To test the hypothesis that room air is superior to 100% oxygen when asphyxiated newborns are resuscitated, 84 neonates (birth weight > 999 g) with heart rate < 80 and/or apnea at birth were allocated to be resuscitated with either room air (n = 42) or 100% oxygen (n = 42). Serial, unblinded observations of heart rates at 1, 3, 5, and 10 min and Apgar scores at 1 min revealed no significant differences between the two groups. At 5 min, median (25th and 75th percentile) Apgar scores were higher in the room air than in the oxygen group [8 (7-9) versus 7 (6-8), p = 0.03]. After the initial resuscitation, arterial partial pressure of oxygen, pH, and base excess were comparable in the two groups. Assisted ventilation was necessary for 2.4 (1.5-3.4) min in the room air group and 3.0 (2.0-4.0) min in the oxygen group (p = 0.14). The median time to first breath was 1.5 (1.0-2.0) min in both the room air and oxygen groups (p = 0.59), and the time to first cry was 3.0 (2.0-4.0) min and 3.5 (2.5-5.5) min in the room air and oxygen groups, respectively (p = 0.19). Three neonates in the room air group and four in the oxygen group died in the neonatal period. At 28 d, 72 of the 77 surviving neonates were available for follow-up (36 in each group), and none had any neurologic sequelae.(ABSTRACT TRUNCATED AT 250 WORDS)