Share this article on:

Extracorporeal Life Support Registry Report 2008: Neonatal and Pediatric Cardiac Cases

Haines, Nikkole M.*; Rycus, Peter T.; Zwischenberger, Joseph B.; Bartlett, Robert H.; Ündar, Akif

doi: 10.1097/MAT.0b013e318190b6f7
Pediatric Circulatory Support

Each year thousands of patients require extracorporeal life support (ECLS) for a variety of respiratory, cardiac, and emergency reasons. The ECLS registry, a Federal Drug Administration approved control group, provides a database of approximately 37,000 ECLS patients from domestic and international hospitals, with details about demographic factors, diagnosis, treatment, and complications. The report is circulated to Extracorporeal Life Support Organization members with the goal of providing feedback on each center’s practices relative to the performance of all of the centers, and to the general public with the goal of enhancing research and improving patient care. In this report, the ECLS Registry Report International Summary for July 2008 is analyzed with a specific focus on neonatal and pediatric cardiac patients. From the data, it is evident that cardiac ECLS patients have a lower chance of survival than respiratory ECLS patients, and furthermore, younger patients have a lower chance of survival. Requirement for inotropes while on ECLS is by far the most common complication, followed by surgical site bleeding, necessitating hemofiltration. The major mechanical complications facing cardiac ECLS patients are oxygenator failure and clots in the circuit. Databases such as this one provide powerful tools for institutions, clinicians, and researchers.

From the *Department of Pediatrics, Pediatric Cardiac Research Laboratories, Hershey, Pennsylvania; †Extracorporeal Life Support Organization, University of Michigan Medical Center, Ann Arbor, Michigan; ‡Department of Surgery, University of Kentucky, Lexington, Kentucky; and §Department of Surgery and Bioengineering, Penn State Milton S. Hershey Medical Center, Penn State College of Medicine, Penn State Hershey Children’s Hospital, Hershey, Pennsylvania.

Submitted for consideration September 2008; accepted for publication in revised form September 2008.

Reprint Requests: Akif Ündar, PhD, Penn State College of Medicine, Department of Pediatrics – H085, 500 University Drive, PO Box 850, Hershey, PA 17033-0850. Email: aundar@psu.edu.

The Extracorporeal Life Support Organization (ELSO) serves to enhance the care of patients receiving extracorporeal life support (ECLS). ELSO provides quality assurance testing for ECLS centers, scientist’s large amounts of data from each ECLS patient, which furnishes treatment guidelines, training sessions for professionals and laymen, and enhances communication between ECLS centers.1,2 Each ECLS patient is registered into the database and classified based on age and etiologic reason for ECLS implementation to make three groups within each age group: respiratory, cardiac, and extracorporeal cardiopulmonary resuscitation. The registry report presents data on overall patient outcomes, number of cases, patient diagnoses, surgery types, procedure types, support modes, and complications. In terms of clinical research, the ECLS registry is a Federal Drug Administration approved control group and the large sample size of this database provides immense statistical power.

As of July 2008, 115 active centers comprised the ELSO membership and 37,717 patient cases have been recorded in the registry from over 170 centers within and outside of the United States.3 The neonatal (<31 days of age) and pediatric (under 18 years of age) age groups totaled 35,030 registered patients as of July 2008.

In this article, we will focus predominantly on neonatal and pediatric cardiac cases. Brown and Goldman4 noted that recent years have shown increasing numbers of cardiac ECLS patients and decreasing numbers of respiratory ECLS patients. This could be associated with increased access to ECLS centers, and more complex and diverse uses for ECLS with regard to cardiac cases, and improved conventional and innovative neonatal care with regard to respiratory ECLS cases.2,4,5 Furthermore, as more high-risk patients, especially neonatal patients, are treated with complex surgeries, ECLS durations have the potential to increase because these patients may have poorer cardiac and pulmonic functions after surgery.5 This article focuses on recent trends in ECLS with regard to survival, duration, support mode, diagnosis, procedures, surgeries, complications, and age.

Back to Top | Article Outline

Methods

The ELSO registry began collecting information on ECLS patients in 1989.2 Since then, the database has expanded to include data from 170 institutions, both in United States and abroad, and currently the ELSO has 115 active members.3 Each active institution provides information on patients with only their birth date and dates of treatment as identifiers.

