PEDIATRICS Vol. 107 No. 3 March 2001, pp. 549-552

International Experience With Trisomy 21 Infants Placed on Extracorporeal Membrane Oxygenation

W. Michael Southgate, MD, David J. Annibale, MD, Thomas C. Hulsey, MSPH, ScD, and Dilip M. Purohit, MD

From the Department of Pediatrics, Medical University of South Carolina, Charleston, South Carolina.

	ABSTRACT

Top Abstract Methods Results Discussion References

Objective.  To characterize the international experience concerning neonates with trisomy 21 (T21) managed with extracorporeal membrane oxygenation (ECMO), and to compare and contrast this group of patients to the neonatal ECMO population as a whole.

Methods.  Data from the Extracorporeal Life Support Organization for newborn infants placed on ECMO between January 1984 and June 1999 were analyzed. Infants with T21 were compared with the group of infants without T21.

Results.  Fifteen thousand nine hundred forty-six infants, including 91 (n = 91) with the diagnosis of T21, were placed on ECMO for neonatal respiratory failure during the 14.5-year period. T21 infants were overrepresented in the ECMO population by several-fold when compared with the incidence of T21 in the general population. Eighty-seven of the 91 T21 infants were placed on ECMO after 1989. The distribution of primary diagnoses leading to ECMO differed between the groups (T21 vs non-T21): primary persistent pulmonary hypertension, 47.3% versus 13%; meconium aspiration syndrome, 23.1% versus 32.9%; sepsis, 7.7% versus 13.2%; congenital diaphragmatic hernia, 7.7% versus 19.9%; and respiratory distress syndrome, 3.3% versus 7.9%. Although survival to discontinuation of ECMO was similar in the 2 groups, likelihood of survival to discharge was decreased for T21 infants (65.9% vs 75.6%) because of increased post-ECMO mortality.

Conclusions.  Extracorporeal Life Support Organization registry data suggests that T21 infants are at a significantly higher risk of being placed on ECMO for neonatal respiratory failure than the general population, perhaps as a result of delayed extrauterine pulmonary vascular adaptation, as manifested in the high rate of primary persistent pulmonary hypertension as the primary diagnosis. There may have been a shift in attitude regarding the use of ECMO in the T21 patient after 1989. Although most T21 patients placed on ECMO will survive, the prognosis is more guarded in this population when compared with all infants so managed. The long-term neurodevelopmental outcome of this group of T21 ECMO survivors is currently unknown.  Key words:  neonate, extracorporeal membrane oxygenation, trisomy 21, persistent pulmonary hypertension.

Infants with trisomy 21 (T21), comprising 1 of every 588 to 1040 newborns,^1-3 are born to women of all ages. Even if not suspected antenatally, the diagnosis is usually considered in the first hours of life because of the familiar dysmorphic features, and diagnostic certainty by karyotype analysis is readily available. Individuals with T21 are moderately to severely retarded and typically survive into adulthood, although those with congenital heart defects have a more guarded prognosis.¹ There has been a decided change in the medical/surgical approach to the treatment of the T21 infant over the last 25 years. In the mid- to late 1970s, the majority of pediatricians and surgeons in the United States recommended nontreatment in cases of T21 with duodenal atresia,⁴ a relatively easily correctable anomaly. Sondheimer⁵ concluded that children with T21 and complete atrioventricular canal were often being denied standard cardiac care in the late 1970s. Within the next decade, however, attitudes had apparently shifted,⁶ resulting in T21 infants with congenital heart defects receiving equivalent cardiac medical and surgical care to those infants without T21 with similar outcomes.^7-11

Extracorporeal membrane oxygenation (ECMO) is a proven technique¹² used to manage term and near-term neonates with reversible cardiopulmonary failure. Currently, mortality in neonates managed with ECMO is 15% to 20%, and major disability can be expected in 10% to 20% of survivors.^13,14 Accepted contraindications to the use of ECMO in neonates include irreversible cardiopulmonary failure and lethal congenital anomalies (ie, Trisomy 13). Subsequent to the introduction of neonatal ECMO in the 1980s, infants with T21 and reversible respiratory failure have been managed with ECMO in many centers. However, there have been no data regarding the use of ECMO in the T21 population, nor the outcomes of these infants.

