Objective. Early cranioplasty for scaphocephaly has become routine in most countries. In addition to normalizing the shape of the skull, it has been found to decrease intracranial hypertension. Whether corrective surgery benefits the child's cognitive outcome has been poorly documented.
Design. Eighteen children whose sagittal suture showed premature fusion at birth or soon thereafter were operated on at age 1 week to 7 months. All patients healed without complications and were followed-up at regular intervals. At the age of 7.8 to 16.3 years they were examined to clarify their neurocognitive development and to compare the results with their age- and gender-matched healthy controls.
Results. Originally scaphocephalic children, although operated on, had mild deficiencies in auditory short-term memory and language development when examined with the general comprehension, similarities, and digit span subscales of the Wechsler Intelligence Scale for Children–Revised. In all other respects their developmental outcome was equal to that of the controls.
Conclusions. Despite relative early correction of the skull shape, originally scaphocephalic children's neurocognitive performances do not reach the same level in all of the neurocognitive domains as their matched controls at school age. Early operation (≤1 month) may decrease this developmental delay. This implies that impairment of brain function has already taken place in utero. For the same reason an early operation seems to be justified not only for correction of the skull shape, but also to allow unrestricted development for the brain. Postponement of the operation would not serve either of these aims.
- WISC–R =
- Wechsler Intelligence Scale for Children–Revised
Scaphocephaly or dolichocephaly is the most common form of craniosynostosis. Premature fusion of the sagittal suture, without involvement of other sutures, leads to a narrowed and elongated head, with a projection of the frontal and occipital areas. The etiopathogenesis of primary craniosynostosis is primarily unknown.1–3
In the case of an isolated single-suture synostosis such as scaphocephaly without other associated anomalies, the rates of mental retardation are low, ranging from 0% to 2.4%.4–6 Mental retardation seems to appear more often as the consequence of an isolated synosthotic coronal suture.7 If several sutures are involved, the risk of retardation is further increased, as in syndrome patients.8
Reports on the neuropsychologic outcome of scaphocephalic children without any operation, after a late correction, and after an early operation, are partly contradictory.5,6,8,9, The authors have not found a single, randomized, prospective study that would clarify the developmental consequences for the child after these different treatments. Three principal causes of defective neurocognitive development, acting alone or in combination, have been mentioned in craniosynostosis: underlying primary brain malformation, secondary cerebral malformation resulting from brain growth in an abnormally shaped skull, and increased intracranial pressure with hypovascularity.9–13 Assuming that the second and/or third alternative play a major role, few arguments are left for postponement of the operation.
Precise cognitive testing is difficult in infancy, and methodologic differences often impede the comparison of research results.13 ,14 Testing before surgery and soon thereafter is problematic. Minor changes might remain uncertain or beyond recognition. Additionally, some aspects of cognitive development are first detectable several years later. Therefore, evaluation has been performed in most studies at preschool age. We considered that at the earliest school age these children are mature enough to notice minor deviations in their neurocognitive outcome.
PATIENTS AND METHODS
Eighteen children operated on for scaphocephaly from 1979 to 1987 at the Turku University Central Hospital in Finland were included in the study. Patients who had synostosis of adjacent sutures in addition to the sagittal suture or who had other forms of craniosynostosis were excluded. The demographic data of the patients is presented in Table 1. The diagnosis of scaphocephaly was made either at the first examination by a pediatrician or a few weeks later, and confirmed in all cases with a plane and tangential radiograph of the skull. All children were otherwise healthy.
At the age of 5 days to 30 weeks (mean, 11.0 weeks) a correction of the scaphocephaly was made with a partial sagittal craniectomy. The resected part consisted of a 3- to 4-cm-wide piece of parietal bone from the coronal to the lambdoid suture with the sagittal suture in the middle. After bone resection, a 5-mm strip of superficial dural layer was dissected, turned over the raw surface of the parietal bone, and sutured together with its outer periosteum.15 All patients recovered uneventfully and could leave the hospital 6 to 16 days (mean, 7.3 days) after surgery.
