Survival Benefits And Prognostic Factors Of Severe ARDS Patients Receiving Rescue Venovenous Extracorporeal Membrane Oxygenation A Single-centered Propensity Score Matching Study

BackgroundThe morbility and mortality of severe acute respiratory distress syndrome(ARDS)remain high.Conventional lung protective mechanical ventilation(CMV)reduces alveolar injury,but cannot ensure oxygenation in the case of the disease developing to the most severe stage.Venovenous extracorporeal membrane oxygenation(VV-ECMO)has increasingly being used in the treatment of severe ARDS during the past decades due to its advantage of reversing hypoxemia through extracorporeal gas exchange,allowing protective lung ventilation and reducing ventilator-induced lung injury.However,owing to the dilemma between methodology and ethics,and the time-consuming and difficulty to carry out restrict randomized controlled trial studies,there is a lack of strict evidence to support the routine use of VV-ECMO in the treatment of severe ARDS patients,and the indicators,survival benefits and prognostic factors of ARDS patients receiving VV-ECMO are still not clearly elucidated.In this retrospective study,we used propensity score method to match the ARDS patients receiving VV-ECMO(ECMO group)with those only receving CMV(CMV group),which was similar to the experimental group and control group with comparable clinical characteristics in the randomized controlled trial,so as to mimic the study in the real world and to elicit the effect of VV-ECMO on severe ARDS patient.ObjectiveThe study aimed to compare the 28-day mortality and 28-day success rates of weaning from ventilator of severe ARDS patients between the ECMO group and the matched CMV group,and to explore the prognostic factors of severe ARDS patients treated with ECMO.MethodsPropensity score matching method was performed to reduce the confounding effects and the likelihood of selection bias.Each Patient in the ECMO group was matched with a patient in the CMV group.Then,the 28-day mortality and 28-day success weaning rate from ventilator were compared between the two groups.Cox regression analysis was employed to determine independent predictive risk factors for the clinical outcomes.ResultsA total of 167 patients with severe ARDS were included,71 patients received VV-ECMO and 96 patients mechanically ventilated with lung protective stratery only.Fourty six pairs from the two groups were successfully matched.The 28-day mortality in the ECMO group was significantly lower than that in the CMV group(58.7%vs.80.4%,P=0.023),and the 28-day success rate of weaning from ventilator in the ECMO group was significantly higher than that in the CMV group(37%vs.15.2%,P=0.018).The ventilator free days in the ECMO group was significantly shorter than that in the CMV group(P=0.023),and there was an insignificant difference in the duration of ventilation or hospitalization between the two groups.The survival curve and weaning curve of ventilator of ECMO group were better than those of CMV group before and after matching.The cost of hospitalization in ECMO group was significantly higher than that of CMV group(P<0.001).ECMO treatment significantly shortened the length of mechanical ventilation(P=0.023),but had no significant correlation with the length of stay in hospital(P=0.475).Multivariate Cox regression analysis indicated that lower BMI,lower mean albumin concentration,higher APACHE Ⅱ score,and extrapulmonary factors caused ARDS were independent risk factors for the poor outcome of ECMO treatment at the 28th day.The adjusted multifactorial analysis indicated that ECMO treatment may improve the survival rate of severe ARDS patients comparing with CMV treatment(HR=0.539,P=0.017).ConclusionOur finding suggested that the 28-day mortality was significantly lower among severe ARDS patients who receiving VV-ECMO treatment than among those who receiving conventional mechanical ventilation.Moreover,VV-ECMO could improve the 28-day weaning success rate from ventilator of patients and shorten the duration of mechanical ventilation significantly.ECMO treatment did not significantly shorten the length of hospital stay.Lower BMI,lower mean albumin concentration,higher APACHE II score and extrapulmonary factors are associated with increased 28-day mortality in patients placed on VV-ECMO for severe ARDS.Meanwhile,VV-ECMO treatment was significantly associated with higher hospitalization costs.