Management and outcome of low birth weight neonates with congenital heart disease

Management and outcome of low birth weight neonates with congenital heart disease Anthony C, Chang, MD, Frank L. Hanley, MD, James E. Lock, MD, Aldo R. Castaneda, MD, and David L. Wessel, MD From the Departments of Cardiology and Cardiac Surgery, Children's Hospital, and the Departments of Pediatrics, Surgery, and Anesthesia, Harvard Medical School, Boston, Massachusetts In I00 consecutive neonates with birth weights _<2500 gm (range, 540 to 2500 gm; median, 2200 gm), major c o n g e n i t a l heart disease (excluding patent ductus arteriosus, isolated atrial septal defect, and ventricular septal defect) was dia g n o s e d b e t w e e n January 1987 and January 1991; 46 had d u c t u s - d e p e n d e n t lesions. Of the 100 neonates, 30 had g e n e t i c aberrations or significant associated c o n g e n i t a l anomalies. The four most c o m m o n c a r d i a c diagnoses were tetralogy of Fallot with or without pulmonary atresia (n = 16); c o a r c t a t i o n of the aorta (n = 12); transposition of the great arteries (n = 11); and c o m m o n atrioventricular canal (n = 11). The hospital survival rate for the entire group of 100 neonates was 70%. The patients were separated into three groups on the basis of the time of intervention. Group I (early intervention) included 62 infants. These neonates (including 31 with d u c t u s - d e p e n d e n t lesions) had surgical or catheter intervention during the initial hospitalization (median age, 9 days), all at weights _<2500 gm. The hospital survival rate was 81% (50/62); survival rates for palliation (78%, 18/23) and for correction (82%, 32/39) were similar. There were 26 neonates in group 2 (late intervention). These neonates did not have surgical intervention during the initial hospitalization. All were m a n a g e d medically; survivors were discharged and had surgical procedures later (at a median a g e of 4.3 months). Six neonates (23%) died during medical m a n a g e m e n t ; all 20 survivors returned and had surgical procedures, with 90% survival. Overall survival rate for this group was 69% (18/26). The remaining 12 patients (group 3) had c o m p l i c a t i n g features that p r e c l u d e d intervention; none survived. On the basis of these results, we c o n c l u d e that early intervention, even with corrective surgery, can be performed in low birth weight neonates with an a c c e p t a b l e mortality rate. Prolonged medical therapy to a c h i e v e further weight gain did not a p p e a r to improve the survival rate. (J PEDIATR1994;424:461-6)

Although there are many reports that focus on the consequence of patent ductus arteriosus in the low birth weight neonate, 13 there are few studies of the impact of serious congenital heart disease on this population. 4"6 The survival Submitted for publication Oct. 21, 1992; accepted Sept. 28, 1993. Reprint requests: Anthony C. Chang, MD, Cardiac Intensive Care Office, Farley 653, The Children's Hospital, 300 Longwood Ave., Boston, MA 02115. Copyright © 1994 by Mosby-Year Book, Inc. 0022-3476/94 $3.00 + 0 9/23/51721

rate has improved for corrective7-9 and even for complex palliative procedures 1° in neonates, but previous authors CHD LBW PDA SGA VLBW

Congenital heart disease Low birth weight Patent ductus arteriosus Small for gestational age Very low birth weight

have reported that low birth weight alone is a risk factor that contributes to death during or after cardiac surgery, u, 12

461

462

Chang et al.

Thus in the management of an LBW neonate with serious CHD, it is not certain whether it is preferable to defer surgery until the patient has grown (presumably reducing the risk of operative repair) or to operate early despite small size (presumably reducing the risks of prolonged preoperative medical management). We recently began to perform early surgical intervention in LBW neonates, with a preference for corrective surgery when anatomically feasible. This retrospective review describes our experience with 100 consecutive neonates with birth weights _.<2500 gm and diagnoses of significant CHD (other than PDA or isolated atrial and ventricular septal defects) between January 1987 and January 1991. METHODS

