Childhood epilepsy and ADHD comorbidity in an Indian tertiary medical center outpatient population

This study aimed to assess the prevalence of Attention Deficit Hyperactivity Disorder (ADHD) and its characteristics and risk factors in children with epilepsy at a tertiary medical center in New Delhi. Children with active epilepsy, aged 6 to 12 years, were assessed for ADHD using DSM-IV-TR criteria. Epilepsy and psychiatric characteristics, sociodemographic indicators, and use of antiepileptic drugs were analyzed for differences between the ADHD and non-ADHD groups. Among the 73 children with epilepsy, 23% (n = 17) had comorbid ADHD, of whom 59% (n = 10) had predominantly inattentive type, 35% (n = 6) combined type, and 6% (n = 1) predominantly hyperactive-impulsive type. Lower IQ scores, epileptiform EEG activity, not attending school, and male sex were significantly associated with comorbid ADHD in children with epilepsy. Groups were similar in terms of age, socioeconomic indicators, family history of psychiatric disorders, seizure frequency in the last six months, seizure etiology, and seizure type. Epilepsy is a common pediatric neurological condition with frequent psychiatric comorbidities, including ADHD. Specialists should collaborate to optimize treatment for children with epilepsy and ADHD, especially for families in developing countries where the burden of disease can be great.

EEG assessment. Interictal digital EEG recording (Medelec profile; Oxford Instruments, Oxfordshire, UK) using an international 10-20 system of electrode placement was done at AIIMS for participants who did not have a good-quality EEG recording from the previous three months. Both awake and sleep states were recorded. Hyperventilation, photic stimulation, and sleep were used as activation procedures. EEG records were reported by two pediatric neurologists as either normal or showing epileptiform discharges.

Statistics. IBM SPSS version 24 (Chicago, USA) was used for univariate analyses and descriptive statistics.
Independent samples t-tests were used for normally distributed continuous variables. Chi-square (χ 2 ) tests for association were used for categorical variables. The Mann-Whitney U Test was used for epilepsy duration and IQ, which had nonparametric distributions as assessed by Shapiro-Wilk's test (p < 0.05). Post-hoc pairwise comparisons using Fisher's exact tests (2 × 2) were used for seizure frequency.
Ethics. This study was approved by the Institutional Ethics Committee at AIIMS, and the study was carried out in accordance with the relevant guidelines and regulations. Informed consent was obtained from the parents/ caregivers of all participants. Data availability. Data analyzed during this study are included in the Supplementary Dataset.
Sociodemographic and family background information is presented in Table 1. IQ was significantly lower by almost 1.5 standard deviations in the ADHD group compared to participants without ADHD (mean score 84 vs 92, p < 0.0001). The ADHD group had significantly more males. Almost a quarter of the ADHD group did not attend school compared to 1.8% of non-ADHD participants (χ 2 (1) = 9.664, p = 0.0019), with epilepsy given in all cases as a reason for not attending. The group difference in school attendance was not driven by age. Seven participants with ADHD (41.2%) met the diagnostic criteria for an additional behavioral disorder, while 14 participants (25%) in the non-ADHD group did (see Supplementary Dataset). Age, socioeconomic indicators, and family history of psychiatric disorders were similar between groups. Table 2. Age at onset of epilepsy ranged from one month to 11 years of age. Epilepsy duration ranged from 6 months to 10 years 5 months. Epileptiform discharges on EEG were positively associated with ADHD diagnosis (χ 2 (1) = 4.718, p = 0.030), with 52.9% of children with both epilepsy and ADHD showing epileptiform discharges, compared with 25% of those without ADHD. The ADHD and non-ADHD groups were comparable in terms of seizure etiology and seizure type. Post-hoc analysis of seizure frequency in the last six months did not indicate group differences in any of the frequency categories.

Discussion
At a tertiary medical care center in New Delhi, a high prevalence (23.3%) of ADHD was identified in children with epilepsy. In agreement with several previous studies 9, 12 , ADHD inattentive type was more frequent in this clinical group than hyperactive-impulsive or combined types. Most epilepsy characteristics did not differ significantly between the groups with or without ADHD, but abnormal EEG was associated with ADHD diagnosis. Children with both epilepsy and ADHD had lower IQ scores and were significantly less likely to be attending school, with epilepsy being the primary reason. Sociodemographic profiles of the two groups were similar. This prevalence rate is similar to the 23.4% of children with epilepsy and comorbid ADHD reported by Philip et al. 19 in a pediatric neurology outpatient population in Karnataka, India (n = 94). A recent register-based study in Sweden 26 of 1,899,654 individuals found that 13.5% of those with epilepsy also had a diagnosis of ADHD, while 4.3% of the full sample had ADHD, meaning a 3.5-fold increased risk in the comorbid group. In Asia, recent reports from Thailand 27 and China 28 found prevalences of 19% and 42%, respectively, of ADHD comorbidity among children with epilepsy. Studies of ADHD prevalence in India are limited, use different methodologies, and report varying figures. Recently reported prevalence figures of ADHD in children in the general population include 1.3% in a study of children in Bengaluru, Karnataka 29 (n = 3120), 1.7% in a study of families in central India 30 (n = 4278 families), 11.32% in a study of children in Coimbatore, Tamil Nadu 31 (n = 770), and 12.66% in a study of children in the state of Assam 32 (n = 300).
The population of children with comorbid epilepsy and ADHD differs in several ways from those with primary ADHD. For instance, the gender distribution is less skewed towards boys than in primary ADHD, and inattention symptoms are more frequent and severe than impulsivity/hyperactivity 33,34 , in line with the results presented here. In an American study, Dunn et al. 35 found that of children aged 9 to 14 with epilepsy, 24% had ADHD inattentive type, compared to 14% with hyperactive-impulsive or combined types. In a Nigerian tertiary care facility, Chidi et al. 36 found that of the 14% of epileptic children who also had ADHD, the inattentive type was predominant (69%).

Biological mechanisms. Recent neuroimaging research using magnetic resonance imaging (MRI) has sought
to shed light on possible shared neural mechanisms underlying epilepsy and ADHD comorbidity 1,10,37-39 . While some evidence implicates shared neurobiological abnormalities 40,41 , other studies suggest that independent factors are at play 10,40 . Regions of decreased cortical thickness have been reported in groups with comorbid epilepsy and ADHD, compared to epilepsy alone, which could indicate altered development of functional networks 42 . Saute et al. 38 reported that pediatric epilepsy with comorbid ADHD was associated with reduced cortical thickness in bilateral areas of the frontal, parietal, and temporal regions, along with smaller caudate, thalamus, hippocampus, and brainstem volumes, compared to the epilepsy non-ADHD group; smaller cerebellum and thalamus were related to epilepsy generally, but brain volume and surface area did not show major group differences. Similarly, Dabbs et al. 43 reported that ADHD problems assessed with the CBCL were associated with decreased cortical thickness in left superior frontal and lateral occipital regions and right fusiform gyrus. In a functional MRI study of boys with ADHD, Bechtel et al. 44 reported similar patterns of brain activity during block-design working memory tasks in the groups with and without comorbid epilepsy, which differed from healthy controls. Neural mechanisms underlying epilepsy and ADHD, as opposed to ADHD or epilepsy alone, remain to be fully elucidated and will benefit from further neuroimaging research.
Genetic studies, especially at the population level, have also begun to identify risk factors related to these and other neurodevelopmental disorders 45 . Relatives of those with epilepsy have an increased risk of ADHD themselves 26 , suggesting genetic and environmental linkages within families, and evidence of an association between maternal epilepsy and ADHD has also been reported 46 . However, Brikell et al. 26 found only a modest genetic correlation between the disorders. In addition to inherited genetic factors, de novo mutations and other rare variants, such as those involved in synaptic function, neurotransmission, and methylation remodeling, should be investigated as potential risk factors (see review by Lo-Castro and Curatolo 47 ).

