Effect of anti‐attention‐deficit hyperactivity disorder (ADHD) medication on clinical seizures and sleep EEG: A retrospective study of Japanese children with ADHD

Abstract Aims Patients with attention‐deficit hyperactivity disorder (ADHD) often exhibit basic or paroxysmal wave abnormalities on electroencephalography (EEG). Methylphenidate (MPH), an anti‐ADHD stimulant, has been reported to lower the seizure threshold. However, there have been no reports comparing EEG changes before and after administration of the central nervous system (CNS) stimulant MPH, or atomoxetine (ATX) hydrochloride, a non‐CNS stimulant. In this study, we investigated changes in sleep EEG before and after the administration of ADHD treatment drugs. Method With the approval of the ethics committee, the medical records of 28 children with ADHD (23 men and 5 women) who gave consent were retrospectively investigated. The appearance of sudden abnormal waves during a 10‐minute sleep EEG recording was measured in 0.1‐second units, and the duration of these waves was calculated as the paroxysmal index (PI). Results Paroxysmal index did not differ significantly between patients who received MPH and those who received ATX. In addition, there were no exacerbations of clinical seizures. Conclusion It was concluded that ADHD medications do not have an adverse effect on epileptic seizures or abnormal sleep EEGs.

study by Hughes et al,7 EEG findings in patients with ADHD showed that 27.8% of patients exhibited normal findings, whereas the abnormal findings included positive spikes, focal epileptic discharges, slow waves, extreme spindles, and bilateral spike-and-wave patterns. An increase in θ waves and a decrease in β waves are basic wave abnormalities; these waves were compared before and after treatment using the θ/β ratios. [8][9][10] It is also known from a previous report 11 that the distribution of basic waves changes with age. Some studies have failed to replicate θ/β ratios differences between ADHD and non-ADHD. 12,13 Several studies have been conducted on these basic wave abnormalities, and it has been found that there are few α waves and many δ and θ waves with high amplitudes of slow waves and low amplitudes of α waves on EEGs performed in patients who were awake. 14 Moreover, the high activity of θ waves remains until adulthood, 15 suggesting cerebral dysfunction and immaturity in patients with ADHD.
Attention-deficit hyperactivity disorder and autism spectrum disorders (ASD) are neurodevelopmental disorders, and both are onset in childhood. ASD and ADHD often co-occur, with comorbidity rates in the 40%-70% range. 16 ASD is also often associated with EEG abnormalities. Reinhold et al 17 reported that out of 316 children evaluated for ASD, 85 (27%) had EEG abnormalities, and Chez et al 18 reported that 40 of 889 (60.7%) had EEG abnormalities. Another report 19 evaluated EEG abnormalities with ASD were high rates as high as 85.8% (870/1014). In recent study, 20 EEG abnormalities were found in 39.1% of patients, and EEG abnormalities correlated with autism severity, hyperactivity, anger outbursts, aggression, negative or destructive behavior, motor stereotypies, intellectual disability, language impairment, and self-harm, although the rate of EEG abnormalities associated with ASD varies from many reports. However, recent brain function and genetic studies have suggested that ASD is associated with schizophrenia, 21 and ADHD with and without ASD should be considered separately.
Several studies have been conducted on epileptic discharge in ADHD. According to one study, 5%-60% of patients with ADHD demonstrate epileptic discharges, of whom 14% show transition to epilepsy. 22 A study by Holtmann et al, 23 which enrolled 48 children, including 16 children with ADHD and Rolandic spikes (RS), 16 children with ADHD without RS, and 16 control children, showed that RS were detected more frequently in children with ADHD who demonstrated higher impulsivity. In addition, in children with ADHD and RS, RS were detected more frequently in children with ADHD and higher impulsivity. Similarly, children with ADHD and RS showed decreased focus and interference control compared with the control group. Furthermore, sleep polysomnography conducted in 42 children with ADHD showed that 53.1% of children developed epileptic discharges. 24 Hence, it is necessary to consider the possibility of exacerbation of epileptic seizures and EEG abnormalities caused due to central nervous system (CNS) drugs in ADHD treatment. Therefore, we thought that we should consider whether epileptic discharges, not basic wave abnormalities, are affected by ADHD treatments. There is a previous study 25 that elucidates EEG abnormalities in ADHD by analyzing the appearance time of epileptic discharges. We referred their study method, called PI index, in this study.

