O060 Loop gain as a measure of respiratory instability in premature infants

INTRODUCTION: Respiratory control is often unstable in premature infants, with periodic breathing, and intermittent hypo- and hyper-oxemia. Understanding how respiratory control changes after premature birth may lead to fewer and less severe events. The aim of this study was to quantify respiratory pattern instability in preterm infants during their first weeks of life. METHODS: As part of a larger study (PreVent, NHLBI U01 HL133708 & HL133700. Dennery et al., Pediatr Res 2019:85(6),769), 55 infants (born 24w1d-28w6d weeks post-menstrual age, PMA) were studied at one or more timepoints (28, 32, 36, 40 and 52 weeks PMA). Respiratory effort (thorax and abdomen bands) and SpO2% were recorded. Airflow was derived from effort bands. Respiratory instability was quantified by a previously described loop gain (LG) model fit to visually identified artefact-free spontaneous sigh breaths during quiet sleep (Edwards et al., SLEEP, 2018:41(11)). RESULTS: In N=1042 sighs, LG varied considerably within and between individuals across the timepoints studied. However, mixed effects modelling revealed a significant trend for increasing LG at each timepoint (R2=.2, p<0.0001, PMA fixed effect, random intercept per infant). DISCUSSION: On average, loop gain increased across the first ~12 weeks of life in very preterm infants. This pattern is dissimilar to that among term infants, where LG peaks between 2 and 4 postnatal weeks. In very preterm infants, understanding the PMA-specific ventilatory control instability and its typical developmental trajectory will better elucidate the challenges in maintaining SpO2% within target range through better description of expected need for respiratory support.