Effects of inhaled nitric oxide and extracorporeal membrane oxygenation on pulmonary hemodynamics and lymph flow in oleic acid lung injury in sheep

Objective: To compare the effects of inhaled nitric oxide (NO) and extracorporeal membrane oxygenation (ECMO) on oxygenation, hemodynamics, and lymphatic drainage in an oleic acid lung injury model in sheep. Design: Prospective, randomized study. Setting: Animal research laboratory. Animals: Thirty female sheep, weighing 35 to 40 kg. Interventions: Acute lung injury was induced by central venous injection of oleic acid (0.5 mL/kg body weight). A chronic lymph fistula had been prepared through a right thoracotomy 3 days before the experiment. Animals were assigned randomly to the NO group (n = 14) or the ECMO group (n = 16). When a lung injury score of >2.5 was achieved, the animals were given NO in dosage increments of 2, 5, 10, 20, and 40 parts per million (ppm), or placed on ECMO with an Fio(2) of 0.21 (ECMO-21) and then 1.0 (ECMO-100) at the oxygenator. Mechanical ventilator parameters were kept constant to isolate the effects of NO and ECMO on systemic and pulmonary hemodynamics, cardiac output, oxygenation parameters, lymph/plasma protein ratio, and lymph flow. Measurements and calculations were performed after 1 hr at each individual step of NO concentration or Fio(2). Measurements and Main Results: In the ECMO group, PVRI and MPAP did not change and were significantly different from the NO group. In the NO group, there was a dose-dependent decrease in venous admixture, maximal at 10 ppm NO and decreasing from 40 +/- 6% to 23 +/- 10% (p<.05). This decrease was significantly different from the ECMO group, where there was no change. There was a significant increase in Pao(2)/Fio(2) in the NO group, maximal at 10 ppm NO (84 +/- 11 to 210 +/- 90, p<.05), but a greater increase in Pao(2)/ Fio(2) on ECMO-21 (81 +/- 14 to 265 +/- 63) and a further increase on ECMO-100 (398 +/- 100) (p <.05). The lymph/plasma protein ratio remained unchanged in both groups after induction of lung injury by oleic acid. However, lymph flow decreased by 11 +/- 6% in the NO group, whereas it increased by 14 +/- 17% in the ECMO group (p < .05). Conclusions: In an oleic acid-induced sheep model of acute lung injury, there were significant differences between the effects of NO and ECMO on acute pulmonary hypertension, hypoxemia, hypercarbia, and lymph flow. NO significantly decreases pulmonary hypertension, whereas pulmonary hemodynamics were not substantially affected by ECMO. Both interventions reversed hypoxemia, but ECMO did so to a greater degree, and only ECMO improved hypercarbia. Only NO decreased lymph flow, possibly as an effect of decreased microvascular filtration pressure. This study did not attempt to evaluate the impact of these interventions on ventilatory requirements, barotrauma, or outcome. However, this model suggests that NO therapy may moderate pulmonary hypertension and improve lymph flow in acute lung injury. Clinical studies are needed to assess whether NO therapy might be beneficial in treatment of severe acute lung injury in older children and adults.