Pre-cannulation international normalized ratio is independently associated with mortality in veno-arterial extracorporeal membrane oxygenation
Abstract
Objective
To explore whether pre-cannulation international normalized ratio (INR) is associated with in-hospital mortality in veno-arterial extracorporeal membrane oxygenation (VA-ECMO) patients.
Design
Retrospective observational cohort study.
Setting
Quaternary care academic medical center.
Participants
Patients with cardiogenic shock on VA-ECMO for >24 hours.
Interventions
None, observational study.
Measurements and Main Results
One hundred eighty-eight VA-ECMO patients were included over 3 years. Patients were stratified into three groups based on their pre-ECMO INR: INR <1.5, INR 1.5-1.8, and INR >1.8. For all patients, demographics, comorbidities, and ECMO details were recorded. The study’s primary outcome was in-hospital mortality and secondary outcomes included major bleeding, minor bleeding, allogeneic transfusion, ischemic stroke, intracranial hemorrhage, acute renal failure, acute liver failure, gastrointestinal bleeding, and ICU and hospital length of stay. Multivariable logistic regression was used to determine whether pre-cannulation INR was independently associated with in-hospital mortality. In-hospital mortality differed significantly by INR group (51.6% INR >1.8 vs. 42.3% INR 1.5-1.8 vs. 24.3% INR <1.5, p=0.004). In a multivariable logistic regression model, pre-cannulation INR >1.8 was independently associated with increased odds of mortality (OR, 2.48; 95% CI, 1.05-6.04) after controlling for sex, SAVE score, and ECMO indication. INR 1.5-1.8 did not confer increased mortality risk.
Conclusions
INR >1.8 before VA-ECMO cannulation is independently associated with in-hospital mortality. Pre-cannulation INR should be considered by clinicians so that ECMO resources can be better allocated and risks of organ failure and intracranial hemorrhage can be better understood.