PURPOSE: Oral feeding can be physiologically stressful for the preterm infant. Understanding the physiological changes that occur during feeding is essential for developing supportive interventions. Heart rate variability (HRV) is a sensitive indicator of distress and may be an important variable during feeding. The purpose of this preliminary case study was to explore HRV as a marker of physiological stability during feeding in the preterm infant by describing the relationship between HRV and oxygen saturation (SaO2) before, during, and after two physiologically different feedings. Feeding 1 was designated high stress secondary to greater swallowing dysregulation (58.6 vs 19.3% in Feeding 2), breathing dysregulation (25.3 vs 18.8%), and hypoxemia (86% of feeding with SaO2 < 90% vs 5%).
SUBJECT: Former 31-week male infant at 37 weeks post-menstrual age and 10 days of oral feeding experience.
DESIGN: Secondary data analysis of a within-subjects crossover design study (K01 NR007668).
METHODS: Oxygen saturation and HRV data were analyzed for approximately 5 minutes before, 6 minutes during, and 5 minutes postfeeding.
MAIN OUTCOME MEASURES: SaO2 and time domain analysis HRV parameters, including standard deviation of normal to normal intervals (SDNN) and square root of the mean squared differences of successive normal to normal intervals (RMSSD), at baseline, during feeding, and postfeeding. SDNN estimates overall HRV and is correlated with parasympathetic tone. RMSSD estimates short-term components of HRV and is correlated with physiologic stress.
PRINCIPAL RESULTS: Mean SaO2 was similar at baseline (94.1 vs 90.4%), lower during Feeding 1 (84 vs 94.7%), and recovered to near-baseline postfeeding (93.6 vs 94.7%). SDNN was similar at baseline (33.2 vs 28.8 ms); increased to 59.3 ms during Feeding 1, while it decreased to 21.3 ms in Feeding 2; then returned to near-baseline postfeeding (23.0 vs 26.7 ms). In this case, elevated SDNN during Feeding 1 was reflective of a near-bradycardic event, not elevated parasympathetic tone. For Feeding 2, SDNN maintained the expected relationship with SaO2. RMSSD was similar at baseline (7.9 vs 7.3 ms); increased to 22.3 ms during Feeding 1, while it remained similar to baseline (7.4 ms) in Feeding 2; then returned to near-baseline postfeeding (9.3 vs 7.6 ms). The expected relationship between RMSSD and SaO2 was maintained, with the highest amount of physiologic work during Feeding 1.
CONCLUSIONS: HRV correlates with changes in oxygenation, but may be complicated by bradycardic events. Further investigation of HRV during feeding is warranted.