Patients. The following data were gathered for each infant in the study: (1) pregnancy profile (maternal age and associated medical conditions, presence and status of twin gestation, and length of gestation), (2) delivery history (type of delivery, Apgar scores at 1 and 5 minutes, need for resuscitative measures, date of birth, and birth weight), (3) preoperative status (presence of genetic aberrations, congenital anomalies, or systemic diseases [such as sepsis, renal failure, intraventricular hemorrhage, and necrotizing enterocolitis], and need for any noncardiac surgery), (4) preoperative diagnosis (date of diagnosis for CHD, dates and details of preoperative echocardiographic examinations and cardiac catheterizations), and (5) surgical or catheter procedures (age and weight at the time of procedure, complications, duration of ventilatory support, cardiac intensive care stay and total hospital stay, and cause of death). Definitions. Low birth weight was defined as a birth weight of -<2500 gm, and very low birth weight was defined as a birth weight of -< 1500 gm. 13 Prematurity was defined as gestational age of <37 weeks. An infant with a birth weight at less than the 10th percentile for gestational age was specified as small for gestational age. 14 Delayed diagnosis was defined as a diagnosis made after 10 days of life. Early intervention was defined as a palliative or corrective procedure (surgical or catheter) performed during the initial hospitalization. Patients who had late intervention were those who were initially managed medically, with subsequent discharge and later return for a procedure. A corrective procedure was defined as a procedure that corrected physiologic derangements and caused the systemic and pulmonary circulations each to be served by a separate ventricle. Prolonged hospital stay was defined as a total stay of longer than 50 days. Renal failure was defined as a serum creatinine concentration > 1.2 mg/dl associated with decreased urine output. Sepsis was defined as clinical deterioration associated with positive blood culture results. Statistical analysis. Diagnostic sensitivity (defined as the

The Journal of Pediatrics March 1994

true positives/ [true positive + false negative] ratio) was chosen to reflect accuracy of echocardiographic diagnosis. Statistical comparison with chi-square analysis (with onehalf correction) or with the Fisher Exact Test was used for detecting differences between various subgroups. Statistical significance was defined as a p value <0.05. RESULTS

Pregnancy and delivery. Pregnancy was characterized by young maternal age (_< 18 years of age) for 3 patients and older maternal age (>__32 years of age) for 24 patients. Associated maternal medical conditions included hypertension (n = 11), diabetes (n = 4), seizure (n = 1), and collagen vascular disease (n = 1). There were 23 twin gestations. Of the 100 infants, 34 were delivered by cesarean section, and 66 were delivered vaginally. In 26 neonates the Apgar scores were __<5at 1 minute, and in 13 neonates, the scores remained --<5 at 5 minutes; resuscitation was considered necessary in 29 neonates. There were 55 male and 45 female infants. The birth weights ranged from 540 to 2500 gm (median, 2200 gin); 18 neonates had VLBWs. Gestational ages ranged from 26 to 40 weeks (median, 35 weeks). Among the 100 neonates, 69 were premature, 23 were SGA, and 8 were both premature and SGA. Preoperative status. Eight neonates had a genetic aberration (trisomy 21 in 4, trisomy 18 in 1, Noonan syndrome in l, Holt-Oram syndrome in 1, and DiGeorge syndrome in 1), and an additional 22 neonates had significant or multiple congenital anomalies, including 7 with VACTERL syndrome (vertebral anomalies, anal atresia, cardiac anomaly, tracheoesophageal fistula, renal anomalies, and limb anomalies). Sepsis and renal failure (in 20 and 15 patients, respectively) were the most common postnatal medical illnesses. Intraventricular hemorrhage was noted in seven patients. Noncardiac surgery was performed in 19 neonates, including 7 who had repair of a tracheoesophageal fistula and 4 who had surgery for necrotizing enterocolitis; all survived. Diagnosis of congenital heart disease. The anatomic diagnoses are summarized in Table I. Forty-six patients had ductus-dependent lesions, and all received a prostaglandin infusion after birth. The age at the time of diagnosis of CHD ranged from birth to 44 days; the diagnosis of CHD was made for more than half the neonates (n = 52) by 1 day of age. Diagnosis was delayed in 14 patients (coarctation of the aorta, 4; complete atrioventricular canal, 2; tetralogy of Fallot, 2; and 1 each with interrupted aortic arch, aortopulmonary window, total anomalous pulmonary venous connection, corrected transposition of the great arteries, tricuspid atresia, and right atrial mass). Echocardiography and cardiac catheterization. Coarcta-

The Journal of Pediatrics Volume 124, Number 3

Chang et al.

