In a randomized, crossover study, we compared arterial partial pressure of oxygen and of carbon dioxide between consecutive periods of conventional and synchronized intermittent mandatory ventilation (SIMV). We studied spontaneously breathing infants with an endotracheal tube in place. The infants were < 12 hours of age, had a diagnosis of respiratory distress syndrome, and had an arterial/alveolar oxygen ratio of < 0.25. The infants had a mean birth weight of 1077 gm and gestational age of 28 weeks. The mean rate of asynchrony on intermittent mandatory ventilation (IMV) was 52% (range, 36% to 76%), and on SIMV was < 1%. Infants were randomly assigned to IMV or SIMV as their initial ventilator mode and underwent ventilation for four 15-minute periods, and crossed over to the alternate mode after each period. Ventilator settings and the fraction of inspired oxygen were not changed between modes. At the end of each period, arterial blood gas measurements were obtained; 26 paired comparisons were made between modes. The mean arterial partial pressure of oxygen was significantly higher during SIMV than during IMV (mean, 61.5 vs 53.3 mmHg; p < 0.01). The mean arterial partial pressure of carbon dioxide was slightly lower during SIMV than during IMV (mean, 42.7 vs 41.3 mm Hg; p < 0.05). The improvement in oxygenation demonstrated with SIMV may allow a reduction in ventilator pressure or oxygen exposure in this group of infants, who are at risk of having complications of ventilation.