Outcomes of patients with acute respiratory failure on veno-venous extracorporeal membrane oxygenation requiring additional circulatory support by veno-venoarterial extracorporeal membrane oxygenation

Abstract

Objective: Veno-venous (V-V) extracorporeal membrane oxygenation (ECMO) is increasingly used to support patients with severe acute respiratory distress syndrome (ARDS). In case of additional cardio-circulatory failure, some experienced centers upgrade the V-V ECMO with an additional arterial return cannula (termed V-VA ECMO). Here we analyzed short- and long-term outcome together with potential predictors of mortality.

Design: Multicenter, retrospective analysis between January 2008 and September 2021.

Setting: Three tertiary care ECMO centers in Germany (Hannover, Bonn) and Switzerland (Zurich).

Patients: Seventy-three V-V ECMO patients with ARDS and additional acute cardio-circulatory deterioration required an upgrade to V-VA ECMO were included in this study.

Measurements and main results: Fifty-three patients required an upgrade from V-V to V-VA and 20 patients were directly triple cannulated. Median (Interquartile Range) age was 49 (28-57) years and SOFA score was 14 (12-17) at V-VA ECMO upgrade. Vasoactive-inotropic score decreased from 53 (12-123) at V-VA ECMO upgrade to 9 (3-37) after 24 h of V-VA ECMO support. Weaning from V-VA and V-V ECMO was successful in 47 (64%) and 40 (55%) patients, respectively. Duration of ECMO support was 12 (6-22) days and ICU length of stay was 32 (16-46) days. Overall ICU mortality was 48% and hospital mortality 51%. Two additional patients died after hospital discharge while the remaining patients survived up to two years (with six patients being lost to follow-up). The vast majority of patients was free from higher degree persistent organ dysfunction at follow-up. A SOFA score > 14 and higher lactate concentrations at the day of V-VA upgrade were independent predictors of mortality in the multivariate regression analysis.

Conclusion: In this analysis, the use of V-VA ECMO in patients with ARDS and concomitant cardiocirculatory failure was associated with a hospital survival of about 50%, and most of these patients survived up to 2 years. A SOFA score > 14 and elevated lactate levels at the day of V-VA upgrade predict unfavorable outcome.

Keywords: acute respiratory distress syndrome; extracorporeal life support (ECLS); sequential organ failure assessment (SOFA) score; shock; survival analysis; triple cannulation; vasoactive inotropic score.

FIGURE 1

Comparison of Vasoactive-inotropic score (A) and serum lactate (B) before V-VA ECMO upgrade and after 24 h under V-VA ECMO support visualized as boxplots and scatterplots. If patients deceased in the first 24 h, the latest value before discontinuation of life-sustaining therapies was chosen.

FIGURE 2

Stratification of predictive variables at veno-veno-arterial ECMO cannulation. Left: Predictive variables at the time of V-VA ECMO cannulation stratified in survivors and non-survivors according to 60-day ICU mortality. Values are expressed as n (%) or median (interquartile range). Right: Forest-plot and univariate cox regression for 60-day ICU-mortality. Values are expressed as Hazard ratio (HR) with 95% confidence interval (CI) and p-value. ARDS, acute respiratory distress syndrome; CPR, cardiopulmonary resuscitation; FIO₂, fraction of inspired oxygen; Hosp., Hospital; ICU, intensive care unit; iMV, invasive mechanical ventilation; LVEF, left ventricular ejection fraction; PaCO₂, partial pressure of carbon dioxide; PaO₂, partial pressure of oxygen.

FIGURE 3

Survival function from veno-venoarterial ECMO cannulation. (A) Overall survival, (B–D) stratified survival function for 60-day ICU mortality. SOFA, Sequential Organ Failure Assessment; VIS, Vasoactive inotropic score.