The July 2008 ECLS Registry Report International Summary was analyzed with focus on neonatal and pediatric cardiac ECLS cases. A patient is classified as a cardiac case if cardiac dysfunction is the primary reason for which support is implemented.1 For the purposes of this article, age group 1 includes patients age 0–30 days, patients age 31 days to less than 1-year are encompassed in age group 2, and patients age 1 year to <16 year are classified in age group 3.

Back to Top | Article Outline

Results

Overall Outcomes

In total, there were 3,416 neonatal cardiac ECLS cases and 4,181 pediatric cardiac cases as of July 2008, numbers much reduced when compared with the cumulative number of respiratory ECLS cases in the same age groups. In recent years, the registry report has shown a downward trend in the number of respiratory ECLS cases and an upward trend in the number of cardiac cases for all age groups (Figure 1). Importantly, a lower percentage of cardiac ECLS neonatal and pediatric patients survived ECLS when compared with respiratory patients in the same age groups. Also, cardiac neonatal and pediatric patients had lower survival to discharge or transfer percentages than respiratory patients (in neonatal patients, 38% compared to 76%, for cardiac and respiratory cases, respectively; in pediatric patients 45% compared to 55%, respectively).

Figure 1.

Figure 1.

Back to Top | Article Outline

Cases by Age

Patients 0–30 days old, age group 1, were the major population receiving cardiac ECLS with a count of 326 cases in 2007 compared to counts of 164 and 143 in patients 31 days to less than 1 year of age, age group 2, and 1 year to <16 years of age, age group 3, respectively. In 2007, the percent survival of patients <16 years of age improved compared with 2006. The lowest survival rate in 2007 occurred in the youngest patients at 40%, compared with 49% in age group 2, and 59% in patients in age group 3. As of July 2008, the improvement in percent survival echoed only in age group 1, at 44%. The average run time in 2007 for patients in age group 2 was the highest ever recorded at 180 hours, whereas the longest run time was not close to the maximum in this age group. The 2007 average run times for age groups 1 and 3 were comparable to previous years. This data is presented in Table 1.

Table 1

Table 1

Back to Top | Article Outline

Cases by Diagnosis

Looking at cumulative data shown in Table 2, the most common diagnosis in cardiac ECLS patients was a congenital heart defect (86.31% of patients in age group 1, 79.06% of patients in age group 2, and 49.97% of patients in age group 3). The second leading diagnosis was “other” (which increased in representation as patients got older), followed by cardiomyopathy (2.55% of patients in age group 1, 4.46% of patients in age group 2, and 15.35% of patients in age group 3).

Table 2

Table 2

For patients in age group 1, the longest average run time occurred in patients with myocarditis and for age groups 2 and 3 the longest average run time was in patients with cardiomyopathy, with myocarditis patients coming a close second.

Patients in age group 1 had a lower percent survival than older patients with the same diagnosis. When comparing survival between patients in age groups 2 and 3 with the same diagnosis the percent survivals were similar. In age group 1, the patients with cardiomyopathy had the highest survival at 62%, whereas those with cardiac arrest had the lowest survival at 23%. In the age group 2, the patients with cardiac arrest and cardiogenic shock had the lowest survival at 40%, whereas those with myocarditis had the highest survival at 67%. In age group 3, the patients with myocarditis also had the highest percent survival at 66% and patients with cardiac arrest had the lowest survival at 38%.

