Hylke Salverda

140 Chapter 8 SpO2 during OxyGenie control (93% vs 94%) on the steeper part of the oxygenhaemoglobin dissociation curve. We hypothesized that these differences in achieved performance may affect short-term clinical outcome. Both systems have been used as standard of care for over a year in our unit, which prompted us to describe shortterm clinical outcomes of preterm infants in a matched cohort study. Methods Study design We performed a propensity score-matched observational study with electronic patient record data from the LUMC, a tertiary hospital with 25 Neonatal Intensive Care Unit (NICU) beds and around 100 admissions per annum of infants born under 30 weeks of gestation. The medical ethical research committee of Leiden Den Haag Delft provided a statement of no objection for obtaining and publishing the anonymised data. Protocol was filed under reference number G19.075. To ensure equal distribution of patient characteristics that can confound with outcome we matched infants that received respiratory support from the SLE6000 ventilator, born in-hospital between November 1st 2018 and March 15th 2020 (OxyGenie cohort), to infants from a cohort supported by the AVEA ventilator born between 18th of October 2015 and the 31st of April 2018 (CLiO 2 cohort). We matched on sex, birthweight, and gestational age using propensity score matching with a match tolerance of 0.1. All infants were born at a gestational age 24+0 until 29+6 weeks. Infants with major congenital malformations were excluded. Automated oxygen controllers In August 2015 automated oxygen titration by the CLiO2 algorithmwas introduced as standard of care. The CLiO2 algorithm was set to target an SpO2 TR of 90%-95%. In November 2018 the AVEA ventilators were replaced with SLE6000 ventilators with the OxyGenie option for automated oxygen control. The OxyGenie uses a narrower SpO2 TR of 91%-95%. The CLiO2 algorithm is of a rule-based design. To determine a fraction of inspiratory oxygen (FiO2) adjustment CLiO2 incorporates the difference between the SpO2 TR and the measured SpO2, the severity of lung disease and the SpO2 trend. The OxyGenie algorithm is an adaptive proportional-integral-derivative (PID) controller – also employed in automotive cruise-control and rockets – with several enhancements to account for the physiology of a neonate and account for the limitations of pulse oximetry.