INTRODUCTION
Extracorporeal cardiopulmonary resuscitation (ECPR) has emerged as a crucial therapeutic option for pediatric patients who experience refractory cardiac or respiratory arrest. By implementing extracorporeal membrane oxygenation (ECMO) during cardiopulmonary resuscitation (CPR), ECPR offers temporary circulatory support, providing essential time for recovery or additional interventions. Since 1992, the Extracorporeal Life Support Organization (ELSO) registry has been monitoring the use of ECMO during pediatric CPR, underscoring its effectiveness in certain pediatric contexts [1,2]. An analysis of the registry data from 2009 to 2022 showed survival rates to hospital discharge of 44.3% in neonates and 41.1% in children, which are notably higher than the 29.1% observed in adults [3]. These better outcomes in pediatric patients are linked to several factors, including the frequent use of ECPR in intensive care units, which results in shorter time intervals between the onset of cardiac arrest and the start of ECMO, and the preference for neck or central vessel cannulation, which is more efficient than femoral access [4]. However, the limited involvement of Korean centers in the ELSO registry has resulted in significant gaps in our understanding of the use and outcomes of pediatric ECPR in Korea.
Globally, the use of pediatric ECPR is increasing, driven by technological advancements and improvements in treatment protocols [5]. However, research focusing on ECPR outcomes in Korean pediatric patients remains scarce. The management of pediatric ECPR is especially challenging due to the diversity of underlying conditions, such as congenital heart defects and congenital diaphragmatic hernia. Additionally, younger and smaller patients present further difficulties, typically necessitating more complex interventions and experiencing higher rates of complications compared to adults [6-8].
Understanding the frequency, indications, and outcomes of pediatric ECPR is essential for optimizing healthcare resources and enhancing patient outcomes. This study is a subgroup analysis of ECPR cases, extracted from a broader investigation of pediatric ECMO that spanned major Korean centers from 2012 to 2021 [9]. Our previous research showed that ECPR was linked to a twofold increase in the risk of mortality. The aims of this study are to examine trends in the use of ECPR, evaluate survival outcomes, and identify factors associated with mortality.
METHODS
This study was approved by the Institutional Review Board of Pusan National University Yangsan Hospital (IRB No. 04-2022-004) and by all relevant institutional review boards at participating centers. Given the retrospective nature of the study design, a waiver of informed consent was granted.
Study Design and Setting
This multicenter retrospective study was conducted across 12 centers from 2012 to 2021. The study period was divided into two distinct intervals: period 1 (2012–2016) and period 2 (2017–2021), to evaluate trends in pediatric ECPR outcomes.
Patient Population and Definitions
The study involved pediatric patients (<18 years) who underwent ECPR, which is defined as the initiation of ECMO during CPR for sustained cardiac arrest or a transient return of spontaneous circulation that lasted <20 minutes [10,11]. The collected data included demographic characteristics, comorbidities, and clinical variables.
Primary ECMO indications were categorized into respiratory, post-cardiotomy, and cardiac-medical categories. Post-cardiotomy ECMO was initiated for various conditions, including preoperative stabilization, weaning failure from cardiopulmonary bypass, postoperative low cardiac output syndrome, and cardiopulmonary arrest. These cases were further stratified according to the Society of Thoracic Surgeons-European Association for Cardio-Thoracic Surgery (STAT) Congenital Heart Surgery mortality categories, scored from 1 (low mortality risk) to 5 (high mortality risk) [12]. The cardiac-medical ECMO group included cases of heart failure unrelated to post-cardiotomy status, such as acute myocarditis, cardiomyopathy, and sudden cardiac arrest of unknown origin, as well as pulmonary hypertension and sepsis. Neonatal ECMO is defined as cases involving patients who are 28 days old or younger at the time of cannulation. Central ECMO access involves line placement through a sternotomy, whereas peripheral access uses vessels in the neck or femoral region.
Outcome Variables
The primary outcomes of the study were survival to hospital discharge and successful ECMO weaning, which was defined as survival for at least 48 hours after weaning. Secondary outcomes included the incidence of complications, such as intracranial hemorrhage, and the causes of mortality. The causes of death were categorized into four groups: (1) incurable primary diseases necessitating ECMO; (2) complications including cerebral hemorrhage, other significant hemorrhages, cerebral infarction, or sepsis; (3) hypoxic brain damage or cerebral edema resulting from conditions prior to ECMO; and (4) causes not related to ECMO.
