Veno-Venous Extracorporeal Membrane Oxygenation In Sepsis Patients With Acute Respiratory Failure

1.ObjectiveExtracorporeal membrane pulmonary oxygenation in sepsis patients with acute respiratory failure has been one of the research hotspots in recent years.The Purpose of this study is to evaluate whether septic patients with life-threatening hypoxemia can benefit from Veno-venous extracorporeal membrane oxygenation.2.Methods2.1 Study populationInclusion criteria:age>18 years;meeting diagnostic criteria for sepsis;combined acute respiratory failure treated with mechanical ventilation.Exclusion criteria:pregnant or breastfeeding women;patients with haematological tumours;patients with cachexia;patients with severe haemorrhagic shock;patients with less than 72h of conventional mechanical ventilation support or ECMO support;patients with a lot of missing clinical data.V-V ECMO support was performed if one of the following conditions was met:（1）Pa O2/Fi O2<80 mm Hg for more than 6 hours;（2）Pa O2/Fi O2<50 mm Hg for more than3 hours;（3）respiratory rate>30 breaths/min,PH<7.25,Pa CO2>60 mm Hg for more than 6 hours.2.2 Study methodsThe clinical data of sepsis patients with acute respiratory failure treated in the intensive care unit of xxxx hospital between June 2017 and January 2021 were retrospectively analysed.They were divided into mechanical ventilation group and mechanical ventilation+V-V ECMO group（ECMO group）based on whether received veno-venous extracorporeal membrane oxygenation.The 1:1 propensity score matching（PSM）method was used to screen patients matched for age,male,APACHE II score and Pa O₂.The day of diagnosis of sepsis with acute respiratory failure treated with mechanical ventilation is the starting point for observation.Survival analysis was performed using the Kaplan-Meier method to compare the primary outcome（difference in survival at 30and 90 days）between the two groups before and after matching.The difference in the secondary outcome:28 days without organ function support（mechanical ventilation,norepinephrine use,renal replacement therapy）days and the incidence of complications were also observed.3.Results3.1 Comparison of basic conditions and prognosis of septic patients with acute lung injury in two groupsAfter the inclusion and exclusion criteria of the study,255 clinical data were finally included for the study,including a total of 222 patients in the conventional treatment group and 33 patients receiving V-V ECMO support.The median age was 64 in the conventional treatment group and 53 in the ECMO group（P=0.006）;the median APACHE II score was significantly lower in the conventional treatment group than in the ECMO group（17 vs 22,P<0.001）,and the median SOFA scores were 14 and 12respectively,with no statistical difference（P=0.074）.Within 24 hours after enrollment,the median PO2 was 113 mm Hg in the conventional treatment group and 68.2 mm Hg in the ECMO group（P<0.001）.The main pathogenic microorganisms were gram-negative bacteria,53.6%in conventional treatment group and 66.6%in ECMO group,including Klebsiella pneumoniae（15.7%in conventional treatment group and 27.2%in ECMO group）,Acinetobacter baumannii（14.8%in conventional treatment group and 18.1%in ECMO group）,Pseudomonas aeruginosa（9.9%in conventional treatment group and9.0%in ECMO group）,and Escherichia coli（4.9%in conventional group and 6.0%in ECMO group）.A total of 50 cases（22.5%）of Gram-positive bacteriawere found in the conventional treatment group,mainly 22 cases（9.9%）of Enterococcus faecalis,13cases（5.8%）of Staphylococcus aureus,and 10 cases（4.5%）of Streptococcus pneumoniae;2 cases of Staphylococcus aureus（6.0%）were found in the ECMO.The median survival time was 28 days in the conventional treatment group and 46 days in the ECMO group,using Kaplan-Meier survival analysis,there was no statistical difference in survival at 30 days（HR 0.69,%95CI 0.42-1.14,P=0.147）and 90 days（HR 0.80,%95CI 0.50-1.29,P=0.361）.3.2 Comparison of primary outcomes after propensity score matchingA total of 31 pairs of patients were matched using a 1:1 propensity score,and there were no statistical differences between the two groups in terms of age,gender,APACHEII score,or Pa O₂.Using Kaplan-Meier survival analysis,patients in the ECMO group had a significantly higher probability of survival at 30 days compared with the conventional group（HR 0.44,95%CI:0.21-0.91,P=0.027）,and a 90-day Kaplan-Meier survival analysis of these two groups did not show a statistical difference（HR 0.059,95%CI.0.26-1.03,P=0.059）.3.3 Comparison of secondary outcomes after propensity score matchingIn the conventional treatment group,there were 16 patients（51.6%）with acute kidney injury,3 patients（9.7%）with bleeding,and 2 patients（6.5%）with hypoglycemia;in the ECMO group,there were 16 patients（51.6%）with acute kidney injury,7 patients（22.6%）with bleeding,and 5 patients（16.1%）with hypoglycemia,and there was no significant difference in complications between the two groups.Within 28days after the diagnosis of sepsis associated acute lung injury,the median time of renal replacement therapyfree days was longer in the V-V ECMO group than in the conventional treatment group（19days vs.8days;p=0.006）,and there was no significant difference in the median time of mechanical ventilation free days and norepinephrine free days.4.ConclusionCompared with conventional mechanical ventilation,sepsis patients with acute respiratory failure treated with V-V ECMO recovered renal function earlier and had a higher probability of survival at 30 days,but there was no difference in the probability of survival at 90 days between the two groups.