Abstract
Background : Extracorporeal cardiopulmonary resuscitation (ECPR) has emerged as a rescue strategy for non-responders to conventional CPR (CCPR) in cardiac arrest. Definitive guidelines for ECPR deployment do not exist. Prior studies suggest that arrest rhythm and cardiac origin of arrest may be variables used to assess candidacy for ECPR.
Aim : To describe a single center experience with ECPR and to assess associations between survival and physician-adjudicated origin of arrest and arrest rhythm.
Methods : A retrospective review of all patients who underwent ECPR at a quaternary care center over a 7-year period was performed. Demographic and clinical characteristics were extracted from the medical record and used to adjudicate origin of cardiac arrest, etiology, rhythm, survival, and outcomes. Univariate analysis was performed to determine association of patient and arrest characteristics with survival.
Results : Between 2010 and 2017, 47 cardiac arrest patients were initiated on extracorporeal membrane oxygenation (ECMO) at the time of active CPR. ECPR patient survival to hospital discharge was 25.5% (n=12). Twenty-six patients died on ECMO (55.3%) while 9 patients (19.1%) survived decannulation but died prior to discharge. Neither physician-adjudicated arrest rhythm nor underlying origin were significantly associated with survival to discharge, either alone or in combination. Younger age and arresting in the emergency department were significantly associated with survival. Nearly all survivors experienced myocardial recovery and left the hospital with a good neurological status.
Conclusions : Arrest rhythm and etiology may be insufficient predictors of survival in ECPR utilization. Further studies are needed to determine evidenced based criteria for ECPR deployment.
Introduction :
Sudden cardiac arrest in adults has low survival rates despite the widespread implementation of cardiopulmonary resuscitation (CPR)1. Hospital cardiac arrest survival to discharge can be less than 20%2–4. Moreover, the odds of survival decline with increasing duration of CPR5,6.
Extracorporeal membrane oxygenation (ECMO) during CPR, referred to as ECPR, can provide complete cardiopulmonary support and has emerged as a strategy to provide temporary perfusion and oxygenation for patients in cardiac arrest. The data in support of ECPR over conventional CPR (CCPR) remain mixed and are limited by the heterogeneity of a small number of studies.
Propensity matched observational studies comparing ECPR with CCPR suggest improved in-hospital and 1-year survival, as well as improved neurological outcomes with ECPR7–10. A large meta-analysis found that for adult patients with in-hospital cardiac arrest thought to be of cardiac origin, ECPR was associated with significantly improved survival and neurological outcomes compared to CCPR 11. Others, however, have found no clear benefit in survival to hospital discharge with ECPR12.
There are no definitive initiation criteria for ECMO deployment during in-hospital CPR. Among patients who undergo ECPR, approximately 20-40% survive to discharge, with younger age being a potential predictor for favorable neurological outcome13–20. Factors that have been reported to correlate with survival in ECPR include younger age17, shorter CPR duration16,21, normal pre-cannulation renal function22, a simplified acute physiology score II of less than 8023, a serum lactate less than 4.6 mmol per liter24, or a diagnosis of acute myocarditis19.
The prognostic significance of the arrest rhythm, arrest etiology, and cardiac versus non-cardiac origin of cardiac arrest in ECPR remains controversial. A meta-analysis of 856 patients who underwent ECPR for in hospital cardiac arrest demonstrated that an initial shockable rhythm was associated with survival. However, no significant association was found between survival and cardiac versus non-cardiac origin of arrest25. In contrast, a comparison of ECPR and CCPR found a higher rate of intensive care unit survival and long term favorable neurological outcome in ECPR patients with a non-cardiac origin of arrest10. Other studies report survival rates of 20% in patients with non-shockable rhythm at time of ECPR, suggesting a shockable rhythm should not be a solitary prerequisite for ECPR, however no information on cardiac arrest origin or etiology was included26.
The high study heterogeneity and low level of evidence limits the external validity of many meta-analyses. What is defined as cardiac versus non-cardiac origin is often not explicitly stated and may be variable between different studies, and underlying etiologies of cardiac arrest are not always available or reported.
In light of the expanding utilization of ECPR and lack of definitive criteria for optimal patient selection, we performed a single center retrospective observational study of adults with cardiac arrest who received ECPR at a large quaternary care center, determined the predominant arrest rhythm and underlying etiology in each case by physician-adjudicated chart review, and defined each cardiac arrest as cardiac or non-cardiac in origin based on the etiology. The purpose of this study was to report the association between ECPR survival and cardiac versus non-cardiac origin of arrest, both alone and combined with arrest rhythm.