Objective: There is consolidated evidence that stage changes in sleep are c
losely related to spontaneous EEG fluctuations centered on the 20-40 period
icity of the cyclic alternating pattern (CAP:). The present investigation a
imed at assessing the involvement of the different components of CAP in the
process of build-up, maintenance and demolition of deep non-REM (NREM) sle
ep.
Methods: CAP parameters were quantified in the first 3 sleep cycles (SC1, S
C2, SC3), selected from polysomnographic recordings of 25 healthy sound sle
epers belonging to an extensive age range (10-49 years). Only ideal SCs wer
e selected, i.e. the ones uninterrupted by intervening wakefulness and in w
hich all stages were represented and linked in a regular succession of a de
scending branch, a trough and an ascending branch.
Results: Among the first 3 SCs, a total amount of 45 (SCI, 16; SC2, 13; SC3
, 16) met the inclusion requirements. SC1 contained the highest amount of s
low wave sleep (43.7 min) and the lowest values of CAP rate (31.6%). The nu
mber of phase A1 subtypes remained unmodified across the 3 SCs (SC1, 48; SC
2, 48; SC3, 48), whereas both subtypes A2 (SC1, 9; SC2, 14; SC3, 14) and A3
(SCI, 2; SC2, 8; SC3, 10) increased significantly (P <0.028 and P <0.0001,
respectively). The A1 subtypes composed more than 90% of all the A phases
collected in the descending branches and in the troughs, while the A2 and A
3 subtypes were the major representatives (64.3%) of the A phases occurring
in the ascending branches.
Conclusions: Within the dynamic organization of sleep, the non-random distr
ibution of CAP sequences, with their succession of slow (subtypes A1) and r
apid (subtypes A2 and A3) EEG shifts, seem to be responsible for sculpturin
g EEG synchrony under the driving and alternating forces of NREM and REM sl
eep. (C) 2000 Elsevier Science Ireland Ltd. All rights reserved.