A relação entre os parâmetros do potencial evocado P300 e o comprometimento cognitivo na epilepsia: Uma revisão sistemática e meta-análise

Abstract

Abstract: Epilepsy is a chronic neurological disorder frequently associated with cognitive alterations affecting domains such as attention, memory, and information processing speed, significantly impacting the quality of life of individuals. In this context, the P300 evoked potential, obtained through electroencephalography, has been widely used as an objective electrophysiological measure to investigate cognitive performance in individuals with epilepsy. Objectives: To analyze the relationship between P300 evoked potential parameters and cognitive impairment in individuals with epilepsy, through a systematic literature review. Methodology: A systematic review was conducted according to the PRISMA guidelines, searching the PubMed and LILACS databases between January 2020 and June 2025. Study selection was conducted by two independent reviewers. Observational studies and clinical trials that evaluated P300 parameters associated with cognitive performance in individuals with epilepsy were included. Methodological quality was assessed using the Newcastle-Ottawa Scale (NOS). When comparable quantitative data were available, a meta-analysis of P300 latency was performed using standardized mean differences (SMD) under a random-effects model with a Restricted Maximum Likelihood (REML) estimator. Due to the heterogeneity of cognitive instruments and the form of statistical reporting, the associations between P300 parameters and cognitive domains were additionally synthesized through a structured integrative synthesis of the findings. Results: Ten studies were included in the qualitative synthesis. NOS scores ranged from 5 to 9 stars, with recurring limitations in the comparability domain. Four studies provided sufficient data for the meta-analysis of P300 latency. The combined estimate under a random-effects model demonstrated a statistically significant prolongation of latency in individuals with epilepsy compared to controls (SMD = 1.29; 95% CI: 1.03–1.54; p < 0.001), with no evidence of statistical heterogeneity (I² = 0%; τ² = 0.00; Q(3) = 2.01; p = 0.57). Publication bias tests did not indicate significant asymmetry (Begg p = 0.333; Egger p = 0.303), although the small number of studies limits statistical power. In the integrative synthesis of findings, prolonged latency was most frequently associated with poorer performance in global cognition, attention, processing speed, and executive functions. Reduced amplitude was associated with impairments in attention, executive functions, and working memory. Subgroups with greater clinical severity, including drug-resistant epilepsy and polytherapy, presented combined alterations of prolonged latency and reduced amplitude accompanied by worse cognitive performance. Conclusion: The evidence indicates that P300 is a sensitive electrophysiological marker of cognitive impairment in epilepsy, reflecting clinical characteristics of the disease and the effects of treatment, with potential complementary application to neuropsychological assessment.