Ms. Scher et al., CARDIORESPIRATORY BEHAVIOR DURING SLEEP IN FULL-TERM AND PRETERM NEONATES AT COMPARABLE POSTCONCEPTIONAL TERM AGES, Pediatric research, 36(6), 1994, pp. 738-744
Cardiorespiratory behavior during sleep has been investigated by compa
ring visually analyzed minutes of EEG sleep with the digitized values
of these two physiologic variables for each corresponding minute. Cont
inuous 3-h nighttime sleep studies on 37 full-term and preterm neonate
s at comparable postconceptional term ages were acquired under control
led conditions, using a 24-channel computerized monitoring system and
an automated event-marker program. Five thousand, two hundred ninety-f
our minutes were assigned an EEG state by traditional criteria. Eighte
en preterm infants were compared with 19 full-term infants with respec
t to six cardiac and six respiratory measures: two nonspectral calcula
tions tie. average per minute and variance of the means) and four spec
tral calculations of the cardiorespiratory signal (i.e. bandwidth, spe
ctral edge, mean frequency, and ratio of harmonics). The relative capa
bilities of these measures to predict a sleep state change were invest
igated using discriminant analysis. A stepwise selection algorithm in
discriminant analysis was used to identify the order of significance f
or the remaining variables. Eight cardiorespiratory measures were then
submitted to multivariate analysis of variance to assess sleep state
or preterm-full-term differences: mean frequency, bandwidth, average p
er minute, and ratio of harmonics for cardiac signals; and spectral ed
ge, meanfrequency, logarithm of variance, and ratio of harmonics for r
espiratory signals. Differences among the sleep states and between neo
natal groups were highly significant (p < 0.0001). Interaction between
sleep state and neonatal group was also significant (p < 0.034). Two
variables differentiated preterm from full-term respiratory behavior:
ratio (p less than or equal to 0.001) and mean frequency (p less than
or equal to 0.02). Three variables demonstrated differences between pr
eterm and full-term cardiac behavior: average heart rate per minute (p
less than or equal to 0.001), ratio (p less than or equal to 0.05), a
nd bandwidth (p less than or equal to 0.08). Notably, the lowest value
s for most spectral measures were noted during trace alternant quiet s
leep compared with the three other segments of the ultradian sleep cyc
le. Our findings demonstrate sleep state-specific differences in cardi
orespiratory behavior in neonates regardless of prematurity. Differenc
es between preterm and full-term infants reflect altered functional de
velopment of the brain because of adaptation to prematurity, an extrau
terine experience, or both and may contribute to a model of physiologi
c vulnerability of certain infants for sudden infant death syndrome.