Printer Friendly

Long-term follow-up of survivors of neonatal ECMO: what do we really know?

Extracorporeal membrane oxygenation (ECMO) has been used increasingly since the 1970s to treat serious respiratory diseases of infancy. However, much is unknown regarding developmental processes and outcomes in these high-risk children. Since the majority of children treated with ECMO have "normal" intelligence, they are often deemed developmentally appropriate by families and primary care providers in early childhood. Unfortunately, many children may exhibit cognitive deficits later, falling behind their peers in school during middle childhood. Predicted academic success cannot be based on IQ alone. Studies have shown that between 10% and 35% of ECMO survivors who are developing "normally" are in need of special educational services due to attention problems and/or express variability in functional abilities. Although the research data remain inconclusive, parents of children who undergo neonatal ECMO should be alerted to potential problems so that children can receive full assessments and appropriate interventions.

Extracorporeal membrane oxygenation (ECMO) has been used with increasing frequency since the 1970s to treat serious respiratory diseases of infancy. Although there are still those who do not feel that the efficacy has been sufficiently documented by empirical data, most will agree that the procedure has become standard practice in the treatment of life-threatening respiratory illnesses in the neonatal period (Page, Frisk, & Whyte, 1994).

The illnesses leading to treatment with ECMO and the treatment itself can place a child at risk of developmental sequelae. For this reason, follow-up of children treated with ECMO should include assessment for potential specific deficits and functional difficulties identified in research to date. Most research to date on developmental sequelae of ECMO has examined outcomes in children up to 3 years of age. A smaller number of studies have examined outcomes in older children. This article will describe the ECMO procedure, review and critique research studies on developmental outcomes, and make recommendations for follow-up of ECMO survivors

The ECMO Procedure

ECMO has been described as, "prolonged extracorporeal cardiopulmonary bypass achieved by extra thoracic cannulation under local anesthesia" (Bartlett et al., 1986, p. 236). ECMO has been used successfully to treat respiratory failure in the newborn since 1975 (Bartlett et al., 1986). Respiratory distress syndrome, sepsis, meconium aspiration syndrome, persistent pulmonary hypertension, and congenital diaphragmatic hernia are conditions for which ECMO is commonly used. The technology is intended for those infants who are term or near term and unresponsive to conventional medical therapy (Short & Lotze, 1988). Although centers differ slightly, general criteria for ECMO candidates have been established (see Table 1). The primary advantage of ECMO is to allow the lungs an opportunity to rest and recover without the trauma often associated with conventional medical therapies (Bartlett, Andrews, Toomasian, Haiduc, & Gazzaniga, 1982; Keszler & Sira Subramanian, 1991; Short & Lotze, 1988).
Table 1. Typical ECMO Criteria Across Centers

Weight greater than 2,000 g
Gestational age greater than 34 weeks
Reversible lung disease
No more than 7 to 10 days of ventilator therapy
No major intracranial hemorrhage
No congenital abnormality incompatible with life
Failure of maximal medical management
      100% oxygen
      Tolazoline treatment
Some evidence of severe respiratory failure
      Alveolar-arterial oxygen gradient
      [is greater than] 600 for 12 hr or [is greater than]
      610 for 8 hr
      Oxygenation Index [is greater than] 40 on
      3-5 arterial gases 1/2 hr apart
     Barotrauma or air leaks
Evidence of acute deterioration
      Pa[O.sub.2] [is less than] 40 torr for 2 hr

Notes: Pa[O.sub.2] = partial pressure of oxygen; torr = millimeters of mercury. From: Gangarosa, M.E., Hynd, G.W., & Cohen, M.J. (1994). Developmental long-term follow-up of extracorporeal membrane oxygenation survivors: A review. Journal of Clinical Child Psychology, 23, p. 175. Copyright 1994 by Lawrence Erlbaum Associates, Inc. Reprinted with permission.

Veno-arterial (VA) cannulation is the most frequently used approach when placing an infant on ECMO. To use this method, a large-bore catheter is placed in the jugular vein and advanced to the right atrium to remove blood by gravity (Donovan & Spangler, 1993). A second catheter is inserted through the neck into the carotid artery and advanced to the aortic arch for the return of oxygenated blood, which is pumped via the ECMO circuit (see Figure 1). The major disadvantages of VA cannulation have been the need to ligate the right common carotid artery at cannulation and the need for heparinization of the blood in the ECMO circuit (Donovan & Spangler, 1993).


