Hylke Salverda

71 4 Comparison of two AOCs in oxygen targeting in preterm infants during admission Introduction Caregivers in the Neonatal Intensive Care Unit (NICU) must continuously maintain a fragile balance between administering too much and too little supplemental oxygen to preterm infants to prevent neonatal morbidity and mortality. Several neonatal morbidities have been linked to a disturbance in this balance, with intermittent hypoxia being associated with retinopathy of prematurity,1 neurodevelopmental impairment and death,2 and hyperoxia long known to be causative of retinopathy of prematurity.3 Maintaining the balance involves titration of the inspired fraction of oxygen (FiO2). When done manually this often leads to an achieved time within the oxygen saturation (SpO2) target range of 50% or less. 4 Automated targeting of SpO 2 by a device titrating FiO2 can increase the time that preterm infants spend within the target range. 5, 6 In general, automated oxygen titration entails a computer program that automatically adjusts the FiO2 based on the measured SpO2. The magnitude of the FiO2 adjustment is usually determined by several factors, such as the currently administered FiO2 and the difference between the measured SpO2 and the intended SpO2. Several of these devices are available commercially.7-13 The function of all of these devices has been examined in cross-over studies lasting 24 hours or less per arm, and has proved superior to manual titration, but head-to-head comparisons of devices are scarce.12, 14 Algorithm choice may influence how successful titration will be.15 We recently demonstrated the OxyGenie controller (SLE Limited, South Croydon, UK) to be more effective in maintaining SpO2 within target range than the CLiO2 controller (Vyaire, Yorba Linda, California, USA) in a randomised 48 hour crossover study.14 In this trial, infants were studied at a median postnatal age of 19 days at which time the lung disease and response to changes in FiO2 may not be representative of each phase of admission. To date no studies have compared automated oxygen controllers head-to-head over long periods of time. In our centre, we are in the unique position of having used two different automated oxygen controllers as standard of care for a total of 6 years, making a comparison between these two controllers feasible. We used the AVEA ventilators with the CLiO2 automated oxygen control (AOC) algorithm integrated for over three years (since August 2015), after which the ventilators were replaced for SLE6000 ventilators with OxyGenie AOC algorithm in November 2018. In this study we compared the effectiveness of these controllers in very preterm infants receiving AOC as standard of care by either the CLiO2 or the OxyGenie controller