Statistics from Altmetric.com
Neurofibromatosis type 1 (NF1) is an autosomal dominant disorder due to a defect on the long arm of chromosome 17. Since 1986, when Cohen et al 1 first drew attention to changes on MRI in NF1, it has increasingly been recognised that up to 75% of children with NF1 have lesions, seen most often in the brain stem, cerebellum, optic tracts, and basal ganglia.2 They are asymptomatic, usually less prominent with age, and the limited available postmortem data suggest that they involve areas of spongiotic or vacuolar change.3It is unclear how often temporal lobe structures are involved and figures range from 0% to 16%.4 5 We report on a pair of identical twins with NF1, with prominent bilateral changes in hippocampal MRI, one of whom presented with a major amnesic syndrome.
The twins, aged 15, were the only children of non-consanguineous unaffected parents with no known family history. Twin 1 had been considered bright throughout his schooling, until in summer 1999 he was involved in a relatively minor fight when another boy punched him a few times and the two boys fell onto a grass lawn without losing consciousness or otherwise sustaining serious head injuries. Five days later his mother became aware of him having memory difficulties; he twice forgot to take a letter to school and then came home recollecting nothing of an evening spent at a church youth group. On admission to hospital he was noted to have greater than 15 café au lait spots and axillary freckling. Visual acuities were normal and there were no major focal neurological findings. Because of the acute memory problems herpes simplex encephalitis was considered despite a normal EEG and CSF examination and he was given acyclovir, without benefit. His brain MRI showed multiple areas of increased signal in the medulla, pons, midbrain, and internal capsule regions (fig 1 A), prominent bilateral lesions in the hippocampus, amygdala, and medial temporal cortex (fig 1C), and enlargement of the optic chiasm with increased signal strongly suggestive of an optic nerve glioma. Extensive blood tests including treponemal serology and chest radiographs were normal. Psychometric testing confirmed twin 1 to have an IQ in the high average range but to have a dense generalised anterograde amnesia. He was able to recall a story immediately it was told but at 5 minutes could only recollect one out of 20 details. Similarly, his immediate copy of the Rey-Osterrieth complex figure was normal (32/36) but he had almost no recall of the figure at 30 minutes (1/2 out of 36). Repeat MRI and neuropsychological assessment 1 year later showed no significant changes.
Because of the lack of a clear diagnosis on twin 1 we also studied his identical twin brother. Southern blot analysis confirmed their monozygosity. Aged 2, twin 2 developed a right facial plexiform neurofibroma and aged 8 a plexiform neurofibroma involving his tongue and larynx was surgically excised. He also had typical cutaneous lesions. His brain MRI showed remarkably similar changes to those seen in twin 1 in the medulla, pons, midbrain, internal capsule (fig 1 B), and hippocampal regions (fig 1 D) although the changes were less marked in the limbic regions in twin 2 than twin 1. The sole major difference was the normal optic chiasm in twin 2. Neuropsychometry on twin 2 disclosed a full scale IQ in the average range with no evidence of significant amnesic problems.
There are three main areas of interest that arise from these twins. The first is how remarkably similar are the changes on brain MRI. The second is that unlike previous studies6-8 they are discordant for optic nerve gliomas. The third, and perhaps most intriguing, is twin 1's amnesiac syndrome, the cause of which remains unestablished. He does not seem to have any of the conditions known to cause profound amnesia with changes on MRI, which include herpes simplex encephalitis, paraneoplastic limbic encephalitis, and possibly neurosyphilis.9 As he has NF1 associated with bilateral hippocampal changes on MRI it is tempting to speculate that the combination of the minor head injury and the lesions in the limbic cortex are responsible. Less likely explanations are malignant gliomatous transformation of hippocampal hamartoma or spread of malignant tissue from his optic nerve glioma.
Changes in MRI specifically within the hippocampus have apparently not been the subject of detailed analysis in NF1 and this paper suggests the need for further study of hippocampal structure and function in NF1.
We thank Professor M Patton, Department of Clinical Genetics, St. George's Hospital Medical School for help with the Southern blots and Lister Bestcare MRI Unit at Mayday Hospital, Croydon, for financial support with the MRI studies.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.