Recently, efforts have been made to reduce the risks associated with carotid artery ligation and heparinization during ECMO. Carotid artery reconstruction has been proposed as one method to reduce morbidity associated with ligation of the artery (Baumgart et al., 1994; Spector, Wiznitzer, Walsh-Sukys, & Stork, 1991). Veno-venous (W) bypass is another method used to eliminate the need for carotid artery ligation. Using a double-lumen catheter inserted into the external jugular vein, blood is removed by gravity from the proximal lumen located in the superior vena cava and returned to the systemic circulation via a lumen that opens into the right atrium (Donovan & Spangler, 1993). Due to the potential for a decrease in blood flow using this method, not all infants are considered candidates (Donovan & Spangler, 1993).

Heparin-bonded surfaces have been introduced for ECMO circuits to reduce the complications associated with systemic heparinization (Muehrcke et al., 1995; Reh et al., 1995; Von Segesser, 1996). Although these newer techniques are promising, much more data are still needed to adequately evaluate their usefulness.

Potential Risks

Although the exact mechanisms are unclear, children who undergo neonatal ECMO are at risk for developmental delays and disabilities. The factors listed in Table 2 may be associated with adverse outcomes in this population of sick neonates.
Table 2. Factors Associated with Adverse Outcomes
in Children Receiving ECMO

  * severe hypoxia
  * stress
  * acidosis
  * hypotension
  * cardiac arrest
  * hyperventilation-induced alkalosis
  * seizures
  * intracranial hemorrhage
  * heparinization mechanically-induced
blood pumping resulting in altered blood flow patterns
  * complications related to mechanical failures of the equipment
  * complications related to cessation of blood flow through
the carotid artery following surgical ligation

Note: From: Gangarosa, Hynd, & Cohen (1994); Page, Frisk, & White (1994).

Additionally, Revenis, Glass, and Short (1992) reported an increased incidence of developmental delay in 1- to 2-year-old infants who weighed between 2.0 and 2.5 kg over those weighing greater than 2.5 kg at birth. The inclusion criteria for undergoing EMCO in most institutions is a birthweight greater than 2 kg. Infants who undergo ECMO are severely ill, medically fragile neonates prior to being placed on mechanical cardiopulmonary bypass. Based on historical data, these infants have been said to have no more than a 20% chance of survival without the use of ECMO (Gangarosa et al., 1994; Page et al., 1994). For certain infants, mortality estimates without the use of ECMO may decrease somewhat with recent changes in the use of medical therapies such as high frequency and liquid ventilation and artificial surfactant administration. Nonetheless, many of these children would still be expected to die without the treatment. Clearly the invasiveness of the ECMO procedure itself is not the only reason for concern about long-term outcomes in this population.

Follow-Up Studies: Young Children

Many early follow-up studies of infants who underwent ECMO focused primarily on mortality in an effort to examine the use of this new technology (Bartlett et al., 1982; Bartlett et al., 1986). Prior to 1990, little was published about long-term neurodevelopmental outcomes for this population. Andrews, Nixon, Cilley, Roloff, and Bartlett (1986) reported on 14 survivors of neonatal ECMO who were between 1 and 3 years of age. Based on the Bayley Scales of Infant Development, 9 children (64%) had normal motor function and 10 children (71%) had normal cognitive function. Neuropsychological testing was not done. Subsequent reports also involving children less than 3 years of age found similar results (Adolph et al., 1990; Glass, Miller, & Short, 1989; Griffin et al., 1992; Robertson et al., 1995; Walsh-Sukys et al., 1994). Given the degree of initial illness in the children studied, researchers were not surprised by the results.

In a review of developmental follow-up studies on ECMO survivors, Gangarosa and colleagues (1994) suggested that 20% to 30% of the children surviving ECMO during the neonatal period have a poor developmental prognosis. In addition, of those considered to be developing "normally," 10% to 35% are in need of special services and are at risk for learning problems and attention deficit hyperactivity disorder (Gangarosa et al., 1994).

