567 - Respiratory Waveform Variability as a Biomarker of Neuromaturation in Preterm Infants

Poster Number: 567
Publication Number: 567.2926

Background: Preterm infants admitted to Neonatal Intensive Care Units require successful oral feeding for discharge, but often struggle with feeding due to immature coordination of swallowing and breathing. The relationship between the quality of breathing patterns and oral feeding success in preterm infants is unknown. A valid and clinically useful measure of respiratory maturity is needed to quantify the relationship between swallowing and breathing in infants to guide clinical practice in neonatal feeding and reduce length of hospital stay.

Objective: We aimed to study an objective criterion "respiratory waveform variability/breathing patterns (RWV)" calculated using Sample Entropy (SE) to quantify the coordination between breathing and swallowing.

Design/Methods: We enrolled 15 preterm infants and measured and quantified RWV as a potential predictor of days to full oral feeding and subsequent discharge from the hospital. The transthoracic impedance signals were obtained by electrocardiogram electrodes placed on the surface of the infants’ chest while respiratory data were extracted from the bedside cardiopulmonary monitor. The RWV was quantified using SE, a nonlinear statistical measure of waveform complexity. Infant demographics, measures of physiologic status, measures of feeding performance, and clinical outcomes were examined and compared between infants during bottle feeding.

Results: We successfully measured transthoracic impedance signals and RWV in all 15 subjects. Figure 1 shows data collected from a set of triplets with comparison of physiologic metrics and RWV with SE. Similar data sets were obtained from all enrolled infants and analyzed. Linear regression analysis was used to determine the association between RWV and the primary outcomes of postmenstrual age (PMA) at full oral feeding (PO all) and days to discharge adjusting for gestational age. For every unit change in SE there was a 0.571 reduction in PO all and a 0.909 reduction in days to discharge. RWV was compared to established measures used in standard of care (e.g., HR, RR, and SpO2) to examine if SE offers greater clinical utility for outcomes of interest. Our findings showed a significant association between RWV as measured by SE and PMA at full oral feeding (p = 0.04) even with a small sample size (n=15).

Conclusion(s): Our findings demonstrated an association between adaptability of RWV, oral feeding performance, and earlier transition to full oral feeding in preparation for discharge. This association suggests that RWV is an indicator of physiologic status and neuromaturation that is necessary for feeding success for preterm infants.