463

Table I. Diagnoses and hospital survival

Group

t:

early intervention (n

Palliative

=

62)

Corrective

Group 2: late intervention (n = 26)

Group 3: no intervention (n = 12)

Diagnosis

No.

%

No.

%

No.

%

No.

Tetralogy of Fallot (9 with pulmonary atresia) Coarctation of the aorta (3 with ventricular septal defect) Transposition of the great arteries (5 with ventricular septal defect and/or arch obstruction) Common atrioventricular canal (2 complex type) Double-outlet right ventricle (4 with single ventricle) Total anomalous pulmonary venous connection Hypoplastic left heart syndrome Truncus arteriosus (2 with interrupted aortic arch) Aortic stenosis Pulmonary stenosis Interrupted aortic arch Heterotaxy Shones complex Tricuspid atresia Others* TOTAL

2/3

67

4/4

100

6/7

86

I/1

100

6/8

75

2/2

6/8

75

1/2

50

1/1

100

0

4/4

100

1/2

3/3

0

0/1

No.

%

0/2

12/16

75

100

0/1

9/12

75

l/1

100

0

8/11

73

5.8

63

0/2

6/11

55

50

1/ 1

100

0/2

6/9

66

100

2/3

67

0

5/6

83

0

0/5

0/6

0

0

0

3/5

60

0 0 0

4/5

80

4/4 3/3

100 100

0 0 0

3/3 2/3 2/2 3/4 70/100

100 67 100 75 70

0

0

3/5

60

0 0 0

4/5 2/2 3/3

80 100 100

3/3 0 i/1 0/i 18/23

100 100 0 78

0 2/2 0 3/3 32/39

i 00 100 82

0 2/2 0 0 0/1 1/1 0 20/26

100

0 100 0 76

0/12

%

Hospital survivors/total patients

*Right atrial thrombus (a neonate weighing 1050 gm had a right atrial mass [which was later diagnosed as a thrombus] and was included in the study since he had removalof this mass during cardiopulmonary bypass), corrected transposition of the great arteries with subpulmonary stenosis and patent duetus arteriosus (one patient), pulmonary atresia with intact ventricular septum (one patient), and aortopulmonary window (one patient).

tion of the aorta (6/12, sensitivity 50%) and total anomalous pulmonary venous connection (2/6, sensitivity 67%) were the two lesions most often misdiagnosed by the referring physicians on the basis of the initial echocardiogram. Thirty-eight patients underwent diagnostic cardiac catheterizations at their initial hospitalization; an additional 12 patients had an interventional catheterization (balloon atrial septotomy in 6 patients and aortic and pulmonary

valvotomy in 4 and 2 patients, respectively); all 6 patients who had balloon valvotomy avoided surgery. Major complications occurred in 6 of 50 (12%) patients: cardiac arrest in two patients and supraventricular tachycardia, complete heart block, hemopericardium (requiring pericardiocentesis), and inferior vena cava perforation in 1 patient each. Five additional patients bad transient loss of pulse. No patient died as a result of cardiac catheterization.

464

Chang et al.

