# Clinical Characteristics and Outcomes for Neonates with Respiratory Failure Referred for Extracorporeal Membrane Oxygenator (ECMO) Support

## TL;DR

This study examines neonates with respiratory failure referred for ECMO support, identifying key clinical indicators that predict the need for this life-saving treatment.

## Contribution

The study identifies oxygenation indices and pulmonary hypertension as independent predictors of ECMO support in neonates.

## Key findings

- 96 out of 147 neonates required ECMO support.
- Oxygen saturation index (OSI) ≥ 10 predicted ECMO need with 96.8% sensitivity.
- OSI and pulmonary hypertension were independent predictors of ECMO support.

## Abstract

Objective: The aim of this study was to describe the presenting characteristics and outcomes of neonates with respiratory failure referred for extracorporeal membrane oxygenation (ECMO) support, compare those who received ECMO support (ECMO group) to those who did not (non-ECMO group), and evaluate the predictive variables requiring ECMO support. Methods: All neonates (<15 days) with respiratory failure (without congenital diaphragmatic hernia or congenital heart disease) referred to our regional ECMO center from 2014 to 2023 were included in this retrospective study. Patient demographics, birth history, and clinical and outcome variables were analyzed. Oxygenation indices and vasoactive–inotropic scores obtained at PICU arrival and four hours after arrival were compared between the two groups using ROC analysis, with ECMO initiation as an outcome variable. Youden’s index was used for optimal threshold values. Chi-square, Mann–Whitney U, and binary logistic regression were used for comparative analyses. Results: Out of the 147 neonates, 96 (65%) required ECMO support. The two groups significantly differed in the prevalence of pulmonary hypertension (pHTN; systemic or suprasystemic pulmonary pressures), lactate level, and oxygenation indices. Mortality was not different between the two groups. Presence of oxygen saturation index (OSI) ≥ 10 had a sensitivity 96.8% in predicting the need for ECMO support. On regression analysis, OSI and pHTN were independent predictors of ECMO support. Conclusions: Oxygenation indices and echo findings predict the need for ECMO support in neonatal hypoxemic respiratory failure. These findings help non-ECMO centers make appropriate and timely transfers of neonates with respiratory failure to ECMO centers.