SummaryIn order to test the hypothesis that a defect in cardiorespiratory regulation contributes to death of infants from sudden infant death syndrome (SIDS), we analyzed the power spectra of heart rate and respiratory activity during 256-sec epochs of quiet sleep. Data were obtained from pneumogram recordings performed for 12 h at night on eight infants who subsequently died from SIDS and 22 age-matched control infants. We computed the heart rate and respiratory power spectra from a single epoch on each infant using an algorithm developed for an 8085 microprocessor system dedicated to this investieation. There was no statistically sienificant difference betweenvSID~ and controls based on mean-respiratory and heart rates. Spectral analysis revealed enhancement in general has rested on the analysis of naturally occurring fluctuations in system variables (4). These fluctuations contain information concerning the spontaneously changing load on the system as well as the system's response. A useful first step in the quantitative analysis of the control system is the spectral analysis of the system variables. Prior work (1, 7, 9, 11) has emphasized the importance and relevance of the analysis of fluctuations in heart rate to assess autonomic functions. We present here the results of spectral analysis of 256-sec-long segments ofthe instantaneous heart and respiratory waveform in 22 control infants and eight infants who died of SIDS.
MATERIALS AND METHODSof low frequency power in the 0.02 to 0.1 Hz band in the heart rate power spectrum in the SIDS group compared to control ( p Infants who died of SIDS diagnosed after an autopsy failed to < 0.002) and dispersion in respiratory frequency as determined identify the cause of death were evaluated because they all had by the respiratory band width ( p < 0.0000~). These data suggest at least one nocturnal recording of ECG and respiration at home. that a predisposition to SIDS manifests itself in an abnormal The indications for conducting the recording were near-SIDS pattern of fluctuations in heart rate and respiratory activity.(2), choking spells (2), and death from SIDS in one sibling (1) or two siblinns (2). One of the latter also exmrienced a near--. , SIDS event. Thus, three had not experiencedmnear-SIDS; in two Abbreviations of these, GER was identified on barium swallow. Although she SIDS, sudden infant death syndrome had no choking spells, GER was also identified in a sibling with ECG, electrocardiogram near-SIDS (patient 7, Table 1). The shortest interval between a EEG, electroencephalogram spell and the nocturnal recording was 5 days (the first recording GER, gastroesophageal reflux in patient 7). Near-SIDS was defined as sleep onset apnea asso-HR, heart rate ciated with limpness and cyanosis or pallor responding only to vigorous stimulation or resuscitation. Identifiable causes of pathologic apnea were ruled out by complete blood count, ECG, The purpose of this investigation was to test the hypothesis EEG, chest x-ray, serum Na' , K' , C1-, HC03-, and Ca2+, and sugg...