The Journal of Pediatrics March 1994

Table II. Early intervention: Hospital survival rate (by year)

Palliative Corrective

1987

1988

1989

1990

4/4 (100%) 2/3 (67%)

6/6 (100%) 3/5 (609'0)

6/8 (75%) 12/14 (86%)

2/5 (40%) 15/17 (88%)

Management strategy. The 100 neonates were subdivided on the basis of the time of initial intervention (Table I). Group 1. Early intervention (n = 62). Neonates in group 1 underwent either surgical or catheter intervention during their initial hospitalization (at a median weight of 2200 gm and all at a weight <_2500 gin). Half of the patients in this group had lesions that were considered ductus dependent. Although the median age at intervention was 9 days (range, birth to 79 days) and the median interval between diagnosis and intervention was only 5 days, I0 neonates had intervention more than 1 month after the diagnosis of a cardiac anomaly was made. ~Other than late referral from referring institutions (n = 6), additional patients had meningomyelocele requiring ventriculoperitoneal shunt (n = 1), severe bronchopulmonary dysplasia (n = 1), tracheoesophageal fistula repair with prolonged postoperative course (n = 1), and right atrial thrombus that became hemodynamically significant more than 1 month after diagnosis (n = 1). The survival rates for early palliation and correction were similar: 18 of 23 (78%) neonates who underwent early palliative operations survived hospitalization, and 32 of 39 (82%) patients who had early corrective procedures survived. The seven nonsurvivors in the early corrective surgery group included one patient with coarctation of the aorta/ ventricular septal defect who had a cardiac arrest in the operating room before surgery, and three patients with lesions associated with an interrupted aortic arch (two with trun2 cus arteriosus and one with double-outlet right ventricle)] Survival rates for palliative procedures did not improve from 1987 to 1990, but the number of corrective procedures increased (from 3 patients in 1987 to 17 patients in 1990; Table II). The overall survival for this entire group was 81% (50/62). (A table listing all the early intervention patients by their weight at the time of surgery is available from the authors on request.) There was no correlation between weight at the time of surgery and survival. Group 2. Late intervention (n = 26). A decision to defer early surgical or catheter intervention (usually because the lesion was not ductus dependent) was made for these patients. Six (23%) died between 12 and 44 days of age (median, 24 days) during medical management before intervention. The causes of death included necrotizing enterocolitis (n = 2), sepsis (bacterial in two, fungal in one), and severe hypoxemia with respiratory failure (one patient).

The remaining 20 patients survived the initial hospitalization and Were discharged. All 20 patients returned and underwent surgical palliation (n --- 1) or correction (n = 19) at a median age of 4.3 months and a median weight of 5.2 kg; 18 of 20(90%) survived the cardiac surgery. The postoperative hospital stay ranged from 5 to 58 days (median, 8 days). Group 3. NO intervention (n = 12). In the patients in group 3, CHD was allowed to follow its natural course; none of the infants survived. The reasons for forgoing medical therapy and intervention included grave prognosis for specific anatomic lesion (hypoplastic left heart syndrome with diminutive aorta in 4; tetralogy of Fallot with pulmonary atresia and severely hypoplastic or absent pulmonary arteries in 2); early multisystem organ failure (n = 2); and coexistence of multiple and severe congenital anomalies (n = 3) or lethaI genetic syndrome (trisomy 18) (n = 1). Entire group. The overall hospital survival rate was 70%. Hypoplastic left heart syndrome was the sole diagnosis for which there were no survivors, but only one attempt at palliative surgery was made. The survival rate for VLBW neonates was (50%) (9/18) and for all other neonates it was 74% (61/82). Seven VLBW neonates, including two who had interventions at 47 days of age and at weights > 1500 gin, had early intervention; all seven survived hospitalization. Of the five VLBW neonates in group 2 (late intervention), three died during medical management. Duration of hospital stay. The median days of mechanical ventilation and intensive care unit stay were 4 and 13 days, respectively, for the early intervention group (group 1); median hospital stay was 26 days (range, 7 to 101 days). The median hospital stay (both initial and subsequent hospitalizations) for the late intervention group was 23 days (range, 5 to 76 days). There was no significant difference between groups 1 and 2 with regard to hospital stay and the percentage of patients who had a prolonged hospital stay (20% vs 28%, respectively). DISCUSSION Although considerable progress has been made in achieving successful repair of heart disease in the neonatal period, application of this strategy of early repair to the LBW neonate has been slowed by the perception that the LBW is a risk factor in surgical intervention. A previous study indicated a high mortality rate in infants who underwent open intracardiac operations in the first 3 months of life, 15 whereas we have extended our preference for corrective rather than palliative cardiac surgery in neonates ~6 to the LBW neonate. Diagnosis of congenital heart disease in the LBW neonate. The diagnosis of cardiac disease may be elusive in the LBW neonate because chest roentgenograms and the hyperoxia

