Controversy exists regarding the integrity of the nervous system in the newborn with a congenital heart defect who must undergo corrective or palliative open heart surgery. Neurodevelopmental sequelae have been primarily attributed to surgical procedures without standardized evaluation of the preoperative neurologic status.
Objective. To determine whether newborns with congenital heart defects demonstrate abnormalities in neurobehavioral status before surgery.
Study Design. In this prospective study, a standardized neonatal neurobehavioral assessment and a neurologic examination were conducted independently in a consecutive series of 56 neonates referred to our hospital for investigation of open heart surgery.
Results. Neurobehavioral and neurologic abnormalities were documented in greater than half of the cohort and included hypotonia, hypertonia, jitteriness, motor asymmetries, and absent suck. Poor state regulation (62%) and feeding difficulties (34%) also were commonly observed. Furthermore, 3 subjects had seizures, 35.7% were microcephalic, and 12.5% were macrocephalic. The overall likelihood of neurobehavioral abnormalities was not enhanced by indicators of cardiorespiratory compromise. Interestingly, newborns with acyanotic congenital heart defects were more likely to demonstrate neurologic compromise than were those with cyanotic defects.
Conclusions. Findings suggest that the prevalence of neurobehavioral abnormalities before surgery in newborns with congenital heart defects has been underappreciated and would indicate that factors other than intraoperative procedures should be considered in the genesis of brain injury in this population.congenital heart defects, neurologic examination, newborn.
A congenital heart defect (CHD) is defined as a gross structural abnormality of the heart or of the major thoracic vessels.1 The incidence of CHD is estimated to be between 5 and 8 per 1000 live births, and many will require surgical interventions.2,,3 Recent improvements in diagnostic testing and surgical techniques have enabled early correction or palliation of most complex CHD in infancy, resulting in dramatic reduction in mortality.4–6 Surgical intervention early in life often is necessary in infants with CHD because of persistent cardiac failure or hypoxemia.7,,8 Cardiopulmonary bypass (CPB) and deep hypothermia coupled with total circulatory arrest (DHCA) are two surgical techniques that allow accurate repair of the heart in a bloodless, motionless operative field.7,,9 Considerable controversy exists with respect to neurologic morbidity, particularly with reference to intraoperative events such as CPB versus deep hypothermia coupled with total circulatory arrest.5,9–13
The integrity of the developing nervous system is influenced by a complex interaction between preoperative, intraoperative, and postoperative factors in children with CHD. Few studies to date have formally evaluated children with CHD preoperatively. One study14 reported that 10 of 12 subjects had abnormal neurologic status before surgery. In a second study describing preoperative neurologic status on 36 subjects, 5 children had microcephaly, 4 were hemiparetic, and 1 was hypotonic. Similarly, Brunberg and associates7 found that only 6 of 21 subjects assessed were normal neurologically. Psychomotor delay was present in 62%, hypotonia in 43%, and seizures and motor asymmetry in 1 patient. Newburger and colleagues9 reported suspect or definite neurologic abnormalities in 60% before surgery. Miller and colleagues15 reported on the perioperative neurologic findings of 91 infants and noted reduced levels of alertness (19%), seizures (15%), and severe hypotonia (11%) before surgery.
Only one study has described the neurobehavioral status of newborns with CHD. Gillon16 summarized her observations in the context of her nursing care of 82 newborns before surgery. Respiratory difficulties and poor coordination of suck, swallow, and breathing were often documented. Neurologic findings included tone abnormalities, abnormal posturing and activity level, weak cry, and poor auditory and visual orienting.
Neuroimaging and neuropathologic studies have provided evidence that central nervous system (CNS) injury or malformation already may be apparent preoperatively. Congenital malformations of the CNS may co-exist in a subset of individuals with CHD.17,,18 In magnetic resonance imaging studies of children with CHD, malformations such as callosal agenesis, abnormal neuronal migration, temporal lobar hypoplasia, and Chiari type I malformation have been reported.19,,20 Two recent studies reported on ultrasounds in term newborns with CHD, and found that 55% to 59% had findings including cerebral atrophy and linear echodensities in the basal ganglia and thalamus.15,,21 Human neuropathologic studies have demonstrated acquired CNS injury before surgery. Brain injury primarily involving the white matter may be of intrauterine or perinatal origin, or a genetic aberration.22–24 These findings in part may account for an increased frequency of microcephaly.25
Postoperative neurologic complications continue to be reported in the pediatric literature.15,,26 Developmental sequelae include cerebral palsy; seizure disorders; and cognitive, language, behavioral, and social problems.27–32 These neurodevelopmental sequelae may be caused not only by intraoperative events, but by preoperative brain abnormalities as a result of cardiac-related stresses17 (eg, chronic hypoxia, congestive heart failure, cardiac arrest, failure to thrive) and diagnostic or therapeutic procedures (eg, cardiac catheterization, balloon-atrial septostomy).
