小児てんかんにおけるADHDの併存のスクリーニングツールと危険因子の妥当性：系統的レビューとメタアナリシス

Attention-deficit/hyperactivity disorder(ADHD)is the most common neuropsychiatric comorbidity in children with epilepsy and occurs at prevalence rates 2.5 to 5.5 times higher in children with epilepsy than the general population. To date,the largest comprehensive meta-analysis(46 studies)estimated the pooled ADHD prevalence among children with epilepsy to be almost one-third(30.7%),with a greater proportion of the inattentive subtype.Children with epilepsy who also have ADHD have poorer seizure outcomes,lower quality of life,academic underachievement,and increased healthcare utilization.Despite being present in so many of them,ADHD frequently remains underdiagnosed in children with epilepsy because of symptom overlap,diagnostic complexity,and the lack of standardized screening protocols. In addition,both the validity of screening instruments and risk factors associated with comorbid ADHD need to be studied to develop ways to diagnose and treat ADHD in this vulnerable population.

To assess the diagnostic accuracy and discriminant validity of ADHD screening tools for children with epilepsy and comorbid ADHD and to conduct a systematic review and meta-analysis on the risk factors related to ADHD in children with epilepsy.

We conducted this study according to the PRISMA guidelines(PROSPERO:CRD42018094617).A systematic search was performed on PubMed,Embase,PsycINFO,Cochrane Library,and Web of Science from inception until December 2023,based on the developed search strategies.The included studies were appraised for quality using three different approaches according to the three-layer analytical framework,in which diagnostic accuracy of a single instrument(Layer 1)was assessed using QUADAS-2,discriminant validity(Layer 2)was analyzed with the Newcastle-Ottawa Scale,and risk factor/prevalence evidence synthesis(Layer 3)was synthesized narratively.Meta-analyses were carried out using the DerSimonian-Laird random-effects models,and heterogeneity assessment was conducted.

Twelve studies met inclusion criteria for qualitative synthesis.Diagnostic accuracy(Layer 1):The SDQ-Hyperactivity subscale had the highest diagnostic accuracy(sensitivity 86.4%,specificity 97.8%,Youden index 0.84,DOR = 285).Discriminant validity(Layer 2):The overall pooled effect size was g = 0.27(95%CI:0.17-0.38)and the externalizing behavior measures showed the highest discriminant validity(g = 0.54).Risk factor evidence was narratively synthesized. In studies which reported extractable association estimates,earlier seizure onset,neurodevelopmental/cognitive vulnerabilities(executive dysfunction,developmental coordination disorder),male sex,medical comorbidities,or higher antiepileptic drug burden were associated with an increased risk of comorbid ADHD.

The SDQ-Hyperactivity subscale demonstrates superior diagnostic accuracy for ADHD screening in children with epilepsy,supporting current ILAE recommendations.Externalizing behavior measures,particularly the DBD rating scale(g = 0.60),show the strongest discriminant validity.Children with poorly controlled seizures,early onset epilepsy and those receiving polytherapy warrant heightened screening vigilance.Implementation of systematic ADHD screening targeting high-risk subgroups may substantially improve early identification and intervention outcomes.