Lipid peroxidation during initiation of extracorporeal membrane oxygenation after hypoxia in endotoxemic rabbits

Abstract

Initiation of extracorporeal membrane oxygenation (ECMO) in septic children with severe respiratory failure often improves oxygenation but not pulmonary function. The factors affecting pulmonary function following onset of ECMO are not completely understood, but are thought to involve injury mediated, in part, by reactive oxygen species. We hypothesized that induction of ECMO using 100% oxygen as the sweep gas through the oxygenator would increase lipid peroxidation in endotoxin-primed animals after severe hypoxia. We further speculated that provision of oxygenated blood to the pulmonary circulation via venovenous ECMO would promote a greater degree of oxidative damage to the lung as compared to venoarterial ECMO.

Eighteen New Zealand White rabbits were assigned to a control group (control) or two intervention groups subjected to 60 min of venoarterial or venovenous ECMO. ECMO was initiated following an intravenous challenge with 0.5 mg/kg of E. coli endotoxin and a period of global hypoxia leading to an arterial pH of 6.99 ± 0.09, PaCO₂ of 103 ± 31 mmHg and PaO₂ of 27 ± 5 mmHg. Malondialdehyde (MDA), a marker of lipid peroxidation, was measured in lung tissue homogenates and in arterial plasma.

Lung tissue MDA demonstrated a strong trend towards an increase in the venoarterial group (1884 ± 945 nmol/g protein) and in the venovenous group (1905 ± 758 nmol/g protein) in comparison to the control group (644 ± 71 nmol/g protein) (p = 0.1; significance at 95% in Scheffe test). Lung tissue MDA in the venovenous group had a significant correlation with mean PaO₂ during ECMO by regression analysis (r ² = 0.678, p = 0.044). The change in blood MDA concentration between pre-ECMO and post-ECMO values was greater in the venovenous group (pre 1.62 ± 0.61 versus post 5.12 ± 0.2.07 μmol/l, p = 0.043) compared with that seen in the venoarterial group (pre 1.46 ± 0.38 versus post 3.9 ± 0.93 μmol/l).

Our data support the hypothesis that initiation of ECMO with a circuit gas oxygen concentration of 100% after global hypoxia enhances oxidative damage to lipids in endotoxin-challenged animals. During venovenous ECMO this finding is dependent on PaO₂.