632 - Neurobehavioral stress linked to the central nervous system is associated with heart rate variability differences in preterm infants

Publication Number: 4618.632

Elan L.. Donnellan, Children's National Health System, Washington, DC, United States; Julius Ngwa, Children's National Health System, Washington, DC, United States; Nickie Andescavage, Children's National Health System, Washington, DC, United States; Melissa Liggett, Children's National Health System, Washington, DC, United States; Catherine Limperopoulos, Children's National Health System, Washington DC, DC, United States; Adre J. du Plessis, Children's National Health System, Bethesda, MD, United States; Rathinaswamy Govindan, Children's National Hospital, Washington, DC, United States

Background: The Neonatal Intensive Care Unit (NICU) Network Neurobehavioral Scale (NNNS) measures neurologic integrity and neurobehavior, including signs of stress. The stress scale is divided into 7 categories, which include physiological, skin, visual, gastrointestinal, state, the autonomic (ANS) and the central nervous system (CNS). Though the dynamics of the ANS and CNS are instrumental to autonomic regulatory capacity, the relationship between observed and elicited measures of stress and objective measures of regulatory capacity, such as heart rate variability (HRV), are not well studied.

Objective: The aim of the present work is to determine the relationship between HRV and observed signs of stress as measured by the NNNS stress scale as well as its associated ANS and CNS subscales.

Design/Methods: Using continuous beat-to-beat heart rate data, we derived the following HRV measures: short (α1), and long (α2) fractal scaling exponents and short (RMS1) and long-term (RMS2) root mean square fluctuations at 33 weeks postmenstrual age (PMA). Infants were assessed using the NNNS version 1 or 2 at term equivalent age. The total stress score was the sum of the 45 items retained across both versions of the NNNS, spanning all 7 categories. Linear regression models with gestational age at birth and postmenstrual age during the NNNS included as covariates were used to test associations between HRV and neurobehavioral stress signs. In the primary analysis, associations were tested between the total number of stress signs and each HRV metric. The secondary analyses tested associations between alpha2 and the ANS and CNS subscales. Lastly, to determine if infants that display particular stress signs have higher alpha2, items from subscales that were significantly associated with alpha2 and that had more than 10 affirmative responses tested for an association.

Results: Linear regression revealed that higher total stress signs were associated with increased alpha2 (p = .006), indicating reduced autonomic complexity. CNS-specific stress signs, particularly high frequency tremors (p = .018), were significantly associated with elevated alpha2 (p = .032). Neither ANS-specific stress signs (p = .198), nor startles were not associated with alpha2 (p = .096).

Conclusion(s): These findings suggest that observed measures of stress on the NNNS were associated with autonomic immaturity indicated by higher alpha2. Furthermore, systemic stress signs from the NNNS ANS subscale are not a reflection autonomic regulatory capacity.

Associations between the total number of stress signs and HRV
^ p < .10, * p < .05, ** p < .01. Covariates include GA at birth and PMA during the NNNS

Associations between ANS & CNS stress signs and alpha2
^ p < .10, * p < .05, ** p < .01

Associations between CNS items and alpha2
^ p < .10, * p < .05, ** p < .01. Covariates include GA at birth and PMA during the NNNS