- Email: [email protected]
Increased rates of intermittent rhythmic delta and theta activity in the electroencephalographies of adult patients with attention-deficit hyperactivity disorder
Epilepsy & Behavior 75 (2017) 60–65
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Increased rates of intermittent rhythmic delta and theta activity in the electroencephalographies of adult patients with attention-deficit hyperactivity disorder Dominique Endres a,⁎, Simon Maier a, Bernd Feige a, Nora Bel Mokhtar a, Kathrin Nickel a, Peter Goll a, Simon A. Meyer a, Swantje Matthies a, Dieter Ebert a, Alexandra Philipsen b, Evgeniy Perlov a,c,1, Ludger Tebartz van Elst a,1 a
Section for Experimental Neuropsychiatry, Department of Psychiatry, Faculty of Medicine, University of Freiburg, Hauptstr. 5, 79104 Freiburg, Germany Medical Campus University of Oldenburg, School of Medicine and Health Sciences, Psychiatry and Psychotherapy — University Hospital, Karl-Jaspers-Klinik, Hermann-Ehlers-Str. 7, 26160 Bad Zwischenahn, Germany c Clinic for Psychiatry Luzern, Schafmattstrasse 1, 4915 St. Urban, Switzerland b
a r t i c l e
i n f o
Article history: Received 28 January 2017 Revised 22 April 2017 Accepted 30 June 2017 Available online xxxx Keywords: ADHD EEG Adults IRDA IRTA LANI
a b s t r a c t Introduction: Adult attention-deficit hyperactivity disorder (ADHD) is a common neurodevelopmental disorder. In subgroups of patients with a (para)epileptic pathomechanism, this might be due to intermittent rhythmic delta or theta activity (IRDA/IRTA). Participants and methods: Using a fully data-driven analysis, we compared the IRDA/IRTA rates in the resting electroencephalography (EEG) results of 97 adult patients with ADHD and 30 control subjects. The IRDA/IRTA rates before hyperventilation (HV) and for HV difference (difference between IRDA/IRTA rate after and before HV) were compared between groups using a linear model. Results: We detected significantly increased rates of IRDA/IRTA before HV (F = 4.209, p = 0.042) in patients with ADHD but no significant difference between the groups for HV-difference (F = 2.46, p = 0.119). Discussion: The increased IRDA/IRTA rates before HV in the group with ADHD might lead to (para)epileptic shortterm effects (e.g., impulsivity) via local area network inhibition, and to long-term effects (e.g., cognitive deficits) via connectivistic brain restructuring. © 2017 Elsevier Inc. All rights reserved.
1. Introduction 1.1. ADHD in adulthood Adult attention-deficit hyperactivity disorder (ADHD) is a common neurodevelopmental disorder with prevalence rates of 2–4% [1–3]. The main symptoms are inattention, hyperactivity, impulsivity, disorganized behavior, emotional instability, and impaired affect control [1,4]. The clinical presentations of an inattentive (iADHD), hyperactive–
⁎ Corresponding author at: Hauptstr. 5, 79104 Freiburg, Germany. E-mail addresses: [email protected] (D. Endres), [email protected] (S. Maier), [email protected] (B. Feige), [email protected] (K. Nickel), [email protected] (P. Goll), [email protected] (S.A. Meyer), [email protected] (S. Matthies), [email protected] (D. Ebert), [email protected] (A. Philipsen), [email protected] (E. Perlov), [email protected] (L. Tebartz van Elst). 1 Contributed equally.
http://dx.doi.org/10.1016/j.yebeh.2017.06.039 1525-5050/© 2017 Elsevier Inc. All rights reserved.
impulsive (hADHD) and combined subtype (cADHD) can be distinguished (http://www.dsm5.org). Frequent comorbidities are depression, anxiety disorders, and addiction [1,5]. Multimodal treatment options include pharmacological interventions (most often with methylphenidate) and psychotherapy [6–10]. 1.2. Pathophysiology of ADHD Neurochemically, a dopaminergic and norepinephrinergic deficit plays a central role in the pathophysiology of ADHD [1,11]. Therefore, both stimulant treatment with the dopamine reuptake inhibitor methylphenidate and nonstimulant therapy with the norepinephrine reuptake inhibitor atomoxetine are effective [3]. Neuroanatomically, dysfunction of the prefronto-striato-thalomo-reentrant circuits was found in ADHD [12]. These circuits are modulated by the mesolimbic dopaminergic and the interacting glutamatergic system [13–15]. Electrophysiologically, a subtle neuronal network instability via local area network inhibition (LANI-hypothesis) might lead to short-term symptoms (e.g., impulsivity), and impaired neuronal function to longterm deficits (e.g., ongoing cognitive deficits) [13,16,17]. Such hypotheses
D. Endres et al. / Epilepsy & Behavior 75 (2017) 60–65
are supported by the efficacy of antiepileptic medication, especially carbamazepine, on a meta-analytic level [18]. 1.3. ADHD and electroencephalography (EEG) findings Electroencephalography (EEG) alterations in patients with ADHD are frequent. Epileptiform activity was found in about 25% of children with ADHD [19] and is therefore more frequent than the 0.5–8% reported in control groups [13]. Electroencephalography pathologies in patients with ADHD were detected mainly in sleep and sleep deprivation (97.5%) compared to 7% in wake-only records [19]. At least 20% of children with epilepsy are affected by ADHD compared to about 5% in the pediatric control population [20,21]. Epilepsy was associated especially with the inattentive subtype [21]. Most of the earlier EEG studies were performed in children with ADHD, not in adults. 1.4. Rationale for our study Given these observations, the aim of our study was to analyze the role of EEG alterations in adult patients with ADHD as compared to a healthy control group. More precisely, we automatically analyzed the rate of intermittent rhythmic delta and theta activity (IRDA/IRTA) in a large cohort of adult patients with ADHD. IRDA/IRTA was interpreted as pathological EEG activity caused by diverse etiology, including epileptic reasons [22–24]. We earlier put forward the idea that IRDA/IRTA has the potential to induce adaptive homeostatic processes, which might lead to functional alterations of the affected neuronal networks [13,16,17,25]. We hypothesized that we would find (1) increased IRDA/IRTA rates in the resting EEGs of the group with ADHD before hyperventilation (HV) and for HV-difference (i.e., for the difference between after and before HV), and (2) significant correlations between IRDA/IRTA rates and psychometric scores of inattention. 2. Participants and methods The study received approval from the local ethics committee (Faculty of Medicine, Freiburg University, 217/06). The EEG examinations in the group with ADHD were part of the clinical diagnostic routine workup. All patients and controls agreed to the EEG measurements.
61
excluded from the study, as well as controls taking psychotropic drugs. Controls were investigated using semi-structured interviews (i.e., Mini International Neuropsychiatric Interview) and psychometric testing [32]. The psychometrics included the CAARS-S:L, WURS-k, BDI, and MWT-B questionnaires. CAARS-t-scores ≥65, WURS-k scores N 30, and BDI scores N 18 led to exclusion from the study. 2.3. Study sample The patient cohort was recruited from our special consultations for ADHD. We searched for the EEGs in our EEG database. Electroencephalographies were available for 108 patients who fulfilled all inclusion criteria. In seven cases, the time period between EEG measurement and psychometric testing was over 2 months; to avoid inaccurate psychometric scores at the time of the EEG measurement, we excluded these patients. Moreover, four EEGs were excluded due to artifacts. In so doing, we were able to collect EEG data for 97 patients. Electroencephalography examinations were performed in 34 healthy controls; one EEG was excluded due to technical reasons (i.e., too many artifacts), one EEG was excluded due to incomplete psychometrics, and two of them were excluded due to marginal but increased psychometric ADHD scores (i.e., CAARS-t-subscores N 65), resulting in a control group of 30 subjects. 2.4. EEG acquisition Topographical EEG was recorded using all 21 standard locations of the international 10–20 system [33] along with a Schwarzer 29-channel system and the Deltamed ‘Coherence’ acquisition software. The recording reference was Fpz, ground Oz. Analog signals were recorded with a time constant of 0.3 s and a low pass of 70 Hz, sampled at 256 Hz, and continuously stored for further processing. Offline, the digital signals were filtered between 0.3 and 45 Hz and down-sampled to 100 Hz. Electroencephalography monitoring was performed over 11 min including (1) a resting state EEG for 6 min, (2) an HV period for 3 min, and (3) a post-HV period for 2 min. 2.5. EEG analysis
The recruitment of patients with ADHD took place between 2007 and 2010. Experienced senior consultant psychiatrists assessed the patients according to DSM-IV criteria for ADHD. Comorbid schizophrenia, bipolar disorder, borderline personality disorder, antisocial personality disorder, suicidal or self-injurious behavior, autism, motor tics, Tourette's syndrome, and substance abuse/dependence within 6 months prior to screening (not episodic abuse) led to exclusion. Positive drug screening also led to exclusion. All patients were stimulant-free for a minimum of 6 months. Patients with somatic diseases that might cause symptoms of ADHD (e.g., former inflammatory disease of the brain, epilepsy, current hyperthyroidism, etc.) were excluded. Psychometric testing included the Wender Utah Rating Scale (WURS-k) for ADHD symptoms in childhood, the Conners Adult ADHD Rating Self-Report Scale in the Long Version (CAARS-S:L) for current ADHD symptoms, and the Beck Depression Inventory (BDI) for current depressive symptoms [26–29]. The multiple-choice vocabulary intelligence test (MWT-B) was used to quantify crystallized intelligence [30,31].