Long-Term Follow-Up of ECMO Survivors

Several studies including older children are included in Table 3. In the largest, long-term follow-up study of children who are all the same age, Glass and colleagues (1995) evaluated 102 5-year-old children who survived neonatal ECMO and 37 "healthy" controls. The rate of major disability in the ECMO treated group was 17%, which the researchers did not consider high given the mortality risk of this population during the newborn period (Glass et al., 1995). However, the researchers raised a concern over the increased rate of parental reported behavioral problems and the increased risk of school failure. The ECMO treated group had a projected risk of school failure of 48%, more than double the 22% risk among the control group (Glass et al., 1995). Although within the normal range, the Full Scale IQ, Verbal IQ, and Performance IQ scores, using Wechsler Preschool and Primary Scale of Intelligence -- Revised (WPPSI-R), were all significantly lower than the control group. In addition, specific measures of receptive language, expressive language, verbal memory, visual memory, visual perception (spatial and motor), and executive function were significantly different in the ECMO-treated group than the control group. When mental retardation was controlled for in the analysis, the children treated with ECMO were still significantly more likely to fail an academic screening battery.


Attention and memory problems have been identified in another study of children treated with ECMO. Schumacher, Sears, Davis, and Wilkerson (1997) reported an increased risk for deficits in sustained attention and memory functioning in a sample of 42 ECMO survivors who were 7 to 9 years old. In addition, these researchers noted considerable variability in skills within single individuals (Davis, Wilkerson, Schumacker, & Reese, 1997). Even when Full Scale IQ scores were within normal limits, there was a discrepancy between Verbal and Performance Scale Scores and performance varied widely between individual sub-test scores. More data is needed before drawing conclusions from this preliminary data. However, it is clear that many questions remain unanswered regarding long-term developmental outcomes and should be explored in depth before children surviving ECMO are deemed "normal."

Critique of Research to Date

Long-term data on developmental outcomes is scarce for older children surviving neonatal ECMO. More research and research including larger sample sizes are needed. In addition, the data that exists typically has been combined with data from children who are much younger (Hofkosh et al., 1991; Stolar, Crisafi, & Driscoll, 1995; Towne, Lott, Hicks, & Healey, 1985). Due in part to the practice of combining the data from children of multiple ages in a single study, interpretation of individual study results and comparisons across studies is difficult.

In addition to the paucity of data on long-term developmental outcomes, no data are available on the development of cognitive processes in children treated with ECMO. Further, the variability between studies in testing materials makes a comparison of results across studies difficult. For future research, agreement is needed on the specific types of data to be collected. Although standardized measures of cognitive development have been used, comparisons between the many scales are difficult. An additional concern is that no standardized measures have been used to examine other physical findings nor is there any agreement between studies on measures such as speech, hearing, behavior, psychosocial, and neuropsychological evaluations.

The tests themselves are not the only issue to be examined. In many studies, only group mean IQ scores have been reported. The preliminary research of Davis and colleagues (1997), discussed previously, suggests that this way of reporting data overlooks important information about variability within individuals and the group. Because of this, in future research, greater emphasis should be placed on individual performance and trying to determine the sources of the variability in functioning. Presentation of group means could mask the variability that may help to identify important variables related to specific outcomes.

Studies to date suggest that predicted academic success can not be based on IQ alone, since it has been determined that 10%-35% of the survivors who are developing "normally" are in need of special educational services (Glass et al., 1995) and attention problems and within-child variability have been documented in this population (Davis et al., 1997; Schumacher et al., 1997). The recognition of the mechanisms underlying the development of these abilities may provide the needed explanation for academic problems. Studies are needed that explore the underlying processes and mechanisms that ultimately affect cognitive functioning. The examination must include a comprehensive, multidisciplinary assessment of physical, neurological, cognitive, and neuropsychological functions. Neuropsychological testing of specific skills such as language processing, visual-spatial memory, and attention function should be included.

In conclusion, very little is known about the long-term outcome of children who underwent neonatal ECMO. Future research should include larger samples of children who are at least 7 years of age. What we do know, clearly indicates that these children must be evaluated through middle childhood and possibly beyond.

Anticipatory Guidance for Parents

Clearly more data is needed before health care professionals can intervene appropriately with either parent counseling and anticipatory guidance or proper referrals for more comprehensive assessments. Although data from research to date remain inconclusive, the parents of children who survive neonatal ECMO should be alerted to potential problems. Glass and colleagues (1995) reported a relative risk for school failure of 48% in 5year-old children treated as neonates with ECMO, compared to 22% in control children, in addition, the ECMO-treated children scored significantly below controls on 13 out of 15 subtests of neuropsychological functioning involving the following domains: language, verbal memory, visual memory, visual-perception (spatial), and visual-perception (motor). The presence of academic problems should be determined early so the child can be placed in special services, if needed. While the mildly affected child may not need special educational services, parents and teachers must be made aware of any problems and taught strategies to help the child succeed in an academic environment. In addition, the child must be taught to recognize his or her strengths and weaknesses and use these strengths to compensate for the area(s) in which the skills are not optimally functional.

