Biological markers of intellectual disability in tuberous sclerosis

• Published in: Psychological medicine 2007-09, Vol.37 (9), p.1293-1304
• Main Authors: RAZNAHAN, ARMIN, HIGGINS, NICHOLAS P., GRIFFITHS, PAUL D., HUMPHREY, AYLA, YATES, JOHN R. W., BOLTON, PATRICK F.
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Adult and adolescent clinical studies; Biological and medical sciences; Brain; Cerebral Cortex - pathology; Child; Child, Preschool; Female; Frontal Lobe - pathology; Genetic disorders; Humans; Infant; Intellectual deficiency; Intellectual Disability - diagnosis; Intelligence - physiology; Intelligence quotient; Intelligence Tests; Magnetic resonance imaging; Male; Medical sciences; Multivariate analysis; Neurobiological aspects; Neurology; Neurosciences; NMR; Nuclear magnetic resonance; Psychology. Psychoanalysis. Psychiatry; Psychopathology. Psychiatry; Risk factors; Spasms, Infantile - diagnosis; Statistics as Topic; Temporal Lobe - pathology; Tuberous sclerosis; Tuberous Sclerosis - diagnosis; Tumors of the nervous system. Phacomatoses; Young adults
• ISSN: 0033-2917; 1469-8978
• DOI: 10.1017/S0033291707000177

Intellectual disability (ID) is highly prevalent in tuberous sclerosis (TS). Putative neurobiological risk factors include indices of cortical tuber (CT) load and epilepsy. We have used univariate and multivariate analyses, including both CT and epilepsy measures as predictors, in an attempt to clarify the pattern of cross-sectional associations between these variables and ID in TS. Forty-eight children, adolescents and young adults with TS were identified through regional specialist clinics. All subjects underwent thorough history taking and examination, and had brain magnetic resonance imaging (MRI) scans. The number and regional distribution of CTs was recorded. Subjects were assigned to one of nine ordered intellectual quotient (IQ) categories (range 130) using age-appropriate tests of intelligence. On univariate analyses, ID was significantly associated with both a history of infantile spasm (IS) (Z=-2.49, p=0.01) and total CT count (Spearman's rho=-0.30, p=0.04). When controlling for total CT count, the presence of CTs in frontal (regression coefficient=-2.43, p=0.02) and temporal (regression coefficient=-1.60, p=0.02) lobes was significantly associated with ID. In multivariate analyses the association between IS and ID was rendered insignificant by the inclusion of the presence of CTs in temporal and frontal lobes, both of which remained associated (p=0.05 and p=0.06 respectively) with ID. The presence of CTs in specific brain regions as opposed to a history of IS was associated with ID in TS. The significance of these findings is discussed in relation to previous work in TS, and the neural basis of intelligence.