The data analysis was performed using our in-house software avg_q (https://github.com/berndf/avg_q). First, the software marked artifacts, including signal jumps and “blocking” (i.e., sections without signal variation in any channel). Second, the artifact-free data parts were analyzed using independent component analysis (ICA). Only the parts 5 s before and after any artifact marker were studied (extended ICA) [34,35]. Third, the detection and correction of electro-oculographic (EOG) artifacts was done by exclusion of EOG-related ICA components. In the fourth step, the IRDAs/IRTAs were detected. Therefore, empirically determined thresholds were used for the detection of jump and phase artifacts for optimizing detection and minimizing false-positive findings. IRDA/IRTA detection was done by sorting the ICA time series between 2 and 7 Hz and thresholding for maximal amplitude between 25 and 245 μV. Only IRDA/IRTA candidates within the artifact-free EEG parts were considered. IRDA/IRTA was detected in all non-excluded independent components, irrespective of topography. IRDA/IRTA rates were calculated as events per minute for the intervals before and after HV. Moreover the difference between after HV and before HV, abbreviated as “HV-difference”, was computed. The method was published earlier [17].
2.2. Healthy control group assessment
2.6. Statistical analyses
For the creation of the control group, we carried out EEG examinations in healthy control subjects between 2011 and 2012. The control subjects were recruited via public announcements. All control subjects with relevant psychiatric, medical, or neurological diseases were
The statistical analyses were performed using the software R, version 3.2.2 (www.r-project.org), and the Statistical Package for the Social Sciences, version 22 (SPSS 22; www-01.ibm.com/software/analytics/ spss). Group comparisons for continuous variables (age, IQ,
2.1. Patient assessment
62
D. Endres et al. / Epilepsy & Behavior 75 (2017) 60–65
psychometric scores) were performed using two-sided, independent sample t-tests. Group comparisons for gender were calculated using Pearson's two-sided chi-squared test. The group with ADHD and the control group differed in age and IQ; for that reason, IRDA/IRTA rates in both groups were corrected for these variables. A general linear model was computed for each of the target variables (“IRDAs/IRTAs before HV” and “HV-difference”) with age, IQ, and their interaction. Then each value was corrected by the difference between the predicted value for the actual age and IQ of the subject and the predicted value at an age of 40 years and an IQ of 110. Group comparisons for these corrected IRDA/IRTA rates were performed using a linear model with the fixed-factor group (patients with ADHD vs. controls) and the dependent variables “IRDAs/IRTAs before HV (corrected)” and “HV-difference (corrected)”. We included only these two independent variables in the linear model; “IRDAs/IRTAs after HV” was omitted because it is linearly dependent upon the other two. In addition, subgroup analyses were performed using a linear model with the fixed-factor subgroup (controls, patients with iADHD, patients with cADHD) and the dependent variables “IRDAs/IRTAs before HV (corrected)” and “HV-difference (corrected)”. Due to the small sample size of the group with hADHD (N = 4), these patients were added to the subgroup with cADHD for the subgroup analysis. Correlation analyses were performed using the Pearson correlation coefficient. For all statistical analyses, a p-value b0.05 served as the criterion of significance.
Fig. 1. Differences of IRDA/IRTA per minute before hyperventilation between the group with ADHD and the control group. Tukey box plot with notches indicating the approximated 95% confidence interval around the median. Abbreviations: ADHD, attention deficit hyperactivity disorder; IRDA, intermittent rhythmic delta activity; IRTA, intermittent rhythmic theta activity; HV, hyperventilation.
3. Results 3.4. Analysis of dimensional associations 3.1. Demographic and psychometric data The group with ADHD (N = 97) and the control group (N = 30) did not differ in gender relation (p = 0.279); however, both groups differed in age and IQ. As expected, psychometric attention, hyperactivity, and depression scores were significantly increased in the patient group. Most patients suffered from the cADHD (45%) and iADHD (51%) subtypes, and only 4% suffered from hADHD subtype (Table 1).
3.2. EEG findings In a general linear model with age- and IQ-corrected values and the fixed-factor group, we detected significantly increased rates of IRDAs/ IRTAs before HV (p = 0.042; Fig. 1) but no significant difference for HV-difference (p = 0.119; Fig. 2) in the group with ADHD (Table 2).
3.3. Subgroup analyses The subgroup analyses revealed no significant differences in IRDA/ IRTA rates before and after HV between the group with iADHD, the group with cADHD, and the control group (Table 3).
Correlation analysis for the whole study sample and the ADHD sample showed no significant correlations between IRDA/IRTA rates (corrected before HV, corrected HV-difference) and ADHD/depressiveness scores (WURS-k-score, CAARS-subscores, and BDI). 4. Discussion The main findings of our study were the increased IRDA/IRTA rates in the group with ADHD before HV. However, contrary to our hypothesis, we found no significant discrepancies for HV-difference, as well as no correlations between IRDA/IRTA rates and scores of inattention. 4.1. Comparison to previous studies Our results with increased EEG alterations in adult patients with ADHD are in line with frequently found EEG alterations in earlier studies where epileptiform activity was found in about 25% of affected children [19,20,36]. The highest rate of EEG pathologies (97.5%) was found in sleep and sleep-deprived records, and abnormalities in wake-only records were found to a lesser extent (7%) [19,36]. Therefore, our findings might be more pronounced if supplemented by sleep and
Table 1 Demographic and psychometric data for the group with ADHD and the control group.
Age Intelligence quotient Gender CAARS ADHD Index (t-scores) WURS-k BDI ADHD subtypes
Group with ADHD (n = 97) mean ± SD
Controls (n = 30) mean ± SD
Statistics
34.18 ± 9.95 111.87 ± 14.66 53 M:44 F 65.39 ± 12.24 40.01 ± 8.65 12.32 ± 8.57 cADHD: 44 iADHD: 49 hADHD: 4
41.77 ± 9.11 124.23 ± 15.81 13 M:17 F 39.57 ± 6.28⁎
p b 0.001 p b 0.001 p = 0.279 p b 0.001 p b 0.001 p b 0.001
6.50 ± 5.94 2.20 ± 2.52
Abbreviations: ADHD, attention-deficit-hyperactivity disorder; SD, standard deviation; M, male; F, female; CAARS, Conners Adult ADHD Rating Scales — Self Report: Long Version, WURS-k, Wender Utah Rating Scale; BDI, Beck Depression Inventory score; cADHD, combined subtype; iADHD, inattentive subtype; hADHD, hyperactive–impulsive subtype. ⁎ Only available from 28 controls.
D. Endres et al. / Epilepsy & Behavior 75 (2017) 60–65
63
studies, which described the group with iADHD as being closer to healthy profiles [43]. 4.2. The role of IRDA/IRTA
Fig. 2. Differences in HV-difference in the group with ADHD and the control group. Abbreviations: ADHD, attention-deficit-hyperactivity disorder; HV, hyperventilation.
sleep-deprived records. Compatible with our findings, quantitative EEG spectral power analyses showed a consistent but nonspecific increment of the power in the theta band, which was mostly frontocentral [37]. In contrast, the power in the beta bands is often decreased in ADHD. For that reason, the theta–beta ratio developed into the most commonly used EEG index in ADHD. Most older studies described a higher theta– beta ratio in subjects with ADHD compared with control subjects [37,38]. Recent trials were not able to confirm these findings, so that the authors of a meta-analysis concluded that it is not a reliable diagnostic measurement instrument for ADHD [39]. However, a subgroup of patients with ADHD still showed alterations in the theta–beta ratio, supporting the idea that paraepileptic pathomechanisms did not play a role in all ADHD cases but did so in a subgroup of patients with ADHD [39,40]. A high number of IRDA/IRTA were also found in patients with borderline personality disorder [25]. One can speculate that IRDA/IRTA is a correlate of impulsivity and other common symptoms of both diseases. However, we detected no correlation between impulsivity scores or other psychometric scores and IRDA/IRTA rates in our study. Taken together, we found evidence for basic neuronal network instability in adult ADHD. Interestingly, we detected no alterations for HVdifference, which means that HV does not seem to destabilize cerebral neuronal networks in ADHD, as we had earlier found in autism [17]. One might speculate that cerebral networks had already stabilized (perhaps delayed) in our adult cohort with ADHD. Such considerations would fit well with the idea of maturational delay in ADHD [41]. Children with ADHD passed through the same steps of maturation (e.g., measured by peak cortical thickness) as healthy controls, but the process is delayed in ADHD. In contrast, different growth curves were described in other neurodevelopmental disorders, such as autism [42]. With regard to subgroup analyses, we found no differences between the cADHD and iADHD subtypes. These findings are in contrast to earlier
In contrast to earlier EEG studies in ADHD, we focused on measuring IRDA/IRTA rates before and after HV. IRDA/IRTA can be relatively easy detected in routine EEGs [25]. It is a generalized, bisynchronous, intermittent, rhythmic slow activity generated in the thalamus or basal nuclei [22]. IRDAs/IRTAs were interpreted as pathological EEG activity and could be caused by structural, metabolic, inflammatory, traumatic, or epileptic reasons [22–25,44,45]. Unprovoked IRDA/IRTA, as we measured before HV, is associated with either diffuse cortical dysfunction or generalized epilepsy [24]. In order to understand the pathophysiological role of IRDA/IRTA in the development of different clinical symptoms, we earlier suggested the local area network inhibition (LANI) model for short-term effects [17]. Following this idea, the IRDA/IRTA represents one form of pathological excitatory network activity that is likely to trigger a secondary homoeostatic reaction leading to LANI and, therefore, to clinical symptoms [13,16,25]. Excitatory network activity (e.g., IRDA/IRTA) may also cause long-term effects via connectivistic brain restructuring in order to minimize the effects of excitatory activity [17,46]. 4.3. Clinical relevance of IRDA/IRTA in ADHD subgroups We suggest that IRDA/IRTA might serve as a biomarker representing relevant neuronal network instability in a subgroup of patients. Therefore, our findings might help to detect subgroups with ADHD on a pathophysiological level. Detecting such pathophysiological subgroups could help to develop personalized, biomarker-supported therapy approaches [47]. However, such considerations are of rather academic interest. From the patient's perspective, the question of whether such EEG pathologies in patients with ADHD might have therapeutic implications is much more important, especially in patients who could not be treated successfully with methylphenidate or atomoxetine. To answer this question, a stepped research concept is necessary, beginning with case studies and cumulative case studies, followed by uncontrolled open retrospective studies, and finally concluding with prospective, controlled studies. Laporte and colleagues presented a case study of a young patient suffering from cognitive and behavioral ADHD symptoms associated with epileptiform discharges occurring during sleep. The symptoms were successfully treated with carbamazepine and the patient's condition deteriorated after carbamazepine withdrawal [48]. In a case series, Schneebaum-Sender and colleagues reported improvement in attention along with the reduction of pathological EEG activity in four patients with ADHD treated with antiepileptics (sulthiame, lamotrigine, and levetiracetam) [49]. In an uncontrolled study of 46 children with ADHD, EEG improvement with antiepileptic valproate treatment showed a high correlation with behavioral improvements [50]. In a review including seven studies analyzing the effectiveness of carbamazepine in children with ADHD, the authors described significant therapeutic responses over the studies. A meta-analysis of three double-blind placebo-controlled studies also showed the effectiveness of carbamazepine treatment in young patients with ADHD [18]. In the first open and uncontrolled pilot study of adult ADHD, oxcarbazepine
Table 2 IRDAs in the group with ADHD and the control group. Group with ADHD (n = 97)
Controls (n = 30)
Statistics
IRDA/IRTA per min before HV
1.0127 ± 1.1945
0.5117 ± 1.0796
Difference between IRDA/IRTA per min after HV and before HV
0.7520 ± 2.4875
0.0215 ± 0.9658
F = 4.209 p = 0.0423 F = 2.461 p = 0.119
Abbreviations: ADHD, attention-deficit-hyperactivity disorder; IRDA, intermittent rhythmic delta activity; IRTA, intermittent rhythmic theta activity; HV, hyperventilation.
64
D. Endres et al. / Epilepsy & Behavior 75 (2017) 60–65
Table 3 IRDAs/IRTAs in the groups of combined and inattentive ADHD subtypes and the control group. Group with cADHD (n = 48⁎⁎)
Group with iADHD (n = 49)
Controls (n = 30)
Statistics
IRDA/IRTA before HV
0.9770 ± 1.1944
1.0476 ± 1.2060
0.5117 ± 1.0796
HV-Difference⁎
0.7165 ± 2.5778
0.7869 ± 2.4220
0.0215 ± 0.9658
F = 2.133 p = 0.123 F = 1.233 p = 0.295
Abbreviations: ADHD, attention-deficit-hyperactivity disorder; cADHD, combined subtype; iADHD, inattentive subtype; IRDA, intermittent rhythmic delta activity; IRTA, intermittent rhythmic theta activity; HV, Hyperventilation. ⁎ Difference between after HV and before HV. ⁎⁎ The hyperactive subtype was added to the combined subtype.
led to a significant reduction of ADHD symptoms [51]. In summary, there are clear indications for the effectiveness of antiepileptic treatment in children. Further research in adult cohorts with ADHD should analyze the treatment effects in patients with and without EEG alterations. If the EEG alterations (e.g., IRDA) function as a biomarker for a (para)epileptic pathomechanism, one could expect a better response to antiepileptic treatment in these patients. 4.4. Limitations The first major limitation of our study is the small and unmatched (for age and IQ) control group. To exclude age and IQ effects, we retrospectively corrected the IRDA/IRTA rates of each subject before HV and for HV-difference for these variables. We correlated the EEG findings with the psychometric scores. To avoid inaccurate psychometric scores at the time of the EEG measurement, we only included patients who had their testing a maximum of 2 months before or after the EEG. However, short time changes in depressive scores could have been overlooked by this procedure. The algorithms for EEG analysis were verified to work as intended and published earlier [17], but we did not perform an evaluation against human raters. The task is quite demanding for human raters, since all 21 channels have to be monitored for IRTA/IRDA activity that is not an artifact. However, the analysis could be repeated using multiple trained human raters in a future study. In previous studies, patients with ADHD with primary and secondary forms have often been included within one sample. Like in the other big neurodevelopmental disorder, autism-spectrum-disorder, secondary ADHD is defined as a condition in which there is evidence for organic genesis like a history of birth complications, encephalitic brain disorder in the past, a history of severe head trauma with loss of consciousness, seizures, or febrile convulsions, or evidence of a syndromic variant of ADHD, for example, on the background of fragile X syndrome [14,52]. Primary ADHD in contrast represent the large subgroup of patients with ADHD in whom no such secondary causes are present. Still there is often a positive family history illustrating that the cause for this variant is probably multigenetic [53]. In our study cohort, all patients were well investigated and only primary forms of ADHD — without an organic basis — were included. Therefore, our cohort is not representative for secondary, organic forms of ADHD. 5. Summary The main finding of our study is that of increased IRDA/IRTA rates before HV in the primary group with ADHD without any features of organic pathophysiology. These changes might lead to short-term effects (e.g., impulsivity) via LANI, and to long-term effects (e.g., cognitive deficits) via impaired neuronal function. Therefore, IRDA/IRTA might serve as a biomarker representing relevant neuronal network instability in a subgroup of adult patients with ADHD. Drug studies point out that antiepileptic medication could reduce ADHD symptoms. Further research approaches should analyze the treatment effects in patients with and without EEG alterations. If the EEG alterations (e.g., IRDA) function as a biomarker for a (para)epileptic pathomechanism, one could expect a better response to antiepileptic treatment in these patients.
Declarations Competing interests DE, SJM, BF, NBM, KN, SAM, DE, and EP have no competing interests to declare. PG received travel grants from GSK, Boston Scientific, and Otsuka Pharma. SM, in the years 2007–2009, received a speakers' fee from Jansen-Cilag and was involved in clinical trials conducted by Janssen-Cilag and Lilly. AP served on advisory boards, gave lectures, performed phase III studies, and received travel grants within the last 3 years from Eli Lilly, Janssen-Cilag, Medice Arzneimittel Pütter GmbH, Novartis, and Shire. LTvE had lectures, workshops, or travel grants within the last 3 years for Eli Lilly, Medice, Shire, UCB, Servier, and Cyberonics. Funding Parts of the study were funded by the German Federal Ministry of Science and Education (ADHD-NET: 01GV0605, 01GV0606). Authors' contributions AP, LTvE, and EP planned the study. DE, SM, AP, EP, and LTvE were involved in patient recruitment. NB performed the data collection and organized the EEG measurements. DE, SJM, NB, and BF conducted the data analysis. DE, BF, EP and LTvE calculated the statistics. DE wrote the paper. All authors were crucially involved in the theoretical discussion and creation of the manuscript. All authors read and approved the final version of the manuscript. References [1] Philipsen A, Hesslinger B, Tebartz van Elst L. Attention-deficit hyperactivity disorder in adulthood: diagnosis, etiology and therapy. Dtsch Arztebl Int 2008;105(17): 311–7. [2] Biederman J. Attention-deficit/hyperactivity disorder: a selective overview. Biol Psychiatry 2005;57(11):1215–20. [3] Volkow ND, Swanson JM. Clinical practice: adult attention deficit-hyperactivity disorder. N Engl J Med 2013;369(20):1935–44. [4] Ebert D, Krause J, Roth-Sackenheim C. ADHD in adulthood—guidelines based on expert consensus with DGPPN support. Nervenarzt 2003;74(10):939–46. [5] Biederman J. Impact of comorbidity in adults with attention-deficit/hyperactivity disorder. J Clin Psychiatry 2004;65(Suppl. 3):3–7. [6] Castells X, Ramos-Quiroga JA, Rigau D, et al. Efficacy of methylphenidate for adults with attention-deficit hyperactivity disorder: a meta-regression analysis. CNS Drugs 2011;25(2):157–69. [7] Mészáros A, Czobor P, Bálint S, Komlósi S, Simon V, Bitter I. Pharmacotherapy of adult attention deficit hyperactivity disorder (ADHD): a meta-analysis. Int J Neuropsychopharmacol 2009;12(8):1137–47. [8] Philipsen A. Psychotherapy in adult attention deficit hyperactivity disorder: implications for treatment and research. Expert Rev Neurother 2012;12(10):1217–25. [9] Philipsen A, Jans T, Graf E, et al. Effects of group psychotherapy, individual counseling, methylphenidate, and placebo in the treatment of adult attention-deficit/ hyperactivity disorder: a randomized clinical trial. JAMA Psychiat 2015;72(12): 1199–210. [10] Tebartz van Elst L, Maier S, Klöppel S, et al. The effect of methylphenidate intake on brain structure in adults with ADHD in a placebo-controlled randomized trial. J Psychiatry Neurosci 2016;41(6):422–30. [11] Biederman J, Faraone SV. Attention-deficit hyperactivity disorder. Lancet 2005; 366(9481):237–48.
