Neurofibromatosis (NF) is a hereditary autosomal dominant disease with a variable expression, although half of the cases result from de novo mutations. It affects the tissues that support the central and peripheral nervous systems. It is a neurocutaneous disorder that causes changes in skin pigmentation (“café au lait” macules), the cranial nerves and the spinal cord. It can also manifest with additional features, such as neurofibromas (cutaneous, subcutaneous or intraspinal tumours), pain in affected nerves, epileptic seizures, blindness or deafness secondary to gliomas in the optic or vesticulocochlear nerves, or hydrocephalus due to aqueductal stenosis.

There are two distinct forms of NF based on clinical, genetic and pathophysiological features, both of which belong to the broader category of neurocutaneous syndromes or phakomatoses. One is the peripheral form, known as NF type 1 (NF1), formerly von Recklinghausen disease. The other is the central form, also known as NF type 2 (NF2).1 Like NF1, it is a hereditary autosomal dominant disease, which in this case is caused by a gene located in chromosome 22q21.

The incidence of NF1 is of 1 per 3000–4000 individuals. The disease is caused by mutations in the NF1 gene located in chromosome 17q11.2. Approximately half of the cases are de novo mutations.2 The clinical diagnosis of NF1 requires the presence of at least 2 of the criteria established at the National Institute of Health Consensus Development Conference3 held in 1988 (Table 1).

Neurofibromatosis type 1 can affect early developmental processes, increasing the risk of learning difficulties in childhood. Approximately 80% of children with NF1 have cognitive comorbidities, with marked impairment in one or more cognitive functions, which has a negative impact on academic achievement and quality of life. In the past, these children received diagnoses of mental retardation, but more recent studies have confirmed that only 4%–8% of individuals with NF1 have an intellectual quotient (IQ) of less than 70. A study by Hyman et al.4 tested the IQ of 81 children with NF1 and compared it to the IQ of their siblings. Sixty-one percent of children with NF1 had IQ scores that were at least 10 points lower than their siblings. The study concluded that in more than 70% of children with NF1, academic performance is inferior to what would be expected based on their intellectual ability.

Individuals with NF1 usually have neuropsychological deficits in various cognitive domains: executive function, visual perception, memory etc. However, there is no widespread agreement on the definition of the cognitive profile of NF1, as 2 different criteria are currently in use. Some authors believe that the differential factors that distinguish NF1 are deficits in visual-spatial skills and problems in attention and executive functioning.5 Others emphasise the usual presence of language deficits combined with problems in visual-spatial skills, attention and executive functioning. There is evidence in the literature that due to the associated neuropsychological changes, between 30% and 65% of children with NF1 have learning difficulties in school in the areas of reading, writing, spelling, mathematics and language.6,7

The aim of our study was to assess the global intellectual ability of children with NF1 and their performance in different cognitive domains assessed by means of the Wechsler Intelligence Scale for Children (WISC-IV).

Cognitive performance in children with NF1 was inferior to that of children in the CG for the 5 composite indices of the WISC-IV: verbal comprehension, perceptual reasoning, working memory, processing speed and the full scale IQ. The NF1G scored lower than the CG in every WISC-IV subtest save for the coding and similarities subtests, in which we observed no significant differences between groups. Table 2 summarises the results.

In this study, we analysed the cognitive performance of a group of children with NF1 compared to another group of healthy children of similar socioeconomic characteristics. We observed significant differences in all composite indices and in 12 of the 14 subtests of the WISC-IV, with children with NF1 obtaining lower scores than children in the CG.

The IQ of the NF1G was below the normal psychometric threshold, and was significantly lower than the IQ of the CG, with a difference of 31 points. But obtaining an IQ score within the normal range does not suffice to determine that cognitive performance is at a comparable level. Intellectual ability and cognitive performance are not always equivalent, as there may be weak areas in the individual's cognitive profile that hinder performance. On the other hand, there is widespread agreement that in addition to intelligence understood as a global parameter, there are different specific intellectual abilities that develop to different degrees, which is referred to as multiple intelligences.9,10

The first publications on the cognitive ability of children with NF1 described levels of cognitive impairment that were significantly greater than those found in recent studies. Thus, Mulvihill reported an average IQ of 45 for a sample of 20 patients with NF1.11 Studies conducted more recently have not corroborated those differences, but placed the IQ of children with NF1 closer to the normal range. For instance, one study found that only 6% of children with NF1 had IQ scores significantly below average.12

In our study, the performance of children with NF1 was inferior to that of the CG for every cognitive domain assessed by the WISC-IV, with the exception of the coding and similarities subtests. These results are not exactly consistent with those of other studies; Cutting and Levine assert that the cognitive profile of NF1 fits the pattern of a “nonverbal” disorder, as most individuals with NF1 exhibit visual-perceptual abnormalities (specifically in visual-spatial skills).13 However, difficulties in verbal language, learning, reading and writing are also frequent in NF1.14 In our study, we found both verbal and nonverbal impairments that had a generalised impact on cognitive performance in children with a NF1 diagnosis.

Other cognitive domains, such as executive functioning, which encompasses the performance of more complex reasoning processes and the organisation of goal-directed behaviour, are also usually affected in children with NF1. The results of our study are consistent with those of other authors, who also found difficulties in reasoning and abstract concept formation in children with NF1; assessment with the WISC-IV has shown performance deficits in the comprehension, similarities, incomplete pictures, arithmetic or picture arrangement subtests.15 In our study, the perceptual reasoning and working memory indices in the NF1G were lower than those of the CG. Both indices are intimately associated with the efficiency of executive functioning.

Most children with NF1 score lower in many cognitive functioning tests compared to controls.16 However, there are studies that challenge this conclusion. For instance, some authors have not found differences in the performance of children with NF1 in the picture arrangement subtest of the WISC.4 Others have reported on the results of the arithmetic subtest, in which children with NF1 did not have deficits compared to a children unaffected by the disease. Thus, the results of the latter study contradict ours, as we did find a statistically significant difference in the arithmetic subtest.17

Attention is one of the cognitive domains that have been studied most extensively in NF1. The work of Watt et al.18 has confirmed that affected children have difficulties performing tasks that require sustained attention. Our study corroborated this finding, as children in the NF1 group scored lower in the subtests that required a higher degree of sustained attention (incomplete pictures, digits and symbol search). We need to take into account that attention is the mechanism involved in accessing the information stored in the nervous system, so that problems with attention hinder learning and memory consolidation. Other cognitive domains, such as memory and language, are not as affected by NF1.19,20

In conclusion, our study revealed the presence of generalised cognitive deficits in the NF1 group. Along with impaired cognitive performance, assessed by the full scale IQ, we observed significant deficits in every assessed cognitive domain. Based on these findings, it would be advisable to offer individualised attention to children with NF1 during their schooling, as they are at higher risk of having learning difficulties. Early identification of children with NF1, along with their neuropsychological assessment and the development of cognitive stimulation programmes, could improve their academic outcomes and reduce the risk of school failure.

The authors have no conflict of interests to declare.