Thorac Cardiovasc Surg 2018; 66(S 01): S1-S110
DOI: 10.1055/s-0038-1627934
Oral Presentations
Sunday, February 18, 2018
DGTHG: ECMO
Georg Thieme Verlag KG Stuttgart · New York

Risk Prediction in Patients with Venoarterial Extracorporeal Membrane Oxygenation for Cardiopulmonary Support: Insights from a European Nationwide Registry

M. Becher
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
,
C. Sinning
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
,
B. Schrage
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
,
N. Fluschnik
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
,
C. Waldeyer
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
,
M. Seiffert
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
,
A. Bernhardt
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
,
H. Reichenspurner
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
,
S. Blankenberg
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
,
R. Twerenbold
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
,
D. Westermann
1   Universitäres Herzzentrum Hamburg, Hamburg, Germany
› Author Affiliations
Further Information

Publication History

Publication Date:
22 January 2018 (online)

 

    Background: Venoarterial extracorporeal membrane oxygenation (va-ECMO) is a common strategy for prolonged mechanical cardiopulmonary support despite limited outcome data. The aim of this study was to analyze the incidence of va-ECMO therapy as well as describe practice and outcome utilizing complete nationwide data. Moreover, we defined mortality-based predictors for va-ECMO treatment.

    Methods: We analyzed characteristics, complications and 30-day outcomes of all va-ECMO procedures performed in Germany from 2007 to 2015. Multivariate regression analysis identified the following 14 predictor variables of in-hospital mortality: age >65 years (y), non-elective admission, atrial fibrillation, hypertension, hyperlipidemia, pulmonary hypertension, liver disease, coronary artery disease, congestive heart failure, acute coronary syndrome, sepsis, acute respiratory distress syndrome (ARDS), prior cardiopulmonary resuscitation (CPR) and heart transplantation. The ECMO-ACCEPTS-score was established using the 14 variables and internally validated in a 1:1 randomly selected derivation/ validation design.

    Results: A total of 9,258 va-ECMO procedures were performed between 2007 and 2015 (61 ± 17 years of age, male 70.1%). The annual number of va-ECMO procedures steadily increased by more than 30-fold from 93 procedures in 2007 to 2,849 in 2015 with a large increase in 2013. The overall 30-day in-hospital mortality was 58.6% and did not significantly change over time. Strong predictors of 30-day in-hospital mortality were age >65 years, CPR prior to va-ECMO and especially its combination (mortality: no CPR/< 65y, 43.6%; no CPR/≥65 years, 61.9%; CPR/< 65y, 65.5%, CPR/≥65 years, 75.1%). The ECMO-ACCEPTS-score, derived among randomly selected 4,629 patients, showed excellent discrimination of 30-day in-hospital mortality in the validation cohort of the remaining 4,629 patients.

    Conclusion: The incidence of va-ECMO increased by more than factor 30 in Germany between 2007 and 2015, whereas in-hospital mortality remained unaffected. Prior CPR and age > 65 years are critical factors for mortality of patients undergoing va-ECMO. The ECMO-ACCEPTS-score may be a useful tool to improve early risk prediction among candidates for va-ECMO to assist in restoring circulation.


    #

    No conflict of interest has been declared by the author(s).