| Radiologic findings of the head and spine in neurofibromatosis 1 (NF1) in Northern Finland | ||
|---|---|---|
| Prev | Chapter 2. Review of the literature | Next |
Macrocephaly, which is relatively common, has been considered a nonspecific finding in NF1 . Weichert et al. (1973) were the first to suggest that macrocranium may be a manifestation of neurofibromatosis. They measured the breadth, height, length and circumference of the skull on the radiographs of 34 NF children. Cases with intracranial tumours or hydrocephalus were excluded. Eight out of 27 patients (30%) had macrocranium by virtue of their head circumference, and 18 out of 24 (75%) based on Haas´ skull measurement criteria (Haas 1952).
For practical purposes, macrocephaly may be defined as a condition where head circumference is above the 97th percentile (> + 2 SD) for age and sex. Riccardi et al. (1981) found the mean head circumference of their NF1 patients to be around the 70th percentile, the median to be at the 75th centile, and 27% of the patients to have a head circumference above the 97th centile. They suggested that macrocephaly in NF1 has a postnatal onset because, in children between two and six years of age, head circumference increased from the 48th to the 77th percentile, at which level it remained in all the older patients. In a population study of 135 NF1 patients, Huson et al. (1988) found macrocephaly in 52/115 (45%) individuals with NF1. Patients with CNS tumours, aqueductal stenosis and large facial plexiform neurofibromas had been excluded. In the Northern Finnish study of Pöyhönen (2000), clinically verified macrocephaly was present in 47/164 patients (29%), being more common in males (36%) than in females (22%). Patients with hydrocephalus were excluded.
The etiology and pathogenesis of macrocranium and macroencephaly in NF1 are not known, if tumours and aqueductal stenosis are excluded. They have been suggested to be secondary to increased glial cell production, which causes a symmetric or asymmetric increase in brain size (Holt 1978, White et al. 1986, Latchaw 1991).
In order to understand the process of exceptional head growth, the roles of the different parts of the brain have been evaluated morphometrically in several studies
Holt & Kuhns (1978) found that 44% of 52 neurofibromatosis patients had cranial capacities over the 95th percentile (70% were above 50%th percentile). Four patients above the 95%th percentile had normal pneumoencephalograms and angiograms, indicating that the cause of macrocranium might be macrencephaly rather than expansion of the CNF spaces. Volumetric measurement of the sella turcica in 27 patients showed idiopathic enlargement of the sella to be rare in neurofibromatosis.
DiMario et al. (1993) studied selected anthropometric measurements performed on plain skull roentgenograms (29 measurements, 9 qualitative assessments and 3 area/volume calculations), to find out whether these would be useful in the diagnosis of NF1. They compared the findings of 14 patients with NF1, 15 with probable NF1 and 29 healthy controls, both children and adults. Both diagnosed NF1 and probable NF1 patients could be distinguished from the controls by measurements of the height and depth of the sella turcica and the width, length, height and volume of the skull.
Mott et al. (1996) measured the midsagittal area of corpus callosum and brain on T1-weighted MR images of 14 NF1 children and their unaffected siblings. They found that the NF1 children with macrencephaly had an exceptionally large corpus callosum, which was most prominent in the midcallosal region. Dubovsky et al. (2001) also measured the surface area of the corpus callosum and the intracranial skull surface on midsagittal T1-weighted MR images of 43 children with NF1 and 43 healthy control subjects They found a statistically significant increase in the mean corpus callosum surface area in pediatric patients with NF1. The large corpus callosum was never an isolated finding in NF1 patients; all of these patients also had the characteristic T2 signal abnormalities. None of the patients had abnormal signals within the corpus callosum.The midsagittal intracranial skull surface was also significantly increased. Kayl et al. (2000) also showed in their MRI study of 34 children with NF1 that the patients had a significantly larger total corpus callosum area and significantly larger regional measurements in 3/7 areas than the controls.
The increased size of the corpus callosum was further verified by Moore et al. (2000) in 52 NF1 patients.
Said et al. (1996) measured the volumes of cerebral grey and white matter in 22 children with NF1 and 20 controls by using axial brain MRI scans. The overall cerebral hemisphere volume was greater in patients than in controls, and this increase was thought to be primarily due to an increase in the white matter volume, particularly in girls. It was suggested that macrencephaly in NF1 is largely a result of increased white matter volume, possibly due to glial cell proliferation. In a subsequent study on 34 NF1 children and 35 controls (Greenwood et al. 1997), the increase in white matter volume was shown to be greater than that in grey matter volume (23% vs. 10%) in the children with NF1 as compared to the controls. Moore et al. (2000) in their study of 52 children and adults with NF1 concluded that especially the grey matter contributed to the increase in brain volume. This was more prominent in younger subjects. Steen et al. (2001), on the other hand, found macrocephaly to be associated specifically with enlargement of white matter in their study of 18 asymptomatic children with NF1.
In a morphometric study of 27 NF1 children and 43 age- and sex-matched controls with MRI, DiMario et al. (1999) evaluated several ventricular and brain parenchymal parameters. Significant differences between the two groups were observed for 6 of the 24 measures. Patients with NF1 had significantly larger values of mean bicaudate width, biatrial width and biparietal diameter than the controls. They also had significantly increased iter measures and antero-posterior dimensions of the descending sigmoid sinus, and an age-specific increase in brainstem height was observed in NF1 patients. The authors concluded that patients with NF1 experience dynamic changes in brain morphometry, resulting in predominant lateral volume expansion of the supratentorial compartment and an increasing rate of brainstem growth as they age. No significant differences were seen on any measure in a comparison of NF1 patients with and without T2 hyperintense lesions of the brain.
Macrocephaly in NF1 has previously been considered an incidental finding with no correlation with impaired intelligence, seizures or electroencephalic changes (Rubenstein et al. 1985). More recent studies have shown that some features of NF1 could be related to quantitative differences in brain morphology. Said et al. (1996) reported a significant relationship between right hemisphere grey matter volume and visuo-spatial and developmental performance in NF1 patients, with greater grey matter volume signifying better performance. In the study of Moore et al. (2000), 52 children and adolescents with NF1 were studied both neuroanatomically and neuropsychologically. It was found out that the grey matter volume was related to the degree of learning disability, and that diminished academic performance and visuo-spatial and motor skills were associated with a greater regional corpus callosum size. Kayl et al. (2000) in their study of corpus callosum morphology and the attention-deficit hyperactivity disorder in children with NF1, noticed that increased severity of attention problems was associated with smaller total callosal areas. Cutting et al. (2000) did not find a close relationship between the impairment of cognitive functions and macrocephaly. The macrocephalic patients, however, showed significant verbal impairment relative to the normocephalic patients.
The correlation of macrocephaly with T2 hyperintense lesions has also been considered. In the series of DiMario et al. (1999), 15 of the 27 NF1 patients showed T2 hyperintense lesions, but no significant differences were seen on any brain measures. Cutting et al. (2000) also came to the conclusion that macrocephaly does not appear to be related to the presence or absence of T2 hyperintense lesions in NF1.