D. Endres et al. / Epilepsy & Behavior 75 (2017) 60–65 [12] Perlov E, Philipsen A, Matthies S, et al. Spectroscopic findings in attention-deficit/ hyperactivity disorder: review and meta-analysis. World J Biol Psychiatry 2009; 10(4 Pt 2):355–65. [13] Tebartz van Elst L, Perlov E. Epilepsie und Psyche: Psychische Störungen bei Epilepsie — epileptische Phänomene in der Psychiatrie. 1st ed. Stuttgart: Kohlhammer; 2013. [14] Tebartz van Elst L, Maier S, Fangmeier T, et al. Disturbed cingulate glutamate metabolism in adults with high-functioning autism spectrum disorder: evidence in support of the excitatory/inhibitory imbalance hypothesis. Mol Psychiatry 2014;19(12): 1314–25. [15] van Elst LT, Maier S, Fangmeier T, et al. Magnetic resonance spectroscopy comparing adults with high functioning autism and above average IQ. Mol Psychiatry 2014; 19(12):1251. [16] Tebartz van Elst L, Krishnamoorthy ES, Schulze-Bonhage A, et al. Local area network inhibition: a model of a potentially important paraepileptic pathomechanism in neuropsychiatric disorders. Epilepsy Behav 2011;22(2):231–9. [17] Endres D, Maier S, Feige B, et al. Altered intermittent rhythmic delta and theta activity in the electroencephalographies of high functioning adult patients with autism spectrum disorder. Front Hum Neurosci 2017;11:66. [18] Silva R, Munoz D, Alpert M. Carbamazepine use in children and adolescents with features of attention-deficit hyperactivity disorder: a meta-analysis. J Am Acad Child Adolesc Psychiatry 1996;35(3):352–8. [19] Millichap J, Millichap J, Stack C. Utility of the electroencephalogram in attention deficit hyperactivity disorder. Clin EEG Neurosci 2011;42(3):180–4. [20] Kaufmann R, Goldberg-Stern H, Shuper A. Attention-deficit disorders and epilepsy in childhood: incidence, causative relations and treatment possibilities. J Child Neurol 2009;24(6):727–33. [21] Williams A, Giust J, Kronenberger W, Dunn D. Epilepsy and attention-deficit hyperactivity disorder: links, risks, and challenges. Neuropsychiatr Dis Treat 2016;12: 287–96. [22] Cobb WA. Rhythmic slow discharges in the electroencephalogram. J Neurol Neurosurg Psychiatry 1945;8:65–78. [23] Brigo F. Intermittent rhythmic delta activity patterns. Epilepsy Behav 2011;20(2): 254–6. [24] Altunel A. In reply. J Clin Neurophysiol Apr 2014;31(2):180 (2014;31(2):180). [25] Tebartz van Elst L, Fleck M, Bartels S, et al. Increased prevalence of intermittent rhythmic delta or theta activity (IRDA/IRTA) in the electroencephalogram (EEG) of patients with borderline personality disorder. Front Behav Neurosci 2016;10:12. [26] Ward MF, Wender PH, Reimherr FW. The Wender Utah rating scale: an aid in the retrospective diagnosis of childhood attention deficit hyperactivity disorder. Am J Psychiatry 1993;150(6):885–90. [27] Retz-Junginger P, Retz W, Blocher D, et al. Wender Utah Rating Scale (WURS-k) Die deutsche Kurzform zur retrospektiven Erfassung des hyperkinetischen Syndroms bei Erwachsenen. Nervenarzt 2002;73(9):830–8. [28] Conners C, Erhardt D, Sparrow E. Conners' adult ADHD rating scales (CAARS). North Tonawanda: multi-health systems. North Tonawanda: Multi-Health Systems; 1999. [29] Hautzinger M. Das Beck Depressionsinventar II. Deutsche Bearbeitung und Handbuch zum BDI II. Frankfurt a. M.: Harcourt Test Services; 2006 [30] Lehrl S. Mehrfachwahl-Wortschatz-Intelligenztest MWT-B. Erlangen: Straube; 1977. [31] Lehrl S, Triebig G, Fischer B. Multiple choice vocabulary test MWT as a valid and short test to estimate premorbid intelligence. Acta Neurol Scand 1995;91(5): 335–45. [32] Sheehan DV, Lecrubier Y, Sheehan KH, et al. The mini-international neuropsychiatric interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59(Suppl. 20): 22–33 [quiz 34–57]. [33] Klem G, Lüders H, Jasper H, Elger C. The ten-twenty electrode system of the international federation. The International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 1999;52:3–6.
65
[34] Lee W, Girolami M, Sejnowski T. Independent component analysis using an extended infomax algorithm for mixed subgaussian and supergaussian sources. Neural Comput 1999;11(2):417–41. [35] Makeig S, Westerfield M, Jung T, et al. Functionally independent components of the late positive event-related potential during visual spatial attention. J Neurosci 1999; 19(7):2665–80. [36] Millichap J, Stack C, Millichap J. Frequency of epileptiform discharges in the sleepdeprived electroencephalogram in children evaluated for attention-deficit disorders. J Child Neurol 2011;26(1):6–11. [37] Alba G, Pereda E, Mañas S, Méndez L, González A, González J. Electroencephalography signatures of attention-deficit/hyperactivity disorder: clinical utility. Neuropsychiatr Dis Treat 2015;11:2755–69. [38] Snyder S, Hall J. A meta-analysis of quantitative EEG power associated with attention-deficit hyperactivity disorder. J Clin Neurophysiol 2006;23(5):440–55. [39] Arns M, Conners C, Kraemer H. A decade of EEG theta/Beta ratio research in ADHD: a meta-analysis. J Atten Disord 2013;17(5):374–83. [40] Jeste S, Frohlich J, Loo S. Electrophysiological biomarkers of diagnosis and outcome in neurodevelopmental disorders. Curr Opin Neurol 2015;28(2):110–6. [41] El-Sayed E, Larsson J, Persson H, Santosh P, Rydelius P. “Maturational lag” hypothesis of attention deficit hyperactivity disorder: an update. Acta Paediatr 2003;92(7): 776–84. [42] Shaw P, Eckstrand K, Sharp W, et al. Attention-deficit/hyperactivity disorder is characterized by a delay in cortical maturation. Proc Natl Acad Sci U S A 2007;104(49): 19649–54. [43] Clarke A, Barry R, McCarthy R, Selikowitz M. EEG analysis in attention-deficit/ hyperactivity disorder: a comparative study of two subtypes. Psychiatry Res 1998; 81(1):19–29. [44] Accolla E, Kaplan P, Maeder-Ingvar M, Jukopila S, Rossetti A. Clinical correlates of frontal intermittent rhythmic delta activity (FIRDA). Clin Neurophysiol 2011; 122(1):27–31. [45] van Vliet J, Mulleners W, Meulstee J. EEG leading to the diagnosis of limbic encephalitis. Clin EEG Neurosci 2012;43(2):161–4. [46] Tebartz van Elst L, Riedel A, Maier S. Autism as a disorder of altered global functional and structural connectivity. Biol Psychiatry 2016;79(8):626–7. [47] Olbrich S, van Dinteren R, Arns M. Personalized medicine: review and perspectives of promising baseline EEG biomarkers in major depressive disorder and attention deficit hyperactivity disorder. Neuropsychobiology 2015;72(3–4):229–40. [48] Laporte N, Sébire G, Gillerot Y, Guerrini R, Ghariani S. Cognitive epilepsy: ADHD related to focal EEG discharges. Pediatr Neurol 2002;27(4):307–11. [49] Schneebaum-Sender N, Goldberg-Stern H, Fattal-Valevski A, Kramer U. Does a normalizing electroencephalogram in benign childhood epilepsy with Centrotemporal spikes abort attention deficit hyperactivity disorder? Pediatr Neurol 2012;47(4): 279–83. [50] Kanemura H, Sano F, Tando T, Hosaka H, Sugita K, Aihara M. EEG improvements with antiepileptic drug treatment can show a high correlation with behavioral recovery in children with ADHD. Epilepsy Behav 2013;27(3):443–8. [51] Davids E, Kis B, Specka M, Gastpar M. A pilot clinical trial of oxcarbazepine in adults with attention-deficit hyperactivity disorder. Prog Neuropsychopharmacol Biol Psychiatry 2006;30(6):1033–8. [52] Tebartz van Elst L, Pick M, Biscaldi M, Fangmeier T, Riedel A. High-functioning autism spectrum disorder as a basic disorder in adult psychiatry and psychotherapy: psychopathological presentation, clinical relevance and therapeutic concepts. Eur Arch Psychiatry Clin Neurosci 2013;263(Suppl. 2):S189–96. [53] Tebartz van Elst L. Autismus und ADHS Zwischen Normvariante, Persönlichkeitsstörung und neuropsychiatrischer Krankheit. 1st ed. Stuttgart: Kohhammer; 2016.
Contents lists available at ScienceDirect
Epilepsy & Behavior journal homepage: www.elsevier.com/locate/yebeh
Increased rates of intermittent rhythmic delta and theta activity in the electroencephalographies of adult patients with attention-deficit hyperactivity disorder Dominique Endres a,⁎, Simon Maier a, Bernd Feige a, Nora Bel Mokhtar a, Kathrin Nickel a, Peter Goll a, Simon A. Meyer a, Swantje Matthies a, Dieter Ebert a, Alexandra Philipsen b, Evgeniy Perlov a,c,1, Ludger Tebartz van Elst a,1 a
Section for Experimental Neuropsychiatry, Department of Psychiatry, Faculty of Medicine, University of Freiburg, Hauptstr. 5, 79104 Freiburg, Germany Medical Campus University of Oldenburg, School of Medicine and Health Sciences, Psychiatry and Psychotherapy — University Hospital, Karl-Jaspers-Klinik, Hermann-Ehlers-Str. 7, 26160 Bad Zwischenahn, Germany c Clinic for Psychiatry Luzern, Schafmattstrasse 1, 4915 St. Urban, Switzerland b
a r t i c l e
i n f o
Article history: Received 28 January 2017 Revised 22 April 2017 Accepted 30 June 2017 Available online xxxx Keywords: ADHD EEG Adults IRDA IRTA LANI
a b s t r a c t Introduction: Adult attention-deficit hyperactivity disorder (ADHD) is a common neurodevelopmental disorder. In subgroups of patients with a (para)epileptic pathomechanism, this might be due to intermittent rhythmic delta or theta activity (IRDA/IRTA). Participants and methods: Using a fully data-driven analysis, we compared the IRDA/IRTA rates in the resting electroencephalography (EEG) results of 97 adult patients with ADHD and 30 control subjects. The IRDA/IRTA rates before hyperventilation (HV) and for HV difference (difference between IRDA/IRTA rate after and before HV) were compared between groups using a linear model. Results: We detected significantly increased rates of IRDA/IRTA before HV (F = 4.209, p = 0.042) in patients with ADHD but no significant difference between the groups for HV-difference (F = 2.46, p = 0.119). Discussion: The increased IRDA/IRTA rates before HV in the group with ADHD might lead to (para)epileptic shortterm effects (e.g., impulsivity) via local area network inhibition, and to long-term effects (e.g., cognitive deficits) via connectivistic brain restructuring. © 2017 Elsevier Inc. All rights reserved.
1. Introduction 1.1. ADHD in adulthood Adult attention-deficit hyperactivity disorder (ADHD) is a common neurodevelopmental disorder with prevalence rates of 2–4% [1–3]. The main symptoms are inattention, hyperactivity, impulsivity, disorganized behavior, emotional instability, and impaired affect control [1,4]. The clinical presentations of an inattentive (iADHD), hyperactive–
⁎ Corresponding author at: Hauptstr. 5, 79104 Freiburg, Germany. E-mail addresses: [email protected] (D. Endres), [email protected] (S. Maier), [email protected] (B. Feige), [email protected] (K. Nickel), [email protected] (P. Goll), [email protected] (S.A. Meyer), [email protected] (S. Matthies), [email protected] (D. Ebert), [email protected] (A. Philipsen), [email protected] (E. Perlov), [email protected] (L. Tebartz van Elst). 1 Contributed equally.
http://dx.doi.org/10.1016/j.yebeh.2017.06.039 1525-5050/© 2017 Elsevier Inc. All rights reserved.
impulsive (hADHD) and combined subtype (cADHD) can be distinguished (http://www.dsm5.org). Frequent comorbidities are depression, anxiety disorders, and addiction [1,5]. Multimodal treatment options include pharmacological interventions (most often with methylphenidate) and psychotherapy [6–10]. 1.2. Pathophysiology of ADHD Neurochemically, a dopaminergic and norepinephrinergic deficit plays a central role in the pathophysiology of ADHD [1,11]. Therefore, both stimulant treatment with the dopamine reuptake inhibitor methylphenidate and nonstimulant therapy with the norepinephrine reuptake inhibitor atomoxetine are effective [3]. Neuroanatomically, dysfunction of the prefronto-striato-thalomo-reentrant circuits was found in ADHD [12]. These circuits are modulated by the mesolimbic dopaminergic and the interacting glutamatergic system [13–15]. Electrophysiologically, a subtle neuronal network instability via local area network inhibition (LANI-hypothesis) might lead to short-term symptoms (e.g., impulsivity), and impaired neuronal function to longterm deficits (e.g., ongoing cognitive deficits) [13,16,17]. Such hypotheses
D. Endres et al. / Epilepsy & Behavior 75 (2017) 60–65
are supported by the efficacy of antiepileptic medication, especially carbamazepine, on a meta-analytic level [18]. 1.3. ADHD and electroencephalography (EEG) findings Electroencephalography (EEG) alterations in patients with ADHD are frequent. Epileptiform activity was found in about 25% of children with ADHD [19] and is therefore more frequent than the 0.5–8% reported in control groups [13]. Electroencephalography pathologies in patients with ADHD were detected mainly in sleep and sleep deprivation (97.5%) compared to 7% in wake-only records [19]. At least 20% of children with epilepsy are affected by ADHD compared to about 5% in the pediatric control population [20,21]. Epilepsy was associated especially with the inattentive subtype [21]. Most of the earlier EEG studies were performed in children with ADHD, not in adults. 1.4. Rationale for our study Given these observations, the aim of our study was to analyze the role of EEG alterations in adult patients with ADHD as compared to a healthy control group. More precisely, we automatically analyzed the rate of intermittent rhythmic delta and theta activity (IRDA/IRTA) in a large cohort of adult patients with ADHD. IRDA/IRTA was interpreted as pathological EEG activity caused by diverse etiology, including epileptic reasons [22–24]. We earlier put forward the idea that IRDA/IRTA has the potential to induce adaptive homeostatic processes, which might lead to functional alterations of the affected neuronal networks [13,16,17,25]. We hypothesized that we would find (1) increased IRDA/IRTA rates in the resting EEGs of the group with ADHD before hyperventilation (HV) and for HV-difference (i.e., for the difference between after and before HV), and (2) significant correlations between IRDA/IRTA rates and psychometric scores of inattention. 2. Participants and methods The study received approval from the local ethics committee (Faculty of Medicine, Freiburg University, 217/06). The EEG examinations in the group with ADHD were part of the clinical diagnostic routine workup. All patients and controls agreed to the EEG measurements.
61
excluded from the study, as well as controls taking psychotropic drugs. Controls were investigated using semi-structured interviews (i.e., Mini International Neuropsychiatric Interview) and psychometric testing [32]. The psychometrics included the CAARS-S:L, WURS-k, BDI, and MWT-B questionnaires. CAARS-t-scores ≥65, WURS-k scores N 30, and BDI scores N 18 led to exclusion from the study. 2.3. Study sample The patient cohort was recruited from our special consultations for ADHD. We searched for the EEGs in our EEG database. Electroencephalographies were available for 108 patients who fulfilled all inclusion criteria. In seven cases, the time period between EEG measurement and psychometric testing was over 2 months; to avoid inaccurate psychometric scores at the time of the EEG measurement, we excluded these patients. Moreover, four EEGs were excluded due to artifacts. In so doing, we were able to collect EEG data for 97 patients. Electroencephalography examinations were performed in 34 healthy controls; one EEG was excluded due to technical reasons (i.e., too many artifacts), one EEG was excluded due to incomplete psychometrics, and two of them were excluded due to marginal but increased psychometric ADHD scores (i.e., CAARS-t-subscores N 65), resulting in a control group of 30 subjects. 2.4. EEG acquisition Topographical EEG was recorded using all 21 standard locations of the international 10–20 system [33] along with a Schwarzer 29-channel system and the Deltamed ‘Coherence’ acquisition software. The recording reference was Fpz, ground Oz. Analog signals were recorded with a time constant of 0.3 s and a low pass of 70 Hz, sampled at 256 Hz, and continuously stored for further processing. Offline, the digital signals were filtered between 0.3 and 45 Hz and down-sampled to 100 Hz. Electroencephalography monitoring was performed over 11 min including (1) a resting state EEG for 6 min, (2) an HV period for 3 min, and (3) a post-HV period for 2 min. 2.5. EEG analysis
The recruitment of patients with ADHD took place between 2007 and 2010. Experienced senior consultant psychiatrists assessed the patients according to DSM-IV criteria for ADHD. Comorbid schizophrenia, bipolar disorder, borderline personality disorder, antisocial personality disorder, suicidal or self-injurious behavior, autism, motor tics, Tourette's syndrome, and substance abuse/dependence within 6 months prior to screening (not episodic abuse) led to exclusion. Positive drug screening also led to exclusion. All patients were stimulant-free for a minimum of 6 months. Patients with somatic diseases that might cause symptoms of ADHD (e.g., former inflammatory disease of the brain, epilepsy, current hyperthyroidism, etc.) were excluded. Psychometric testing included the Wender Utah Rating Scale (WURS-k) for ADHD symptoms in childhood, the Conners Adult ADHD Rating Self-Report Scale in the Long Version (CAARS-S:L) for current ADHD symptoms, and the Beck Depression Inventory (BDI) for current depressive symptoms [26–29]. The multiple-choice vocabulary intelligence test (MWT-B) was used to quantify crystallized intelligence [30,31].
The data analysis was performed using our in-house software avg_q (https://github.com/berndf/avg_q). First, the software marked artifacts, including signal jumps and “blocking” (i.e., sections without signal variation in any channel). Second, the artifact-free data parts were analyzed using independent component analysis (ICA). Only the parts 5 s before and after any artifact marker were studied (extended ICA) [34,35]. Third, the detection and correction of electro-oculographic (EOG) artifacts was done by exclusion of EOG-related ICA components. In the fourth step, the IRDAs/IRTAs were detected. Therefore, empirically determined thresholds were used for the detection of jump and phase artifacts for optimizing detection and minimizing false-positive findings. IRDA/IRTA detection was done by sorting the ICA time series between 2 and 7 Hz and thresholding for maximal amplitude between 25 and 245 μV. Only IRDA/IRTA candidates within the artifact-free EEG parts were considered. IRDA/IRTA was detected in all non-excluded independent components, irrespective of topography. IRDA/IRTA rates were calculated as events per minute for the intervals before and after HV. Moreover the difference between after HV and before HV, abbreviated as “HV-difference”, was computed. The method was published earlier [17].
2.2. Healthy control group assessment
2.6. Statistical analyses
For the creation of the control group, we carried out EEG examinations in healthy control subjects between 2011 and 2012. The control subjects were recruited via public announcements. All control subjects with relevant psychiatric, medical, or neurological diseases were
The statistical analyses were performed using the software R, version 3.2.2 (www.r-project.org), and the Statistical Package for the Social Sciences, version 22 (SPSS 22; www-01.ibm.com/software/analytics/ spss). Group comparisons for continuous variables (age, IQ,
2.1. Patient assessment
62
D. Endres et al. / Epilepsy & Behavior 75 (2017) 60–65
psychometric scores) were performed using two-sided, independent sample t-tests. Group comparisons for gender were calculated using Pearson's two-sided chi-squared test. The group with ADHD and the control group differed in age and IQ; for that reason, IRDA/IRTA rates in both groups were corrected for these variables. A general linear model was computed for each of the target variables (“IRDAs/IRTAs before HV” and “HV-difference”) with age, IQ, and their interaction. Then each value was corrected by the difference between the predicted value for the actual age and IQ of the subject and the predicted value at an age of 40 years and an IQ of 110. Group comparisons for these corrected IRDA/IRTA rates were performed using a linear model with the fixed-factor group (patients with ADHD vs. controls) and the dependent variables “IRDAs/IRTAs before HV (corrected)” and “HV-difference (corrected)”. We included only these two independent variables in the linear model; “IRDAs/IRTAs after HV” was omitted because it is linearly dependent upon the other two. In addition, subgroup analyses were performed using a linear model with the fixed-factor subgroup (controls, patients with iADHD, patients with cADHD) and the dependent variables “IRDAs/IRTAs before HV (corrected)” and “HV-difference (corrected)”. Due to the small sample size of the group with hADHD (N = 4), these patients were added to the subgroup with cADHD for the subgroup analysis. Correlation analyses were performed using the Pearson correlation coefficient. For all statistical analyses, a p-value b0.05 served as the criterion of significance.
Fig. 1. Differences of IRDA/IRTA per minute before hyperventilation between the group with ADHD and the control group. Tukey box plot with notches indicating the approximated 95% confidence interval around the median. Abbreviations: ADHD, attention deficit hyperactivity disorder; IRDA, intermittent rhythmic delta activity; IRTA, intermittent rhythmic theta activity; HV, hyperventilation.
3. Results 3.4. Analysis of dimensional associations 3.1. Demographic and psychometric data The group with ADHD (N = 97) and the control group (N = 30) did not differ in gender relation (p = 0.279); however, both groups differed in age and IQ. As expected, psychometric attention, hyperactivity, and depression scores were significantly increased in the patient group. Most patients suffered from the cADHD (45%) and iADHD (51%) subtypes, and only 4% suffered from hADHD subtype (Table 1).
3.2. EEG findings In a general linear model with age- and IQ-corrected values and the fixed-factor group, we detected significantly increased rates of IRDAs/ IRTAs before HV (p = 0.042; Fig. 1) but no significant difference for HV-difference (p = 0.119; Fig. 2) in the group with ADHD (Table 2).
3.3. Subgroup analyses The subgroup analyses revealed no significant differences in IRDA/ IRTA rates before and after HV between the group with iADHD, the group with cADHD, and the control group (Table 3).
Correlation analysis for the whole study sample and the ADHD sample showed no significant correlations between IRDA/IRTA rates (corrected before HV, corrected HV-difference) and ADHD/depressiveness scores (WURS-k-score, CAARS-subscores, and BDI). 4. Discussion The main findings of our study were the increased IRDA/IRTA rates in the group with ADHD before HV. However, contrary to our hypothesis, we found no significant discrepancies for HV-difference, as well as no correlations between IRDA/IRTA rates and scores of inattention. 4.1. Comparison to previous studies Our results with increased EEG alterations in adult patients with ADHD are in line with frequently found EEG alterations in earlier studies where epileptiform activity was found in about 25% of affected children [19,20,36]. The highest rate of EEG pathologies (97.5%) was found in sleep and sleep-deprived records, and abnormalities in wake-only records were found to a lesser extent (7%) [19,36]. Therefore, our findings might be more pronounced if supplemented by sleep and
Table 1 Demographic and psychometric data for the group with ADHD and the control group.
Age Intelligence quotient Gender CAARS ADHD Index (t-scores) WURS-k BDI ADHD subtypes
Group with ADHD (n = 97) mean ± SD
Controls (n = 30) mean ± SD
Statistics
34.18 ± 9.95 111.87 ± 14.66 53 M:44 F 65.39 ± 12.24 40.01 ± 8.65 12.32 ± 8.57 cADHD: 44 iADHD: 49 hADHD: 4
41.77 ± 9.11 124.23 ± 15.81 13 M:17 F 39.57 ± 6.28⁎
p b 0.001 p b 0.001 p = 0.279 p b 0.001 p b 0.001 p b 0.001
6.50 ± 5.94 2.20 ± 2.52
Abbreviations: ADHD, attention-deficit-hyperactivity disorder; SD, standard deviation; M, male; F, female; CAARS, Conners Adult ADHD Rating Scales — Self Report: Long Version, WURS-k, Wender Utah Rating Scale; BDI, Beck Depression Inventory score; cADHD, combined subtype; iADHD, inattentive subtype; hADHD, hyperactive–impulsive subtype. ⁎ Only available from 28 controls.
D. Endres et al. / Epilepsy & Behavior 75 (2017) 60–65
63
studies, which described the group with iADHD as being closer to healthy profiles [43]. 4.2. The role of IRDA/IRTA
Fig. 2. Differences in HV-difference in the group with ADHD and the control group. Abbreviations: ADHD, attention-deficit-hyperactivity disorder; HV, hyperventilation.
