- Top of page
- Supporting Information
Purpose: Children with epilepsy have a significant risk for attention-deficit/hyperactivity disorder (ADHD), which is often accompanied by deficits in working memory performance. However, it is not yet clear whether there are specific differences in the underlying mechanisms of working memory capability between children with epilepsy-related ADHD and those with developmental ADHD. There is evidence that methylphenidate can improve the behavioral difficulties in children with developmental ADHD. Whether this medication has the same effect on ADHD symptoms in patients with epilepsy is not yet well understood. The aim of the present study is, therefore, to evaluate whether boys with epilepsy-related ADHD and developmental ADHD share a common behavioral, pharmacoresponsive, and neurofunctional pathophysiology.
Methods: Seventeen boys with diagnosed combined epilepsy/ADHD, 15 boys with developmental ADHD, and 15 healthy controls (aged 8–14 years) performed on working memory tasks (N-back) while brain activation was recorded using functional magnetic resonance imaging. Each patient was tested twice: once after the intake of methylphenidate and once without in a counterbalanced order.
Key Findings: On a behavioral level, we show that boys with epilepsy-related ADHD as well as those with developmental ADHD performed similarly poorly on tasks with high cognitive load when compared to healthy controls, and that intake of methylphenidate improved performance almost to normal levels in both ADHD groups. On the functional level, both patient groups showed similar reductions of activation in all relevant parts of the functional network of working memory when compared to controls. Of interest, intake of methylphenidate did not significantly alter this activity pattern.
Significance: Our data show strong similarities between epilepsy-related and developmental ADHD on the behavioral, pharmacoresponsive, and neural level, favoring the view that ADHD with and without epilepsy shares a common underlying neurobehavioral pathophysiology.
Children with epilepsy often experience attentional problems such as attention-deficit/hyperactivity disorder (ADHD). Approximately one-third of children with epilepsy, in addition to their diagnosis of epilepsy, also receive a diagnosis of ADHD (Dunn et al., 2003; Hermann et al., 2007) and a potential bidirectional association of those two disorders has become the focus of recently published studies. On one hand it has been shown that children with developmental ADHD have an incidence of electroencephalography (EEG) abnormalities significantly higher than in the normal pediatric population (Richer et al., 2002; Davis et al., 2010) and on the other hand, attention problems including ADHD have also been reported to exist before the first recognized seizure (Dunn et al., 1997; Austin et al., 2001; Hesdorffer et al., 2004). Various mechanisms that may account for this association are discussed. These include effects of antiepileptic medications, common genetic predisposition, biochemical factors, subclinical epileptiform discharges, or even the fact that ADHD and epilepsy are both common childhood disorders (Hamoda et al., 2009; Kaufmann et al., 2009; Parisi et al., 2010). However, most of the studies are exclusively based on behavioral and neuropsychological examinations and are limited by their methodologic procedures. Whether the associated ADHD in epilepsy and developmental ADHD share a common underlying neurobiology remains an open question.
Hermann et al. (2007) published the first results on the neurobiology of ADHD in children with idiopathic epilepsy using voxel-based morphometry (VBM). The authors found significantly decreased gray matter in the frontal lobes and smaller brainstem volumes in children with epilepsy/ADHD compared to healthy controls. However, they did not match the epileptic patients with a sample of patients with developmental ADHD and, therefore, no comparison could be drawn between these two clinical samples. In a recent published study, we compared a sample of patients with epilepsy/ADHD to those with developmental ADHD and a group of healthy controls using diffusion tensor imaging (DTI) (Bechtel et al., 2009). Results revealed deficient cerebellar connections in both patient groups compared to controls, suggesting that patients with epilepsy and/or ADHD may have similar cerebellar pathology. However, this study was restricted to cerebellar regions and permits conclusions exclusively on a structural level. Functional data on the topic are still missing.
In patients with developmental ADHD a substantial literature confirms the presence of executive dysfunctions, including working memory deficits (Willcutt et al., 2005), which are also supported by functional magnetic resonance imaging (fMRI) studies. Although healthy controls showed robust increased activation patterns in frontal, parietal, and cerebellar regions during working memory tasks (Owen et al., 2005), several studies reported decreased activation in this functional network in patients with ADHD (Valera et al., 2005; Ehlis et al., 2008; Kobel et al., 2009). Working memory tasks seem to be sensitive in differentiating between healthy controls and patients with ADHD and may also give valuable insights into the neurofunctional pathology of patients with epilepsy-related ADHD.
Methylphenidate (MPH) is the best studied and widely prescribed stimulant for the treatment of ADHD (Weber & Lutschg, 2002), and a number of studies have shown that it is highly effective in alleviating the symptoms of the disorder (Solanto, 1998). Recent controlled trials of MPH in patients with well-controlled epilepsy and ADHD have shown significant improvements in ADHD symptoms without an exacerbation of seizures (Gross-Tsur et al., 1997; Baptista-Neto et al., 2008). In higher doses and in patients with frequent seizures, the safety of MPH needs further evaluation (Gonzalez-Heydrich et al., 2010). Although MPH is the most prominent medication to improve the symptoms of ADHD, studies examining the effect of stimulants in children with epilepsy by functional neuroimaging techniques have not yet been reported.
The aim of the present study was to clarify whether ADHD in epilepsy shares the pathophysiology of ADHD seen in nonepileptic populations. To answer this question, we examined behavioral differences in working memory performance, pharmacologic efficiency profiles of MPH, and functional brain organization in children with epilepsy/ADHD, children with developmental ADHD, and healthy controls.
