The AASM manual recommends to apply these scoring rules in infants aged 0–2 months post-term (37–48 weeks CA). Sleep and wakefulness in infants 38–48 weeks CA are scored based on behavioral observation; regularity or irregularity of respiration; and EEG, EOG, and chin EMG patterns. We can distinguish four behavioral stages:

Stage N (NREM) can be scored if four or more of the following features are present for more than half the length of the epoch: (a) eyes closed with no eye movements; (b) chin EMG tone present but lower than during W; (c) regular respiration; (d) tracé alternant (TA), high voltage slow (HVS), or sleep spindles; and (e) reduced movement relative to W. During this stage, sucking can occur.

Stage R (REM) can be scored if four or more of the following criteria are present: (a) low chin EMG; (b) eyes closed with at least one rapid eye movement; (c) irregular respiration; (d) mouthing, sucking, twitches, or brief head movements; and (e) continuous EEG pattern, including low voltage irregular (LVI), high voltage slow (HVS), and mixed (M) without sleep spindles.

The AASM manual recommended applying the following scoring rules in children 2 months post-term or older.

The scoring rule for stage W is when more than 50 % of the epoch contains age-appropriate posterior dominant rhythm over the occipital region and/or if there is eye blinks (0.5–2 Hz), reading eye movements, or rapid eye movements associated with normal or high chin muscle tone.

Stage N3 begins when ≥20 % of an epoch consists of SWA, irrespective of age. Sleep spindles may persist in stage N3 sleep while eye movements are not typically seen. An example of this sleep stage in a 4-year-old infant can be seen in Fig. 6.2.

Stage R is characterized by the presence of rapid eye movements (REM), low chin EMG tone, sawtooth waves, transient muscle activity, and continuous, low-amplitude, mixed-frequency EEG activity. This EEG activity of stage R in infants and children looks like that of adults, although the dominant frequencies increase with age: 3 Hz activity at 7 weeks post-term, 4–5 Hz activity at 5 months, 4–6 Hz at 9 months, and 5–7 Hz theta activity at 1–5 years of age. By the age of 5–10 years, the low-amplitude, mixed-frequency activity in stage R is similar to that of adults [11].

CAP is an endogenous rhythm present in NREM sleep characterized by a periodic EEG activity with sequences of transient electrocortical activations (phase A of the cycle) that are distinct from the background EEG activity (phase B of the cycle). These sequences are repeated several times during the night and organized in a cyclic pattern interrupted by the presence of a stable sleep, without oscillations, called non-CAP (NCAP), longer than 60 s. CAP A phases have been subdivided into different subtypes: A1, A2, and A3, based on their frequency content [25, 26]. The A1 subtypes are composed prevalently by slow waves and the A3 subtype is generally composed of fast EEG activities, with subtype A2 presenting a combination of both.

In newborn and infants, CAP appears in a rudimentary form at 46–55 weeks CA, related to the emergence of an oscillating pattern of slow EEG activities, but the attainment of mature sleep EEG patterns is essential to score CAP. Miano et al. [27] suggested that two sleep EEG patterns are required to score CAP: (a) high-voltage slow activity (HVS) and rudimentary spindles and (b) SWS and spindles. In fact, CAP rate was 6.83 ± 3.58 S.D. in infants with the sleep EEG pattern (a) and increased to 12.9 ± 2.21 S.D. in children with pattern (b). The percentage of A1, A2, and A3 showed nonsignificant variations with age, but an increase of A1 index (number of events/hour) was observed in children with pattern (b). The duration of CAP events was similar in all age groups considered, and similarly, the arousal indexes were not statistically different [27].

In the preschool age, CAP rate clearly increases with highest values during SWS [28]. In comparison with infants and school-aged children, preschool subjects show a lower percentage of A1 and a corresponding increase in percentage of A2; this finding might represent an indirect marker of a more disturbed sleep or of the maturational processes of sleep [26].

The sleep structure in 6- to 10-year-old children is very stable and can be considered to be the “gold standard” for sleep quality because of its length, continuity, and restorative features [29]. Within this pattern of stable sleep architecture, CAP rate is higher than at younger ages and shows a progressive increase with the deepness of sleep, with the highest values during SWS [30]. In school-aged children, A1 subtypes are predominant (84.45 % of total) and occur mainly during SWS, followed by A3 (9.14 %) and A2 (6.44 %) subtypes. Similarly to preschool children and adults, CAP time structure shows the same periodicity of A1 subtypes, with a peak at around 25 s. The almost identical periodicity and time interval distribution of CAP A1 subtypes from infants to adults indicates that the periodicity of CAP components can be considered very stable during development [28, 31].

The way from preadolescence to adolescence is characterized by peculiar changes of the sleep EEG mainly represented by a decline of low EEG frequencies (theta and delta activities) in NREM [18]. These changes are convoyed by a great instability of sleep EEG that consequently affects CAP parameters. In fact, an important increase of CAP rate has been found in a group of peripubertal children (age 8–12 years; Tanner stage 2 and 3) who showed a CAP rate of 62.1 %; CAP A1 subtypes were the most numerous (85.5 %), whereas A2 were 9.1 % and A3 were 5 % [32]. Peripubertal and adolescents show the highest CAP rate among all life periods, if we exclude the elderly period [33] (Table 6.1).