The purpose of this study was to investigate and describe the international ECMO community's experience with T21 infants, and compare and contrast this experience with that of the neonatal ECMO population as a whole, using data collected by the Extracorporeal Life Support Organization (ELSO).

	METHODS

Top Abstract Methods Results Discussion References

Data from the ELSO for newborn infants with respiratory failure placed on ECMO between January 1984 and June 1999 were analyzed. ELSO receives data from ECMO centers in the United States and abroad by means of a standardized data capture form. This includes information regarding patient demographics, primary and secondary diagnoses, criteria used to determine the need for ECMO, pre-ECMO and ECMO course, mechanical and patient complications, and outcomes to hospital discharge. Infants with T21 were compared with the group of infants without T21 using ² analysis.

	RESULTS

Top Abstract Methods Results Discussion References

Incidence of T21 Infants in the ECMO Population

A total of 15 946 infants were entered into the ELSO registry from January 1984 through June 1999, including 91 diagnosed with T21; a rate of 5.7 T21 infants per 1000 ECMO patients. Based on the expected rate of T21 in the general population (0.96 to 1.7 per 1000 live births), T21 infants were overrepresented in the ECMO population by 5.9- to 3.6-fold. The frequency of T21 infants in the ECMO population increased after 1989, with more than 95% of T21 patients being registered after that year versus 77% of the non-T21 infants (Fig 1). As a result, during the years 1990 to 1998, T21 infants were overrepresented by a factor of 7.4- to 4.2-fold. Gender and racial distribution did not differ between the 2 groups (data not shown).

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Cumulative percentage of both T21 and non-T21 patients placed on ECMO during the study period.

Primary Diagnoses

The distribution of primary diagnoses leading to ECMO was markedly different between the 2 groups (Fig 2). Nearly half of T21 patients were registered as having a primary diagnosis of primary persistent pulmonary hypertension (PPPHN), whereas only 13% of the non-T21 were so coded. Because persistent pulmonary hypertension is seen as an associated finding in the majority of neonates placed on ECMO regardless of their underlying primary diagnosis (respiratory distress syndrome, congenital diaphragmatic hernia [CDH], meconium aspiration syndrome, sepsis), a review of listed secondary diagnoses was conducted. This review did not reveal an increased likelihood of possible miscoding the primary diagnosis as PPPHN in the T21 group when compared with the non-T21 group. In the group of T21 infants, 9% of those registered with PPPHN as their primary diagnosis were also registered as having 1of the other 4 diagnoses commonly seen in neonates placed on ECMO, whereas 28% of the non-T21 PPPHN infants carried 1 of these additional diagnoses. Of the 91 T21 patients, 16 (17.6%) were registered as having a congenital heart defect, including 8 (8.8%) with an atrioventricular canal defect.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Distribution of primary diagnoses in T21 and non-T21 patients. * P = .001; **P = .047; ***P = .004.

Outcome

Over the 14.5 years of this review, discontinuation of ECMO attributable to recovery was as likely in the T21 infant as in the non-T21 infant (87.8% vs 84.9%) but, because of a high post-ECMO mortality, the likelihood of survival to discharge was significantly less (65.9 vs 95.7%, P = .03; risk ratio = 1.6, 95% confidence interval: 1.04-2.5) in the T21 population than in the non-T21 patients. This increased post-ECMO mortality was attributable neither to a specific primary diagnosis (eg, CDH) nor the presence of congenital heart disease. As stated, there was a greater incidence of CDH among the non-T21 infants. Controlling for CDH with logistic regression, T21 infants were at a 4.2 times (95% confidence interval: 2.5-7.2) greater risk for post-ECMO mortality. For comparative purposes, controlling for T21, the post-ECMO mortality for infants with CDH was 5.5 times that of non-CDH patients.