The Journal of Pediatrics Volume 124, Number 3

test are not helpful in the presence of hyaline membrane disease. There is only limited information on the accuracy of echocardiographic diagnosis of CHD in the LBW neonate. 17 Coarctation of the aorta and total anomalous pulmonary venous connection were the lesions most often missed on the initial echocardiogram; the favorable outcome of surgery for both of these lesions increases the importance of accurate echocardiographic diagnosis. 18, 19 The advent of color Doppler studies and the development of 7.5 and 10 MHz imaging may further increase the sensitivity of echocardiography in the LBW neonate. Potential limitations in cardiac catheterization include lack of vascular access as well as a decreased amount of contrast medium and limited sizes of catheters tolerated by the LBW neonate. Because morphologic information may be more valuable than precise physiologic measurements, echocardiography alone may serve as the definitive imaging modality in many neonates2°, 21; only one third (38/100) of our LBW neonates underwent diagnostic cardiac catheterization. Although previous studies indicated a correlation between low patient weight and complications of catheterization,22, 23 the incidence of complications in our LBW neonates was not higher than in these previously reported series. Management of congenital heart disease in the LBW Neonate. The LBW neonate with CHD often has the characteristic problems of premature or SGA neonates related to the inadequate maturation of organ systems. These problems include metabolic disturbances such as electrolyte imbalances, hypothermia, and hypoglycemia; pulmonary diseases such as apnea; hyaline membrane disease and bronchopulmonary dysplasia; pulmonary hemorrhage; and others (including intraventricular hemorrhage, necrotizing enterocolitis, and renal insufficiency). For the VLBW neonates, the challenges of medical management are even greater.24,25 Genetic evaluation for lethal syndromes or chromosomal aberrations and assessment for major congenital anomalies are also important because these conditions are common in LBW neonates (30% in our series). The advantage of performing early corrective surgery is to decrease the potential secondary effects of chronic cyanosis and of pressure and volume overload26-2s; we therefore prefer to reserve palliative procedures for those neonates whose anatomy and physiology are not amenable to a two-ventricular repair. Utilization of cardiopulmonary bypass, necessary for corrective cardiac surgery, may also be contraindicated in the presence of certain medical conditions (such as intraventricular hemorrhage), and such patients will require surgical palliation instead. An alternative approach, if feasible, is catheter intervention. Our results for interventional catheterization in LBW neonates were good: palliative balloon atrial septotomy was effective in all six neonates, and balloon valve dilations were success-

Chang et al.

465

ful in obviating surgery in all six patients in whom they were attempted. Few studies have focused on the consequences of low patient weight on survival after cardiac surgery. One study 1l reported decreased survival rates for the arterial switch operation when birth weight was <3.0 ks. Another study compared infants with ventricular septal defects who weighed >4 kg and those who weighed <4 ks, and showed that there was no difference in surgical survival or complication rates. 29 Our group 79 previously reported survival rates for neonatal repair of tetralogy of Fallot, truncus arteriosus, and transposition of the great arteries to be 82% to 91%; the overall outcome of early corrective surgery in our LBW neonates is comparable (82% survival). The only other reported series on surgical outcome of LBW neonates with CHD reported a mortality rate of 17% in 60 patients. 6 Although it is encouraging that the incidence of corrective surgery in our LBW neonates increased and surgery seemed to have resulted in improved outcome from 1987 to 1991, the management of cardiac lesions that are not ductus dependent may not be straightforward, especially because there are few data on the progression of pulmonary vascular disease in LBW neonates with left-to-right shunts. Neonates with a common atrioventricular canal may benefit from an initial period of medical management because maturation of the atrioventricular valve tissue could increase the likelihood of a more stable surgical repair later in life. Conversely, delaying surgery for truncus arteriosus increases the severity of pulmonary hypertension postoperatively,3° and there may be no theoretical advantage in delaying surgical intervention.31 Although there are undoubtedly technical limitations of corrective surgery in the VLBW neonate, our series included an infant who weighed 1050 gm and underwent resection of a right atrial mass with cardiopulmonarybypass support and a neonate who weighed 1550 gm underwent successful corrective repair for truncus arteriosus. Overall, prolonged efforts to achieve medical stability and promote weight gain may not yield superior results compared with early surgical intervention; risks in medical management in our series included a 23% mortality rate as a result of sepsis and necrotizing enterocolitis. Despite the encouraging short-term results of early intervention for a heterogeneous group of cardiac lesions in LBW neonates, follow-up of these patients must be meticulous in the search for significant postoperative residua. Finally, growth in these children requires close follow-up because growth retardation is common even after early corrective surgery. 32 With appropriate screening for lethal genetic aberrations, major congenital anomalies, and coexisting medical illnesses, the LBW neonate with CHD can be successfully managed with early corrective surgery. Surgical correction in LBW neonates has increased, with good survival rates, in

466

Chang et al.

recent years. A l t h o u g h the experience with V L B W neonates is limited, we believe t h a t L B W > 1500 g m should not be a contraindication to a two-ventricular correction. Prolonged medical therapy to achieve further weight gain does not improve survival rates. We thank Dr. Richard A. Jonas and Dr. John E. Mayer, the surgeons for these children, and Dr. Donald Fyler and Ms. Cheryl King for their assistance. REFERENCES

1. Kitterman JA, Edmunds LH, Gregory GA. Patent ductus arteriosus in premature infants: incidence and relation to pulmonary disease and management. N Engl J Med 1972;287:473-7. 2. Friedman WE, Hirschklau M J, Printz MP. Pharmacologic closure of the patent ductus arteriosus in the premature infant. N Engl J Med 1976;295:526-9. 3. Ellison RC, Peckham G J, Lang P. Evaluation of the preterm infant for patent ductus arteriosus. Pediatrics 1983;71:364-72. 4. Fyler D. Report of the New England Regional Infant Cardiac Program. Pediatrics 1980;65:377-461. 5. Huhta JC. Structural heart disease in the low birth weight neonate. In: Adams FH, Emmanouilides GC, Riemenschneider TA, eds. Heart disease in infants, children, and adolescents. Baltimore: Williams & Wilkins, 1989:419-24. 6. Pawade A, Waterson K, Laussen P, Karl TR, Mee RB. Cardiopulmonary bypass in neonates weighing less than 2.5 kg: analysis of the risk factors for early and late mortality. J Cardiac Surg 1993;8:1-8. 7. DiDonato RM, Jonas RA, Lang P, Rome J J, Mayer JE, Castaneda AR. Neonatal repair of tetralogy of Fallot with and without pulmonary atresia. J Thorac Cardiovasc Surg 1991; 101:126-37. 8. Trowitzsch E, Sluysmans T, Parness IA. Repair of truncus arteriosus during infancy. J Am Coil Cardiol 1991;17:110A. 9. Norwood WI, Dobell AR, Freed MD, Kirklin JW, Blackstone EH. Intermediate results of the arterial switch repair. J Thorac Cardiovasc Surg 1988;96:854-63. 10. Murdison KA, Baffa JM, Farrell PE, etal. Hypoplastic left heart syndrome: outcome after initial reconstruction and before modified Fontan procedure. Circulation 1990;82(suppl IV):IV199-207. 11. Trusler GA, Castaneda AR, Rosenthal A, Blackstone EH, Kirklin JW. Current results of management in transposition of the great arteries with special emphasis on patients with associated ventricular septal defect. J Am Coil Cardiol 1987; 10:1061-71. 12. Quaegebeur JM, Rohmer J, Ottenkamp J, et al. The arterial switch operation: an eight year experience. J Thorac Cardiovasc Surg 1986;92:361-81. 13. Lantos JD, Miles SH, Silverstein MD. Survival after cardiopulmonary resuscitation in babies of very low birth weight. N Engl J Med 1988;318:91-5. 14. Kramer HH, Trampisch HJ, Rammos S, Giese A. Birth weight of children with congenital heart disease. Eur J Pediatr 1990; 149:752-7. 15. Kirklin JK, Blackstone EH, Kirklin JW, McKay R, Pacifico AD, Bargeron LM. Intracardiac surgery in infants under age

