# Autonomic Function and Cerebral Autoregulation in Children Receiving Extracorporeal Life Support

## TL;DR

In children on ECMO, heart rate variability and brain blood flow regulation are weakly linked and each independently predict neurological outcomes.

## Contribution

The study shows that autonomic function and cerebral autoregulation are distinct but both independently predict neurological outcomes in ECMO patients.

## Key findings

- Heart rate variability and cerebral autoregulation showed weak coupling in ECMO-supported children.
- NN skewness and COx were independently associated with neurological outcomes.
- Impaired cerebral autoregulation was present in 16% of patients.

## Abstract

What are the main findings?
In children supported with ECMO, heart rate variability and cerebral autoregulation showed weak coupling.Both NN skewness and COx were independently associated with neurological outcomes without evidence of interaction.

In children supported with ECMO, heart rate variability and cerebral autoregulation showed weak coupling.

Both NN skewness and COx were independently associated with neurological outcomes without evidence of interaction.

What are the implications of the main findings?
Autonomic function and cerebrovascular regulation represent distinct physiologic domains that independently contribute prognostic information regarding neurological outcomes.These findings support the need for multimodal monitoring approaches and larger multicenter studies incorporating high-resolution data to better characterize neurological risk.

Autonomic function and cerebrovascular regulation represent distinct physiologic domains that independently contribute prognostic information regarding neurological outcomes.

These findings support the need for multimodal monitoring approaches and larger multicenter studies incorporating high-resolution data to better characterize neurological risk.

Background/Objectives: Heart rate variability (HRV) and cerebral autoregulation (CAR) reflect physiologic processes that may influence neurological injury in children supported with extracorporeal membrane oxygenation (ECMO). Although abnormalities in both have been associated with adverse neurological outcomes, their physiologic relationship during ECMO remains unclear. Methods: This retrospective single-center study evaluated the association between HRV and CAR during the first 24 h of ECMO support and assessed their independent relationships with neurological outcome. Patients with at least two hours of simultaneous HRV and CAR monitoring within 24 h of ECMO initiation were included. HRV metrics were derived from artifact-free NN intervals across time, frequency, and nonlinear domains, while CAR was quantified using the cerebral oximetry index (COx), with impaired CAR defined as COx > 0.3. Associations between HRV indices and COx were examined using Spearman correlations at hourly and 24 h resolutions. Unfavorable outcome was defined as death or a Pediatric Cerebral Performance Category (PCPC) score ≥3 at discharge with deterioration from baseline. Results: Eighty-nine patients met inclusion criteria, and 16% demonstrated impaired CAR. HRV measures were reduced relative to age-adjusted norms in both CAR groups without significant differences between groups. Correlations between HRV indices and COx were consistently weak. Overall, 50% experienced unfavorable neurological outcomes. In adjusted logistic regression models, NN skewness and COx were independently associated with outcome, although only NN skewness remained significant in interaction analyses. Conclusions: HRV and CAR exhibited limited physiological coupling during early ECMO support, while each measure provided independent prognostic information with respect to neurological outcome.