Performance on measures of 'executive function' following pediatric traumatic brain injury

Objective: To investigate the relationships among age at injury, neuroanatomic lesion location, and measures of executive function (EF) following paediatric traumatic brain injury (TBI). Methods: EF was assessed in 68 children (aged 7-15) with moderate-to-severe TBI 1 year post-injury. EF tests included: (1) Tower of Hanoi (TOH), a measure of problem solving ability, (2) Wisconsin Card Sorting Test (WCST), a measure of categorization and ability to shift cognitive strategies, (3) Letter Fluency (LF), a measure of novel lexical search and rule-governed word generation. EF variables included number of moves needed to achieve a 3-ring solution on the TOH, number of perseverative and non-perseverative errors on the WCST, and number of words generated on LF. Intellectual functioning was also assessed using the Verbal Intellectual Quotient (VIQ) from the Wechsler Intelligence Scale from Children-3rd edn (WISC-III). Data from standardized MRIs, performed at 3-months post-injury, were available for all subjects and were used to determine lesion location, lesion volumes, and total number of lesions. The relationships among EF, lesion variables (frontal lesion volume, extrafrontal lesion volume, total number of lesions) and age at injury were examined. Pre-injury special education services and attention deficit hyperactivity disorder (ADHD) were controlled for. Results: Younger age at injury was associated with more perseverative errors on the WCST and worse performance on LF. Frontal lesion volume was not predictive of performance on any measures of EF. Greater extrafrontal lesion volume and total number of lesions were predictive of worse performance on LF. When controlling for pre-injury special education placement and pre-injury ADHD, there was little change in the results. Conclusions: Younger age at injury places children at greater risk of impairment on measures of EF. Performance on measures of EF depends on brain variables other than frontal lobes including extrafrontal cortical brain areas and total number of lesions. The relationship between extrafrontal brain regions and EF suggests that domain-specific cognitive content (i.e. language or visuospatial analysis), mediated by the parietal or temporal lobes, may disrupt underlying cognitive processes necessary for successful performance on measures of EF. In addition, the association between total number of lesions and EF may be related to disconnections and disruption of frontal/subcortical systems.