Oxygen targets following immediate percutaneous coronary intervention in comatose survivors of out-of-hospital cardiac arrest - a BOX sub-study

Abstract

Abstract Funding Acknowledgements Type of funding sources: Foundation. Main funding source(s): The Novo Nordisk Foundation The Lundbeck Foundation. Background Treatment with supplemental oxygen in patients with acute myocardial infarction (AMI) has previously been associated with poorer outcomes. However, in patients with AMI resuscitated from out-of-hospital cardiac arrest (OHCA) and undergoing immediate revascularization, evidence regarding specific oxygen targets is sparse. Purpose To assess the impact of a restrictive versus a liberal oxygen target on infarct size in OHCA patients undergoing immediate percutaneous coronary intervention (PCI). Methods The current study is a sub-study of the BOX trial, a randomized trial with a 2-by-2 factorial design, in which comatose OHCA patients with a presumed cardiac cause of arrest were randomly assigned to either a restrictive oxygen target (partial pressure of arterial oxygen (Pao2) of 9 to 10 kPa) or a liberal oxygen target (Pao2 of 13 to 14 kPa). In this sub-study we included patients who underwent immediate revascularization with PCI. The primary outcome was myocardial infarct size as assessed by the cardiac enzyme troponin-T (TnT). TnT was measured repeatedly during the first 72 hours and analyzed with a repeated measures variance component model with baseline adjustment. Secondary outcome was one-year all-cause mortality assessed by Kaplan-Meier and Cox proportional hazards analyses. Results A total of 329 patients underwent immediate PCI, whereof 172 patients were randomized to the restrictive oxygen target and 157 to the liberal oxygen target. The oxygen intervention was successful with a significantly higher PaO2 in the liberal oxygen target group during the first 72 hours (p<0.0001). Baseline characteristics and findings on coronary angiography (CAG) were similar in the two oxygen target groups. Thus, median age was 63 (interquartile range (IQR) 55–71), 81% were male, and estimated left ventricular ejection fraction on arrival to hospital was 35% (IQR 25–45%). Further, 74% of the patients had an initial postresuscitation electrocardiogram with ST-segment elevations, and the left anterior descending artery was deemed as culprit lesion in 49% of the patients. There was no difference in peak TnT between the restrictive and liberal oxygen target group (median 2540 (IQR 925–6420) versus 3410 (792–8630), ng/l, p=0.23), nor did TnT differ between the groups during the first 72 hours after admission (Figure 1, p=1.00). Within one year from admission, 120 patients (36.5%) had died. There was no difference in mortality between the restrictive (34%) and liberal (39%) oxygen target group (hazard ratio, 0.80; 95% CI, 0.56-1.14; p=0.21) (Figure 2). Conclusion Targeting a restrictive versus a liberal oxygen level in comatose survivors of OHCA who underwent immediate revascularization did not affect infarct size nor did it result in a significant difference in one-year mortality rate.Figure 1:Troponin-T levelsFigure 2:Kaplan-Meier curves