To assess the possibility that infants with T21 may have been more likely to have been electively removed from support once the diagnosis of T21 was confirmed by karyotype analysis sometime into the course of ECMO, the reason for discontinuation of ECMO was determined for the 2 groups. One infant was removed from ECMO in response to parental request in the T21 group, which was not different from expected when compared with the non-T21 infants.

	DISCUSSION

Top Abstract Methods Results Discussion References

These data indicate that infants with T21 are at a significantly higher risk of being placed on ECMO than the general population of newborns, perhaps as a result of delayed extrauterine pulmonary vascular adaptation, as manifested by the high rate of PPPHN as the primary diagnosis leading to the use of ECMO. Over the past 30 years, there has been considerable debate regarding the prevalence of hypertensive pulmonary vascular disease in children with T21, including those with and without congenital heart disease. In 1972, Sharer¹⁵ reported that half of T21 infants with congenital heart defects <1 year old were found to have pulmonary hypertension during cardiac catheterization. Yamaki^16,17 has described a significantly increased incidence of both structural and hemodynamic evidence of pulmonary vascular disease in T21 patients with both simple cardiac anomalies and complete atrioventricular canal when compared with non-T21 patients. Patients as young as 3 months old had complete occlusion of the lumen of the small pulmonary arteries as a result of acute intimal fibrous proliferation. Fixed elevated pulmonary resistance is seen in >10% of T21 infants with complete atrioventricular canal during the first year of life.^18,19 To the contrary, based on histologic findings of tissue obtained at autopsy, Wilson and colleagues²⁰ concluded that patients with T21 and congenital heart malformations have no predisposition to develop precocious or severe hypertensive pulmonary vascular disease. Likewise, in a more recent report, Minich²¹ actually found less evidence of pulmonary hypertension in T21 infants undergoing atrioventricular septal defect repair than in non-T21 patients.

Fetal and transitional pulmonary vascular tone is influenced by a number of endogenous vasoactive mediators (nitric oxide, prostaglandins, endothelin) and neurohumoral agents, as well as inherent structural components of the developing vasculature.²² Although the data in this study strongly suggest the T21 infant is at risk for an abnormal transition to extrauterine life, the underlying mechanism for such a problem, if it exists, is unknown.

Congenital heart disease is found in 40% to 60% of infants with T21.^3,23 Eighty percent of T21 infants with congenital heart disease have either an atrioventricular canal (45%) or a ventricular septal defect (35%). Less than 18% of T21 patients placed on ECMO in this review were diagnosed as having a structural heart defect. It is possible that the wider population of T21 infants with congenital heart disease and pulmonary hypertension were successfully medically managed by accepting expected relatively low PaO₂ until the pulmonary vasculature relaxed enough to allow decreased support.

On the other hand, we speculate that the finding of a congenital heart defect in an infant with other features suggestive of T21 might have dissuaded some ECMO physicians and/or families from placing these infants on ECMO. If such speculation is true, despite the disproportionately large number of T21 infants placed on ECMO during the study period, it is quite possible, if not likely, that a number of infants who were failing conventional therapy were denied ECMO based on a clinical or karyotype diagnosis of T21. Discussions with ECMO physicians from centers in North America and Europe at a conference in February 1999 where these findings were presented in abstract form revealed a wide range of attitudes regarding the use of ECMO for T21 infants. Reactions ranged from surprise that physicians would contemplate not offering ECMO, to shock that anyone would consider using this extraordinary form of support in this population.