There is increasing evidence to suggest that newborns with CHD are at high risk for neurologic abnormalities; however, there is a paucity of studies that have described the baseline neurologic status of newborns with CHD using objective clinical assessments of CNS integrity. The primary aim of our study was to examine in a standardized and systematic manner the neurologic status of newborns with CHD before open heart surgery. Therefore, we hypothesized that a proportion of newborns with CHD will manifest neurologic abnormalities before surgery.
This study is the first phase of an ongoing prospective study that is examining the neurodevelopmental status of young children with CHD before and after open heart surgery.33 This project was scientifically and ethically approved by the hospital's Scientific Review Committee and the Institutional Review Board. A consecutive series of newborns referred to the Division of Cardiology at the Montreal Children's Hospital for investigation of CHD were recruited (fall 1994 to spring 1997). Neurologic, neurobehavioral, and cardiorespiratory assessments were performed within the first month of life before cardiac surgery.
Inclusion criteria were 1) gestational age >36 weeks, and 2) a diagnosis of a CHD requiring palliative or corrective open heart surgery. All subjects had CPB ± circulatory arrest during surgery. Subjects were excluded because of 1) hypoplastic left heart syndrome (because of the higher prevalence of neurologic morbidity in this subgroup than in others);34 2) closed surgical procedures; 3) a language barrier (ie, parents who did not speak either English or French); 4) known extracardiac anomalies involving the CNS (eg, Down syndrome); or 5) known insult to the CNS (eg, perinatal asphyxia) not directly attributed to the heart defect. Criteria for perinatal asphyxia included criteria established by the American Academy of Obstetrics and Gynecology and the American Academy of Pediatrics35 and included 1) profound metabolic or mixed acidosis (pH < 7.10); 2) Apgar scores of ≤3 at 5 or more minutes; 3) clinical evidence of neonatal encephalopathy; and 4) evidence of multiorgan dysfunction in the immediate neonatal period. It should be noted that CNS insults or malformations were limited to those recognized clinically (by neonatology or cardiology attending physicians) by physical examination and by neuroimaging. Furthermore, patients with clinically recognized syndromes that are associated with developmental disability (and confirmed by genetic testing) were excluded. Therefore the population of interest was newborns with CHD who were not known to have any disorder or impairment of the CNS attributable to causes other than complications of the CHD at the time of testing.
Once informed parental consent was obtained, the subjects were assessed within the newborn period (ie, first month of life). The Einstein Neonatal Neurobehavioral Assessment Scale (ENNAS) was performed by an occupational therapist. Data from 37 healthy control subjects obtained for a previous study36 were used to compare performance between subjects with CHD and full-term, healthy newborns. The control group was recruited at a well-baby nursery and were term neonates (>36 weeks' gestational age) with an Apgar score of 8 to 10 at 5 minutes, a birth weight appropriate for age, and no history of prenatal or perinatal complications.
One of two pediatric neurologists examined the subjects, and a pediatric cardiologist reviewed the medical charts to document cardiorespiratory status at the time of assessment. All inpatients were assessed in the neonatal intensive care unit (NICU). All subjects were evaluated clinically by a neurologist and/or an occupational therapist who were blinded to the cardiac diagnosis and to each other's clinical findings.
The ENNAS is a formal neurologic assessment that evaluates neurologic status and behavioral organization of the newborn infant. Standardized procedures and methods are used in performing each individual item; however, the order of presentation of each item is such that state is optimized. The areas evaluated include muscle tone, passive and active movements, primitive reflexes, and responses to visual and auditory stimuli.37 The ENNAS comprises 20 individual items and 4 summary items and takes 20 to 30 minutes to administer. Interrater reliability has been determined to be 0.97 by Kurtzberg and co-workers.38 Each item is scored on an ordinal scale sequenced from minimum to maximum response. Cut-offs for normal/abnormal for each item is based on established criteria with the suspect or “mildly abnormal” rating recently being added (C. McCarton, personal communication, 1994). A deviant score (ie, number of items failed) between 0 and 2 is considered normal, whereas a deviant score between 3 and 6 is a suspect examination and greater than 6 is considered abnormal. Neonates with “suspect” findings demonstrate subtle or mild abnormalities (eg, mildly hypotonic, poor orienting responses) and would be considered clinically abnormal. For the purpose of analysis, suspect or abnormal examinations were clustered as abnormal. This test discriminates well among groups of healthy and high-risk newborns.36 Two prospective studies have examined the predictive value of the ENNAS in preterm and full-term high-risk newborns. The ENNAS had very good negative predictive value and sensitivity (83% to 100%) for cognitive and motor performance at school entry.39 In particular, the auditory and visual composite scores correlated with intelligence quotients at 6 years.40
The occupational therapist also completed an observation sheet that included information about the infant's perinatal status. General feeding status was recorded as described by the primary caregiver and included the method of feeding and feeding efficiency (ie, amount of intake per feed, duration of feeding, need for nasogastric supplementation). Abnormalities in behavioral state organization (ie, lethargy, excessive irritability/agitation, inability to maintain a quiet alert state) during the assessment, muscle tone, and the overall impression regarding the infant's neurobehavioral status also were documented at assessment.
A pediatric neurologist examined all subjects based on criteria outlined by Volpe.41 The examination took ∼15 minutes and included assessment of head circumference, muscle bulk and tone, cranial nerves, deep tendon reflexes, and activity level and the presence of any abnormal movement patterns such as posturing or tremor. Subjects were categorized as normal or abnormal.
A pediatric cardiologist reviewed the medical chart of each subject on the day of assessment and documented the transcutaneously measured arterial oxygen saturation, respiratory rate, whether newborns were intubated, medications in use, and the presence of cyanotic/acyanotic lesions and/or congestive heart failure. The cardiologist also recorded whether the subjects were evaluated as in- or outpatients because, presumably, outpatients would be more stable from a cardiorespiratory viewpoint. These parameters were used as indicators of cardiorespiratory compromise.
Descriptive statistics were used to characterize the sample with respect to perinatal and cardiorespiratory factors. Proportions were used to describe the presence and type of neurologic abnormalities. The performance on the ENNAS was compared between subjects and healthy control subjects using χ2 Fisher exact test (for items) and the Mann–Whitney test (distribution of items). χ2(Fisher exact test) analysis also was used to compare neurobehavioral abnormalities with cardiorespiratory factors. A multiple regression model was applied with deviant score and classification of normal/abnormal on the ENNAS and neurologic examination as the outcomes. Microcephaly and cyanotic/acyanotic lesion were the primary independent variables, adjusting for use of prostaglandins, congestive heart failure, ventilation, oxygen saturation.
Of 135 newborns who presented in the neonatal period to Cardiology and required cardiovascular surgery for a CHD, 57 were excluded because of prematurity (35); “closed” surgical procedures not requiring CPB (9); language barrier (5); CNS anomalies (3); hypoplastic left heart syndrome (3); and syndromes (2). Of the remaining 78 newborns, 60 families were approached for consent and 56 (93.3%) agreed to participate in the study. Eighteen families were not approached for consent because of lack of availability of the caregivers or investigators. Of the 56 subjects recruited, 33 (58.9%) were males. All subjects were full-term, and birth weights were appropriate for gestational age in all subjects. Apgar scores at 5 minutes ranged from 7 to 10 in all but 1 subject, whose score was 4 (this subject's 10-minute Apgar score was 7) (Table 1). Nine of 56 (16.1%) subjects were evaluated as outpatients, whereas the remainder were assessed in the NICU. Subjects fell into one of the following diagnostic categories of CHD: transposition of the great arteries (12), complex (6), tetralogy of fallot (6), tetralogy of fallot with pulmonary atresia (5), coarctation of the aorta (with aortic arch hypoplasia) (5), ventricular septal defect (3), aortic stenosis (3), interrupted aortic arch (3), double outlet right ventricle with subaortic ventricular septal defect (3), double outlet right ventricle with subpulmonary ventricular septal defect (3), atrioventricular septal defect (2), truncus arteriosus (2), pulmonary atresia with intact ventricular septum (2), and total anomalous pulmonary venous drainage (1).
Cardiorespiratory status was documented by the cardiologist in all 56 newborns at the time of neurobehavioral assessment. Twelve (21.4%) subjects were intubated and ventilated, 2 (3.6%) subjects were not intubated but were receiving supplemental inspired oxygen, and the remainder (75.0%) were breathing spontaneously in room air. Thirty-nine newborns (69.6%) had cyanotic heart defects and 17 (30.4%) had acyanotic heart lesions. Twenty-two (39.3%) had arterial oxygen saturations <85%, 17 (30.4%) were tachypneic (ie, respiratory rate ≥55 breaths per minute), and 13 (23.2%) had clinical signs of congestive heart failure. Twelve subjects were not receiving any medications, whereas 33 were receiving prostaglandin E2, 15 furosemide, 5 digoxin, 2 hydrochlorothiazide, 1 spironolactone, 1 ethacrymic acid, 1 dobutamine and dopamine, and 4 fentanyl/midazolam.
Description of Test Performance of Newborns
Of 56 subjects recruited, the neurologist was available to examine 50 before surgery at a mean age of 13.9 ± 11.8 days (range, 1 to 44 days), with outpatient evaluations conducted somewhat later (mean, 25.4 days). In 1 subject, although recruited before 30 days of life, medical complications delayed surgery and therefore testing was conducted at 44 days. All other subjects were younger than 1 month of age. Fifty-six percent (28 of 50) demonstrated one or more abnormal neurologic findings including diffuse hypotonia (20), hypertonia (6), jitteriness (4), no suck (3), motor asymmetry (2), decreased muscle power in the upper and lower extremities (5), and cranial nerve abnormalities (2). Altered states of consciousness such as lethargy or stupor also were documented (7), as well as restlessness and agitation (4). Preoperative seizures were clinically reported in 4 subjects. In addition, microcephaly (ie, head circumference at or below the third percentile) was documented in 35.7% (20) of subjects (Fig 1). Two thirds of these neonates with microcephaly had abnormal neurologic examinations. There was a negative effect (odds ratio [OR]: 1.72; confidence interval: 0.42–7.12) of microcephaly on neurologic status; however, this did not reach significance. Results were comparable with the ENNAS as the outcome variable (odds ratio: 1.86).
Of the 56 subjects in our sample, all but 6 were evaluated by an occupational therapist using the ENNAS, at a mean age of 12.6 ± 11.5 days (range, 1 to 44 days). Twenty-one (42%) subjects had a normal examination (deviant score, 0 to 2); 19 neonates (38%) had a deviant score between 3 and 6, indicating a suspect examination; and 10 (20%) had an abnormal ENNAS (>6). Abnormalities observed and recorded on suspect or abnormal ENNAS examinations included hypotonia (22), hypertonia (5), jitteriness (4), absent suck (4), and motor asymmetry (2). Furthermore, poor state regulation was documented in 31 (62%) subjects. This was characterized by lethargy defined as an inability to arouse or maintain an optimal state of arousal (14), and/or excessive irritability and abrupt swings between sleep and crying states with no quiet alert periods (17). Poor state modulation limited the ability to evaluate orienting responses in some subjects. Nonetheless, in those assessed, poor orienting responses were observed often, which consisted of poor visual fixation and tracking (15/30) as well as poor auditory alerting (29/44). It should be considered that clinical assessment of the very ill neonates was limited at times and incomplete because of the presence of central or peripheral lines, ventilators, etc, restricting the amount of handling and position changes. As a result, the prevalence of neurobehavioral abnormalities may be underestimated because some items could not be completed.
Of 56 subjects, 26 were fed orally (breast or bottle); however, 5 required nasogastric supplementation, and 30 were on total parenteral feedings. Thirteen (26%) subjects had a weak suck, and 4 had no suck as documented on the ENNAS (item 4). All newborns with a weak nonnutritive suck also were found to be hypotonic on neurobehavioral assessment. Feeding efficiency was documented informally on subjects who were fed orally, based on information provided by the primary caregiver (eg, primary nurse or mother) regarding the infant's feeding behaviors. Decreased feeding efficiency was noted in 34% and was characterized by a longer feeding time and more frequent feedings to ensure adequate oral intake and sufficient weight gain.
Comparison of Performance on the ENNAS Between Subjects and Healthy Controls
The overall deviant score was significantly different between newborns with CHD (mean, 3.86; standard deviation [SD], 2.4; median, 4.0) and healthy controls (mean, 0.5; SD, 0.7; median, 0). There also were significant differences in test performance on several subtests, including orienting responses (both visual and auditory subtests) and passive and active movements (eg, head extension, head lag, muscle tone) (Table 2).
Significant differences in the distribution of scores were documented in the majority of items, with the exception of the following: lateral head preference (which evaluated head position in supine at rest), popliteal angle (which determines the degree of flexion at the knees), asymmetric tonic neck reflex, postrotatory nystagmus, and the presence of tremor (Table 2).
Relationship Between Neurobehavioral Abnormalities and Cardiorespiratory Status
χ2 analyses (Fisher's exact test) demonstrated a significant association (P < .05) between neurobehavioral status (normal/abnormal) and cyanotic versus acyanotic CHDs, in which newborns with acyanotic CHDs were more likely to be neurologically abnormal (78.6%) than those with cyanotic defects (47.2%) (Fisher exact P = .06). Multiple logistic regression demonstrated further that acyanotic lesions were associated with an abnormal neurologic examination (odds ratio, 4.10; CI, 0.98–17.2). However, there was no significant association (P > .05) between neurobehavioral performance and any other indicators of cardiorespiratory compromise including the presence of congestive heart failure, tachypnea, oxygen saturation <85%, use of prostaglandin E2, mechanical ventilation, or inpatient versus outpatient status. Furthermore, an unpaired t test showed no significant difference (P = .19) in oxygen saturation in those subjects with normal ENNAS scores (mean oxygen saturation, 88.2% ± 8.4) and those with abnormal ENNAS (mean oxygen saturation, 91.0% ± 8.0).
In addition, specific medical factors that may influence neurologic status were reviewed in selected patients. Hypertension, a complication that may be associated with neurologic abnormalities, was not a feature in our newborns with coarctation of the aorta, because all were normotensive before surgery. Similarly, hypocalcemia may be associated with jitteriness and seizures41; however, calcium levels were normal in this subset of subjects. It should be noted that only 1 of the 4 children on fentanyl/midazolam was hypotonic; 2 were hypertonic (despite fentanyl), and 1 had a normal examination. Of 9 subjects assessed as outpatients, 5 were abnormal neurologically (55.6%: similar to the inpatients evaluated).
The results of this study demonstrated that neurobehavioral abnormalities on standardized testing were common in newborns with CHD before open heart surgery. There was excellent agreement (crude agreement, 96.9%; κ statistic = .94) between the two independent evaluations of overall neurologic status conducted by the neurologist and occupational therapist.42 This reinforces the validity of the findings, because the two approaches in neurologic evaluation of these newborns yielded very similar results. Neurobehavioral findings were characterized by seizures, muscle tone abnormalities, motor asymmetries, jitteriness, and poor orienting responses. These findings support Gillon's16 clinical observations.
More than one third of newborns with CHD also were microcephalic despite birth weights appropriate for gestational age, suggesting possible disruption in antenatal brain development, when the brain is undergoing the most rapid rate of cell division.43Moreover, 66.7% of newborns with microcephaly in this sample also had abnormal neurologic examinations. This subgroup of children may be at increased risk for later developmental disability. Although there was a nonsignificant association between microcephaly and abnormal neurologic status, a larger sample would be needed to confirm this finding.
Neurobehavioral state organization is defined as an infant's ability to demonstrate well defined states, and to make smooth and organized transitions between states (eg, sleep to arousal, to alert, to crying). The infant who can not be aroused, who is excessively irritable, or who swings abruptly between states with no alert periods may be demonstrating CNS immaturity or a pathologic condition.44Thirty one (62%) newborns with CHD demonstrated “poorly modulated” behavioral state organization profiles. Als and associates45 proposed that poor state organization in preterm infants may serve as a necessary protective reflex for shutting out excessive stimulation in an attempt to maintain physiologic homeostasis. However, this may have a secondary consequence of sensory and interactional deprivation. Poor state regulation in our subjects may serve as an intrinsic protective mechanism in an attempt to obliterate additional physiologic instability exacerbated by the environment. One also could postulate that the poor state regulation may be attributable to acute hypoxic–ischemic injury in a subset of subjects. It should be noted that poor state regulation alone (or poor orienting responses) did not constitute an abnormal neurobehavioral assessment, but occurred in concert with other neuromotor abnormalities. Although poor state control may affect orienting responses, it could not explain the neuromotor abnormalities noted in our cohort.
Feeding difficulties also were noted frequently. Thirteen (26%) subjects had a weak nonnutritive suck, 4 had no suck, and 17 (34%) presented with decreased feeding efficiency. Moreover, all newborns with a weak suck also were found to be hypotonic. Therefore, poor oral–motor skills did not appear to be related exclusively to decreased endurance, but possible CNS injury or transient perturbation may have impacted on feeding patterns. More standardized measures of feeding performance are needed to validate these findings.
Comparison of performance on the ENNAS between subjects with CHD and healthy term control subjects demonstrated significant differences in test performance between the two groups, particularly in the distribution of scores. Poor visual and auditory responses were characteristic of the vast majority of newborns with CHD; however, performance may have been influenced by poor state regulation, which was a hallmark feature in this cohort. Newborns with CHD also manifested motor deficits such as decreased limb traction and recoil, poor antigravity movements, limited spontaneous movements, as well as tone abnormalities. The high prevalence of acute neurobehavioral abnormalities is comparable with the findings of other NICU populations and would suggest that careful surveillance of developmental progress may be justified. The literature suggests that many of these neonates with neurobehavioral abnormalities may be at greater risk for long-term neurodevelopmental sequelae.39,,46
The relationship between specific cardiorespiratory factors and neurobehavioral performance was examined to determine whether neurobehavioral status was influenced by the acuity of illness. Analysis revealed no significant association between individual medical parameters examined (ie, oxygen saturation, tachypnea, congestive heart failure, intubation, administration of prostaglandins) and an increased risk for neurologic findings with the exception of cyanotic versus acyanotic heart defects. This perhaps may be attributed to decreased systemic blood flow and potentially cerebral blood flow in those patients with acyanotic heart defects exhibiting congestive heart failure. Additionally, a number of acyanotic heart defects result in impaired blood flow in the ascending aorta and thus may limit cerebral perfusion. This compromise in cardiovascular physiology may have an antenatal origin. Brain injury may ensue after alterations in cerebral perfusion, possibly resulting in impaired autoregulation of cerebral blood flow.7 Conversely, newborns with cyanotic CHD may demonstrate sufficient adaptability to chronic hypoxemia. Metabolic needs may be reduced or blood flow may be redistributed to organ systems with the greatest oxygen requirements to accommodate for the oxygen insufficiency in the blood. These hypotheses would need to be substantiated by future experimental studies. Interestingly, van Houten and co-workers21 reported that newborns with coarctation of the aorta and ventricular septal defects (acyanotic lesions) had a higher incidence (71%) of ultrasound abnormalities such as intraventricular hemorrhage, cerebral atrophy, and echodensities in the basal ganglia and thalamus before surgery. The association between acyanotic heart lesions and a greater risk for neonatal neurologic abnormalities was borderline significant and would therefore need to be validated on a larger sample.
As mortality rates after open heart surgery continue to decline, the morbidity of survivors has come under increasing scrutiny. Our findings suggest that the prevalence of neurobehavioral abnormalities in the infant cardiac population before surgery has been underappreciated, and that these infants may require regular developmental screening. Moreover, preoperative neurologic dysfunction in turn may render infants more susceptible to additional neurologic injury during cardiac surgery. Studies are needed to assess the determinants of brain injury so that neurologic sequelae of CHD may be reduced. Given the multiple needs of this population, interdisciplinary efforts among cardiovascular surgeons, cardiologists, pediatricians, neurologists, neonatalogists, rehabilitation specialists, and nutritionist experts are critical to best meet the needs of this high-risk group. Such collaborative efforts are essential to prevent or minimize neurodevelopmental morbidity through early identification of infants at risk and permit the initiation of early intervention programs to enhance the outcome of survivors.
This study was funded by the National Health Research and Development Program (Health Canada), and the Montreal Children's Hospital—Research Institute.
We thank the attending staff of the Division of Newborn Medicine, Dr Marie Beland and Dr Luc Jutras from the Division of Cardiology, and Johanne Therrien for their assistance in recruitment of subjects. Special thanks to Joseph Falworth for coordination of the project, data entry, and manuscript preparation, as well as to Dr Michel Abrahamowicz and the Biostatistical Consultating Service at the Montreal Children's Hospital for statistical consultation. We are indebted to the families who participated in the study.
