A . G . Beshish et al .: J Extra Corpor Technol 2024 , 56 , 174 – 184 175
Introduction
The use of extracorporeal life support ( ECLS ) following cardiac surgery is utilized in 3 – 5 % of all neonatal and pediatric cardiac surgeries with a 43 – 52 % survival [ 1 , 2 ]. Neonates with univentricular physiology who undergo a Norwood operation represent an especially high-risk population [ 3 – 6 ]. The tenuous nature of their postoperative physiology requires significant cardiopulmonary support and at times ECLS to maintain adequate hemodynamics and oxygen delivery . Previous reports indicate that 8 – 22 % of neonates will require ECLS post- Norwood operation [ 4 , 5 , 7 – 9 ], with an overall survival rate of 30 – 60 % [ 4 , 5 , 7 – 9 ].
When ECLS is used postoperatively , patients are supported with venoarterial-ECLS . These ECLS circuits utilize an efficient oxygenator resulting in high partial pressure of oxygen ( PaO 2 ) that can exceed 400 mmHg . Exposure to supranormal levels of oxygen is termed hyperoxia . Hyperoxia has been well studied in different clinical situations , including after resuscitation from cardiac arrest , perinatal asphyxia , myocardial infarctions , traumatic brain injury , and following cardiopulmonary bypass ( CPB ). Studies in both adults and children have demonstrated an association with increased morbidity and mortality [ 1 , 10 – 18 ]. Furthermore , minimizing PaO 2 while on CPB in cyanotic patients with complete mixing congenital heart lesions has been shown to result in less end-organ damage , inflammation , and oxidative stress when compared to those exposed to higher oxygen concentrations [ 1 , 19 , 20 ]. Although the negative effect of hyperoxia and its association with adverse outcomes is known , adequate oxygen delivery is necessary and the level at which PaO 2 may become deleterious may differ depending on the clinical situation , duration of exposure , patient ’ s age , underlying pathophysiology , and disease process [ 1 , 10 , 14 , 21 – 23 ].
Given the lack of a clear definition of hyperoxia , we aimed to evaluate a high-risk homogenous patient population , specifically neonates with univentricular physiology who underwent Norwood operation and required ECLS in the postoperative period . Our primary aim was to determine if hyperoxia while on ECLS is associated with mortality using a derived cut-point within our cohort . Our secondary aim was to determine if hyperoxia during ECLS was associated with morbidity including Functional Status Scale ( FSS ), acute kidney injury ( AKI ), and prolonged postoperative length of stay ( PPLOS ).
Materials and methods
This is a single-center retrospective cohort study including all neonates with univentricular physiology who underwent a Norwood operation and required ECLS in the postoperative period between January 1st , 2010 , and December 31st , 2020 , at Children ’ s Healthcare of Atlanta ( CHOA ), a large quaternary children ’ s hospital . An internal surgical and ECLS database / registry were queried , and eligible patients were identified . Procedures were followed in accordance with the ethical standards of the CHOA Institutional Review Board ( IRB # 00001119 , approval 07 / 19 / 2021 ), and the Helsinki Declaration of 1975 . Informed consent was waived .
Data and definitions
All consecutive patients who underwent Norwood operation and required ECLS in the index postoperative ICU hospitalization were included . Patients who failed to separate from CPB and initiated ECLS in the operating room were excluded ( n = 6 ). Also , patients who underwent a hybrid procedure prior to undergoing a Norwood ’ operation were excluded from our study analysis ( n = 3 ). These 2 exclusion criteria were chosen to make the patient population in the study as homogenous as possible . None of our patients underwent a concomitant AVV repair at the time of Norwood ’ s operation . Demographic features and clinical characteristics , including chromosomal abnormalities , genetic syndrome , primary cardiac diagnosis , type of systemic ventricle , source of pulmonary blood flow [ modified Blalock-Taussig-Thomas ( m-BTT ) shunt vs . Sano shunt / right ventricle to pulmonary artery ( RV-PA ) conduit ], preoperative respiratory support , creatinine , preoperative and intraoperative echocardiogram [ assessing atrioventricular valve regurgitation ( AVVR ), systemic ventricular function , and ascending aorta diameter ], preoperative and postoperative vasoactive inotropic scores ( VIS ) at dedicated postoperative times points ; operative variables , including CPB , aortic crossclamp ( XC ), and circulatory arrest times ; and ECLS variables ( cannulation site , ECLS duration , and ECLS complications ) were collected . All arterial blood gases ( ABGs ) were obtained from the patient ’ s right radial arterial line ( institutional standard of care includes placement of a right radial arterial line to monitor hemodynamics and arterial blood gases ) during and after the Norwood operation . ABG data was obtained prior to ECLS initiation , and during the first 48 hours while on ECLS .
Outcomes
The primary outcome was all-cause ECLS mortality . The secondary outcomes included FSS ( preoperative compared to discharge ), AKI ( Stage II or III , as defined by the KDIGO scoring criteria ) [ 24 ], and PPLOS , defined a-priori as the fourth quartile of the postoperative length of stay ( 55 days ). Length of stay may be biased by mortality , because those who die may have a shorter length of stay , which would erroneously appear as a good outcome . Two methods were used to control for this bias . In standard logistic regression , the length of stay for mortality was defined as prolonged , regardless of duration . In a second analysis , a composite rank-based outcome was created for days alive ICU-free ( AIF ). In this method , the AIF composite endpoint equals the number of ICU-free days over the first 28 postoperative days . Those with mortality were assigned a score of �1 [ 25 ]. Functional Status Scale ( FSS )
The FSS consists of six main domains : mental status , sensory , communications , motor function , feeding , and respiratory . Functional status for each domain was categorized from normal ( 1 ) to very severe dysfunction ( 5 ), with total FSS scores ranging from 6 to 30 [ 26 ]. Functional status scoring for this study involved retrospectively scoring baseline status ( i . e ., on