In the absence of data regarding the effectiveness of ECMO in the T21 patient, such diverse opinions may be expected. Despite this paucity of data, there was a sharp upswing in the use (lack of denial?) of ECMO for the T21 patient in the beginning of the 1990s. This may represent either a shift in physicians' acceptance of ECMO as proven therapy or a reflection of a change in the perception of the balance of quality of life of the T21 patient versus the potential harm of ECMO. It is also possible that as the number of ECMO centers expanded in the early 1990s, a more accurate spectrum of physicians' opinions was reflected. Finally, in the United States, neonatologists' concerns regarding legal obligations in the aftermath of the "Baby Doe" regulations of the 1980s may have influenced decision-making.²⁵

This study provides physicians with information regarding the short term outcome of T21 infants placed on ECMO because of pulmonary failure. T21 infants are as likely to survive to the point of removal from bypass as their non-T21 counterparts, but for reasons that we were not able to determine, are at a significantly higher risk of death after decannulation. In contrast, Klein²⁴ reported comparable survival rates between T21 and non-T21 infants who were placed on ECMO after corrective heart surgery because of persistent low cardiac output or life-threatening pulmonary vasoreactive crisis.

Long-term outcome of infants treated with ECMO is comparable to those infants with persistent pulmonary hypertension not treated with bypass. However, Mean Full Scale, Verbal and Performance IQs are lower, and rates of behavioral problems are higher in infants treated with ECMO¹³ than healthy controls. The impact of this therapy in a group of children who, in the best of circumstances, are destined to have at least a moderate degree of mental retardation is not known. The American Academy of Pediatrics "believes that parents and physicians should make reasoned decisions together about critically ill infants using the principles of informed parental permission."²⁶ Although the information provided by this investigation is helpful, before physicians can make truly enlightened recommendations and parents can give genuinely informed consent, the long-term neurodevelopmental outcome of T21 infants treated with ECMO merits additional investigation.

	ACKNOWLEDGMENT

We thank Myla Ebeling, BS, for her patient assistance in data recovery and compilation.

	FOOTNOTES

Received for publication Apr 21, 2000; accepted Jun 28, 2000.

Address correspondence to W. Michael Southgate, MD, Department of Pediatrics, Division of Neonatology, Medical University of South Carolina, 165 Ashley Ave, Box 250917, Charleston, SC 29425. E-mail: southgaw{at}musc.edu