In addition to cognitive functioning, children surviving ECMO have been shown to have an increased risk of behavioral problems (Davis et al., 1997; Glass et al., 1995), which may be primary deficits or secondary problems related to alterations in their functional abilities. Whether primary or secondary, behavior problems can also interfere with a successful academic performance. Trying to perform tasks that are expected by teachers and parents, but for which a child is not equipped, can lead to frustration in children with unrecognized learning disabilities. Such frustration often manifests itself in specific behavioral problems. Without proper evaluation and treatment, the child may be mis-labeled or simply go unnoticed until the problem becomes more severe as the child falls farther behind his or her peers. Early recognition and intervention can help the child have a successful academic experience in spite of limitations in functioning.

Parents must be informed of the need for long-term neuropsychological evaluation to identify potential problems unique to each individual. When parents present at clinics and other health care facilities with complaints of behavior and/or academic problems, they must receive both accurate information and referrals for more comprehensive evaluations. Many hospitals and universities offer follow-up services for families of children treated with ECMO. However, many families are lost to follow-up because the child appeared "normal" until reaching second or third grade when higher level cognitive skills are needed. Health care professionals should be aware of potential challenges children may face and should help families meet the needs of these children.


Adolph, V., Ekelund, C., Smith, C., Starrett, A., Falterman, K., & Arensman, R. (1990). Developmental outcome of neonates treated with extracorporeal membrane oxygenation. Journal of Pediatric Surgery, 25, 43-46.

Andrews, A.F., Nixon, C.A., Cilley, R.E., Roloff, D.W., & Bartlett, R.H. (1986). One- to three-year outcome for 14 neonatal survivors of extracorporeal membrane oxygenation. Pediatrics, 78, 692-698.

Bartlett, R., Andrews, A., Toomasian, J., Haiduc, N., & Gazzaniga, A. (1982). Extracorporeal membrane oxygenation for newborn respiratory failure: Forty-five cases. Surgery, 92, 425-433.

Bartlett, R., Gazzaniga, A., Toomasian, J., Corwin, A., Roloff, D., & Rucker, R. (1986). Extracorporeal membrane oxygenation (ECMO)in neonatal respiratory failure. Annals of Surgery, 204, 236-245.

Baumgart, S., Streletz, L.J., Needleman, L., Merton, D.A., Wolfson, P.J., Desai, S.A., McKee, L.M., Desai, H., Spitzer, A.R., & Graziani, L.J. (1994). Right common carotid artery reconstruction after extracorporeal membrane oxygenation: Vascular imaging, cerebral circulation, electroencephalographic, and neurodevelopmental correlates to recovery. Journal of Pediatrics, 125, 295-304.

Davis, D.W., Wilkerson, S.A., Schumacher, E., & Reese, A.H. (1997). Seven years after neonatal ECMO: Preliminary analyses. Manuscript submitted for publication.

Donovan, E.F. & Spangler, L.L. (1993). New technologies applied to the management of respiratory dysfunction. In C. Kenner, A. Brueggemeyer, & L.R Genderson (Eds.), Comprehensive neonatal nursing: A physiological perspective (pp. 313-336). Philadelphia: W.B. Saunders Company.

Gangarosa, M.E., Hynd, G.W., & Cohen, M.J. (1994). Developmental long-term follow-up of extracorporeal membrane oxygenation survivors: A review. Journal of Clinical Child Psychology, 23, 174-192.

Glass, R, Miller, M., & Short, B. (1989). Morbidity for survivors of extracorporeal membrane oxygenation: Neurodevelopmental outcome at 1 year of age. Pediatrics, 83, 72-76. Glass, R, Wagner, A., Papero, R, Rajasingham, S., Civitello, L., Kjaer, M., Coffman, C., Getson, R, & Short, B. (1995). Neurodevelopmental status at age five years of neonates treated with extracorporeal membrane oxygenation. Journal of Pediatrics, 127, 447-457.