The follow-up of the study patients took place 1 month, 6 months, and 1 year, postoperatively. All children were doing well without any noticeable dysfunction. Within half a year after the operation the head had gained its final shape as measured with the cephalic index.
In 1995, at the age of 7.8 to 16.3 years (Table 1), all these children were asked to come to the outpatient department for long-term follow-up to carefully check their neurocognitive development as described below in detail. All children were right-handed.
Eighteen average-level pupils from Nummi's Comprehensive School in Turku, chosen by teachers, served as the control group. Teachers were asked to select pupils with average achievement. Their gender and age (from 8.7 to 15.8 years) matched those of the study group (Table 2). The inhabitants in the Nummi's school district represent well the inhabitants of the Turku district. Concerning the educational level of Nummi's School, all children in the study group could have studied in that school.
The neurocognitive functions were assessed in four central domains: fine motor, language, visual-motor-spatial, and memory. In addition, behavior and school achievement were assessed. One of the authors (J.F.) administered the neurocognitive tests to all patients and controls at the pediatric outpatient department. During the same visit, parents (usually mothers) rated the behavior of the patients. Teacher ratings of their school achievement were obtained through a questionnaire mailed to the teachers. The neurocognitive and behavioral tests were as follows.
The Purdue Pegboard Test17 measures fine-motor coordination and dexterity. The patients were asked to place pegs in holes and to form assemblies of a peg, washers, and a collar as fast as possible.
To measure verbal conceptual thinking and reasoning, the subscales similarities and general comprehension of the Finnish adaptation of the Wechsler Intelligence Scale for Children–Revised (WISC–R) were applied. Scaled scores with a mean of 10 and a standard deviation of 3 were used according to the manual of the test.18 Reading comprehension was measured with an unstandardized but widely used clinical test with local norms.
The block design subtest of the WISC–R18 was used to measure visual-spatial constructive skills. The Developmental Test of Visual-Motor Integration of Beery,19 which required the child to copy complex designs, was used to assess fine-motor coordination and visual-perceptual organization.
The digit span subtest of the WISC–R18 was applied to measure short-term auditory memory. The Visual Retention Test, Multiple Choice Form I by Benton et al,20 was used to measure visual short-term memory.
The Revised Parent Rating Scale21 ,22 was used to measure children's behavior. It consists of descriptions of 48 symptoms. Parents rated the appearance of these using a 4-point scale, in which: 1 = not at all present, 2 = a little present, 3 = pretty much present, and 4 = very much present. The questionnaire included items on six factors: defiant or aggressive conduct disorders, a learning problem factor, a psychosomatic factor, an impulsive-hyperactivity factor, and an anxiety factor.
Teachers gave an estimation of general school achievement and separately of the skills in reading, spelling, mathematics, and first foreign language. The ratings were made using a 4-point scale (1 = belongs to the best quarter of the class, 4 = belongs to the poorest quarter of the class). In addition, the teachers reported if the children had received special teaching or other special support during the school year. In most cases, the teachers were unaware of the child's scaphocephaly.
Statistically significant differences between the study group and controls were found using t tests in the areas of language and auditory memory (Table 3). The control children scored higher than the study group in tests measuring verbal conceptual thinking and reasoning (similarities and general comprehension subtests) and verbal short-term memory (digit span). On the other hand, the results of the reading comprehension test did not differ between the groups. Furthermore, there were no differences between the groups in test results of visual-motor, visual-spatial, or fine-motor performances, or in the test results of visual short-term memory. According to the results of the WISC–R subtests, all children in the study group were in the average or low-average range of general intelligence.
Correlations between the neurocognitive test results and the cephalic index at the time of operation varied from 0.01 to 0.18 and were not significant. When the study group was divided into an early-operation group (≤1 month; median, 0.7 months; n = 9 and a late-operation group (>1 month; median, 3 months; n = 9) the results favored the early operated group on eight neurocognitive measures and the late operated group on three measures. However, the differences between the subgroups in analysis of variance neurocognitive test results were not significant. The mean age of the groups at the time of testing did not differ (145 months vs 146 months).
