Clinical sensitivity and specificity of multiple T2-hyperintensities on brain magnetic resonance imaging in diagnosis of neurofibromatosis type 1 in children: diagnostic accuracy study

Abstract

AimTo determine the prevalence, number, and location of multiple (≥2) T2-hyperintensities on brain magnetic resonance imaging (MRI) in children with neurofibromatosis type 1 (NF1) and their correlation with age, and to establish their sensitivity, specificity, and accuracy for the diagnosis of NF1 in children, especially in the early age (2-7 years).MethodsWe performed a cross-sectional study of 162 patients with NF1 from Croatian Neurofibromatosis Association Database and 163 control children between the ages of 2 and 18 years who underwent brain MRI between 1989 and 2009.ResultsMultiple T2-hyperintensities were present in 74% of NF1 patients and 1.8% of controls. They were mainly located in the basal ganglia, brainstem, and cerebellum and were significantly decreased in prevalence and number in the older age. T2-hyperintensities had excellent diagnostic accuracy with the area under the receiver operating characteristic (ROC) curve of 0.849 and 95% confidence interval (CI) of 0.805-0.886. The diagnostic sensitivity, specificity, and accuracy rate of T2-hyperintensities for NF1 were highest in the youngest age (2-7 years): 81% (95% CI 71%-89.1%), 99% (95% CI 92.3%-100%), and 85.8 (95% CI 83.3-93.8), respectively.ConclusionThis study strongly suggests the inclusion of T2-hyperintensities on brain MRI on the list of diagnostic criteria for NF1, especially in children of early age, when the clinical penetration of the NF1 gene has not yet been completely finished.

Cognitive and Psychosocial Phenotype of Young Children with Neurofibromatosis-1

“…Combining careful phenotyping with brain-imaging techniques may ultimately prove fruitful to gain a more nuanced sense of the neurocognitive profile of individual children with NF1. Recent studies have highlighted the potential diagnostic utility of MRI for young children with NF1 (Sabol et al, 2011), with high prevalence of UBOs in young children with NF1, however brain-based markers of cognitive risk within the NF1 population have not been definitively identified. There is good reason to expect neurocognitive difficulties in children with NF1, as there is an identified role for neurofibromin in regulating GABA release, which is critical to prefrontal-striatal communication and long-term potentiation in the hippocampus (Shilyansky et al, 2010), in turn affecting learning, attention, working memory, and processing speed, and general recruitment of brain areas for cognitive tasks (Costa & Silva, 2002; Cui et al, 2008; Genova, Hillary, Wylie, Rypma, & Deluca, 2009; Schneider et al, 2010).…”

Synthetic MRI for Clinical Neuroimaging: Results of the Magnetic Resonance Image Compilation (MAGiC) Prospective, Multicenter, Multireader Trial

“…[22][23][24][25] Statistical variations in diagnostic classifications may be centrally attributable to small samples of certain pathologies in the present study, meriting further study of these pathologic subgroups. Synthetic scanning is performed in the axial view only, and some clinical cases may be limited by spatial resolution in this section direction.…”

Annual review of children with neurofibromatosis type 1

“…There is debate about the role of brain MRI in aiding the diagnosis of NF1 and in screening for optic pathway gliomas (OPGs) 5–7. ‘Unidentified bright objects’, alternatively called focal areas of signal intensity or focal abnormal signal intensity , seen on brain MRI imaging may aid diagnosis but currently these do not form part of the diagnostic criteria 8. A retrospective cohort study by King et al 6 showed no difference in outcome between patients with OPG, who presented because of symptoms or ophthalmic signs, and asymptomatic patients identified as a result of MRI.…”