Mechanical ventilation during extracorporeal life support (ECLS): a systematic review

In patients with acute respiratory distress syndrome (ARDS), extracorporeal life support (ECLS) has been utilized to support gas exchange and mitigate ventilator-induced lung injury (VILI). The optimal ventilation settings while on ECLS are unknown. The purpose of this systematic review is to describe the ventilation practices in patients with ARDS who require ECLS. We electronically searched MEDLINE, EMBASE, CENTRAL, AMED, and HAPI (inception to January 2015). Studies included were randomized controlled trials, observational studies, or case series (a parts per thousand yen4 patients) of ARDS patients undergoing ECLS. Our review focused on studies describing ventilation practices employed during ECLS for ARDS. Forty-nine studies (2,042 patients) met our inclusion criteria. Prior to initiation of ECLS, at least one parameter consistent with injurious ventilation [tidal volume > 8 mL/kg predicted body weight (PBW), peak pressure > 35 cmH(2)O (or plateau pressure > 30 cmH(2)O), or FiO(2) a parts per thousand yen0.8] was noted in 90 % of studies. After initiation of ECLS, studies reported median [interquartile range (IQR)] reductions in: tidal volume [2.4 mL/kg PBW (2.2-2.9)], plateau pressure [4.3 cmH(2)O (3.5-5.8)], positive end-expiratory pressure (PEEP) [0.20 cmH(2)O (0-3.0)], and FiO(2) [0.40 (0.30-0.60)]. Median (IQR) overall mortality was 41 % (31-51 %). Reduction in the intensity of mechanical ventilation in patients with ARDS supported by ECLS is common, suggesting that clinicians may be focused on reducing VILI after ECLS initiation. Future investigations should focus on establishing the optimal ventilatory strategy for patients with ARDS who require ECLS.