiratory activity, expiratory time, and the calculated inspiratory slope (PEMG/TI) were determined by overlaying individual breath traces of the time course of MAdi during tonic REMS to form a composite tracing. Next, the time course of the EMG during individual breaths in slow-wave sleep (SWS) and a complete period of consecutive breaths in REMS (both tonic and phasic) were compared with this tonic REMS composite. The number of eye movements per breath was tabulated as an index of phasic activity. The inspiratory slopes during SWS and tonic REMS were similar. However, during phasic REMS, many breaths displayed either increases (excitation) or decreases (inhibition) in slope compared with the "typical" breaths seen in tonic REMS. The occurrence of these altered slopes increased with the frequency of phasic events. TI was inversely related to the slope of the EMG, which tended to minimize changes in PEMG.(ABSTRACT TRUNCATED AT 250 WORDS)