Griffin, M., Minifee, R, Landry, S., Allison, P., Swischuk, L., & Zwischenberger, J. (1992). Neurodevelopmental outcome in neonates after extracorporeal membrane oxygenation: Cranial magnetic resonance imaging and ultrasonography correlation. Journal of Pediatric Surgery, 27, 33-35.

Hofkosh, D., Thompson, A., Nozza, R., Kemp, S., Bowen, A., & Feldman, H. (1991). Ten years of extracorporeal membrane oxygenation: Neurodevelopmental outcome. Pediatrics, 87, 549-556.

Keszler, M., & Sira Subramanian, K.N. (1991). Reversing respiratory failure in newborns. Contemporary OB/GYN, 36, 99-112.

Muehrcke, D.D., McCarthy, RM., Stewart, R.W., Seshagiri, S., Ogella, D.A., Foster, R.C., & Cosgrove, D.M. (1995). Complications of extracorporeal life support systems using heparinbound surfaces: The risk of intracardiac clot formation. Journal of Thoracic & Cardiovascular Surgery, 110, 843-851.

Page, J., Frisk, V., & Whyte, H. (1994). Developmental outcome of infants treated with extracorporeal membrane oxygenation (ECMO): Is the evidence all in? Paediatric & Perinatal Epidemiology, 8, 123-139.

Reh, D., Rossaint, R., Streich, R., Gerlach, H., Pappert, D., Kramer, H., & Falke, K.J. (1995). Extracorporeal membrane oxygenation with heparincoated systems in a 13-month-old infant with acute hypoxic respiratory failure after correction of tetralogy of Fallot. European Journal of Cardio-Thoracic Surgery, 9, 226-229.

Revenis, M.E., Glass, R, & Short, B.L. (1992). Mortality and morbidity rates among lower birth weight infants (2000 to 2500 grams) treated with extracorporeal membrane oxygenation. Journal of Pediatrics, 121, 542548.

Robertson, C., Finer, N., Sauve, R., Whitfield, M., Belgaumkar, T., Synnes, A., & Grace, M. (1995). Neurodevelopmental outcome after neonatal extracorporeal membrane oxygenation. Canadian Medical Association Journal, 152, 1981-1988.

Schumacher, E.A., Sears, L.L., Davis, D.W., & Wilkerson, S.A. (1997, March). Neuropsychological outcome of neonates treated with extracorporeal membrane oxygenation (ECMO): An examination of right hemisphere deficits at school-age. Presentation at the 12th Annual Children's National Medical Center ECMO Symposium, Keystone, CO.

Short, B.L. & Lotze, A. (1988). Extracorporeal membrane oxygenator therapy. Pediatric Annals, 17, 516-523.

Spector, M.L., Wiznitzer, M., Walsh-Sukys, M.C., & Stork, E.K. (1991). Carotid reconstruction in the neonate following ECMO. Journal of Pediatric Surgery, 26, 357-361.

Stolar, C., Crisafi, M., & Driscoll, Y. (1995). Neurocognitive outcome for neonates treated with extracorporeal membrane oxygenation: Are infants with congenital diaphragmatic hernia different? Journal of Pediatric Surgery, 30, 366-372.

Towne, B., Lott, I., Hicks, D., & Healey, T. (1985). Long-term follow-up of infants and children treated with extracorporeal membrane oxygenation (ECMO): A preliminary report. Journal of Pediatric Surgery, 20, 410-414.

Von Segesser, L.K. (1996). Heparin-bound surfaces in extracorporeal membrane oxygenation for cardiac support. Annals of Thoracic Surgery, 6, 330-335.

Walsh-Sukys, M., Bauer, R., Cornell, D., Friedman, H., Stork, E., & Hack, M. (1994). Severe respiratory failure in neonates: Mortality and morbidity rates and neurodevelopmental outcomes. Journal of Pediatrics, 125, 104-110.

Deborah Winders Davis, DNS, ARNP, is Associate Director, Neonatal Follow-up Program; Assistant Professor, Department of Pediatrics, University of Louisville, Louisville, KY.
COPYRIGHT 1998 Jannetti Publications, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:extracorporeal membrane oxygenation
Author:Davis, Deborah Winders
Publication:Pediatric Nursing
Date:Jul 1, 1998
Previous Article:Operationalizing Donna Wong's principle of atraumatic care: pain management protocol in the NICU.
Next Article:Preventing lead poisoning and its consequences.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters