Extracorporeal life support. A "breath-taking" technology?

Abstract

Each year, hundreds of thousands of critically ill patients worldwide undergo mechanical ventilation (1). Ventilation is vital for these patients, but also has strong potential to harm the lungs, a phenomenon frequently referred to as ventilator-induced lung injury (2). It has been theorized that ventilator-induced lung injury is caused, at least in part, by energy transferred from the ventilator to the lungs. As more energy is transferred with the use of larger tidal volumes and higher respiratory rates, strategies that allow use of lower tidal volumes (3, 4) or lower respiratory rates (5, 6) could mitigate ventilator-induced lung injury. Extracorporeal life support (ECLS) is one of several terms used for an extracorporeal circuit that employs a membrane for oxygenation and elimination of carbon dioxide (7, 8). The “veno–venous” approach to ECLS uses a blood pump, in contrast to the “arterio–venous” approach that uses intrinsic arterial blood pressure to drive blood through the extracorporeal circuit. When the primary need is oxygenation (extracorporeal membrane oxygenation or ECMO), larger membranes are mandatory. When the goal is primarily or entirely the elimination of carbon dioxide (extracorporeal CO2 removal or ECCO2R), ECLS can be achieved using smaller membranes. ECLS has conventionally been used as a salvage strategy for patients with severe acute respiratory distress syndrome (ARDS). Use of ECLS in these patients may also enhance protective ventilation, by enabling further reduction in tidal volumes delivered by the ventilator. In this issue of AnnalsATS, Munshi and coworkers (pp. 802–810) report a systematic review and meta-analysis of four randomized controlled trials and six observational studies comparing a strategy of augmenting mechanical ventilation with ECLS to conventional ventilation alone for patients with ARDS (9). In the overall analysis, ECLS was not associated with a reduction in in-hospital mortality (relative risk [RR], 1.02; 95% confidence interval [CI], 0.79–1.33]; P . 0.05). However, ECLS was associated with reduced mortality in studies of veno– venous ECLS (RR, 0.64; 95% CI, 0.51–0.79; P , 0.05) and in studies that used lung– protective ventilation with lower tidal volumes (6 ml/kg) (RR, 0.53; 95% CI, 0.53– 0.80; P , 0.05). Heterogeneity among the studies included in the meta-analysis may explain, at least in part, why no advantage was found in the overall analysis. Notably, the included randomized controlled trials had important limitations related to quality, with some trials lacking allocation concealment and several not following an intention-to-treat analysis. The finding that ECLS was associated with reduced mortality in studies that used lung-protective ventilation may be explained in part by the lower heterogeneity of these trials. It could also be that use of ECLS was associated with a further reduction of tidal volume size (i.e., below 6 ml/kg), or respiratory rate, or both, thereby improving outcomes. Unfortunately, Munshi and colleagues (9) did not analyze specific ventilator settings. The meta-analysis by Munshi and coworkers (9) highlights the need for better studies of ECLS. In particular, we need well-powered, high-quality, randomized controlled trials in which ECLS is compared with the current standard of ventilation care (i.e., lung-protective ventilation with lower tidal volumes). We also need a better understanding for why ECLS could benefit patients with ARDS: is it the use of even lower tidal volumes (i.e., lower than 6 ml/kg), or lower respiratory rates, or maybe both? Indeed, the most appropriate ventilator and ECLS settings for patients with severe ARDS who receive ECLS are largely unknown (7). Large databases like the ELSO registry (www.elsonet.org) may shine additional light on this, but in the end we need randomized controlled trials. Is there merit for using ECLA in patients without ARDS? Several studies showed that conventional mechanical ventilation is far from a safe strategy for patients without ARDS, and that the lungs of these patients can be protected by using lower tidal volumes (10). One could thus speculate that there is a role for ECLS in patients without ARDS, and there is some published evidence in support of this hypothesis. Indeed, use of ECCO2R may avoid intubation and invasive mechanical ventilation in patients with acute on chronic respiratory