All children had studied in regular school classes and had begun school at the normal age, and none of the children had repeated a class. In the study group, 3 pupils (nos. 12, 13, and 17; Table 1) had received special teaching, in addition to the regular class teaching, in reading and/or spelling. The corresponding number in the control group was 2. In both groups, 2 children had received speech therapy (nos. 13 and 17 in the study group). The teacher ratings of general school success and of skills in reading, spelling, mathematics, and first foreign language (usually English) did not result in significant differences between the groups (Table 4). For 5 children in the study group and for 1 child in the control group, the teacher ratings were not obtained. According to the reports of the parents, these children had no problems at school.
According to the ratings of the mothers, the behavior of the children was very similar in the study group and in the control group. There were no differences between the groups in rates of conduct disorders, learning problems, psychosomatic problems, impulsivity and hyperactivity, anxiety, or other miscellaneous problems. In both groups the most commonly reported problems were defiant conduct-disorders, and the least reported were aggressive-conduct disorders. As usual in behavior ratings, the standard deviations were relatively robust. In the study group, standard deviations, from 19% (other problems) to 34% (anxiety) of the mean. In the control group, the standard deviation varied from 20% (conduct problem 1) to 53% (impulsive-hyperactive) of the mean. Except for the extreme standard deviation for impulsive-hyperactive problems in the control group, the standard deviations did not differ significantly between the groups.
The objective of the study was to assess, compared with a well-matched control group, the neurocognitive development of school-aged children operated on in infancy for scaphocephaly. The main finding was that, except for performance in the domains of language and short-term auditory memory, the neurocognitive development and behavior of the former patients were not different from the matched control group. On the tests of verbal conceptual thinking and reasoning and on the test of auditory short-term memory, the study group performed significantly less well than the control group. On the other hand, on tests of fine-motor skills, visual-motor skills, and visual-spatial skills, the study group and the control group did not differ. All children were in the average or low-average range of general intelligence. Similarly, school performances, as rated by the teachers, did not differ substantially between the groups. Furthermore, the number of different behavior problems, as rated by parents, was practically identical in both groups.
In all, the problems in language and memory found in the test results of the study group seem not to have had severe deleterious effects on the everyday life of the children. This apparently paradoxical finding becomes obvious when it is noted that the mean scores of the study group on those tests that differentiated the groups are only approximately one standard deviation less than the mean of the control group. Although the differences found are statistically significant, their clinical importance may not be very significant. The results of the present study are consistent with the results of Noezler et al4 and Kapp-Simon et al,13 suggesting that simple craniosynostosis has no major unfavorable effects on neurocognitive development. However, the significance of language, especially for academic success, is notable. Some functions of the brain in children will continue to develop up to the second decade; therefore, cognitive problems may first appear when the poorly functioning parts of the brain mature.23 Because the age range of the children was wide, it is not certain that academic achievement will be favorable for all children in the future, as well. Therefore, an additional follow-up is needed in later years before final conclusions are possible.
The origin of the language and memory impairment found in the present study remains uncertain. The finding of Renier et al24that there was a correlation between preoperative intracranial pressure and later low cognitive performance suggests that a brain damage etiology is possible. Geschwind and Galaburda25 suggested that, in general, the left hemisphere has a slower maturation rate than the right hemisphere and could therefore be more vulnerable to damage than the right hemisphere. Most of the language functions are localized in the left hemisphere. However, prenatal and perinatal damage seems to affect nonverbal functions more than verbal functions.26 A possible explanation for this is the transfer of verbal functions from the damaged left hemisphere to the right hemisphere, where, according to the crowding hypothesis, it may cause some perceptual deficits.27 However, this may take place only when the damage is relatively extensive,28 which might explain why there are so many mild language problems in children. It also leaves open the possibility that mild brain damage in the left hemisphere may take place in utero. Neurocognitive performance was not related to the cephalic index. Instead, the results obtained with the subgroup operated at the age of <1 month, speak in favor of the early operation. However, because of the small number of children in the subgroups, caution should be taken when interpreting these results.
