WIP 99 - Episodes of Intermittent Hypoxia with Oro/Nasogastric Feeds in Preterm Infants and the Effect of Continuous Positive Airway Pressure (CPAP)

Poster Number: WIP 99
Publication Number: WIP 99.2359

Background: Intermittent hypoxemia (IH) in preterm infants are frequent and associated with long-term respiratory and neurological morbidities, retinopathy of prematurity, and poor growth. Anecdotally, IH are associated with oro/nasogastric (OG) feeds but this has not been quantified. Nasal continuous positive airway pressure (nCPAP) is now the standard of care for preterm infants, yet its association with IH after OG feeds is unknown. Gastric distention from OG feeds may alter the pressure relationship of abdominal and thoracic compartments leading to IH, which may be exaggerated in preterm infants due to a compliant chest wall. CPAP promotes lung recruitment and maintains functional residual capacity, improving oxygenation.

Objective: Characterize the association of OG feeds and episodes of IH in preterm infants and evaluate the effect of nCPAP on the occurrence of IH after OG feeds.

Design/Methods: Prospective substudy in preterm infants enrolled in the NHLBI supported trial: NCT04295564. Stable preterm infants born at ≤32 weeks’ gestation who meet specific respiratory stability criteria are randomized to two additional weeks of CPAP (eCPAP) versus CPAP discontinuation (dCPAP) to room air. Continuous pulse oximeter data with a 2-second sample rate is collected during these two weeks that allows granular quantification of IH. All feed intervals in this two-week period will be analyzed for the last 30 patients in the parent RCT. The primary outcome will be percentage of time with oxygen saturation < 90%. The three comparisons of interest will be: pre-feed versus combined during and after feed in the dCPAP group, dCPAP versus eCPAP groups during the feed, and dCPAP versus eCPAP in the hour post-feed. Fractional regression with clustering by patient will be used to account for the distribution of the data and the repeated measures. The secondary outcome of percentage of time with oxygen saturation < 80% and number of IH episodes to < 90% and < 80% will be examined using negative binomial models. Data analysis will be complete by December 15, 2023.