Multicenter crossover study of automated control of inspired oxygen in ventilated preterm infants.

OBJECTIVE: To determine the efficacy and safety of automated adjustment of the fraction of inspired oxygen (Fio(2)) adjustment in maintaining arterial oxygen saturation (Spo(2)) within an intended range for mechanically ventilated preterm infants with frequent episodes of decreased Spo(2). METHODS: Thirty-two infants (gestational age [median and interquartile range]: 25 weeks [24-27 weeks]; age: 27 days [17-36 days]) were studied during 2 consecutive 24-hour periods, one with Fio(2) adjusted by clinical staff members (manual) and the other by an automated system (automated), in random sequence. RESULTS: Time with Spo(2) within the intended range (87%-93%) increased significantly during the automated period, compared with the manual period (40% ± 14% vs 32% ± 13% [mean ± SD]). Times with Spo(2) of >93% or >98% were significantly reduced during the automated period (21% ± 20% vs 37% ± 12% and 0.7% vs 5.6% [interquartile ranges: 0.1%-7.2% and 2.7%-11.2%], respectively). Time with Spo(2) of <87% increased significantly during the automated period (32% ± 12% vs 23% ± 9%), with more-frequent episodes with Spo(2) between 80% and 86%, whereas times with Spo(2) of <80% or <75% did not differ between periods. Hourly median Fio(2) values throughout the automated period were lower and there were substantially fewer manual Fio(2) changes (10 ± 9 vs 112 ± 59 changes per 24 hours; P < .001), compared with the manual period. CONCLUSIONS: In infants with fluctuations in Spo(2), automated Fio(2) adjustment improved maintenance of the intended Spo(2) range led to reduced time with high Spo(2) and more-frequent episodes with Spo(2) between 80% and 86%.