30-Day Outcomes Post Veno-Arterial Extra Corporeal Membrane Oxygenation (VA-ECMO) After Cardiac Surgery and Predictors of Survival

Background

Veno-arterial extracorporeal membrane oxygenation (VA-ECMO) is sometimes needed for post-cardiotomy cardiogenic shock (PCCS). There is little data regarding outcomes in the Australian context, particularly in a non-cardiac transplant centre. Our aim was to report on 30-day outcomes after patients with PCCS treated with VA-ECMO in an Australian non-cardiac transplant tertiary centre, and to determine risk factors for non-survival in this population.

Methods

A retrospective analysis was performed on all adults treated with VA-ECMO for PCCS between August 2001 and September 2016 at our centre. Univariate analysis with adjustment for multiplicity identified risk factors for non-survival. Area under the receiver operating characteristics (AUROC) method was used to assess their predictive value.

Results

We identified 64 patients out of 5,502 open-heart surgery cases of which three patients did not meet inclusion criteria. Mean (SD) age was 63 (14) years. Survival to hospital discharge or 30 days post VA-ECMO occurred in 27/61 (44%) patients. VA-ECMO was able to be weaned in 44/61 patients (72%); 54/61 patients (89%) had at least one major complication. Prior to VA-ECMO initiation, no statistically significant differences between survivors and non-survivors could be determined. After VA-ECMO initiation, only 24-hour nadir lactate and 48-hour nadir lactate levels were significantly different between survivors and non-survivors (1.50 mmol/L vs 3.20 mmol/L p=0.001; and 1.20 mmol/L vs. 1.90 mmol/L p=0.001 respectively). For mortality prediction, 24- and 48-hour nadir lactate levels had AUROCs of 0.775 and 0.782, respectively.

Conclusions

VA-ECMO is associated with acceptable survival rates but significant morbidity. Nadir lactate levels in the first 24 and 48 hours after VA-ECMO initiation may be useful in predicting early survival.

Introduction

Post-cardiotomy cardiogenic shock (PCCS) refractory to pharmacological and intra-aortic balloon pump (IABP) therapy is uncommon following cardiac surgery [[1], [2], [3], [4], [5]]. In these patients, salvage therapy in the form of veno-arterial extra-corporeal membrane oxygenation (VA-ECMO) can provide life-sustaining, temporary mechanical circulatory support, allowing time for recovery of cardiopulmonary function, and time for further treatment decisions to be made [1,6,7]. However, in these patients, survival rates to hospital discharge have been reported at only 30–40% over the last 20 years, and significant morbidity and prolonged hospital stays are associated with VA-ECMO in this setting [1,4,8,9].

For these reasons, the use of VA-ECMO following cardiac surgery remains controversial. There are no universally-accepted guidelines for the institution or withdrawal of this resource-intensive intervention in this setting [9]. Previous retrospective studies have suggested various risk factors for in-hospital mortality in this group of patients, including advanced age, renal insufficiency, and blood lactate levels before or after institution of VA-ECMO [1,3,4,8,9]. However, data are often conflicting, with incomplete accounting for important risk factors. Risk stratification in these patients therefore remains very difficult. Additionally, there is a paucity of data regarding outcomes from Australia and New Zealand, particularly from centres that do not have access to cardiac transplantation or long-term ventricular support. The aim of this study was to describe outcomes up to 30 days after VA-ECMO in an Australian non-cardiac transplant tertiary centre and to explore risk factors for predicting early non-survival in patients in this context, which has not been done before.