The relationship between cognitive impairment and unidentified bright objects (UBOs) on the MRI in children with neurofibromatosis 1 (NF-1) was studied at the Kennedy Krieger Institute, and the Johns Hopkins University, Baltimore, MD. The data set included WISC-R-derived FSIQ for 20 pairs of children (NF-1 and unaffected sibling of NF-1); a “lesion count” for number of locations in which UBOs were seen; and the ratio of total volume of UBOs divided by total brain tissue volume. The number of locations occupied by UBOs accounted for IQ lowering in children with NF-1, whereas the total UBO volume was not associated with a discrepancy of IQ of NF-1 patients compared to their unaffected siblings (D-SIQ). The child’s age, familiality, or summed UBO volume did not strengthen the regression model based on number of UBO locations. The mean discrepancy (D-SIQ) for the NF-1 affected children was 13 points. The mean number of UBO-occupied locations was 3, with basal ganglia the most frequent site. 
COMMENT. In this study, the number of locations occupied by UBOs correlated with the lowering of IQ in children with neurofibromatosis-1. Previous studies have provided conflicting findings, some showing significant correlations between a lowered IQ and the presence of UBOs, and others failing to demonstrate a significant relation between intelligence and the number or location of T2 weighted foci in the brain. (see Ped Neur Briefs Jan 1996;10:3-4).
IQ correlated with UBOs in thalamus. A most recent paper from the University of Texas MD Anderson Cancer Center, and the UTMS, Houston, TX, reports that hyperintensities located in the cerebral hemispheres, basal ganglia, brainstem, or cerebellum show no correlation with neuropsychological functioning, whereas hyperintensities in the thalamus were significantly correlated with a lowered IQ. Mean scores for IQ, memory, motor, distractibility, and attention performance in children with UBOs in the thalamus were significantly lower than scores for those with UBOs located elsewhere. The presence or absence of UBOs and the number of UBOs were not significantly correlated with IQ. 
Somatic mosaicism in neurofibromatosis-1 is reported from the University of Florida, Gainesville, FL, and may possibly explain the discrepancies in patient selection and results of the above studies.