	ABBREVIATIONS

T21, trisomy 21; ECMO, extracorporeal membrane oxygenation; ELSO, Extracorporeal Life Support Organization; PPPHN, primary persistent pulmonary hypertension; CDH, congenital diaphragmatic hernia.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Hayes C, Johnson Z, Thornton L, Ten year survival of Down syndrome births. Int J Epidemiol 1997; 26:822-829 [Abstract/Free Full Text]
2. Penrose LS, Smith GF. Down's Anomaly. Boston, MA: Little, Brown & Co; 1966
3. Freeman SB, Taft LF, Dooley KJ, Population-based study of congenital heart defects in Down syndrome. Am J Med Genet 1998; 80:213-217 [CrossRef][Medline]
4. Todres ID, Krane D, Howell MC, Pediatrician's attitudes affecting decision-making in defective newborns. Pediatrics 1977; 60:197-201 [Abstract/Free Full Text]
5. Sondheimer HM, Byrum CJ, Blackman MS Unequal cardiac care for children with Down's syndrome. Am J Dis Child 1985; 139:68-70 [Abstract/Free Full Text]
6. Todres ID, Guillemin J, Grodin MA, Batten D Life-saving therapy for newborns: a questionnaire survey in the state of Massachusetts. Pediatrics 1988; 81:643-649 [Abstract/Free Full Text]
7. Schneider DS, Zahka KG, Clark EB, Neill CA, and the Baltimore-Washington Infant Study Group Patterns of cardiac care in infants with Down Syndrome. Am J Dis Child 1989; 143:363-365 [Abstract/Free Full Text]
8. Teddell JS, Litwin SB, Berger S, Twenty-year experience with repair of complete atrioventricular septal defects. Ann Thorac Surg 1996; 62:419-424 [Abstract/Free Full Text]
9. Campbell RM, Adatia I, Gow RM, Webb GD, Williams WG, Freedom RM Total cavopulmonary anastomosis (Fontan) in children with Down's syndrome. Ann Thorac Surg 1998; 66:523-526 [Abstract/Free Full Text]
10. Hanley FL, Fenton KN, Jonas RA, Surgical repair of complete atrioventricular canal defects in infancy. Twenty-year trends. J Thorac Cardiovasc Surg 1993; 106:387-397 [Abstract]
11. Amark K, Sunnegardh J The effect of changing attitudes to Down's syndrome in the management of complete atrioventricular septal defects. Arch Dis Child 1999; 81:151-154 [Abstract/Free Full Text]
12. Anonymous UK collaborative randomised trial of neonatal extracorporeal membrane oxygenation. UK Collaborative ECMO Trial Group . Lancet 1996; 348:75-82 [CrossRef][Medline]
13. Glass P, Wagner AE, Papero PH, Neurodevelopmental status at age five years of neonates treated with extracorporeal membrane oxygenation. J Pediatr 1995; 127:447-457 [CrossRef][Medline]
14. Glass P, Bulas DI, Wagner AE, Severity of brain injury following neonatal extracorporeal membrane oxygenation and outcome at age 5 years. Dev Med Child Neurol 1997; 39:441-448 [Medline]
15. Sharer RM, Farina MA, Porter IH, Bishop M Clinical aspects of congenital heart disease in mongolism. Am J Cardiol 1972; 29:497-503 [CrossRef][Medline]
16. Yamaki S, Horiuchi T, Sekino Y Quantitative analysis of pulmonary vascular disease in simple cardiac anomalies with the Down syndrome. Am J Cardiol 1983; 51:1502-1506 [CrossRef][Medline]
17. Yamaki S, Yasui H, Kado H, Pulmonary vascular disease and operative indications in complete atrioventricular canal defect in early infancy. J Thorac Cardiovasc Surg 1993; 106:398-405 [Abstract]
18. Clapp S, Perry B, Farooki Z, Down's syndrome, complete atrioventricular canal, and pulmonary vascular obstructive disease. J Thorac Cardiovasc Surg 1990; 100:115-121 [Abstract]
19. Hals J, Hagemo PS, Thaulow E, Sorland SJ Pulmonary vascular resistance in complete atrioventricular septal defect. A comparison between children with and without Down's syndrome. Acta Paediatr 1993; 82:595-598 [Medline]
20. Wilson SK, Hutchins GM, Neill CA Hypertensive pulmonary vascular disease in Down syndrome. J Pediatr 1979; 95:722-726 [CrossRef][Medline]
21. Minich LL, Tani LY, Pagotto LT, Hawkins JA, McGough EC, Shaddy RE Size of ventricular structures influences surgical outcome in Down syndrome infants with atrioventricular septal defect. Am J Cardiol 1998; 81:1062-1065 [CrossRef][Medline]
22. Ziegler JW, Ivy DD, Kinsella JP, Abman SH The role of nitric oxide, endothelin, and prostaglandins in the transition of the pulmonary circulation. Clin Perinatol 1995; 22:387-403 [Medline]
23. Greenwood RD, Nadas AS The clinical course of cardiac disease in Down's syndrome. Pediatrics 1976; 58:893-897 [Abstract]
24. Klein MD, Shaheen KW, Whittlesey GC, Pinsky WW, Arciniegas E Extracorporeal membrane oxygenation for the circulatory support of children after repair of congenital heart disease. J Thorac Cardiovasc Surg 1990; 100:498-505 [Abstract]
25. Kopelman LM, Irons TG, Kopelman AE Neonatologist judge the "Baby Doe" regulations. N Engl J Med 1988; 318:677-683 [Abstract]
26. American Academy of Pediatrics, Committee on Bioethics Ethics and the care of critically ill infants and children. Pediatrics 1996; 98:149-152 [Abstract/Free Full Text]