The effect of different oxygen concentrations on oxidative stress in anesthetized horses

Abstract

Anesthesia-associated mortality in horses is high when compared with other species, and may be related to development of hypoxemia during recumbency. To overcome this, high fractions of inspired oxygen (FiO2 >95%) are used. However, this may lead to an increase in oxidative stress, free radical formation and tissue injury. Isoprostanes are a marker of oxidative stress and their quantification is a relatively new and reliable approach to evaluate oxidative stress status. The aims of this research were to evaluate changes in urine isoprostanes (IsoP), serum cardiac troponin I (cTnI), arterial blood parameters, and serum chemistry in horses anesthetized with FiO2 of 0.45, 0.75, or >0.95, and also evaluate the quality of recover utilizing accelerometry. Twenty-four healthy adult horses were anesthetized with isoflurane for 120 minutes. Horses were randomly assigned to receive a FiO2 of 0.45 (n=8), or 0.75 (n=8), or >0.95 (n=8) during anesthesia. Horses were anesthetized with a standard anesthesia induction and maintenance protocol and placed in dorsal recumbency. Horses were mechanically ventilated during throughout anesthesia. After 120 minutes of anesthesia, horses were allowed to recovery unassisted in a dedicated recovery stall wearing an accelerometer. Urine, venous, and arterial blood samples were collected at baseline prior to anesthesia, after 60 minutes of anesthesia, at the end of anesthesia, after recovery, and at 24 hours after induction. Urine IsoP and creatinine concentrations were measured. Myocardial injury was assessed utilizing cTnI. Oxygenation status was evaluated by arterial blood gas analysis. All parameters were evaluated by time and by group effect using 2-way ANOVA, followed by Friedman test, and Kruskal-Wallis followed by Dunn’s test. A p < 0.05 was used for significance. During anesthesia, all physiologic parameters monitored remained within normal range without differences among groups. No difference was found over time or among groups for isoprostanes or cTnI. Measured serum chemistry variables, except for serum iron, returned to baseline values at T24h. Serum iron was lower at T24h compared with T0 in all groups (G95 T0 = 148.00 [103.00; 171.00] and T24h = 60.50 [39.00; 102.00] mcg/dL; G75 T0 = 153.00 [130.00; 198.00] and T24h = 53.50 [41.00; 96.00] mcg/dL; and G45 T0 = 149.50 [103.00; 213.00] and T24h = 63.50 [35.00; 102.00] mcg/dL). PaO2 was different among groups during anesthesia, but the PaO2/FiO2 relationship was not different among groups. End tidal CO2 was not different among groups, however, PaCO2 was different during anesthesia at specific time points between G95 and G45 (G95 T60 = 59.30 [50.50; 67.50]; T120 = 55.65 [52.20; 63.30] mmHg; G45 T60 = 50.20 [43.10; 55.70]; T120 = 46.90 [41.90; 55.90] mmHg). There was no different in recovery score between groups. The fractions of oxygen used during anesthesia in horses does not appear to have an impact on oxidative stress measured by urine isoprostane concentrations or myocardial injury measured by cTnI. Additionally, the fractions of oxygen tested did not have an effect on recovery scores of healthy horses anesthetized with isoflurane for two hours in dorsal recumbency.