Morales Karina, Harvey Danielle, Dunn David, Jones Jana, Byars Anna, Austin Joan, Hermann Bruce, Oyegbile-Chidi Temitayo
Department of Neurology, University of California, Davis, Sacramento, California, USA.
Department of Public Health Sciences, University of California, Davis, Sacramento, California, USA.
Epilepsia. 2025 Jan;66(1):265-278. doi: 10.1111/epi.18176. Epub 2024 Nov 11.
Behavioral problems in children with new onset epilepsies have been well established in the literature. More recently, the literature indicates the presence of unique behavioral patterns or phenotypes in youth with epilepsy that vary significantly in vulnerability and resilience to behavioral problems. This study contrasts the interpretation of behavioral risk as inferred from cross-sectional versus latent group analytic perspectives, as well as the presence, consistency, stability, and progression of behavioral phenotypes in youth with new onset epilepsy and sibling controls over 3 years.
Three hundred twelve participants (6-16 years old) were recruited within 6 weeks of their first recognized seizure along with 223 unaffected siblings. Each child's behavior was recorded by parents and teachers frequently over 36 months using the Child Behavior Checklist (CBCL), and each child completed self-report measures of depression symptoms over 36 months. Measures were evaluated cross-sectionally and longitudinally to identify clusters with prototypical behavioral trajectories.
Cross-sectional analyses exhibited a pattern of generalized and undifferentiated behavioral problems compared to sibling controls at baseline and prospectively. In contrast, latent trajectory modeling identified three distinct behavior phenotype clusters across all raters (parents, teachers, and youth) over baseline and longitudinal assessments. CBCL Cluster 1 (30% of youth with epilepsy) exhibited behavior similar to/better than controls, Cluster 2 (50%) exhibited moderate behavior issues, and Cluster 3 (~20%) exhibited the most pronounced/problematic behavior, falling into Achenbach's clinically relevant behavior range. Behavior within clusters remained stable and consistent. Teachers' and children's behavior assessments corresponded to these cluster groupings consistently over 36 months. Predictors of cluster membership include seizure syndrome type and social determinants of health.
This study demonstrates the varying public health perspectives of behavioral risk in youth with epilepsy that result as a function of analytic approach as well as the presence of distinct latent behavioral trajectory phenotypes over time in youth with new onset epilepsy.
新发癫痫患儿的行为问题在文献中已有充分记载。最近，文献表明癫痫青少年存在独特的行为模式或表型，其在行为问题的易感性和恢复力方面存在显著差异。本研究对比了从横断面与潜在群体分析角度推断的行为风险解读，以及新发癫痫青少年和同胞对照在3年期间行为表型的存在、一致性、稳定性和进展情况。
在首次确诊癫痫发作的6周内招募了312名参与者（6至16岁）以及223名未受影响的同胞。在36个月内，家长和教师频繁使用儿童行为清单（CBCL）记录每个孩子的行为，每个孩子在36个月内完成抑郁症状的自我报告测量。对这些测量进行横断面和纵向评估，以识别具有典型行为轨迹的聚类。
横断面分析显示，与同胞对照相比，在基线和前瞻性研究中，存在广泛且无差异的行为问题模式。相比之下，潜在轨迹模型在基线和纵向评估中，在所有评估者（家长、教师和青少年）中识别出三个不同的行为表型聚类。CBCL聚类1（约30%的癫痫青少年）表现出与对照相似/优于对照的行为，聚类2（约50%）表现出中度行为问题，聚类3（约20%）表现出最明显/最成问题的行为，属于阿肯巴克临床相关行为范围。聚类内的行为保持稳定和一致。在36个月内，教师和儿童的行为评估与这些聚类分组始终一致。聚类成员的预测因素包括癫痫综合征类型和健康的社会决定因素。
本研究表明，癫痫青少年行为风险因分析方法不同而具有不同的公共卫生视角，以及新发癫痫青少年随时间存在明显的潜在行为轨迹表型。
