Study objectives: To evaluate the characteristics of the cyclic alternating pattern (CAP) in the sleep of preschool-aged children in order to obtain normative parameters in this age range.
Design: Prospective study.
Settings: University sleep laboratory.
Participants: Ten normal healthy subjects (6 girls and 4 boys, mean age 4.6 years; range 3-6 years) underwent polysomnography recordings for 2 consecutive nights in a standard laboratory setting. Sleep data were stored on a computer using a polysomnography digital system (Embla N7000, Medcare, Iceland). Sleep structure was visually scored according to the Rechtschaffen and Kales criteria. The cyclic alternating pattern was visually scored following the criteria by Terzano et al. These criteria were modified for the purposes of this study because it was noticed that, at the age of the group under analysis, most electroencephalogram arousals, often accompanied by electromyogram activation, are expressed at the level of the electroencephalogram by theta frequencies and not alpha or higher frequencies.
Measurements and results: The CAP rate in preschool-aged children (25.93%) showed a progressive increase with the deepness of sleep, with highest values during non-rapid eye movement (NREM) stage 3 (44.0%) and 4 sleep (46.08%) and lowest values in NREM stage 2 (17.26%). The CAP time showed its longest duration during stage 2 sleep, followed by stage 4, stage 3, and NREM stage 1. The CAP cycle duration showed no differences across NREM stages. The cyclic alternating pattern phase A was longer and phase B was shorter during stage 1 than during stages 2, 3, and 4. A1 phases were the most numerous (63.2%), followed by A2 (21.5%) and by A3 (15.3%). The distribution of A-phase subtypes across NREM sleep stages (A index) showed significant differences for the A1 subtypes that occurred more frequently during stage 3 and 4 sleep than during stages 1 and 2. The A2 index showed no significant differences across NREM sleep stages, while the A3 index was significantly higher during stage 1 sleep than during stages 2, 3, and 4. The analysis of the A1 interval distribution showed a log-normal-like distribution with a peak around 25 seconds for the A1 phases and no clear peak for A2-A3 phases.
Conclusions: The analysis of CAP in preschool-aged children is characterized by an increase of CAP rate during slow-wave sleep and a high percentage of A1 phases and A2 phases. However, the lower percentage of A1 paralleled by an increase of A2 could represent a signal of higher sleep instability in this age group as compared with prepubertal school-aged children.