sleep-deprived records. Compatible with our findings, quantitative EEG spectral power analyses showed a consistent but nonspecific increment of the power in the theta band, which was mostly frontocentral [37]. In contrast, the power in the beta bands is often decreased in ADHD. For that reason, the theta–beta ratio developed into the most commonly used EEG index in ADHD. Most older studies described a higher theta– beta ratio in subjects with ADHD compared with control subjects [37,38]. Recent trials were not able to confirm these findings, so that the authors of a meta-analysis concluded that it is not a reliable diagnostic measurement instrument for ADHD [39]. However, a subgroup of patients with ADHD still showed alterations in the theta–beta ratio, supporting the idea that paraepileptic pathomechanisms did not play a role in all ADHD cases but did so in a subgroup of patients with ADHD [39,40]. A high number of IRDA/IRTA were also found in patients with borderline personality disorder [25]. One can speculate that IRDA/IRTA is a correlate of impulsivity and other common symptoms of both diseases. However, we detected no correlation between impulsivity scores or other psychometric scores and IRDA/IRTA rates in our study. Taken together, we found evidence for basic neuronal network instability in adult ADHD. Interestingly, we detected no alterations for HVdifference, which means that HV does not seem to destabilize cerebral neuronal networks in ADHD, as we had earlier found in autism [17]. One might speculate that cerebral networks had already stabilized (perhaps delayed) in our adult cohort with ADHD. Such considerations would fit well with the idea of maturational delay in ADHD [41]. Children with ADHD passed through the same steps of maturation (e.g., measured by peak cortical thickness) as healthy controls, but the process is delayed in ADHD. In contrast, different growth curves were described in other neurodevelopmental disorders, such as autism [42]. With regard to subgroup analyses, we found no differences between the cADHD and iADHD subtypes. These findings are in contrast to earlier
In contrast to earlier EEG studies in ADHD, we focused on measuring IRDA/IRTA rates before and after HV. IRDA/IRTA can be relatively easy detected in routine EEGs [25]. It is a generalized, bisynchronous, intermittent, rhythmic slow activity generated in the thalamus or basal nuclei [22]. IRDAs/IRTAs were interpreted as pathological EEG activity and could be caused by structural, metabolic, inflammatory, traumatic, or epileptic reasons [22–25,44,45]. Unprovoked IRDA/IRTA, as we measured before HV, is associated with either diffuse cortical dysfunction or generalized epilepsy [24]. In order to understand the pathophysiological role of IRDA/IRTA in the development of different clinical symptoms, we earlier suggested the local area network inhibition (LANI) model for short-term effects [17]. Following this idea, the IRDA/IRTA represents one form of pathological excitatory network activity that is likely to trigger a secondary homoeostatic reaction leading to LANI and, therefore, to clinical symptoms [13,16,25]. Excitatory network activity (e.g., IRDA/IRTA) may also cause long-term effects via connectivistic brain restructuring in order to minimize the effects of excitatory activity [17,46]. 4.3. Clinical relevance of IRDA/IRTA in ADHD subgroups We suggest that IRDA/IRTA might serve as a biomarker representing relevant neuronal network instability in a subgroup of patients. Therefore, our findings might help to detect subgroups with ADHD on a pathophysiological level. Detecting such pathophysiological subgroups could help to develop personalized, biomarker-supported therapy approaches [47]. However, such considerations are of rather academic interest. From the patient's perspective, the question of whether such EEG pathologies in patients with ADHD might have therapeutic implications is much more important, especially in patients who could not be treated successfully with methylphenidate or atomoxetine. To answer this question, a stepped research concept is necessary, beginning with case studies and cumulative case studies, followed by uncontrolled open retrospective studies, and finally concluding with prospective, controlled studies. Laporte and colleagues presented a case study of a young patient suffering from cognitive and behavioral ADHD symptoms associated with epileptiform discharges occurring during sleep. The symptoms were successfully treated with carbamazepine and the patient's condition deteriorated after carbamazepine withdrawal [48]. In a case series, Schneebaum-Sender and colleagues reported improvement in attention along with the reduction of pathological EEG activity in four patients with ADHD treated with antiepileptics (sulthiame, lamotrigine, and levetiracetam) [49]. In an uncontrolled study of 46 children with ADHD, EEG improvement with antiepileptic valproate treatment showed a high correlation with behavioral improvements [50]. In a review including seven studies analyzing the effectiveness of carbamazepine in children with ADHD, the authors described significant therapeutic responses over the studies. A meta-analysis of three double-blind placebo-controlled studies also showed the effectiveness of carbamazepine treatment in young patients with ADHD [18]. In the first open and uncontrolled pilot study of adult ADHD, oxcarbazepine
Table 2 IRDAs in the group with ADHD and the control group. Group with ADHD (n = 97)
Controls (n = 30)
Statistics
IRDA/IRTA per min before HV
1.0127 ± 1.1945
0.5117 ± 1.0796
Difference between IRDA/IRTA per min after HV and before HV
0.7520 ± 2.4875
0.0215 ± 0.9658
F = 4.209 p = 0.0423 F = 2.461 p = 0.119
Abbreviations: ADHD, attention-deficit-hyperactivity disorder; IRDA, intermittent rhythmic delta activity; IRTA, intermittent rhythmic theta activity; HV, hyperventilation.
64
D. Endres et al. / Epilepsy & Behavior 75 (2017) 60–65
Table 3 IRDAs/IRTAs in the groups of combined and inattentive ADHD subtypes and the control group. Group with cADHD (n = 48⁎⁎)
Group with iADHD (n = 49)
Controls (n = 30)
Statistics
IRDA/IRTA before HV
0.9770 ± 1.1944
1.0476 ± 1.2060
0.5117 ± 1.0796
HV-Difference⁎
0.7165 ± 2.5778
0.7869 ± 2.4220
0.0215 ± 0.9658
F = 2.133 p = 0.123 F = 1.233 p = 0.295
Abbreviations: ADHD, attention-deficit-hyperactivity disorder; cADHD, combined subtype; iADHD, inattentive subtype; IRDA, intermittent rhythmic delta activity; IRTA, intermittent rhythmic theta activity; HV, Hyperventilation. ⁎ Difference between after HV and before HV. ⁎⁎ The hyperactive subtype was added to the combined subtype.
led to a significant reduction of ADHD symptoms [51]. In summary, there are clear indications for the effectiveness of antiepileptic treatment in children. Further research in adult cohorts with ADHD should analyze the treatment effects in patients with and without EEG alterations. If the EEG alterations (e.g., IRDA) function as a biomarker for a (para)epileptic pathomechanism, one could expect a better response to antiepileptic treatment in these patients. 4.4. Limitations The first major limitation of our study is the small and unmatched (for age and IQ) control group. To exclude age and IQ effects, we retrospectively corrected the IRDA/IRTA rates of each subject before HV and for HV-difference for these variables. We correlated the EEG findings with the psychometric scores. To avoid inaccurate psychometric scores at the time of the EEG measurement, we only included patients who had their testing a maximum of 2 months before or after the EEG. However, short time changes in depressive scores could have been overlooked by this procedure. The algorithms for EEG analysis were verified to work as intended and published earlier [17], but we did not perform an evaluation against human raters. The task is quite demanding for human raters, since all 21 channels have to be monitored for IRTA/IRDA activity that is not an artifact. However, the analysis could be repeated using multiple trained human raters in a future study. In previous studies, patients with ADHD with primary and secondary forms have often been included within one sample. Like in the other big neurodevelopmental disorder, autism-spectrum-disorder, secondary ADHD is defined as a condition in which there is evidence for organic genesis like a history of birth complications, encephalitic brain disorder in the past, a history of severe head trauma with loss of consciousness, seizures, or febrile convulsions, or evidence of a syndromic variant of ADHD, for example, on the background of fragile X syndrome [14,52]. Primary ADHD in contrast represent the large subgroup of patients with ADHD in whom no such secondary causes are present. Still there is often a positive family history illustrating that the cause for this variant is probably multigenetic [53]. In our study cohort, all patients were well investigated and only primary forms of ADHD — without an organic basis — were included. Therefore, our cohort is not representative for secondary, organic forms of ADHD. 5. Summary The main finding of our study is that of increased IRDA/IRTA rates before HV in the primary group with ADHD without any features of organic pathophysiology. These changes might lead to short-term effects (e.g., impulsivity) via LANI, and to long-term effects (e.g., cognitive deficits) via impaired neuronal function. Therefore, IRDA/IRTA might serve as a biomarker representing relevant neuronal network instability in a subgroup of adult patients with ADHD. Drug studies point out that antiepileptic medication could reduce ADHD symptoms. Further research approaches should analyze the treatment effects in patients with and without EEG alterations. If the EEG alterations (e.g., IRDA) function as a biomarker for a (para)epileptic pathomechanism, one could expect a better response to antiepileptic treatment in these patients.
Declarations Competing interests DE, SJM, BF, NBM, KN, SAM, DE, and EP have no competing interests to declare. PG received travel grants from GSK, Boston Scientific, and Otsuka Pharma. SM, in the years 2007–2009, received a speakers' fee from Jansen-Cilag and was involved in clinical trials conducted by Janssen-Cilag and Lilly. AP served on advisory boards, gave lectures, performed phase III studies, and received travel grants within the last 3 years from Eli Lilly, Janssen-Cilag, Medice Arzneimittel Pütter GmbH, Novartis, and Shire. LTvE had lectures, workshops, or travel grants within the last 3 years for Eli Lilly, Medice, Shire, UCB, Servier, and Cyberonics. Funding Parts of the study were funded by the German Federal Ministry of Science and Education (ADHD-NET: 01GV0605, 01GV0606). Authors' contributions AP, LTvE, and EP planned the study. DE, SM, AP, EP, and LTvE were involved in patient recruitment. NB performed the data collection and organized the EEG measurements. DE, SJM, NB, and BF conducted the data analysis. DE, BF, EP and LTvE calculated the statistics. DE wrote the paper. All authors were crucially involved in the theoretical discussion and creation of the manuscript. All authors read and approved the final version of the manuscript. References [1] Philipsen A, Hesslinger B, Tebartz van Elst L. Attention-deficit hyperactivity disorder in adulthood: diagnosis, etiology and therapy. Dtsch Arztebl Int 2008;105(17): 311–7. [2] Biederman J. Attention-deficit/hyperactivity disorder: a selective overview. Biol Psychiatry 2005;57(11):1215–20. [3] Volkow ND, Swanson JM. Clinical practice: adult attention deficit-hyperactivity disorder. N Engl J Med 2013;369(20):1935–44. [4] Ebert D, Krause J, Roth-Sackenheim C. ADHD in adulthood—guidelines based on expert consensus with DGPPN support. Nervenarzt 2003;74(10):939–46. [5] Biederman J. Impact of comorbidity in adults with attention-deficit/hyperactivity disorder. J Clin Psychiatry 2004;65(Suppl. 3):3–7. [6] Castells X, Ramos-Quiroga JA, Rigau D, et al. Efficacy of methylphenidate for adults with attention-deficit hyperactivity disorder: a meta-regression analysis. CNS Drugs 2011;25(2):157–69. [7] Mészáros A, Czobor P, Bálint S, Komlósi S, Simon V, Bitter I. Pharmacotherapy of adult attention deficit hyperactivity disorder (ADHD): a meta-analysis. Int J Neuropsychopharmacol 2009;12(8):1137–47. [8] Philipsen A. Psychotherapy in adult attention deficit hyperactivity disorder: implications for treatment and research. Expert Rev Neurother 2012;12(10):1217–25. [9] Philipsen A, Jans T, Graf E, et al. Effects of group psychotherapy, individual counseling, methylphenidate, and placebo in the treatment of adult attention-deficit/ hyperactivity disorder: a randomized clinical trial. JAMA Psychiat 2015;72(12): 1199–210. [10] Tebartz van Elst L, Maier S, Klöppel S, et al. The effect of methylphenidate intake on brain structure in adults with ADHD in a placebo-controlled randomized trial. J Psychiatry Neurosci 2016;41(6):422–30. [11] Biederman J, Faraone SV. Attention-deficit hyperactivity disorder. Lancet 2005; 366(9481):237–48.
