The impact of slow wave sleep proximity on evoked K-complex generation

During human stage 2 non-rapid eye movement (NREM) sleep, spontaneous K-complexes are more likely to occur prior to transitions to stage 3 or stage 4 sleep (referred to as slow wave sleep or SWS) compared to transitions to REM sleep, suggesting that the K-complex may be the 'forerunner' of SWS. The present study investigated the impact of SWS or REM sleep proximity on the probability of evoking a K-complex (pKC) during stage 2 and on components of the NREM sleep averaged evoked potential. Ten subjects spent three nights in the laboratory. On either the second or third night tones were presented continuously during sleep. Evoked K-complexes and sleep-evoked potentials were assessed for the 10 min of stage 2 prior to SWS (SWS-10) or REM (REM-10) sleep episodes as well as for all of SWS. pKC did not differ between SWS-10 (0.88) and SWS (0.91) but was significantly larger in SWS-10 than REM-10 (0.63). Amplitude effects were seen for the P2, N350, P900 NREM sleep-evoked potential components but not for the K-complex related N550. In each case where amplitude effects were found, SWS-10 was larger than REM-10. No latency differences were seen between conditions for the earlier components (P2, N350) however, both N550 and P900 were significantly shorter during SWS-10 compared to REM-10. These results are consistent with previous spontaneous K-complex studies and are supportive of a relationship between the K-complex and delta activity. They also indicate that stage 2 may consist of a continuum of microstates between SWS and REM sleep that are indicative of different brain stem, diecephalic and cortical patterns of activation.