The Journal of ExtraCorporeal Technology No 56-4 | Page 44

182 A . G . Beshish et al .: J Extra Corpor Technol 2024 , 56 , 174 – 184

Table 6 . Outcomes of neonates requiring extracorporeal life support post-Norwood operation using univariable and multivariable regression analysis .

a Adjusted for source of pulmonary blood flow , CPB time , and post-Norwood VIS-score at 48 h . b Third quartile of postoperative length of stay ( 55 ) was applied as cutoff .

Abbreviations : PaO 2 : Partial Pressure of Oxygen ; AKI : Acute Kidney Injury ; PPLOS : Prolonged Postoperative Length of Stay .

higher serum lactate prior to ECLS initiation . It is difficult to account for these differences , however they support the benefits of a prospective and controlled study .

Because there is no generally accepted definition of hyperoxia in this population , we used an ROC curve analysis to determine which PaO 2 values may be associated with an adverse outcome . A similar strategy was employed in previous reports . Sznycer-Taub et al . evaluated hyperoxia in infant cardiac patients supported on VA-ECMO and found that a PaO 2 of 193 Torr in the first 48 h was determined to have an association with 30-day mortality [ 1 ]. Using a similar strategy , Beshish showed that a PaO 2 of 313 Torr for infants undergoing cardiac surgery utilizing CPB was associated with 30-day mortality [ 14 ]. Our cut-off definition of hyperoxia was very close to that identified by Sznycer-Taub although the patient population was slightly different , as the former captured all infants on postoperative VA-ECMO . Our study supports the findings of this prior report as we evaluated a more homogenous and larger patient population who required ECLS following Norwood operation ( n = 65 ) using a similar hyperoxia definition .

Figure 3 . Scatterplot illustrating the association of average PaO 2 , ECLS duration , and mortality in neonates requiring ECLS support post-Norwood operation .

risk of morbidity and mortality [ 37 – 42 ]. In patients who have experienced cardiac arrest or resuscitation aftershock , an increase in ROS may deplete plasma antioxidant potential which may lower the threshold for subsequent oxidative injury [ 32 ]. This effect may be more pronounced in neonates and infants as they are known to have immature antioxidant defenses and thus may be more susceptible to ROS [ 1 ]. The effect of hyperoxia may be further pronounced in patients with cyanotic heart diseaseastheyhavesignificantly lower PaO 2 at baseline . It is not known if the antioxidant systems of the body are downregulated in patients with lower baseline PaO 2 . Although unknown , it is plausible that these patients are more vulnerable to supraphysiologic oxygen .

Interestingly we found that the hyperoxia cohort did not have longer times from the OR / CICU arrival to initiation of ECLS , and 4 h after initiation on ECLS , they had a higher median rate of flow ( Table 2 ). This suggests that the hyperoxia cohort did not have a delay in support or inadequate ECLS 0support . However , the hyperoxia cohort conversely had a

Limitations

Our findings are subject to all limitations inherent to singlecenter retrospective cohort studies . Although PaO 2 levels were obtained at dedicated time intervals , it is not possible to discern the effect of time spent in a hyperoxia state as opposed to the effects of acutely high PaO 2 levels . Additionally , there may be some bias as to which patients are exposed to hyperoxia . Our center does not have a standard protocol dictating goal PaO 2 levels , however , this study builds support for such a practice . The majority of our cohort had a PaO 2 level over 200 mmHg while on ECLS limiting our ability to study the relationship between lower oxygen tension levels and outcomes . This data may not be generalizable to centers that target lower PaO 2 values . The retrospective nature of this study prevented our description of other surrogate markers of cardiac output such as ECLS flows , pulsatility , or Qp : Qs . There may be concerns about generalizability , as our incidence of ECLS ( 24 %) was higher than the 8 – 20 % in other reports [ 4 , 5 , 7 – 9 ]. This may be due to a lower threshold for ECLS at our center , higher complexity of patients , or variations in surgical strategy . Notably , despite the higher rate of ECLS , Norwood operation survival in our center ( 86 %) is similar to STS benchmark public outcome reporting ( 88 %) [ 43 , 44 ]. It is important to note that some