- Received November 26, 1997.
- Accepted August 13, 1998.
Reprint requests to (A.M.) School of Physical and Occupational Therapy, McGill University, 3654 Drummond, Montreal, Quebec H3G 1Y5 Canada.
- CHD =
- congenital heart defects •
- CPB =
- cardiopulmonary bypass •
- CNS =
- central nervous system •
- ENNAS =
- Einstein Neonatal Neurobehavioral Assessment Scale •
- NICU =
- neonatal intensive care unit •
- OR =
- odds ratio •
- SD =
- standard deviation
- ↵Perry LW, Neill CA, Ferencz C, et al. Infants with congenital heart disease: the cases. In: Ferencz C, Rubin JD, Loffredo CA, Magee CA, eds. Epidemiology of Congenital Heart Disease: The Baltimore–Washington Infant Heart Study 1981–1989. Mount Kisco, NY: Furuta; 1993:33–61
- ↵Clark EB, ed. Epidemiology of congenital cardiovascular malformations. In: Emmanouilides GC, Riemenschdeider TA, Allen HD, Gutesell HP, eds. Heart Disease in Infants, Children and Adolescents. 5th ed. 1995;1:60–70
- Stuart AG
- Brunberg JA,
- Reilly D,
- Doty DB
- Newburger JW,
- Jonas RA,
- Wernovsky G,
- et al.
- Royston D
- Sanderson JM,
- Wright G,
- Sims FW
- Taylor R,
- Burrows F,
- Bissonette B
- Clarkson PM,
- MacArthur BA,
- Barratt-Boyes BG,
- Whitlock RM,
- Neutze JM
- ↵Newburger JW. Central nervous system sequelae of congenital heart disease. In: Flyer DC, ed. Nadas Pediatric Cardiology. Philadelphia, PA: Hanley & Belfus; 1992:77–82
- ↵Burn J. Aetiology of congenital heart disease. In: Anderson RH, Macartney FJ, Shinebourne EA, Tynan M, eds. Pediatric Cardiology. Edinburgh, Scotland: Churchill Livingstone; 1987:2–14
- Miller G,
- Mamourian AC,
- Tesman JR,
- Baylen BG,
- Myers JL
- Schilr M,
- Brierley J
- Cohen M
- Gilles FH,
- Leviton A,
- Jammes J
- Bellinger DC,
- Wernovsky G,
- Rappaport LA,
- Mayer JE,
- Castaneda AR,
- Farell DM
- Haka-Ikse K,
- Blackwood MJ
- Hesz N,
- Clark EB
- ↵Majnemer A, Limperopoulos C, Shevell M, Rosenblatt B, Rohlicek C, Tchervenkov C. Neurologic status prior to and following open heart surgery in newborns and infant. Circulation. 1997;96(suppl 1):130. Section I
- ↵American Academy of Pediatrics/American College of Obstetricians and Gynecologists. Relationship between perinatal factors and neurologic outcome. In: Poland RC, Freeman RK, eds. Guidelines for Perinatal Care. 3rd ed. Elk Grove Village, IL: American Academy of Pediatrics; 1992:221–224
- ↵Daum C, Grellong B, Kurtzberg D, Vaughan HG. The Albert Einstein Neonatal Neurobehavioral Scale. 1977. Unpublished manual
- ↵Volpe JJ, ed. Neurology of the Newborn. 3rd ed. Philadelphia, PA: WB Saunders; 1995
- Limperopoulos C,
- Majnemer A,
- Rosenblatt B,
- Shevell M,
- Rohlicek C,
- Tchervenkov C
- Winick M,
- Ross P
- ↵Hunter JG. The neonatal intensive care unit. In: Case-Smith J, Allen AS, Pratt PN, eds. Occupational Therapy for Children. St Louis, MO: Mosby; 1996:583–647
- ↵Als H, Lester B, Brazelton TB. Dynamics of the behavioral organization of the premature infant: a theoretical perspective. In: Field TM, ed. Infants Born at Risk. New York, NY: Spectrum Publications; 1979:173–192
- Allen ME,
- Capute AJ
- Copyright © 1999 American Academy of Pediatrics