Because of the large percentage of patients diagnosed with a congenital heart defect and multiple possible defects with varying prognostic factors, the registry distinguishes between types of congenital heart defects. In all patients under 16 years of age, the majority of congenital cardiac diagnoses were classified as other. In patients in age group 1, most patients (25.21%) were classified as other, but 24.02% were diagnosed with a hypoplastic left heart, and 14.87% were diagnosed as cyanotic with increasing pulmonary congestion. In patients in age group 2, the other congenital defect diagnosis fit 37.55% of patients, whereas a left to right shunt diagnosis encompassed 23.85% of patients and a hypoplastic left heart diagnosis included 10.09% of patients. In patients in age group 3, the other category fit 38.79% of patients with congenital heart defects, whereas patients with left to right shunts, at 19.14%, and cyanotic patients with decreased pulmonary flow, at 13.82%, also made up a large portion of the diagnoses. In patients under a year old (age groups 1 and 2), the shortest average run time occurred in patients with a diagnosis of a right-sided obstructive defect. In patients in age group 3, the shortest average run time occurred in patients diagnosed as cyanotic with increasing pulmonary congestion. The longest average run time occurred in cyanotic patients with decreasing pulmonary flow in age group 1 at 154 hours, in age group 2 patients with a left to right shunt and cyanotic patients with decreasing pulmonary flow at 153 hours, and in patients with a hypoplastic left heart at 151 hours in age group 3. In age group 1, cyanotic patients with increasing pulmonary congestion were the most likely to survive with a 44% survival, whereas those with a hypoplastic left heart were the least likely to survive with a percent survival of 28%. In age group 2, cyanotic patients with increasing pulmonary flow and those listed as other had the highest percent survival at 47%, whereas those with a hypoplastic left heart again had the lowest at 36%. Patients in age group 3 with a left-sided obstructive defect had the highest percent survival at 52%, and cyanotic patients with increased pulmonary flow had the lowest at 13%, but this diagnostic group had a small sample size comparatively (n = 8).

Back to Top | Article Outline

Cases by Surgical Type

The cumulative data presented in Table 3 shows that for children in all age groups, the most common surgery leading to cardiac ECLS was recorded as “other postoperative and not bridged to transplant,” with 2,727 cases or 75.58% of patients in age group 1. The second most common event that led cardiac ECLS in children <16 years of age were events unrelated to surgery and in patients not bridged to transplant, with 758 cases or 21.01% in patients in age group 1.

Table 3

Table 3

Patients recorded as “other postoperative and bridged to transplant” had the longest average run time for all age groups. The shortest average run time in all patients under 16 years of age occurred in patients classified as other postoperative and not bridged to transplant.

In age group 1, the highest percent survival, at 41%, occurred in patients not undergoing surgery and not bridged to transplant, whereas the lowest percent survival in the this age group, at 32%, was in patients undergoing cardiac transplant. In age group 2, patients who did not undergo a previous surgery but were bridged to transplant had the highest survival percentage at 53%, whereas those classified as other postoperative and bridged to transplant had the lowest survival percentage at 37%. Cardiac transplant patients in age group 3 had the highest percent survival at 56%, whereas those recorded as other postoperative and not bridged to transplant had the lowest at 44%. With all surgery types the older the patient population, the higher the percent survival.

Back to Top | Article Outline

Cases by Procedure

Because the majority of cardiac ECLS cases receive the diagnosis of a congenital heart defect, the type of procedure preformed to alleviate the congenital malformation is also recorded in the registry. The most common three procedures preformed in cardiac ECLS patients in age group 1 were Stage I Norwood procedures with 707 procedures, 691 unlisted cardiac procedures, and arterial switches with 317 procedures. Stage I Norwood procedures had a 30% survival. Patients undergoing an Arterial switch had a survival of 43%. Thirty-seven percent of patients survived “unlisted cardiac” procedures.

In age group 2, the most common procedures were unlisted cardiac procedures with 573 procedures, tetralogy repair with 170 procedures, and atrioventricular (AV) canal repair with 117 procedures. In this age group, the percent survival for unlisted cardiac procedures was 38%, whereas 51% of tetralogy repair patients survived, and AV canal repair patients had a 49% survival.

In age group 3, unlisted cardiac procedures were most common with 367 procedures, heart transplants were the second most common with 209 procedures, and the Simple Fontan was third with 92 procedures. Unlisted cardiac procedures had a survival of 39%, 56% of patients survived their heart transplant, and patients undergoing a Simple Fontan procedure had a survival of 30%.

Back to Top | Article Outline

Support Mode

In the past year, most cardiac patients in these three age groups received veno-arterial (VA) support, whereas a small percentage received VA with retrograde venous drainage (VA+V). This is shown in Figure 2. If looking at the cumulative picture, VA is the most common support mode, with VA+V the second most common. In age group 1, 92.16% of patients received VA support, whereas 4.49% of patients received VA+V support. VA supported 89.90% of patients in age group 2 and 7.20% of patients in this group received VA+V support. In age group 3, 86.55% of patients received VA support, whereas 10.64% received VA+V support.

Figure 2.

Figure 2.

Patients receiving VA support had percent survivals of 37, 43, and 47% for patients in age groups 1, 2, and 3, respectively. Patients receiving VA+V supports had survival percentages of 40, 46, and 55% for patients in age groups 1, 2, and 3, respectively.

Back to Top | Article Outline

Complications

ECLS patients suffer a long list of complications, as illustrated by an abbreviated list in Table 4. Patients necessitating inotropes while on ECLS (classified as a cardiovascular complication) was the most common complication in patients of all three age groups. A total of 61.3% of patients in age group 1, 63.6% of patients in age group 2, and 62.4% of patients in age group 3 reported this complication. Survival with this complication was better the older the patient, with 33% survival in age group 1, 39% in age group 2, and 43% in age group 3. In all age groups, the second most common complication in cardiac ECLS patients was surgical site bleeding, and the third most common complication was obligatory renal hemofiltration. The survival percentages with these complications also increased with age.

Table 4

Table 4

The second most common cardiovascular complication associated with cardiac ECLS patients was cardiac arrhythmia, which occurred in 15% of patients in age group 1, 18.2% of patients in age group 2, and 21.2% of patients in age group 3. The percent survival for these patients was 26, 37, and 38% for age groups 1, 2, and 3, respectively.

The most common mechanical complications in cardiac ECLS patients occurred in oxygenators, either failure or clots. In age group 1, the most common mechanical problem was clots in other circuit locations, the second most common was clots in the oxygenator and tied for the third most common were oxygenator failure and clots in the bladder. Oxygenator issues created 18.3% of the problems in this age group. The most common mechanical complication in age group 2 was oxygenator failure, the second was clots in other locations, and the third was oxygenator clots. Therefore, the oxygenator caused 15% of the problems in this age group. The most common mechanical problem in patients in age group 3 was oxygenator failure. Clots in other locations and cannula problems were the third most common and oxygenator clots were the fifth most common mechanical problem in this age group. In total, the oxygenator created 16% of the complications for age group 3.

Back to Top | Article Outline

Discussion

The ECLS registry report is an invaluable resource to active centers who are ELSO members as well as other centers that do not actively participate but still treat patients with ECLS. For active centers the registry provides a direct evaluation of their achievement compared to many other centers. For other institutions it can provide information on complications that must be addressed in different age groups, which patients are high risk, and prognostic values, such as average run time, which physician can use when delivering news to parents.

From the July 2008 International Summary, it is evident from survival percentages that cardiac ECLS patients are high-risk patients. Furthermore, the younger the patients, for the three age groups we looked at, the lower the percent survival. Patients in these age groups are most likely affected by congenital defects, especially in the youngest patients. Currently, for all three age groups the major support mode is VA, whereas VA+V is used for almost all of the remaining cases. Patients in all three age groups were likely to require inotropes while on ECLS, and the majority of patients in these age groups do. The second most common adverse event was surgical site bleeding. In terms of mechanical complications, oxygenator function played a large role. Oxygenator failure rates may be related to long run times that occur as a result of complexity cardiac cases.6 Clots throughout the circuit play a large role in mechanical complications, but management of these must be carefully weighed against the risk of bleeding, a large problem previously mentioned.

As techniques improve, new complications and risks will arise. With advancement always occurring, treating sicker patients with more complex cases will change the patient demographic. This will alter the picture of ECLS and the complications associated with it.5 This database provides a powerful up-to-date tool that will help ensure that patients receive the best care possible.

Back to Top | Article Outline

References

1. Conrad SA, Rycus PT, Dalton H: Extracorporeal life support registry report 2004. ASAIO J 51: 4–10, 2005.
2. Bartlett RH: Extracorporeal life support registry report 1995. ASAIO J 43: 104–107, 1997.
3. ELSO Membership Information. Available at: http://www.elso.med.umich.edu/Membership.htm. Accessed July 28, 2008.
4. Brown KL, Goldman AP: Neonatal extra-corporeal life support: Indications and limitations. Early Hum Dev 84: 143–148, 2008.
5. Roy BJ, Rycus P, Conrad SA, Clark RH: The changing demographics of neonatal extracorporeal membrane oxygenation patients reported to the Extracorporeal Life Support Organization (ELSO) Registry. Pediatrics 106: 1334–1338, 2000.
6. Tracy TF, DeLosh T, Bartlett RH: Extracorporeal Life Support Organization 1994. ASAIO J 40: 1017–1019, 1994.
Copyright © 2009 by the American Society for Artificial Internal Organs