Background: Neonatal sepsis is the leading cause of mortality in preterm newborns. The autonomic nervous system modulates the response to sepsis through the cholinergic anti-inflammatory reflex. However, premature neonates exhibit immaturity of the autonomic nervous system, which could increase the risk of sepsis. Kangaroo Care (skin-to-skin contact) may promote autonomic nervous system modulation and maturation in preterm newborns with sepsis. The objective of this study is to determine the effect of Kangaroo Care on heart rate variability in preterm newborns with late-onset clinical sepsis.

Methods: A cross-over randomized clinical trial will be conducted, including 20 preterm infants with late-onset sepsis. The autonomic nervous system will be assessed using heart rate variability analysis. The study interventions consist of routine care in an incubator and Kangaroo Care. Randomization will be performed using a four-block permuted design for the two intervention sequences AB: Kangaroo Care - incubator care, or BA: incubator care - Kangaroo Care. Heart rate variability will be recorded using a Polar Rs800 monitor and analyzed with Kubios software.

Discussion: This study will provide information on the relationship between Kangaroo Care and autonomic nervous system activity in preterm neonates with late-onset sepsis. These data will contribute to the understanding of the cholinergic anti-inflammatory reflex in neonates and the capacity of skin-to-skin contact to modulate autonomic activity in neonatal infection. Thus, the study seeks to provide initial evidence for the use of skin-to-skin contact as a non-pharmacological therapeutic intervention in neonatal sepsis.

• The autonomic nervous system (ANS) plays a fundamental role in modulating the immune inflammatory response to infection.
• Premature neonates have immaturity in the development of the ANS at birth, which may lead to altered responses or inadequate modulation of the immune inflammatory response to infection.
• Kangaroo Care modulates and matures the ANS, especially the parasympathetic branch.
• Skin-to-skin contact could be used as a non-pharmacological intervention in sepsis.

Objective To determine the effect of Kangaroo Care (skin-to-skin contact) on heart rate variability in preterm newborns with late-onset clinical sepsis.

Specific Objectives

• Describe the observed changes in heart rate variability during Kangaroo Care (skin-to-skin contact) and incubator care.
• Describe changes in the stability of the cardiorespiratory system, using the Stability of the Cardiorespiratory System in Premature Infant (SCRIP) score, during Kangaroo Care and incubator care.
• Compare the effect of skin-to-skin contact (Kangaroo Care) and incubator care on HRV metrics in preterm neonates with late-onset clinical sepsis.

Study Design A cross-over randomized clinical trial will be conducted.

Participants The study will include 20 preterm newborns hospitalized in the participating neonatal units, diagnosed with late-onset sepsis. The inclusion criteria are preterm newborns with Ballard gestational age less than 37 weeks, older than 72 hours, hospitalized since birth in the neonatal unit, diagnosed with late-onset clinical sepsis, and weighing between ≥1200 and ≤2500 g. Colombian Association of Neonatology (ASCON) criteria will be used to define late-onset clinical sepsis 24. Neonates with cardiovascular or neurological congenital anomalies, respiratory and/or hemodynamic clinical instability, including the need for invasive mechanical ventilation, respiratory lability, hypotension, and/or the need for vasopressor medications, will be excluded.

Given the exploratory nature of the clinical trial, a control group will be included to perform HRV analysis during incubator care and Kangaroo Care. This control group will consist of 20 preterm neonates with the same characteristics as the late-onset sepsis cohorts (Ballard <37 weeks, >72 hours of life, weight between ≥1200 and ≤2500 g, hospitalized since birth in the neonatal unit), but without a diagnosis of sepsis and in the nutritional recovery phase with all chronic medical conditions stabilized and controlled.

Definitions

Late-onset clinical sepsis. ASCON criteria 24:

• Clinical manifestations of systemic inflammatory response (see below).
• Age greater than 72 hours.
• Presence of risk factors: prematurity, weight less than 2500 g, prolonged use for more than 7 days of mechanical ventilation and/or central venous catheter and/or parenteral nutrition, permanent nasogastric tube, prolonged hospitalization, use of steroids and/or H2 blockers, and prior exposure to broad-spectrum antibiotics.
• Laboratory abnormalities: leukopenia, leukocytosis, immature/total ratio greater than 0.2, thrombocytopenia; positive C-reactive protein (CRP) value and/or identification of microorganisms by Polymerase Chain Reaction.

Outcomes

Primary outcome: Change in the mean HRV analysis indices between the two interventions.

• Secondary outcomes:

o Change in the Stability of the Cardiorespiratory System in Premature Infant (SCRIP) score with each intervention.

Study Procedures Once the newborn's enrollment in the study is confirmed, procedures for HRV recording will be initiated, following the procedure below. Randomization of the newborn to the intervention sequence will be performed accordingly: sequence AB as Kangaroo Care - incubator care, or sequence BA as incubator care - Kangaroo Care.

The HRV signal will be obtained using a Polar H10 chest strap connected via Bluetooth to the Polar RS800 monitor. In the RR function, the accuracy is less than 1 millisecond per beat. The chest strap will be placed on the anterior region of the newborn's chest (Figure 1). Before the start of data recording, a stabilization period of 30 minutes in the assigned intervention sequence will be allowed. After this period, HRV recording will begin using the Polar monitor for 60 minutes for each intervention. Before changing the intervention sequence, a washout period of three hours will be allowed

Data from the Polar monitor will be downloaded and stored on a computer using Polar Protrainer 5 software. The data will then be exported to an Excel file and further analyzed using Kubios software (Heart Rate Variability (HRV) Software) for detailed HRV analysis. The software will provide metrics in the time domain: average NN interval, standard deviation of NN intervals (SDNN), and root mean square of successive differences of RR intervals (RMSSD). Metrics in the frequency domain will include total power (TP), very low frequency (VLF), low frequency (LF), high frequency (HF), and LF/HF ratio. Non-linear analysis will involve SD1 (Poincaré plot) and SD2 26.

In addition to HRV analysis, clinical characteristics of the newborns will be collected for further analysis and characterization.

Sample Size No previous studies have compared HRV analysis between Kangaroo Care and incubator care. Information from the literature comparing HRV in preterm neonates with sepsis versus no sepsis 27and comparing preterm neonates in an incubator versus Kangaroo Care 23 was used to estimate the sample size. The main HRV metric to assess parasympathetic activity in preterm neonates is HF (high frequency) and RMSSD, and the sympathetic/parasympathetic modulation is represented by the LF/HF ratio. The sample size was estimated considering the highest reported values for these metrics. Based on this data, with an expected mean difference in LF of 5.3, standard deviation in group 1 of 4.9, and standard deviation in group 2 of 6.5, a confidence level of 95%, and a power of 80%, a sample size of 20 pairs of preterm newborns is expected. No loss to follow-up is anticipated.

Inclusion Criteria:

• Newborns with Ballard gestational age less than 37 weeks
• Older than 72 hours
• Hospitalized since birth in the neonatal unit
• Diagnosed with late-onset clinical sepsis
• Weighing between ≥1200 and ≤2500 g.
• Colombian Association of Neonatology (ASCON) criteria will be used to define late-onset clinical sepsis

Exclusion Criteria:

• Neonates with cardiovascular or neurological congenital anomalies,
• Respiratory and/or hemodynamic clinical instability, including the need for invasive mechanical ventilation, respiratory lability, hypotension, and/or the need for vasopressor medications.

Masking will not be applied due to the nature of the interventions. However, blinding of the intervention sequence will be maintained for the researchers responsible for data analysis.