Sleep organization of infants may be influenced by differences in nutrient intakes from human milk and formula. Because sleep/awake and sleep stage patterns affect energy expenditure, we hypothesized that differences in sleep organization between breast-fed and formula-fed infants might account in part for differences in energy expenditure between feeding groups. Sleep stages and cycling of 4-mo-old breast-fed (n = 10) formula-fed (n = 10) infants were studied with simultaneous measurements of energy expenditure. EEG, electrooculogram, body movement by triaxial accelerometry, heart rate, and oxygen saturation were monitored during an overnight sleep session. Sleep stages, nonrapid eye movement (NREM), and rapid eye movement (REM) were determined. Behavioral observations were recorded by video tape and by a technologist. Oxygen consumption and carbon dioxide production were measured with an indirect calorimeter. Total number and duration of sleep cycles, REM latency, number of NREM and REM epochs, and duration of NREM epochs did not differ between feeding groups. Sleep latency was shorter (p < 0.05) and duration of REM epochs longer (p < 0.01) in the formula-fed group. Formula-fed infants spent a higher percentage of sleep time in REM compared with the breast-fed infants (42 versus 34%) (p < 0.003). Conversely, breast-fed infants spent a higher percentage of sleep time in NREM sleep and their heart rates during sleep were lower (114 versus 126 bpm; p < 0.01). Energy expenditure during REM sleep was 13.0 +/- 4.4% higher than during NREM sleep (p < 0.001).(ABSTRACT TRUNCATED AT 250 WORDS)