Impact of extracorporeal cardiopulmonary resuscitation on neurological prognosis and survival in adult patients after cardiac arrest: an individual pooled matched patient data meta-analysis

Sudden cardiac arrest (CA) is a leading cause of cardiovascular death worldwide, with approximately 500,000 cases per year in North America (1). Despite significant advancements in recent decades, survival rates for both in-hospital cardiac arrest (IHCA) and out-of-hospital cardiac arrest (OHCA) remain disappointingly low (2). Several therapeutic strategies have been introduced, including target temperature management (TTM) following return of spontaneous circulation (ROSC) and immediate coronary angiogram with percutaneous coronary intervention (PCI) after ROSC. However, the effectiveness of these interventions in improving patient outcomes remains controversial and the identification of suitable patient populations for these treatments is still uncertain (3-5). Conventional cardiopulmonary resuscitation (CCPR) focuses on initiating early resuscitation maneuvers, particularly chest compressions, to reduce "no-flow time" and enhance the chances of successful recovery and survival (6). However, even with proper execution of CCPR, achieving ROSC is not guaranteed for a significant number of patients (7). Furthermore, among those who do achieve ROSC, prolonged resuscitation attempts significantly decrease the probability of favorable long-term neurological outcomes (8,9). In cases of refractory CA unresponsive to CCPR, the initiation of veno-arterial extracorporeal membrane oxygenation (ECMO), i.e. extracorporeal cardiopulmonary resuscitation (ECPR), during resuscitation has shown promise. ECPR extends the available time for resuscitation, preserves organ function, reduces the need for inotropic and vasopressor drugs, lowers myocardial oxygen consumption and potentially facilitates cardiac recovery (10). Additionally, when combined with a heat exchanger, ECPR enables rapid induction and stable maintenance of hypothermia, which may have neuroprotective effects in this context (11). ECPR also allows for investigation and treatment of potential underlying causes, such as coronary occlusion or pulmonary thromboembolism (12). Despite its potential benefits, ECPR is an invasive and potentially traumatic intervention, associated with various complications, including hemolysis, coagulopathy and bleeding, acute kidney injury, limb ischemia, and severe pulmonary edema (13). Successful vascular cannulation for ECPR requires technical expertise and surgical skills. Consequently, three randomized trials have shown different effects of ECPR in OHCA patient (14-16). Observational studies have reported more conflicting results (17,18) and limitations in study design have hindered comprehensive meta-analysis of available data (19). An individual participant data (IPD) meta-analysis offers advantages over aggregate meta-analysis, allowing for the inclusion of individual patient characteristics in the analyses. The aim of this study was therefore to evaluate the impact of ECPR on neurological outcomes and mortality compared to CCPR, with a specific focus on predefined subgroups, using a matched IPD meta-analysis