Integrating Mechanical Ventilation and Extracorporeal Membrane Oxygenation in Severe Acute Respiratory Distress Syndrome.

To the Editor:

We read with great interest the study of Araos and colleagues (1), who elegantly demonstrated that near-apneic ventilation decreased lung injury and early fibroproliferation in an animal model of extracorporeal membrane oxygenation (ECMO)-supported acute respiratory distress syndrome (ARDS). Although minimizing risks of ventilator-induced lung injury on venovenous ECMO is paramount, the risks/benefits of strategies employed to minimize ventilator-induced lung injury also merit due consideration.

First, a demonstration that near-apneic ventilation with moderate positive end-expiratory pressure does not promote atelectasis and worsen intrapulmonary shunt fraction in that study may have been particularly helpful. Blood flow through the diseased pneumonic lung or lungs or parts of the lung that have collapsed will contribute to intrapulmonary shunting, also referred to as venous admixture. The shunt fraction is the calculated estimate of how much hypoxic blood should return to the arterial side after passing through the shunt to produce the measured arterial oxygen results, for a given . The lung is a mixture of heterogeneous units, each with a different ratio that can be severely affected by the loss of hypoxic vasoconstriction on venovenous ECMO. In addition, hypoventilation induced by extracorporeal carbon dioxide removal can lower the global ratio of the native lungs and results in reabsorption atelectasis, therefore worsening hypoxemia (2). During near-apneic ventilation in severe ARDS, the contribution of native lungs to oxygenation is obviously significantly reduced. This makes the patient near-total ECMO dependent for oxygenation, often requiring high ECMO blood flows in the setting of a high state.

Second, as reported by the authors, the very low respiratory rate contributed significantly to the marked decrease in mechanical power observed in the near-apneic group. It should be noted that our understanding of the complex heart–lung–ventilator ECMO interactions are still evolving. Given that venovenous ECMO is a therapy delivered over weeks to months, the benefits of extreme lung protection should be balanced against risks associated with such strategies. We need to ensure that ventilation strategies employed on ECMO do not limit our ability to provide evidence-based supportive measures such as fluid restriction, minimization of sedation, and pharmacologic paralysis and early rehabilitation. To date, potential clinical benefits have only been demonstrated with a ventilation strategy that employed moderate positive end-expiratory pressure and limited plateau pressures ≤24 cm H2O (3, 4). Last, moving forward, there is a sound physiologic rationale to reinforce the ventilator strategy employed in the EOLIA (ECMO to Rescue Lung Injury in Severe ARDS) trial with prone positioning. Although ECMO provides lung protection, prone positioning may further improve respiratory system compliance and matching (5, 6). This may minimize reliance on higher ECMO blood flows with such extreme mechanical ventilation reduction strategies until the risk/benefit ratio of such strategies is clearly established.