5: Long-term Follow-up of Pre-operative Infant Event-related Potential in Children with Craniosynostosis

Abstract

Purpose: Event-related potentials (ERPs) provide a real-time safe, noninvasive method to assess neural networks in infancy. Limited data exist, however, as to the predictive value of these results with eventual neurocognitive outcomes at school age. The purpose of this study was to investigate whether ERP recordings at infancy also translated into neurocognitive deficits at school age. Methods: 16 infants who underwent evaluation of auditory processing in infancy were followed up for neurocognitive testing. ERPs were extracted from electroencephalographic data evoked in response to language stimuli, and were found to have a positive inflection peaking 150 to 300msec (P150) after presentation of the stimuli, followed by a negative wave 300 to 700msec (N450) after. The P150 and N450 components were examined over the right and left frontotemporal scalp areas. Patients underwent a battery of neurocognitive tests including assessments of IQ (Weschler Abbreviated Scale of Intelligence 2nd edition), academic achievement (Kaufman Test of Educational Achievement 3rd edition), and motor coordination (Beery-Buktenica Development Test of Visual-Motor Integration 6th edition). Absolute deviation of ERP results from age-matched controls were computed. Spearman’s correlations were used to assess the strength of correlation between neurocognitive scores and ERP values, as well as other sociodemographic variables. Results: 11 patients with ERP in infancy were neurocognitively tested. Average time to follow-up neurocognitive testing was 10.1 ± 1.7 years. 7 (63.6%) were male, 9 (81.8%) were white. 7 (63.6%) underwent whole vault cranioplasty while 4 (36.4%) underwent endoscopic strip craniectomy. Mean age at surgery was 7.4±2.5 months. A greater absolute difference from age-matched controls in the left-sided P150 component was correlated with worse outcomes in the following domains: full-scale IQ (r=-0.709, p=0.015), performance IQ (r=-0.673, p=0.023) and fine motor skills (r=-0.654, p=0.029). While aberrancy in the P150 component was also correlated with math achievement (r=-0.065, p=0.028) and reading comprehension (r=-0.603, p=0.036), significant correlations were also found between math achievement and mother and father’s highest level of education, and between reading comprehension and mother’s highest level of education and household income. There was no correlation, however, between any demographic variables and full-scale IQ or performance IQ. Various socioeconomic factors were found to correlate with math and reading achievement, but not IQ or motor skills. Conclusion: Brain activity measured at infancy has predictive value for eventual neurocognitive outcome, with greater deviation from normal being associated with greater IQ deficits. Early ERP assessment may help manage patient expectations about neurocognitive outcomes following intervention.