Statistical Analysis
Statistical analyses were conducted using IBM SPSS version 28 (IBM Corp.). Continuous variables were reported as medians with interquartile ranges, while categorical variables were examined using the chi-square test. Logistic regression models were used to identify factors associated with mortality, with findings reported as odds ratio (OR) and 95% CI. Statistical significance was established at p<0.05.
RESULTS
Patient Characteristics and Indications
Over the 10-year study period, 206 ECPR runs, accounting for 27.3% of the 754 ECMO cases, were performed (Table 1). Of these, neonatal CPR was involved in 24.6% (n=61) and pediatric CPR in 28.7% (n=145) of the cases. The median age of the patients was 0.3 years, with a median ECMO duration of 7 days. The median duration of CPR prior to ECMO initiation was 60 minutes. Central ECMO access was utilized in 67% of the cases. Cardiac indications were predominant, comprising 93.2% of the cases, with post-cardiotomy ECMO (53.9%) and cardiac-medical ECMO (39.3%) being the most prevalent categories. The highest ECPR rates were noted in cardiac-medical ECMO, representing 52.9% of neonatal ECMO and 36.2% of pediatric ECMO cases.
ECMO Utilization and Trends
The number of annual ECPR cases declined over time, with 114 instances in period 1 and 92 in period 2. The percentage of post-cardiotomy ECMO cases decreased from 58.8% in period 1 to 47.8% in period 2. Conversely, the incidence of cardiac-medical ECMO rose from 32.5% to 47.8% during the same periods (Table 2).
Survival Outcomes
The overall survival rate for hospital discharge was 31.1%, showing an increase from 28.1% in period 1 to 34.8% in period 2; however, this improvement was not statistically significant (p=0.301) (Table 2). Neonatal ECMO, which accounted for 29.6% of the cases, had lower survival rates compared to pediatric ECMO (21.3% vs. 35.2%, p=0.459). Survival rates for post-cardiotomy ECPR significantly improved from 23.9% in period 1 to 43.2% in period 2 (p=0.032). Among individual diseases, hypoplastic left heart syndrome had the lowest survival rate, with only 11.8% (2/17) of patients surviving.
Successful weaning from ECMO occurred in 38.8% (n=80) of cases; however, 23.8% (n=19) of these patients subsequently died. Among the 142 non-survivors, 62.7% (n=89) died due to an incurable primary disease, while 17.6% (n=25) succumbed to hypoxic brain injury or cerebral edema related to conditions present before ECMO initiation. Other causes of death included intracranial hemorrhage (n=13), extracranial hemorrhage (n=7), cerebral infarction (n=2), and sepsis (n=1). Patients who died from hypoxic brain injury or cerebral edema experienced longer CPR durations prior to ECMO initiation, although the difference was not statistically significant (80 minutes vs. 86 minutes, p=0.661).
Comparisons between Survivors and Non-survivors
Table 3 illustrates the differences between survivors and non-survivors. Survivors were generally older than non-survivors, although the differences in age and neonatal status did not reach statistical significance. However, survivors experienced significantly shorter durations of ECMO (p=0.038) and CPR times prior to ECMO initiation (p=0.018). Survival rates were significantly higher when the duration of CPR before ECMO initiation was ≤30 minutes (47.4% vs. 27.4%, p=0.016; OR, 2.39; 95% CI, 1.160–4.911) or ≤60 minutes (38.0% vs. 23.4%, p=0.025; OR, 2.00; 95% CI, 1.087–3.664). Dialysis was required in 61.2% of cases, most frequently in those undergoing post-cardiotomy ECMO. In this group, the need for dialysis due to acute kidney injury (AKI) was significantly associated with higher mortality (p<0.001).
DISCUSSION
This multicenter retrospective study provides a comprehensive analysis of pediatric ECPR, drawing on data from 12 major centers collected over a decade. It identifies trends in the utilization of ECPR, evaluates survival rates, and examines factors influencing mortality. These findings offer valuable insights that can help advance pediatric ECPR practices and optimize treatment strategies. Compared to the ELSO registry data, previous research on pediatric ECMO utilization in Korea has shown distinctive patterns, including higher use for cardiac conditions and lower use in neonates [9]. Similarly, this study identified cardiac indications as the primary reason for ECPR. Despite the global increase in pediatric ECPR usage, this study noted a declining trend in annual ECPR cases, primarily due to reductions in post-cardiotomy ECMO cases, which constituted the largest proportion of this cohort. This decrease likely reflects a shrinking pediatric population and fewer cases of complex congenital heart disease, attributed to declining birth rates over the past decade [13-16]. However, as this study is limited to data from 12 centers and covers a specific time period, further research using a nationwide dataset is necessary to confirm these trends and provide a more comprehensive understanding.
