The study was designed to evaluate whether results of a single overnight recording session are sufficient for the study of sleep profiles and detection of apneas in infants, or whether it would be beneficial to extend the recording period. Nineteen infants were recorded during successive nights. Eight of the 19 infants were studied after an idiopathic apparent life-threatening event, whereas the other 11 were healthy. There were 13 boys and six girls, with a median age of 11 weeks (range 5-36 weeks). All infants were recorded polygraphically during 2 nights, and 11 were recorded during 3 successive nights. No significant difference was observed between any of the following variables, regardless of the number of nights for which the recording was performed: total recording time, total sleep time, delay in sleep onset, time awake, percent of rapid eye movement or nonrapid eye movement sleep, mean respiratory rates, density and duration of central, obstructive or mixed apneas. The frequency of obstructed breathing events for each infant did not differ significantly from 1 night to the next. The present study indicates that under adequate study conditions, recordings of a single night can reliably describe the frequency of central and obstructive apneas in infants.