Another possible explanation for the differences between the groups is related to the parents' educational level, which has been found to explain children's later language development more strongly than perinatal risk factors.26 In the present study, the mothers' education in the two groups did not differ statistically, even if there was a tendency for mothers in the control group to have higher education than the mothers in the study group. However, a detailed analysis of the test scores of children with highly educated mothers did not support the possibility that the mothers' education could explain the group differences in the language domain in the present study.
The majority of the earlier follow-up studies of scaphocephaly have applied global scores of mental development, such as the IQs of the children (nos. 13, 4). However, overall intelligence may be normal although specific cognitive dysfunctions may be found in more detailed assessment. The cognitive structure of children is not sufficiently differentiated for a detailed neurocognitive assessment until near school age. Before that, age-specific cognitive problems may be masked by attentional problems.29 Therefore, in the present study specific neurocognitive assessment methods were chosen to assess four main neurocognitive areas in school-aged children: motor, visual-motor-spatial, language, and memory. All these domains have been shown to have relevance for children's academic performance.30 The cognitive tests used in the present study are, except for the reading comprehension test, well-known methods with good reported reliability values. Attention, which is an important cognitive factor related to school performance, was assessed only with the tests of short-term memory, but school success and parents' behavior ratings for the children did not support the existence of attentional problems in the study group.
To summarize, except for the possibility of slight problems in language, the neurocognitive development in school-aged children operated on in infancy for scaphocephaly was within the average range. All children studied in regular school classes without an increased need for special teaching and without increased behavior problems. It is not possible to conclude that the minor language problems found in the present study are related to scaphocephaly, because it is also reasonable that they may be attributed to some causes unrelated to scaphocephaly. The results of Kapp-Simon et al13 and Turtas et al31 support the latter possibility. In all, the results of the present study suggest that there are no marked cognitive and behavioral problems in school-aged children operated on in infancy for simple scaphocephaly.
- Received January 27, 1998.
- Accepted October 16, 1998.
Reprint requests to (R.V.) Department of Pediatric Surgery, Turku University Central Hospital, FIN-20520 Turku, Finland.
- ↵Cohen MM Jr. The etiology of craniosynostosis. In: Cohen MM Jr, ed. Craniosynostosis: Diagnosis, Evaluation and Management. New York, NY: Raven Press; 1986:59–79
- ↵Kokich VG. The biology of sutures. In: Cohen MM Jr, ed. Craniosynostosis: Diagnosis, Evaluation and Management. New York, NY: Raven Press; 1986:81–103
- Shillito J Jr,
- Matson DD
- ↵Camfield PR, Camfield CS. Neurologic aspects of craniosynostosis. In: Cohen MM Jr, ed. Craniosynostosis: Diagnosis, Evaluation and Management. New York, NY: Raven Press; 1986:215–226
- Laitinen L
- Freeman JM,
- Borkowf S
- van der Werf AJM
- ↵Tiffin J. Purdue Pegboard. Examiners Manual. Chicago, IL: Science Research Associates; 1968
- ↵Wechsler D. Wechsler Intelligence Scale for Children–Revised. A Finnish adaptation. Helsinki, Finland: Psykologien kustannus Oy; 1984
- ↵Beery KE. Developmental Test of Visual-Motor Integration. 3rd rev. Cleveland, OH: Modern Curriculum Press; 1989
- ↵Benton AL, Hamsher K deS, Stone FB. Visual Retention Test: Multiple Choice I. Iowa City, IA: The University of Iowa Hospitals and Clinics; 1977
- Conners CK
- Kinsbourne M
- ↵Geschwind N, Galaburda AM. Cerebral lateralization: biological mechanisms, associations, and pathology: I-III. A hypothesis and a program for research. Arch Neurol. 1985;42:428–459, 521–552, 634–654
- ↵Rourke BP, Bakker DJ, Fisk JL, Strang JD. Child Neuropsychology: An Introduction to Theory, Research and Clinical Practice. New York, NY: Guilford Press; 1983
- ↵Rourke BP, Fisk JL, Strang JD. Neuropsychological Assessment of Children: A Treatment-Oriented Approach. New York, NY: Guilford Press; 1986
- Copyright © 1999 American Academy of Pediatrics