Effects of Supplemental Oxygen During Mechanical Ventilation on Diaphragmatic Blood Flow

Abstract

Introduction During mechanical ventilation (MV), supplemental oxygen is commonly administered to critically-ill patients to prevent hypoxemia. However, in ICU patients, hyperoxia during MV is associated with increased mortality and a decrease in ventilator free days. Hyperoxic MV can elicit a paradoxical reduction in O2 delivery to the heart, brain, and kidneys. Whereas MV in normoxia diminishes diaphragm perfusion and O2 delivery, the effect of MV with supplemental oxygen on diaphragm perfusion is unknown. Previous studies demonstrate that MV with hyperoxia substantially impairs diaphragm contractile function. Therefore, hyperoxia during MV may potentiate reductions in diaphragm perfusion with MV and predispose patients to problematic weaning. Purpose Using an established animal model of MV, we tested the following hypotheses: 1) MV supplemented with hyperoxic gas (FIO2 0.99) would reduce diaphragmatic blood flow compared to spontaneously breathing (SB) animals, and 2) hyperoxic MV would decrease diaphragmatic blood flow to a greater extent compared to MV with room air (normoxia, FIO2 0.21). Methods Using fluorescent microspheres, tissue blood flow was measured in adult female Sprague-Dawley rats (n = 4) during spontaneous breathing (SB), mechanical ventilation + hyperoxia (MV +O2), and mechanical ventilation + room air (MV + RA). Following microsphere infusion during SB, animals were intubated and immediately ventilated with supplemental hyperoxic gas for 30 min and an arterial blood sample (~100 ul) was collected to confirm hyperoxemia prior to the second microsphere infusion. Thereafter, animals were subjected to MV using room air for 30 min followed by a third infusion of microspheres. Following the final microsphere infusion, tissues were harvested for blood flow analysis. The diaphragm was sectioned into costal (ventral, medial, dorsal) and crural portions to determine blood flow distribution within the diaphragm. Results During MV + O2, average arterial PO2 (PaO2) was 370 ± 39 mmHg. Compared to SB, total diaphragm blood flow was significantly decreased with MV + O2 (50 ± 2 vs 17 ± 3 ml/min/100g; P<0.05) and MV + RA (30 ± 1 ml/min/100g; P<0.05 vs SB). Compared to SB, blood flow to the medial costal diaphragm was reduced during both MV + O2 (59 ± 5 vs 18 ± 4 ml/min/100g; P<0.05) and MV + RA (59 ± 5 vs 29 ± 6 ml/min/100g; P<0.05). During MV + RA, compared to MV + O2, blood flow to the whole diaphragm was higher (30 ± 1 vs 17 ± 3 ml/min/100g; P<0.05). Although not significant, medial costal diaphragm flow tended to be higher with MV + RA versus MV + O2 (P=0.09). Conclusion These data demonstrate that supplementing short term MV with hyperoxia versus room air reduces total diaphragmatic blood flow to a significantly greater degree, suggesting that hyperoxia during MV reduces diaphragmatic O2 delivery. This raises the likelihood that prolonged bouts of MV (i.e. 6+ h) with hyperoxia may accelerate MV-induced vascular dysfunction in the quiescent diaphragm thus exacerbating downstream contractile dysfunction.