Abstract
Objectives. The purpose of this study was to assess the neurodevelopmental status of children after Fontan repair of functional single ventricle and to examine the relationship between cognitive function and selected patient characteristics.
Study Design. Neurodevelopmental tests including the Stanford-Binet Intelligence (IQ) scale and the Developmental Test of Visual Motor Integration (VMI) were administered to 32 children (26 months to 16 years of age) with complex single ventricle. The mean and distribution of IQ and VMI scores were compared with population norms. The relationship between test scores and patient characteristics was examined utilizing analysis of variance and correlational methods.
Results. The majority of children had intellectual function within the normal range (mean, 97.5 ± 12.1). Below average VMI scores were found in 21.4% of children. There were no significant correlations between intellectual function or visual motor integration ability and preoperative oxygen saturation or age at Fontan. Children who had deep hypothermic circulatory arrest during a prior Norwood procedure tended to have a lower IQ score.
Conclusions. Intellectual development in children with Fontan repair of complex heart defects is essentially within the normal range. Visual motor integration deficits may be more prevalent in these children. In our population, the duration and degree of preoperative hypoxemia had no apparent effect on cognitive function.
- VMI =
- Developmental Test of Visual Motor Integration •
- IQ =
- Stanford-Binet intelligence scale
For many children with complex congenital heart disease, the Fontan procedure has become the definitive surgical treatment. Since its introduction more than 25 years ago for patients with tricuspid atresia, modifications of the Fontan—including the lateral baffle technique, staging via a superior cavopulmonary (hemi-Fontan) anastomosis, and fenestration of the Fontan baffle—have allowed increased application of the Fontan approach to treat most forms of functional single ventricle.1-3 Evolution of these surgical techniques and other treatment strategies has contributed to improved overall survival of these children.4 At our institution, overall survival is 87% at 5 and 10 years after the Fontan operation. Although existing data regarding survival has justified extension of the Fontan procedure to younger patients with increasingly complex disease, little is known about the functional outcomes in these early survivors with complex congenital heart defects.
In children with Fontan correction, multiple preoperative, perioperative, and postoperative factors may affect developmental and neurologic outcome.5 6 Potential preoperative factors include coexisting neurologic problems, periods of acidosis or low cardiac output, poor nutrition, and chronic congestive heart failure or hypoxemia. In these children, who usually undergo more than one cardiac surgical procedure, prolonged circulatory arrest time, multiple or prolonged cardiopulmonary bypass periods, profound hypothermia, and systemic to pulmonary artery collateral circulation may contribute to cerebral injury.5 6 Postoperative factors including arrhythmias, cerebral edema and elevated central venous pressure, and thromboembolic events may adversely affect neurodevelopmental outcomes. The purpose of this study was to evaluate the neurodevelopmental status of children after the Fontan procedure. An additional objective was to examine the association between cognitive function and the child's age at Fontan and preoperative oxygen saturation.
METHODS
Patient Population
The target patient population consisted of 68 survivors of Fontan procedures (n = 74) performed between January 1986 and December 1994 who were less than 16 years of age at operation and more than 6 months post-Fontan. Fifteen patients who were non-English speaking, 10 patients who no longer reside in California, and 5 who are followed by other institutions closer to their homes, were excluded from the study. One patient with Down syndrome was also not included. Of the 37 remaining eligible children, 32 patients/families (86%) agreed to participate in the study. There were 17 girls and 15 boys. Based on parental occupation/employment, 9 (28%) families were classified as upper, 15 (47%) as middle, and 8 (25%) as low socioeconomic status. Age at testing ranged from 26 months to 16 years (median, 5.3 years). Eight children (25%) were more than 8 years old.
Preoperative/Operative Characteristics
The median age at Fontan for the study group was 30 months (range, 19 to 114 months). As shown in Table 1, the primary cardiac diagnoses were tricuspid atresia in 6 patients (19%), pulmonary atresia with intact septum in 6 (19%), double inlet left ventricle in 4 (12.5%), hypoplastic left heart syndrome in 4 (12.5%), complex double outlet right ventricle in 4 (12.5%), and other forms of functional single ventricle in 7 (22%). Sixteen patients had a systemic left ventricle and 14 had a systemic right ventricle. Most patients (97%) had prior palliative surgery including an intermediate stage bidirectional Glenn or hemi-Fontan connection in 17 patients (53%). Oxygen saturation measured at cardiac catheterization before Fontan ranged from 71% to 91% (mean, 82%). Hemoglobin, reflecting the level of chronic hypoxemia, ranged from 11.7 to 18.2 g (mean, 16.1 g). Types of Fontan included a lateral baffle in 18 patients (56%), a Bjork modification in 2 patients, and an extra cardiac conduit in 1 patient. Eleven early patients had a direct atriopulmonary connection. Cardiopulmonary bypass time at Fontan operation ranged from 65 to 243 minutes (mean, 129 minutes).
Patient Population—Preoperative/Operative Characteristics (n = 32)
Measures
Neurodevelopmental tests, including assessment of the child's level of intellectual functioning and visual motor integration skill, were administered by a trained psychometrist, supervised by a pediatric psychologist (A.L.). The Stanford-Binet Intelligence Scale,7 standardized for ages 2 to adult, was used to assess intelligence level. The Bayley Scales of Infant Development were substituted in 1 child only 27 months of age at testing. The Stanford-Binet yields a total mental processing composite based on subtests measured from four domains: verbal reasoning, numeric reasoning, abstract visual reasoning, and short term memory. The Developmental Test of Visual Motor Integration (VMI)8 was used to assess visual spatial and visual motor integration ability. The test, a developmental sequence of 24 geometric forms to be copied with paper and pencil, is designed to identify potential learning problems. The psychometric tests are age-normed to have a mean of 100. Modest correlations between the VMI and intelligence scores have been reported, ranging from .37 to .59.8 Parents completed the Achenbach Child Behavior Checklist,9 which was designed to record in a standardized format the behavioral problems and competencies of children age 2 through 16 years, as reported by their parents. The Child Behavior Checklist provided information regarding the child's exercise activities and school performance.
Statistical Analyses
Psychometric test scores were examined in relation to population norms and standard deviations. To examine the relationships between test scores and categorical variables, analysis of variance with the Kruskal-Wallis procedure was utilized. Pearson product moment correlations were computed to assess the relationship between test scores and continuous variables such as oxygen saturation and age at Fontan.
RESULTS
The mean intelligence quotient in children after Fontan repair of functional single ventricle was 97.5 (range, 70–131). Most children scored within the normal range (Fig 1). Only 1 child was below the normal range and would be classified as a slow learner. Analysis of Stanford-Binet subscale scores showed a mean of 98.1 ± 14.3 for verbal reasoning, 95.8 ± 11.4 for abstract visual reasoning, 97.3 ± 9.7 for numeric reasoning, and 98.2 ± 13.6 for short-term memory (Table 2).
Distribution of intelligence scores.
Neurodevelopmental Outcomes
The mean score on the VMI (n = 24) was 94.8 with a median score of 93, suggesting a relative mild weakness in visual motor integration. As show in Fig 2, 6 (21.4%) of 28 children had below-average VMI scores, with 15% being the expected number below the average range. The correlation between overall IQ score and VMI score was .436, P < .05.
Distribution of visual motor integration scores.
In school-aged children, performance in academic subjects was above average in 45%, average in 32%, and below average in 23%, by parental report. No child had repeated a grade.
No significant correlation was found between intellectual function (IQ) or visual motor integration and age at Fontan, pre-Fontan oxygen saturation, preoperative hemoglobin, or duration of cardiopulmonary bypass at Fontan surgery (Table 3). There was no significant difference in IQ in children with a systemic left ventricle (mean, 98.8) as compared with children with a systemic right ventricle (mean, 96.1). Children who had cardiac surgery (a Norwood procedure) utilizing deep hypothermic circulatory arrest before the Fontan procedure (n = 5) tend to have a lower mean IQ score (88.4) than children who had not undergone circulatory arrest (99.1),P = .06. Mean circulatory arrest time at Norwood was 47.8 ± 8.2 minutes.
Correlations Between Patient Preoperative/Operative Characteristics and Neurodevelopmental Outcomes
With respect to physical functional status, children participated in age-appropriate activities with the exception of competitive contact sports. Recreational activities included bike riding, swimming, dancing, skating, baseball, basketball, karate, and gymnastics. One child has a resolving hemiparesis, related to a thromboembolic event that occurred 14 months after a fenestrated Fontan. One child has asthma. No other health problems, including neurologic complications, were identified.