Survival rates for pediatric ECPR in this study remain low, especially among neonates (21.3% compared to 44.3% in the ELSO registry) and children (35.2% compared to 41.1%). However, there was a significant improvement in post-cardiotomy ECPR survival between period 1 and period 2, with rates increasing from 23.9% to 43.2%. This positive trend is likely due to advancements in the surgical management of congenital heart defects in Korea [16,17], as well as enhanced perioperative care, refined ECMO protocols, and improved post-arrest intensive care practices.
ECPR is associated with an increased risk of mortality, influenced by factors such as non-cardiac diagnoses, preexisting renal failure, prolonged time from CPR to ECMO initiation, interruptions during CPR, and complications like AKI [18-20]. In this study, STAT mortality categories 4 and 5, which represent the highest mortality risk, were associated with increased mortality risk. Factors contributing to this include the need for dialysis due to AKI and longer CPR durations before ECMO initiation. The duration of pre-ECMO resuscitation, the severity of the primary disease, and clinical conditions before and after arrest were critical determinants of survival. This underscores the importance of timely ECMO initiation and optimized post-resuscitation care.
Delayed initiation of ECMO following cardiac arrest significantly increases the risk of brain damage, multi-organ failure, and mortality [18,19,21,22]. In this study, hypoxic brain injury was responsible for 17.6% of the deaths among ECPR patients, which is a notable increase from the 6.7% observed in pediatric ECMO patients in prior studies [9]. Although statistical significance was not reached, these findings likely reflect the consequences of prolonged cardiac arrest on brain injury.
Survival outcomes are closely linked to the quality of conventional CPR, ECPR practices, and post-cardiac arrest care. The 2021 ELSO guidelines for pediatric ECPR offer a comprehensive management framework, emphasizing the need for a highly coordinated team of specialized professionals to ensure rapid and effective implementation [4]. However, it is necessary to adapt these guidelines to the unique demographic and clinical characteristics of Korean healthcare. Variations in ECMO access and ECPR timing underscore the importance of developing tailored protocols and focusing on continuous quality improvement. Additionally, significant challenges within the Korean healthcare system, such as the limited availability of ECMO machines, circuits, and catheters specifically designed for neonates and small children, need to be addressed as a priority.
This study has several limitations. Its retrospective design and reliance on medical records from multiple centers introduce the potential for information bias. Additionally, the relatively small sample size of 206 cases and the inclusion of only 12 centers further limit the generalizability of the findings and reduce statistical robustness.
In addition, the study relied solely on subgroup data related to ECPR from previous research, leaving significant gaps in critical information. Notably, it fails to distinguish between in-hospital cardiac arrest (IHCA) and out-of-hospital cardiac arrest (OHCA), as well as the specific locations where ECPR was administered, such as the emergency room, intensive care unit, or operating room. This distinction is crucial because pediatric IHCA generally has a better prognosis than OHCA, with survival rates of 41.1% versus 11.4%, respectively [23]. The improved outcomes for IHCA can likely be attributed to the immediate access to care in controlled settings like ICUs, whereas OHCA often involves asphyxial arrest and delayed initiation of CPR [24,25].
Although speculative, the respiratory and post-cardiotomy ECPR cases in this study likely represent instances of IHCA, while the cardiac-medical cases may include a mix of IHCA and OHCA. However, considering the current medical infrastructure in Korea, it is presumed that the majority of cases are IHCA, as performing pediatric ECPR in emergency rooms poses significant challenges. The absence of such critical data significantly limits a comprehensive understanding of ECPR outcomes.
Furthermore, the absence of neurological outcome assessments prevents the evaluation of long-term prognosis and the overall impact of ECPR on recovery. Future research should address these limitations by conducting nationwide, prospective studies that investigate risk factors, prognostic indicators, and long-term outcomes. These studies should focus on differentiating between IHCA and OHCA and examining the effects of the ECPR location on outcomes.
In conclusion, this study underscores significant trends, survival outcomes, and risk factors linked to pediatric ECPR. Prompt initiation of ECMO, enhanced CPR quality, and optimized management during resuscitation are crucial for improving survival rates. There is a need for tailored guidelines and additional research to refine practices in pediatric ECPR and enhance patient outcomes.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation
