HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Bhagawan Koirala Brian McCrindle John Coles Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chow, G. Articles by de Veber, G. Search for Related Content PubMed PubMed Citation Articles by Chow, G. Articles by de Veber, G. Related Collections Cerebral protection Extracorporeal circulation

Eur J Cardiothorac Surg 2004;26:38-43
© 2004 Elsevier Science NL

Predictors of mortality and neurological morbidity in children undergoing extracorporeal life support for cardiac disease

^a Division of Neurology, The Hospital for Sick Children, University of Toronto, Toronto, Ont., Canada
^b Division of Cardiovascular Surgery, The Hospital for Sick Children, University of Toronto, Toronto, Ont., Canada
^c Division of Neuroradiology, The Hospital for Sick Children, University of Toronto, Toronto, Ont., Canada
^d Division of Cardiology, The Hospital for Sick Children, University of Toronto, Toronto, Ont., Canada
^e Department of Critical Care Medicine, The Hospital for Sick Children, University of Toronto, Toronto, Ont., Canada
^f Department of Cardiovascular Perfusion, The Hospital for Sick Children, University of Toronto, Toronto, Ont., Canada

Received 8 September 2003; received in revised form 8 April 2004; accepted 9 April 2004.

^* Corresponding author. Address: University Hospital, Queen's Medical Centre, Nottingham, NG7 2UH, UK. Tel.: +44-115-924-9924x43328; fax: +44-115-970-9763
e-mail: gcschow{at}hotmail.com

	Abstract

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 
Objectives: The objective of this study was to determine the incidence and risk factors for death and adverse neurological outcomes in children receiving extracorporeal life support (ECLS) for cardiac indications. Methods: A retrospective single centre consecutive cohort study was conducted in children who received ECLS for cardiac indications between January 1990 and June 2000. Health records and neuroimaging films were assessed, and long-term outcomes were obtained by standardized telephone follow-up or by assessments performed in outpatient clinic. Clinical, neuroimaging and surgical predictors of outcome were tested. Results: Of 90 children studied, short-term clinical neurological events (during hospitalization) occurred in 20 children (22%) during or following ECLS. Long-term neurological sequelae were present in 11 of 31 children discharged alive, after a mean follow-up interval of 4.5 years (range 4 months to 9 years). Death occurred in 59 children (66%) during hospitalisation, and in 3 following discharge. Of the 28 long-term survivors, only 15 children (17%) survived without neurological sequelae. Abnormal neuroimaging was associated with short-term neurological events (P=0.03, OR 10.5), and the use of CPR prior to ECLS (P=0.02, OR 2.9) was the only significant predictor of death. There were no significant predictors of long-term neurological sequelae. Conclusions: More than two-thirds of the children receiving ECLS died, and 39% (11/28) of long-term survivors had neurological deficits. Although mortality is close to 100% without this type of support, there is still a significantly high morbidity and mortality with this type of support.

Key Words: Extracorporeal life support • Predictors • Neurological sequelae

	1. Introduction

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 
Extracorporeal Life Support (ECLS) provides circulatory support during periods of severe myocardial dysfunction. There is now increasing literature regarding neurological sequelae post ECLS in children [1–4]. In neonates and older children receiving ECLS for cardiac conditions, adverse outcomes may be even greater with reports of short-term neurological morbidity in 28–44% and death in 33–61% for neonates and older children [5–7]. The long-term neurological outcomes and predictors of outcome after ECLS have not been established.

The aim of our study was to determine the incidence and risk factors for adverse neurological outcomes and death in children receiving extracorporeal life support (ECLS) for cardiac indications.

	2. Methods

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 
This was a single centre consecutive cohort, outcome study using standardized assessment and testing for outcome predictors. Research ethics board approval was obtained prior to initiation of the study. The study included all children who received ECLS at the Hospital for Sick Children, Toronto, Canada between January 1990 and June 2000 for underlying cardiac conditions.

Patients were identified from the ECLS register. Detailed clinical, surgical and radiological data were obtained including the health record charts, operative records, perfusion and pediatric intensive care unit data and autopsy reports.

2.1. ECLS
ECLS consisted of a Biomedicus Portable Bypass System (Medtronic Canada) incorporating a centrifugal pump console and a water pump system. Circuits were either uncoated or Carmeda (Medtronic) coated. Oxygenators used were Avecor Silicone membrane lungs or Carmeda coated Minimax oxygenator/heat exchangers (Medtronic). The heat exchanger used in the standard circuit was a Dideco Heliox (Sorin Biomedica). The patient was heparinized with a loading dose of 50–75 units per kg. An ACT of 160–180 s was desirable. Cannulation was either through the right side internal jugular vein and carotid artery or direct thoracic cannulation of the aorta and right atrium if required. Flows in the range of 100–150 ml/kg per min for patients under ten kilograms were desirable, or 2.4–3.2 l/min per m² are ideal for patients greater than 10 kg. Femoral cannulation was also used in some cases. Alpha stat blood gas management was employed routinely.

Echocardiographic evidence of myocardial improvement is the most important predictor of successful weaning from ECLS. Under direct vision the heart was assessed with reduced ECLS flows and restored pharmacologic support. ECLS was discontinued by gradually opening the bridge and clamping the inflow and outflow lines to the patient. The cannulae were left in place until haemodynamic stability is achieved. Intermittent flushing and transfer of the heparin infusion to the patient reduced the risk of clot formation in the cannulae. ECLS was terminated if there is no return of cardiac function and when the option of cardiac transplantation does not remain. When there was evidence of multi organ failure, overwhelming sepsis or profound neurological impact, termination of ECLS was considered. Norwood patients were not put on ECLS electively, but only for acute circulatory collapse, and therefore required a short period of full support followed by a left venticular assist device if needed.

2.2. Neuroimaging and autopsies
All the original neuroimaging was evaluated by the study neuroradiologist (D.A.) who was blinded to the patients' clinical details and outcomes. For cranial ultrasound studies, clinically generated reports were reviewed. Autopsy reports of the brain were classified as for neuroimaging data as focal or diffuse parenchymal abnormalities or extra-parenchymal haemorrhage.

2.3. Outcome assessment
Short-term neurological events and deaths were determined from standardized review of the health record in all cases. Short-term neurological events during hospitalization consisted of seizures, and the following as previously reported by Fallon et al. [8]:

• Pyramidal signs (hemiparesis, quadriparesis)
  
• Extrapyramidal signs (dystonia, chorea)
  
• Coma (including fluctuating level of consciousness)
  
• Neuro-ophthalmic defects (gaze palsies and visual field defects)
  
• Miscellaneous (e.g. brachial plexus palsy)
  

The long-term outcome in patients surviving to discharge was determined by contacting families of survivors and conducting standardized questionnaire by telephone in 19 patients, in cardiac outpatient clinic in 5 patients, and for 1 patient each in neurodevelopmental clinic, and during elective hospital admission. Sequelae were divided into motor impairment, cognitive/behavioural impairment, other impairments, and seizures. Standardized questions were asked.

2.4. Outcome predictors
The association between the putative predictors and three outcomes (presence or absence of short-term neurological events, late neurological sequelae and death) were sought using ² and Fisher's exact tests (Table 1) . Logistic regression was used to analyse factors associated with short-term events Cox regression analysis was used to analyse factors associated with long-term neurological sequelae and with survival. Mann–Whitney test was used to assess associations with age. The Kruskal–Wallis test was used to compare the duration of ECLS for non surgical patients or pre-operative, peri-operative or post-operative patients. All analyses were performed using SPSS^TM statistical software Version 10.

	3. Results

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

3.2. Cardiac diseases: indications for ECLS
The patients had either:

(1) Primary myocardial failure from myocarditis (n=10) or cardiomyopathy (n=9), or

(2) Underlying structural congenital heart disease (n=71), 58 had biventricular repairs, and 7 univentricular repairs, 6 did not have surgical intervention (Table 2) .

Of the children with underlying structural congenital heart disease, 40 children were unable to wean off cardiopulmonary bypass (CPB) after surgery, so proceeded to ECLS. Three patients used ECLS as a bridge to heart transplantation. 8 patients had multiple ECMO runs, 7 patients had 2 runs and 1 patient had 3 ECMO runs.

3.3. CPB and ECLS data
The duration of ECLS was 6–394 h, median 90 h. In the 70 children who had received CPB, in addition to ECLS, cardiopulmonary bypass time was 40–554 min, median 169 min.

3.4. Other interventions
Prior to ECLS, 34 children (37%) received CPR for cardiac arrest. Twenty-five of the children subsequently died. While on ECLS, 13 children received heart transplantation (2 died), and 13 had intracardiac repair (11 died).

3.5. Neuroimaging and autopsy results
Eight children had imaging before ECLS, 1 had a subdural haemorrhage and 1 an infarct. Six of these 8 children died, 2 of whom had imaging abnormalities as above. Thirty-eight children underwent neuroimaging during or after ECLS (35 ultrasounds (US), 17 CT and 2 MRI scans). Twenty-four had abnormalities, (9 parenchymal or extraparenchymal haemorrhages, 4 infarcts, 7 ventriculomegaly, and 4 with atrophy. None had hydrocephalus.

In 11 children who died, brain autopsy findings were available and consisted of 7 haemorrhages, 3 had infarcts, and 1 had cerebellar tonsillar herniation. Only 1 child who died had both neuro-imaging and autopsy.

3.6. Outcomes
Of the 90 children included in the study, outcomes were available for 88 children. For survivors, the mean interval from ECLS to long-term outcome assessment was 4.5 years (range 4 months to 9 years). A summary of the overall outcomes is presented in Fig. 1 .

View larger version (12K): [in this window] [in a new window]   Fig. 1. Chart showing the outcome of all the children studied.

Overall 15/90 children (17% of the total in the study) were neurologically normal, and 11 had long-term neurological sequelae. Overall 39% (11/28) of long-term survivors had long-term neurological sequelae.

Pre ECLS, neurological examinations were performed on 35 children. Twenty-seven were normal, 5 had developmental delay, 2 had epilepsy and 1 was hypotonic. Of the 27 children with previously normal examinations, 14 died, of those with epilepsy 1 died, 4 with developmental delay died and the child with hypotonia survived with no sequelae.

During the study interval, 62 children died (69%). In 45 children, deaths occurred during or within 24 h of ECLS. A survival plot is shown in Fig. 2 . In all these children, ECLS was discontinued as part of the decision to withdraw life support due to terminal cardiovascular failure with a hopeless prognosis. Nine children received ECLS >208 h, of these, 5 died during hospitalisation, 3 were discharged home, and 1 died a late death. The mortality was 68% for those who had congenital heart defects versus 32% of those with myocarditis or cardiomyopathy (P=0.03). Late deaths after discharge occurred in 3 children, all had cardiac transplantation. One died of a lymphoma, one bowel obstruction, one of unknown cause.

View larger version (8K): [in this window] [in a new window]   Fig. 2. Survival plot.

3.7. Predictors of outcome
Univariate analysis was performed with short-term neurological events and mortality while in hospital as outcomes (see Table 1). The findings were that the presence of congenital heart disease (²=4.7, DF=1, P=0.03), previous CPR (²=5.4, DF=1, P=0.02), and abnormal neuroimaging (Fisher's Exact=0.02) predicted short-term neurological sequelae. The presence of congenital heart disease (²=4.7, DF=1, P=0.03) and the presence of short-term neurological events (Fisher's Exact=0.04) was associated with death.

Logistic regression identified that only abnormal neuroimaging was associated with short-term neurological events (P=0.03, OR 10.5, 95% CI 2.2–50.3). Non-significant factors adjusted for in the regression analysis were age, sex, previous ECMO, previous CPR, type of ventricular repair, and whether the child had congenital heart disease or myocarditis.

Cox regression was used to identify factors associated with long-term neurological sequelae and with long tem mortality. For both outcomes the following factors were entered into the equations: age, sex, previous ECMO, previous CPR, short-term neurological event, type of ventricular repair, abnormal neurorimaging during admission, and whether the child had congenital heart disease or myocarditis. Multivariate analysis was performed for the three outcomes. Lack of long-term survival was significantly predicted only by previous CPR (P<0.001, OR 2.6, 95% CI, 1.47–4.47). There were no significant predictors of long-term neurological sequelae.

	4. Discussion

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 
ECLS is used to wean off cardiopulmonary bypass due to poor cardiac output, hypoxaemia, or myocardial stunning. It is also used for cardiac failure associated with cardiomyopathy or myocarditis, or as a bridge to transplantation. Duration of ECLS of greater than 6–8 days [5,6] may predict increased mortality.

During the course of their illness and treatment, children undergoing ECLS are exposed to multiple risk factors for cerebral damage including: hypoxia, hypercapnia, poor tissue perfusion and acidosis, ligation of the carotid artery, and treatment with heparin [9]. In children with congenital heart disease, cerebral lesions can be acquired as part of the primary cardiac illness, a cardiovascular collapse, or cardiac surgery in addition to the ECLS. There is significant risk of renal, hepatic and neurological dysfunction for children on ECLS [10].

This study reports the outcome in 90 consecutive children undergoing ECLS for cardiac conditions. This was a retrospective study and so not all the children had detailed neurological examinations or neuroimaging prior to ECLS, which is an obvious limitation to our study. However, trying to define predictors of death or adverse neurological outcomes is important in order to design strategies to reduce these complications.

Previous reports of mortality with ECLS in children with congenital heart disease have ranged from 40–52% in hospital to 51–71% after discharge [5,10–13]. Our study had an in-hospital mortality of 66%, with a post discharge mortality of 3%.

The ECLS Registry International Summary of neonatal, pediatric and adult ECLS cases for cardiac conditions [7] has reported that 3.1% of children have central nervous system (CNS) infarction by US/CT and 3.5% have CNS haemorrhage by US/CT. These data are selected in that not all children underwent neuroimaging. Other studies have reported a frequency of 15% of neuroimaging abnormalities [3]. In our study, neuroimaging and autopsy reports revealed a 17% incidence of haemorrhage and an 8% incidence of infarction.

What was learnt from this study was that short-term neurological events were associated with abnormal neuroimaging. There were no predictors of long-term neurological sequelae. The presence of congenital heart disease prior to ECLS was a predictor of death in hospital. There was no clear association between aortic arch surgery or complex cyanotic heart disease and adverse sequelae.

However, the true frequency of short-term neurological events is likely to be greater than this since in our study, many severely compromised children were receiving medication in the intensive care unit which would sedate or paralyse the child and therefore potentially prevent clinical neurological signs from being apparent.

ECLS after unsuccessful weaning from open heart surgery has been associated with good results. Del Nido [14] studied 11 children who received ECLS for cardiac arrest after open heart surgery, and found that 7 survived (64%). Our results however showed that CPR prior to ECLS was a predictor for death and that this was therefore a very high risk group.

In one previous report [5] of 64 paediatric cardiac patients receiving ECLS, predictors of poor survival included a diagnosis of a single ventricle, starting ECLS >50 h after surgery, and a duration of ECLS of >208 h. In our study, the presence of a univentricular heart was not associated with a worse outcome, and one child survived ECLS of 346 h duration with no apparent neurological sequelae.

There is a high incidence of adverse neurological events associated with ECLS. In this study of 90 patients, only 28 (31%) children survived, and only 15 children without neurological sequelae. There were short-term neurological events in 22% (20/90) and long-term neurological sequelae in 12% (11/90), accounting for 39% (11/28) of survivors. We have learnt that the presence of abnormal neuroimaging is associated with short-term neurological events, and CPR prior to ECLS is a significant predictor of mortality. It is unlikely however that these factors are preventable. While ECLS is a life-saving intervention in children, given the significant neurological morbidity, prospective studies are needed which will further assess incidence, risk factors including the condition of the patient prior to ECLS, factors during cardiopulmonary bypass, and potential strategies to prevent these adverse outcomes.

We are indebted to Dr Jacqueline Collier, Professor of Health Services Research, Queen's Medical Centre, Nottingham for her assistance in statistical analysis of our data.

	References

Top Abstract 1. Introduction 2. Methods 3. Results 4. Discussion References

 

1. Desai S.A., Stanley C., Gringlas M., Merton D.A., Wolfson P.J., Needleman L., Graziani J., Baumgart S. Five-year follow-up of neonates with reconstructed right common carotid arteries after extracorporeal membrane oxygenation. J Pediatr 1999;1344:428-433.
2. Glass P., Wagner A.E., Papero P.H., Rajasingham S.R., Civitello L.A., Kjaer M.S., Coffman C.E., Getson P.R., Short B.L. Neurodevelopmental status at age five years of neonates treated with extracorporeal membrane oxygenation. J Pediatr 1995;127:447-457.[CrossRef][Medline]
3. Meliones J.N., Custer J.R., Snedecor S., Moler F.W., O'Rourke P.P., Delius R.E. Extracorporeal life support for cardiac assist in pediatric patients. Review of ELSO Registry data. Circulation 1991;84:168-172.[Abstract/Free Full Text]
4. Jarjour I.T., Ahdab-Barmada M. Cerebrovascular lesions in infants and children dying after extracorporeal membrane oxygenation. Pediatr Neurol 1994;10:13-19.[CrossRef][Medline]
5. Kulik T.J., Moler F.W., Palmisano J.M., Custer J.R., Mosca R.S., Bove E.L., Bartlett R.H. Outcome associated factors in the pediatric patients treated with extracorporeal membrane oxygenator after cardiac surgery. Circulation 1996;94:63-68.
6. Raithel S.C., Oennington G., Boegner E., Fiore A., Weber T.R. Extracorporeal membrane oxygenation in children after cardiac surgery. Circulation 1992;86:305-310.
7. ECMO Registry of the Extracorporeal Life Support Organization (ELSO). Ann Arbor, MI; 2000.
8. Fallon P., Aparicio J.M., Elliott M.J., Kirkham F.J. Incidence of neurological complications of surgery for congenital heart disease. Arch Dis Child 1995;72:418-422.[Abstract/Free Full Text]
9. Lago P., Rebsamen S., Clancy R.R., Pinto-Martin J., Kessler A., Zimmerman R., Schmelling D., Bernbaum J., Gerdes M., D'Agostino J.A., Baumgart S. MRI, MRA, and neurodevelopmental outcome following neonatal ECMO. Pediatr Neurol 1995;12:294-304.[CrossRef][Medline]
10. Black M.D., Coles J.G., Williams W.G., Rebeyka I.M., Trusler G.A., Bohn D., Gruenwald C., Freedom R.M. Determinants of success in pediatric cardiac patients undergoing extracorporeal membrane oxygenation. Ann Thorac Surg 1995;60:133-138.[Abstract/Free Full Text]
11. Jaggers J.J., Forbers J.M., Shah A.S., Meliones J.N., Kirshbom P.M., Miller C.E., Ungerleider R.M. Extracorporeal membrane oxygenation for infant postcardiotomy support: significance of shunt management. Ann Thorac Surg 2000;69:1476-1483.[Abstract/Free Full Text]
12. Jacobs J.P., Ojito J.W., McConaghey T.W., Boden B.D.C.A.C., Aldousany A., Zahn E.M., Burke R.P. Rapid cardiopulmonary support for children with complex congenital heart disease. Ann Thorac Surg 2000;70:742-750.[Abstract/Free Full Text]
13. Anderson H.L., III, Attorri R.J., Custer J.R., Chapman R.A., Bartlett R.H. Extracorporeal membrane oxygenation for pediatric cardiopulmonary failure. J Thorac Cardiovasc Surg 1990;99:1011-1019.[Abstract]
14. Del Nido P., Dalton H.J., Thompson A.E., Siewers R.D. Extracorporeal membrane oxygenator rescue in children during cardiac arrest after cardiac surgery. Circulation 1992:300-304.

This article has been cited by other articles:

				L. Lequier, A. R. Joffe, C. M.T. Robertson, I. A. Dinu, Y. Wongswadiwat, N. R. Anton, D. B. Ross, I. M. Rebeyka, and Western Canadian Complex Pediatric Therapies Progr Two-year survival, mental, and motor outcomes after cardiac extracorporeal life support at less than five years of age. J. Thorac. Cardiovasc. Surg., October 1, 2008; 136(4): 976 - 983.e3. [Abstract] [Full Text] [PDF]

				M. Tajik and M. G. Cardarelli Extracorporeal membrane oxygenation after cardiac arrest in children: what do we know? Eur. J. Cardiothorac. Surg., March 1, 2008; 33(3): 409 - 417. [Abstract] [Full Text] [PDF]

				A. Hoskote, D. Bohn, C. Gruenwald, D. Edgell, S. Cai, I. Adatia, and G. Van Arsdell Extracorporeal life support after staged palliation of a functional single ventricle: Subsequent morbidity and survival J. Thorac. Cardiovasc. Surg., May 1, 2006; 131(5): 1114 - 1121. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Bhagawan Koirala Brian McCrindle John Coles Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chow, G. Articles by de Veber, G. Search for Related Content PubMed PubMed Citation Articles by Chow, G. Articles by de Veber, G. Related Collections Cerebral protection Extracorporeal circulation