DISCUSSION
Therapeutic advances that have resulted in improved survival and surgical outcomes for children with congenital heart disease during the past 3 decades may have also enhanced the neurodevelopmental outcomes in these children. In our study, the majority of children with Fontan correction of complex heart disease and functional single ventricle had intellectual function within the normal range. In early studies, developmental delay and lower IQ were more common in children with congenital heart disease.10 11 These lower scores were more evident in the first 3 years of life, especially in cyanotic children, when decreased physical capacity or symptoms would have negatively influenced gross motor function.12 Age at cardiac repair was also older than current practice at most cardiac centers.
In contrast to early studies, our data revealed no significant correlation between IQ or visual motor integration ability and degree of hypoxemia or age at Fontan. Previous studies have reported lower intelligence for cyanotic than acyanotic children, attributed to the degree of hypoxemia.13 14 It is further suggested that these differences persist even when the effect of neurologic abnormalities and definitive surgery is removed and remain despite the severity of illness or the child's age at testing.13Newburger and colleagues14 also reported that postponing repair of a cyanotic congenital heart defect is associated with progressive impairment of cognitive function. The degree or duration of hypoxemia had no apparent effect on cognitive outcome in our patients. Consistent with our findings, Oates et al15 has recently reported no evidence of detrimental effect of older age at operation in children with cyanotic heart disease as assessed by neuropsychologic measures.
As previously mentioned, the likelihood of cerebral injury or poor neurodevelopmental outcome is also believed to be related to cardiopulmonary bypass, duration of hypothermic circulatory arrest, and patient age at time of surgery.5 16 Although longer cardiopulmonary bypass at Fontan had no apparent adverse effects on intelligence or visual motor integration ability, children who had prior deep hypothermic circulatory arrest with the Norwood procedure had a somewhat lower IQ. In a study of infants who underwent an arterial switch, surgery performed with circulatory arrest as the predominant support strategy has been associated with a higher risk of delayed motor development and neurologic abnormalities at 1 year of age.17 The score on the Psychomotor Development index17 was inversely related to the duration of circulatory arrest. In an earlier report, these investigators reported that the intraoperative variable most strongly associated with cognitive outcome was the duration of core cooling on cardiopulmonary bypass before deep hypothermic circulatory arrest.18 Our patients with prior circulatory arrest during Norwood procedures had intelligence quotients ranging from 70 to 100, in contrast to a recent report of major developmental disabilities (4 moderate and 3 severe-to-profound mental retardation) in 7 (64%) of 11 children who had prior Norwood procedures.19 Although the number and type of palliative or staged procedures before Fontan varied among our patients, the Fontan operation was the definitive surgical procedure. It is beyond the scope of our study to isolate the effects of these palliative procedures on intellectual functioning from the status after cardiac repair. The normal intellectual functioning in the majority of our patients after Fontan is encouraging, however, in view of the reported stability between preoperative child IQ and postoperative adult IQ in patients with cyanotic tetralogy of Fallot.20
Our data suggested a mild weakness in visual motor integration in these children. Poor visual motor integration can result from slow maturation, inadequate educational experiences, socioeconomic disadvantage, and/or neurologic and chronic health problems.8 Three children with borderline to low VMI scores had significant behavior problems and 1 had below average academic performance. Significant improvements in VMI test scores and similar sensory-motor measures after various types of teaching programs have been reported.8 Recognition is important to help prevent learning and behavioral problems through early intervention.
CONCLUSION
We conclude that intellectual development in children with Fontan repair of complex congenital heart defects is generally within the normal range. Visual motor integration deficits may be more prevalent in these children and require further evaluation in association with behavioral and learning problems. Long-term studies are needed to identify factors associated with adverse neurodevelopmental outcomes, especially the role of intraoperative factors such as deep hypothermic circulatory arrest. Despite the presence of complex heart defects, preoperative hypoxemia, and multiple staged surgical interventions in most instances, these children have cognitive abilities comparable to their peers as they face life's challenges after the Fontan procedure.
ACKNOWLEDGMENT
This study was supported in part by a grant from Children's Hospital Research Center.
Footnotes
- Received May 1, 1997.
- Accepted August 12, 1997.
Reprint requests to (K.U.) Children's Heart Institute, Children's Hospital & Health Center, 3020 Children's Way, San Diego, CA 92123.
REFERENCES
- Copyright © 1998 American Academy of Pediatrics