Treatment challenges and AEDs.
A clearer understanding of medication effects on the symptomatology and progression of combined epilepsy and ADHD is needed to tailor medical management in this population 34 . It is possible that epilepsy alters the ADHD symptoms in children who have both, based on possible shared etiology and pathophysiology 44,48 . For example, children with ADHD are predisposed to epileptic seizures 40 and have an increased rate of EEG abnormalities without a history of epilepsy 49 .
Determining an effective treatment regimen for children with epilepsy and ADHD is challenging, as clinicians must consider a patient's seizure frequency, cognitive function, drug interactions, and medication adherence. Use of AEDs, in particular polytherapy, has in some studies been associated with cognitive deficits in children including attention problems and hyperactivity [50][51][52] . Sleep disturbances due to AEDs, along with subclinical seizures and learning disabilities, can contribute to inattention 51 , which was the ADHD subtype diagnosed in 59% of children with ADHD in this study. Several studies suggest that use of methylphenidate appears to be effective in children with comorbid epilepsy and ADHD 49,52,53 . A 2015 Brazilian study 54 , which administered low to moderate doses of methylphenidate, observed reduced seizure frequency and severity along with improved quality of life among adolescents with ADHD and difficult-to-treat epilepsy. However, increased risk of seizures with high dosage of osmotic release oral system methylphenidate 55 and lowered seizure threshold associated with tricyclic antidepressants and bupropion use has also been reported 1,41,52,56 . Most studies of medication for ADHD in children with epilepsy have had small numbers of participants and/or short study duration 52 , and clinical trials in developing countries have been limited 57,58 . Research on other medications, such as amphetamines, atomoxetine, and alpha-2 adrenergic agonists, has also been limited in this population. Medication effects on attentive behavior and seizure threshold should therefore be closely monitored. EEG monitoring is a useful tool in this population to identify whether staring episodes or repetitive movements are related to seizure activity or ADHD. Children with epilepsy in this study experienced ADHD at a high rate, and further research in larger populations should therefore clarify the risk-benefit ratios associated with various medications. Moreover, Gonzalez-Heydrich et al. 59 report that children with epilepsy and ADHD often have additional comorbidities, such as anxiety or oppositional defiant disorder, making treatment decisions complex 33 . AED polytherapy may also create financial stress for some families 3 . Holistic management of children with epilepsy would include screening and management of psychiatric problems, including ADHD. School failure and drop-out was a problem for a number of children in this study and underscores the need to engage the education community in this region to improve academic outcomes 60 . Further studies are required to evaluate the efficacy of behavior therapy, drug therapy, and other treatment modalities for individuals with epilepsy and ADHD, as well as their effect on improving quality of life 61 . Strengths and limitations. Limited research in India has examined ADHD comorbidity in children with epilepsy using diagnostic criteria. In this study, standard criteria from the DSM-IV-TR were used for diagnosing ADHD. Participants were recruited from the outpatient population of a government hospital which serves children from all socioeconomic backgrounds.
The study population was drawn from the tertiary care center outpatient population, and all participants in this study had a history of epilepsy, making it difficult to generalize to primary care settings and to the typically developing pediatric population. Future studies should therefore also include a healthy matched group without epilepsy. Within-group analysis, for example more detailed analysis of seizure frequency related to comorbidity, would be strengthened by a larger number of participants. While Malin's Indian adaptation of the WISC has been widely used across India for cognitive testing in children, future studies should consider using newer assessments, such as the WISC-IV India. Finally, the revised DSM-V was released after the data collection in this study. Using the updated diagnostic criteria for ADHD may have classified additional children with ADHD, meaning that the prevalence figures reported here are likely conservative. The DSM-V designation of mild, moderate, or severe ADHD would have provided extra detail to the current study.

Conclusion
Prevalence of comorbid ADHD was 23% in children with epilepsy attending outpatient services at a tertiary medical center in New Delhi, with inattentive type being predominant. In this study, children with comorbid ADHD had lower IQ scores on average, and four of the five children who did not attend school had both ADHD and epilepsy. Clinicians should be sensitive to identify ADHD in children with epilepsy since this may lead to more effective intervention and improved quality of life 62 . Further research examining potential shared neurological causes as well as strategies to optimize medication and treatment are needed to better serve this pediatric population.