Methylphenidate (MPH), a commercially available CNS stimulant
used to treat ADHD in Japan, has been reported to lower seizure threshold. However, there have been no comparative studies on EEG changes before and after the administration of ADHD treatment drugs, including atomoxetine (ATX), a non-CNS stimulant. Therefore, we examined the changes in sleep EEGs before and after the administration of anti-ADHD medication.

| ME THODS
The medical records of patients who gave consent were retrospectively examined with the approval of the Ethics Committee of NCNP (approval number A2014-114). Paroxysmal abnormalities were obtained from sleep EEGs, assuming that the record of sleep stages 1-2 occurred for at least 10 minutes. EEG recordings were performed in all patients under sedation with oral medication, pentobarbital, or triclofos sodium. No complications were observed by sedation in all patients.
Electroencephalography assessment was performed by measuring spikes, slow waves, and spike-and-wave complexes detected in a 10-minute EEG recording during light sleep in 0.1-second units, as previously described by Altunel et al 25 The wave durations were calculated as the paroxysmal index (PI; Figure 1). PI was calculated visually by a pediatric neurologist. The PIs were then compared before and after the oral administration of ADHD drugs. In addition, the ADHD Rating Scale (ADHD-RS) and Clinical Global Impression-Improvement (CGI-I) scale were distributed to parents to evaluate the clinical symptoms of ADHD; similarly, the changes before and after the oral administration of ADHD drugs based on these scales were compared with the PIs in patients who responded to the questionnaires.
Statistical analysis was performed using the Wilcoxon signedrank test to detect PI changes before and after treatment. PI changes before and after treatment were evaluated using the t test in each of the MPH and ATX groups. The U test was used with or without epilepsy and the change in PI before and after treatment.

| Clinical findings
We included 26 patients (21 men and 5 women) but finally performed the analysis in 28 patients (23 men and 5 women) because two patients were taking MPH and ATX at different times, they were included in this study (Table 1). A total of 18 patients (16 men and 2 women) received oral MPH, and 10 (7 men and 3 women) received oral ATX. The mean age of the patients at the start of MPH or ATX was 10.2 ± 2.0 (mean ± standard deviation) years; the mean age of epilepsy patients only was 10.5 ± 2.0 (mean ± standard deviation) years. The mean age at the time EEG evaluation of the patients before the start of oral administration of ADHD treatment drugs was 10 8 ± 2.7 (mean ± standard deviation) years, and the mean age at the time EEG evaluation of after the start of that was 11.2 ± 2.8 years.  (Table 2). In addition, nine patients were taking AEDs to improve the emotional aspects of their ADHD (VPA in six, CBZ in two, and CLB in one patient). In patients with epilepsy who were receiving AEDs, they were already being administered prior to ADHD F I G U R E 1 Paroxysmal index: PI (referential derivation). Above is shown for example of PI. PI was calculated epileptic discharge occurrence duration treatment. In patients with comorbid epilepsy, no exacerbation of epileptic seizures was observed. These nine patients were started on AEDs after the initial EEG recording. On the other hand, none of the patients with epilepsy had added or changed the amount of AEDs before and after the administration of anti-ADHD drugs.

