Apnea is a common problem that causes significant parental anxiety. The aim of this study was to describe polysomnographic findings in infants who were referred over a 4-year period for an apnea and/or cyanotic event. Our hypothesis was that most infants with apnea or cyanosis events will have normal polysomnography (PSG). In total, 320 patients younger than 2 years old were recruited sequentially and prospectively. Patients underwent a day or overnight PSG by computerized polysomnograph; 78% of studies were performed with pH probe in situ. Subjects' ages ranged from 10 days to 21 months (55% male and 84% full-term babies); 55% and 74% were younger than 3 and 6 months, respectively. The average total sleep time was 473.4 min (SD, 52.3), with mean sleep efficiency of 83.5%. The distribution of sleep stages was 56.5% active, 38.5%, quiet and 5.1% indeterminate sleep. Sixty-nine percent (n = 220) of PSG studies were done overnight. There were significant differences in average sleep efficiency (78.1% vs. 83.3%) and REM sleep time (63.6% vs. 59.1%) between PSGs done during the day and overnight (P < 0.05). In total, 34 patients with apneas were studied. The median apnea index for the whole study population was 1.01 (range, 0.1-9.1). In conclusion, our study showed a high prevalence of normal polysomnographic findings in infants referred because of apnea and/or cyanotic events. New clinical prospective studies should be conducted to evaluate whether there is a correlation between PSG findings and outcome at follow-up of infants with a history of apnea and cyanosis.