The Journal of Pediatrics March 1994

3 months: incremental risk factors for hospital mortality. Am J of Cardiol 1981;48:500-6. 16. Castaneda AR, Mayer JE, Jonas RA, Lock JE, Wessel DL, Hickey PR. The neonate with critical congenital heart disease: repair--a surgical challenge. J Thorac Cardiovasc Surg 1989; 98:869-75. 17. Pieroni DR, Gingell RL, Roland JM, Chung CY, Broda JJ, Subramanian S. Two dimensional echocardiographic recognition and surgical management of aortopulmonary septal defect in the premature infant. J Thorac Cardiovasc Surg 1982; 30:180-3. 18. Morrow R, Huhta JC, Murphy D J, McNamara DG. Quantitative morphology of the aortic arch in neonatal coarctation. J Am Coil Cardiol 1986;8:616-20. 19. Chin AJ, Sanders SP, Sherman F, Lang P, Norwood WI, Castaneda AR. Accuracy of subcostal 2-dimensional echocardiography in prospective diagnosis of total anomalous pulmonary venous connection. Am Heart J 1987;113:1153-9. 20. Huhta J, Glasgow P, Murphy D. Surgery without catheterization for congenital heart defects: management of 100 patients. J Am Coil Cardiol 1987;9:823-9. 21. Chang AC, Vetter JM, Gill SE, Franklin WH, Chin AJ. Accuracy of prospective two-dimensional Doppler echocardiography in the assessment of reparative surgery. J Am Coil Cardiol 1990;16:903-12. 22. Burrows PE, Benson LN, Williams WG. Iliofemoral arterial complications of balloon angioplasty for systemic obstructions in infants and children. Circulation 1990;82:1697-704. 23. Fellows KE, Radtke W, Keane JF, Lock JE. Acute complications of catheter therapy for congenital heart disease. Am J Cardiol 1987;60:679-83. 24. Hart SM, McNair M, Gamsu HR, Price JF. Pulmonary interstitial emphysema in very low birth weight infants. Arch Dis Child 1983;58:612-5. 25. Sadiq HF, Devaskar S, Keenan W J, Weber TR. Broviac catheterization in low birth weight infants: incidence and treatment of associated complications. Crit Care Med 1987;15:47-50. 26. Rabinovitch M, Herrera-deLeon V, Castaneda AR, Reid L. Growth and development of the pulmonary vascular bed in patients with tetralogy of Fatlot with or without pulmonary atresia. Circulation 1981;64:1234-49. 27. Flanagan MF, Fujii AM, Colan SD, Lock JE. Inhibitory effects on myocardial perfusion in pressure overload hypertrophy in immature lambs [Abstract]. Pediatr Res 1988;23:218A. 28. Newberger JW, Silbert AR, Buckley LP, Fyler DC. Cognitive function and age at repair of transposition of the great arteries in children. N Engl J Med 1984;310:1495-9. 29. Hardin JT, Muskett AD, Canter CE, Martin TC, Spray TL. Primary surgical closure of large ventricular septal defects in small infants. Ann Thorac Surg 1992;53:397-401. 30. Hanley FL, Heinemann MK, Jonas RA, et al. Repair of truncus arteriosus in the neonate. J Thorac Cardiovasc Surg 1993; 105:1047-56. 31. Bore EL, Lupinetti FM, et al. Results of a policy of primary repair of truncus arteriosus in the neonate. J Thorac Cardiovasc Surg 1993;105:1057-66. 32. Weintraub RG, Menahem S. Early surgical closure of a large ventricular septal defect: influence on long-term growth. J Am Coll Cardiol 1991;18:552-8.