| Changes before and after drug administration
The duration of abnormal waves tended to decrease after MPH and Similarly, changes in PI values with epilepsy showed a significant negative correlation with patient age (age at the first measurement: Spearman's ρ = −0.26, P = 0.462; age at the second measurement: Spearman's ρ = −0.65, P = 0.041). There were no significant correlation changes in PI values with patient age at the first and second in patients without epilepsy. Spearman's ρ = 0.08, P = 0.842; age at the second measurement: Spearman's ρ = −0.56, P = 0.143). In patients without ASD, changes in PI values showed a significant negative correlation with patient age (age at the first measurement: Spearman's ρ = 0.26, P = 0.260; age at the second measurement: Spearman's ρ = −0.56, P = 0.009). For the ADHD-RS, the difference in values before and after oral administration was used to signify the change in symptoms (Figures 8 and   9). Overall, the decrease in PI values was significantly correlated with hyperactivity (ρ = 0.49; P = 0.032), but there was no correlation with inattention. In particular, the MPH group showed a significant corre- The PI values and CGI-I scores showed no correlation in any analysis.

| D ISCUSS I ON
The EEGs of patients with ADHD showed basic wave and sporadic abnormalities that were investigated in this study. Another study showed that MPH administration normalized the EEG in 56.9% of patients, with no change in 33.8%, and worsened the EEG in 9.3% of patients, indicating an inconsistency in EEG changes due to ADHD treatment drug administration. 26 However, another study indicated that in cases where the EEG was normalized, there was a decrease in θ waves and an increase in β waves. 27 In addition, there are few reports on the changes in basic waves due to ATX administration, with one report demonstrating an increase in β waves. 28 In the pre- In the present study, we investigated the correlation between ADHD symptoms and PI using a questionnaire administered to patients. The results revealed that there was a correlation between decreased PI values and improved hyperactivity in the MPH group, although there was no correlation between decreased PI and hyperactivity in the ATX group. While there was a difference between the two drugs, (MPH is a CNS agent, while ATX is a non-CNS agent), both MPH and ATX improve inattention and hyperactivity/impulsivity. One study showed that MPH improved hyperactivity, learning disability symptoms, and heat-induced seizures in epileptic rats, 36 and another study showed that MPH improved ADHD symptoms in 75% of patients with epilepsy. 37 Thus, the results of the present study are consistent with those of previous studies on MPH.
Regarding ATX and changes in ADHD symptoms and EEG basic Eight patients with ASD were included in present study. There were no significant PI changes between with and without ASD be-   In this study, we considered that there were no adverse effects on the abnormal sleep EEG in ADHD patients following administration of ADHD treatment, especially MPH. Although there was a concern that MPH administration in patients with ADHD with a history of epilepsy would exacerbate seizures, the results of this study suggest that it can be safely administered to patients with ADHD F I G U R E 8 PI change and ADHD-RS. PI decrease and reduction of hyperactivity were negative significant correlation (A) (ρ = 0.49, P =0.032). No significant correlation PI change and hyperactivity of ADHD-RS whether epilepsy, AEDs, ASD or not (B-G). ADHD, attentiondeficit hyperactivity disorder; ADHD-RS, ADHD Rating Scale; AED, antiepileptic drugs; ASD, autism spectrum disorders; PI, paroxysmal index symptoms. For patients with ADHD, alleviating ADHD symptoms is also involved in quality of life 39 ; therefore, appropriate treatment is desired even for patients with ADHD with epilepsy.

ACK N OWLED G M ENTS
This study was supported in part by an Intramural Research Grant

CO N FLI C T O F I NTE R E S T
Author EN has received speaker honoraria from Eisai Co., Ltd. The other authors declare no competing interests.

AUTH O R CO NTR I B UTI O N S
HY, EN, YKita, and MI were involved in study design and data interpretation. HY, Kita, and MI were involved in the data analysis. All authors critically revised the report, commented on drafts of the manuscript, and approved the final report.

A PPROVA L O F TH E R E S E A RCH PROTO CO L BY A N I N S TITUTI O N A L R E V I E WER B OA R D
Reviewer Board: approval of the Ethics Committee of NCNP (approval number A2014-114).

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.