130 A. G. Beshish et al.: J Extra Corpor Technol 2025, 57, 129 – 136 to the body contains high partial pressure of oxygen( PaO 2) that can exceed 400 – 500 mmHg. When cells are exposed to supraphysiologic levels of oxygen in the bloodstream, the term hyperoxia is used. Hyperoxia has been well studied in various clinical scenarios, including after resuscitation from cardiac arrest( CA), perinatal asphyxia, myocardial infarction, traumatic brain injury, and following cardiopulmonary bypass( CPB). Several studies in both adults and children have demonstrated an association between hyperoxia exposure and increased morbidity and mortality [ 2, 4 – 14 ]. Although the association of hyperoxia with adverse outcomes has been shown previously, the level at which PaO 2 becomes deleterious may differ depending on the clinical situation, including the duration of exposure, the patient’ s age, underlying physiology, and disease process [ 4, 5, 9, 13 – 17 ].
Given the lack of a clear definition of hyperoxia, we sought to evaluate hyperoxia exposure in a high-risk patient population who required ECLS for cardiopulmonary failure. Our primary aim was to determine if hyperoxia while on ECLS was associated with mortality using a derived cut-point within our cohort. Our secondary aim was to determine if hyperoxia during ECLS was associated with greater odds of new morbidity including Functional Status Scale( FSS), and the development of complications while on ECLS, such as acute kidney injury( AKI).
Materials and methods
This is a single-center retrospective cohort study in a highvolume ECLS center. All patients who required VA-ECLS due to cardiopulmonary failure between January 1, 2014, and December 31, 2019, at Children’ s Healthcare of Atlanta( CHOA), a free-standing, university-affiliated quaternary children’ s hospital. An internal ECLS database was queried, and eligible patient encounters were identified. The study was approved by the CHOA Institutional Review Board( IRB # 00001239, approval date: 10 / 11 / 2022). Informed consent was waived.
Data and definitions
All consecutive patients who required VA-ECLS support in index hospitalization were included. Demographic features, clinical characteristics, and ECLS variables were collected. All arterial blood gases were obtained from the patient’ s arterial line during the first 48 h while on ECLS. The primary outcome was defined as all-cause ECLS mortality. The secondary outcome variables included FSS, AKI( Stage II or Stage III, as defined by the KDIGO criteria) [ 18 ] and major complications. Major complications were defined as the presence of either cardiovascular, renal or mechanical complications.
Functional Status Scale( FSS)
The FSS consists of six main domains: mental status, sensory, communication, motor function, feeding, and respiratory. Functional status for each domain was categorized from a normal score of 1 to very severe dysfunction with a score of
5, giving total FSS scores ranging from 6 to 30 as previously described [ 19 ]. Functional status scoring for this study involved retrospectively scoring baseline status( prior to admission functional status according to caregivers) and again at hospital discharge by examining the detailed history and physical exam performed by the primary medical team. FSS score determination was blinded from hyperoxia status. Newborns who had never achieved a stable baseline FSS were assigned a score of 6. This was operationalized by assigning a baseline FSS score of 6 to all admissions for infants 0 – 2 days old and to transfers from another facility for infants 3 – 6 days old, as previouslyreported [ 20 – 23 ]. New morbidity was defined as an increase in total FSS score of 3 points, and unfavorable functional outcomes were defined as an increase of 5 [ 24 ].
Clinical management
All circuits were blood primed before the start of ECLS with packed red blood cells, 25 % albumin, sodium bicarbonate, calcium gluconate, and heparin for patients < 40 kg. It is common practice for ABGs to be obtained at the discretion of the clinical team, most typically 30 min after initial ECLS-cannulation, and then hourly for the first 3 h. Subsequently, they are typically obtained every 3 – 6 h and 30 min after an adjustment in ECLS support. Target gas exchange parameters are not dictated by protocol at our center. Goal PaO 2 ranges have not established as normal and the variation we describe is derived from measurements occurring during clinical care. Goal PaCO 2 was 35 – 45 mmHg, and goal pH was 7.35 – 7.45. Once patients are placed on ECLS, the ventilator is placed on“ rest settings” of the following: ventilator mode pressure control, peak inspiratory pressure 20 cm H 2 O, peak end-expiratory pressure 10 cm H 2 O, respiratory rate 20 / minute, inspiratory time 1 second, and FiO 2 30 %.
Statistical analysis
Statistical analysis was conducted using SAS version 9.0 software, with a significance level set at p < 0.05. The diagnostic utility of mean PaO 2 in predicting mortality was evaluated using Youden’ s index( J = sensitivity + specificity � 1) and receiver operating characteristic( ROC) curves. The study population was stratified into hyperoxia and non-hyperoxia groups based on the optimal cut-off value for mean PaO 2, determined by maximizing the J-value. Fisher’ s exact test was employed for comparing categorical variables, while Student’ s t-test and the nonparametric Wilcoxon rank-sum test were used for continuous variables, as appropriate. Additionally, a scatterplot was generated to examine the relationship between mean PaO 2, duration of ECLS run, and survival, with Spearman’ s correlation coefficient reported. To assess the impact of hyperoxia on mortality and AKI, univariable and multivariable logistic regression analyses were performed, adjusting for BSA, age group, and indication for ECLS in the multivariable analysis, which were determined a priori. The results are presented as odds ratios( OR) with corresponding 95 % confidence intervals( CI).