- Top of page
- Supporting Information
Although there is an increasing interest in the clinical diagnosis of ADHD in epilepsy, this is to our knowledge the first study evaluating its mechanisms using neuropsychological and functional imaging methods. When we applied working memory tasks, our data showed that behavioral and functional performance of patients with epilepsy/ADHD were comparable to patients with developmental ADHD but differed from healthy controls.
Our behavioral results are consistent with those of studies that reported working memory deficits in nonepileptic ADHD (Martinussen et al., 2005; Willcutt et al., 2005; Kobel et al., 2009). In children with idiopathic epilepsy, cognitive problems have been reported before as well (MacAllister & Schaffer, 2007); however, with respect to working memory performance, findings are contradictory. Pascalicchio et al. (2007) have found deficits in digit span in children with juvenile myoclonic epilepsy; in contrast, patients in the study of Myatchin et al. (2009) performed as well as controls. Nevertheless, few studies have controlled for the confounding effect of the comorbidity ADHD. The importance of this issue is nicely illustrated by Hermann et al. (2007) who compared patients with epilepsy with and without ADHD. The authors found significant differences in motor/psychomotor speed, as well as in executive functions between patient groups implying, that ADHD has a disruptive effect on cognition in children with epilepsy. In the present study, patients with epilepsy/ADHD performed as poorly as boys with developmental ADHD, leading to the assumption that both patient groups show a comparable impairment in working memory functioning. No differences in performance were found due to the different subtypes of ADHD. This finding is in accordance with an earlier study by Pasini et al. (2007).
MPH-treatment led to an improvement in performance in both patient groups, which became similar to the one of healthy controls in the more difficult tasks. In addition, within-subject comparisons for the patient groups showed the effect of MPH on performance, in particular for the most difficult task. From these results it can be concluded, that more demanding tasks seem to be especially sensitive to the improving effect of MPH. This improvement has been already shown for a variety of cognitive domains in children with ADHD without epilepsy (Pietrzak et al., 2006; Kobel et al., 2009). Our data showed no difference between the two patient groups in performance, implying that all patients clearly benefited from MPH to the same extent during tasks requiring higher cognitive processes. Data therefore support a clear effect of this medication also in epilepsy-related ADHD.
The evaluation of working memory using fMRI is well established and induces prominent blood oxygen level dependent (BOLD) activations in frontal, parietal, and cerebellar regions (Owen et al., 2005). For the first time in the epilepsy literature, functional brain correlates of ADHD were revealed using working memory tasks. Boys with epilepsy/ADHD recruited substantially less cortical regions that are crucial for working memory performance compared to healthy controls and showed in addition no involvement of the cerebellum. Of interest, these activation patterns did not differ from the ones of boys with developmental ADHD and replicate the robust findings of previous studies reporting hypoactivation in the functional network of working memory in developmental ADHD (Valera et al., 2005; Dickstein et al., 2006; Cherkasova & Hechtman, 2009; Kobel et al., 2009; Wolf et al., 2009). Due to the functional similarity, data support the idea that ADHD with or without epilepsy show a common aberrant network of working memory. It has to be mentioned though, that our results are based on a block-design and, therefore, it was not possible to exclude error trials. In the future an event-related approach could lead to even more sensitive findings.
Concerning the intake of MPH, we did not find any alterations in activation patterns in patients. These findings of improved performance but not normalized brain activity have been already reported for patients with ADHD (Schweitzer et al., 2004; Kobel et al., 2009). However, in the literature, findings are heterogeneous, reporting also normalization of brain activation (Rubia et al., 2011a,b). Such discrepancies may result from heterogeneity in patient populations, such as different mean age, familiarity with medication, and methodologic differences such as different cognitive paradigms and different MRI scanners/scanning parameters. In addition, Bush et al. (2005) mentioned that especially fMRI data with a strong focus on lateral frontal cortex is much more inconsistent and produces high variability.
Altogether the present study revealed behavioral, pharmacoresponsive, and functional similarities between patients with isolated ADHD and epilepsy-related ADHD. Our findings, therefore, clearly support the assumption that ADHD with and without epilepsy share a common underlying neurobehavioral pathophysiology. This stands in contrast to postulations indicating that the symptoms of ADHD in epilepsy might rather originate from the epilepsy itself and its related neurologic damage or use of AEDs causing functional impairment within the brain (Kaufmann et al., 2009). In our study we cannot rule out subclinical effects of AEDs on attention (Glauser et al., 2010), but we assume that they are not the determining factors for the attentional problems in our sample of patients with epilepsy. First, all patients were seizure free for at least half a year, thereby minimizing the effect of electrical bursts on cognition, and secondly, patients were treated with AEDs that are considered not to have a substantial influence on cognitive, especially attention functions (Nadkarni et al., 2005; Donati et al., 2007), as diagnosed by the clinical diagnostic procedure. Moreover, it has been shown that behavioral problems including ADHD often exist before the first recognized seizure and, therefore, before the intake of AEDs (Austin et al., 2001; Hesdorffer et al., 2004). We assume that the above-mentioned factors might have an aggravating effect on ADHD symptomatology but may not be the underlying cause. The assumption of a common underlying pathophysiology is also supported by the work of Gonzalez-Heydrich et al. (2007). The authors found a strong similarity in comorbidity and clinical presentation between ADHD with and without epilepsy, suggesting that attentional/hyperactive and impulsive behavior commonly observed in boys with epilepsy truly constitute ADHD. To keep our sample as homogeneous as possible, we included only boys. Whether these data can be transferred unchanged to girls remains to be shown. In conclusion, our data support the assumption that both conditions might represent epiphenomena of a common underlying functional as well as neurobiologic network abnormality. Therefore, in terms of clinical relevance, our findings support the idea that attention deficits that emerge in the context of a developmental disorder are directly comparable to attention deficits in epileptic children and that treatment with MPH shows equivalent effectiveness in both patient samples.