256 - Cardiorespiratory markers of hemodynamic instability in very low birth weight infants

Publication Number: 4252.256

Nicholas Heitkamp, University of Virginia School of Medicine, Crozet, VA, United States; Douglas Lake, University of Virginia, Charlottesville, VA, United States; Brynne Sullivan, University of Virginia School of Medicine, Charlottesville, VA, United States

Background: The postnatal period represents a critical phase of physiological adaptation for preterm infants. Very low birth weight (VLBW, < 1500 g) infants transition from fetal to neonatal circulation with immature myocardial function, vascular stability, and autonomic control. Many experience hemodynamic instability, including hypotensive shock requiring vasoactive support. Continuous cardiorespiratory monitoring data have identified illness signatures in VLBW infants related to sepsis and later adverse outcomes, but markers of unstable hemodynamic transition remain poorly characterized.

Objective: Characterize differences in heart rate (HR) and oxygen saturation (SpO2) features among VLBW infants with and without shock requiring treatment with fluids and/or vasoactives

Design/Methods: We conducted a single-center, retrospective cohort analysis of VLBW infants admitted between 2018 and 2023 within 72 hours of birth. Continuous HR and SpO2 data were routinely archived, sampled at 0.5 Hz from the bedside monitor. Infants were stratified as those with and without hypotension treated with vasoactive inotropes in the first 72 hours after birth. HR and SpO2 data were summarized hourly as mean, standard deviation, skewness, kurtosis, and cross-correlation. Summary metrics were retrieved using a previously developed data sharing and visualization platform, and data on demographics and shock were extracted from a clinical database.

Results: We analyzed 580 infants with median gestational age (GA) 28.0 weeks and birth weight 997 g (Table 1). Among 147 (25%) infants with shock, mean GA and BW were lower, and mortality was higher. Infants with shock had higher mean HR and lower HR standard deviation (HR variability) in the 72 hours after birth (Figure 1). Infants with shock also had a lower mean SpO₂ and higher SpO₂ skewness (indicating more desaturations) (Figure 2).

Conclusion(s): Continuous HR and SpO2 features differed among VLBW infants with and without shock in the transitional period, providing insight into early hemodynamic adaptation of premature infants. Those who developed shock demonstrated higher average HR, reduced HR variability, and lower SpO₂ during the first 72 hours after birth. Next steps include a time-based analysis of HR and SpO₂ features before, during, and after instability episodes, as well as analyzing the associations between HR-SpO₂ features and outcomes.

Table 1
Demographic information for VLBW infants ( <32 weeks GA) at UVA, 2018-2023 (n=580).

Figure 1
Heart rate differences in the cohort comparing those who received shock and those who did not; the red line represents infants without shock, and the blue line represents infants with shock. Each panel shows hourly group averages over time since birth for (A) mean heart rate, (B) heart rate kurtosis, (C) heart rate skewness, and (D) heart rate standard deviation.

Figure 2
SpO2 differences in the cohort comparing those who received shock and those who did not; the red line represents infants without shock, and the blue line represents infants with shock. Each panel displays hourly group averages over time since birth for (A) mean SpO2, (B) SpO2 kurtosis, (C) SpO2 skewness, and (D) SpO2 standard deviation.