D. Endres et al. / Epilepsy & Behavior 75 (2017) 60–65 [12] Perlov E, Philipsen A, Matthies S, et al. Spectroscopic findings in attention-deficit/ hyperactivity disorder: review and meta-analysis. World J Biol Psychiatry 2009; 10(4 Pt 2):355–65. [13] Tebartz van Elst L, Perlov E. Epilepsie und Psyche: Psychische Störungen bei Epilepsie — epileptische Phänomene in der Psychiatrie. 1st ed. Stuttgart: Kohlhammer; 2013. [14] Tebartz van Elst L, Maier S, Fangmeier T, et al. Disturbed cingulate glutamate metabolism in adults with high-functioning autism spectrum disorder: evidence in support of the excitatory/inhibitory imbalance hypothesis. Mol Psychiatry 2014;19(12): 1314–25. [15] van Elst LT, Maier S, Fangmeier T, et al. Magnetic resonance spectroscopy comparing adults with high functioning autism and above average IQ. Mol Psychiatry 2014; 19(12):1251. [16] Tebartz van Elst L, Krishnamoorthy ES, Schulze-Bonhage A, et al. Local area network inhibition: a model of a potentially important paraepileptic pathomechanism in neuropsychiatric disorders. Epilepsy Behav 2011;22(2):231–9. [17] Endres D, Maier S, Feige B, et al. Altered intermittent rhythmic delta and theta activity in the electroencephalographies of high functioning adult patients with autism spectrum disorder. Front Hum Neurosci 2017;11:66. [18] Silva R, Munoz D, Alpert M. Carbamazepine use in children and adolescents with features of attention-deficit hyperactivity disorder: a meta-analysis. J Am Acad Child Adolesc Psychiatry 1996;35(3):352–8. [19] Millichap J, Millichap J, Stack C. Utility of the electroencephalogram in attention deficit hyperactivity disorder. Clin EEG Neurosci 2011;42(3):180–4. [20] Kaufmann R, Goldberg-Stern H, Shuper A. Attention-deficit disorders and epilepsy in childhood: incidence, causative relations and treatment possibilities. J Child Neurol 2009;24(6):727–33. [21] Williams A, Giust J, Kronenberger W, Dunn D. Epilepsy and attention-deficit hyperactivity disorder: links, risks, and challenges. Neuropsychiatr Dis Treat 2016;12: 287–96. [22] Cobb WA. Rhythmic slow discharges in the electroencephalogram. J Neurol Neurosurg Psychiatry 1945;8:65–78. [23] Brigo F. Intermittent rhythmic delta activity patterns. Epilepsy Behav 2011;20(2): 254–6. [24] Altunel A. In reply. J Clin Neurophysiol Apr 2014;31(2):180 (2014;31(2):180). [25] Tebartz van Elst L, Fleck M, Bartels S, et al. Increased prevalence of intermittent rhythmic delta or theta activity (IRDA/IRTA) in the electroencephalogram (EEG) of patients with borderline personality disorder. Front Behav Neurosci 2016;10:12. [26] Ward MF, Wender PH, Reimherr FW. The Wender Utah rating scale: an aid in the retrospective diagnosis of childhood attention deficit hyperactivity disorder. Am J Psychiatry 1993;150(6):885–90. [27] Retz-Junginger P, Retz W, Blocher D, et al. Wender Utah Rating Scale (WURS-k) Die deutsche Kurzform zur retrospektiven Erfassung des hyperkinetischen Syndroms bei Erwachsenen. Nervenarzt 2002;73(9):830–8. [28] Conners C, Erhardt D, Sparrow E. Conners' adult ADHD rating scales (CAARS). North Tonawanda: multi-health systems. North Tonawanda: Multi-Health Systems; 1999. [29] Hautzinger M. Das Beck Depressionsinventar II. Deutsche Bearbeitung und Handbuch zum BDI II. Frankfurt a. M.: Harcourt Test Services; 2006 [30] Lehrl S. Mehrfachwahl-Wortschatz-Intelligenztest MWT-B. Erlangen: Straube; 1977. [31] Lehrl S, Triebig G, Fischer B. Multiple choice vocabulary test MWT as a valid and short test to estimate premorbid intelligence. Acta Neurol Scand 1995;91(5): 335–45. [32] Sheehan DV, Lecrubier Y, Sheehan KH, et al. The mini-international neuropsychiatric interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998;59(Suppl. 20): 22–33 [quiz 34–57]. [33] Klem G, Lüders H, Jasper H, Elger C. The ten-twenty electrode system of the international federation. The International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol Suppl 1999;52:3–6.
65
[34] Lee W, Girolami M, Sejnowski T. Independent component analysis using an extended infomax algorithm for mixed subgaussian and supergaussian sources. Neural Comput 1999;11(2):417–41. [35] Makeig S, Westerfield M, Jung T, et al. Functionally independent components of the late positive event-related potential during visual spatial attention. J Neurosci 1999; 19(7):2665–80. [36] Millichap J, Stack C, Millichap J. Frequency of epileptiform discharges in the sleepdeprived electroencephalogram in children evaluated for attention-deficit disorders. J Child Neurol 2011;26(1):6–11. [37] Alba G, Pereda E, Mañas S, Méndez L, González A, González J. Electroencephalography signatures of attention-deficit/hyperactivity disorder: clinical utility. Neuropsychiatr Dis Treat 2015;11:2755–69. [38] Snyder S, Hall J. A meta-analysis of quantitative EEG power associated with attention-deficit hyperactivity disorder. J Clin Neurophysiol 2006;23(5):440–55. [39] Arns M, Conners C, Kraemer H. A decade of EEG theta/Beta ratio research in ADHD: a meta-analysis. J Atten Disord 2013;17(5):374–83. [40] Jeste S, Frohlich J, Loo S. Electrophysiological biomarkers of diagnosis and outcome in neurodevelopmental disorders. Curr Opin Neurol 2015;28(2):110–6. [41] El-Sayed E, Larsson J, Persson H, Santosh P, Rydelius P. “Maturational lag” hypothesis of attention deficit hyperactivity disorder: an update. Acta Paediatr 2003;92(7): 776–84. [42] Shaw P, Eckstrand K, Sharp W, et al. Attention-deficit/hyperactivity disorder is characterized by a delay in cortical maturation. Proc Natl Acad Sci U S A 2007;104(49): 19649–54. [43] Clarke A, Barry R, McCarthy R, Selikowitz M. EEG analysis in attention-deficit/ hyperactivity disorder: a comparative study of two subtypes. Psychiatry Res 1998; 81(1):19–29. [44] Accolla E, Kaplan P, Maeder-Ingvar M, Jukopila S, Rossetti A. Clinical correlates of frontal intermittent rhythmic delta activity (FIRDA). Clin Neurophysiol 2011; 122(1):27–31. [45] van Vliet J, Mulleners W, Meulstee J. EEG leading to the diagnosis of limbic encephalitis. Clin EEG Neurosci 2012;43(2):161–4. [46] Tebartz van Elst L, Riedel A, Maier S. Autism as a disorder of altered global functional and structural connectivity. Biol Psychiatry 2016;79(8):626–7. [47] Olbrich S, van Dinteren R, Arns M. Personalized medicine: review and perspectives of promising baseline EEG biomarkers in major depressive disorder and attention deficit hyperactivity disorder. Neuropsychobiology 2015;72(3–4):229–40. [48] Laporte N, Sébire G, Gillerot Y, Guerrini R, Ghariani S. Cognitive epilepsy: ADHD related to focal EEG discharges. Pediatr Neurol 2002;27(4):307–11. [49] Schneebaum-Sender N, Goldberg-Stern H, Fattal-Valevski A, Kramer U. Does a normalizing electroencephalogram in benign childhood epilepsy with Centrotemporal spikes abort attention deficit hyperactivity disorder? Pediatr Neurol 2012;47(4): 279–83. [50] Kanemura H, Sano F, Tando T, Hosaka H, Sugita K, Aihara M. EEG improvements with antiepileptic drug treatment can show a high correlation with behavioral recovery in children with ADHD. Epilepsy Behav 2013;27(3):443–8. [51] Davids E, Kis B, Specka M, Gastpar M. A pilot clinical trial of oxcarbazepine in adults with attention-deficit hyperactivity disorder. Prog Neuropsychopharmacol Biol Psychiatry 2006;30(6):1033–8. [52] Tebartz van Elst L, Pick M, Biscaldi M, Fangmeier T, Riedel A. High-functioning autism spectrum disorder as a basic disorder in adult psychiatry and psychotherapy: psychopathological presentation, clinical relevance and therapeutic concepts. Eur Arch Psychiatry Clin Neurosci 2013;263(Suppl. 2):S189–96. [53] Tebartz van Elst L. Autismus und ADHS Zwischen Normvariante, Persönlichkeitsstörung und neuropsychiatrischer Krankheit. 1st ed. Stuttgart: Kohhammer; 2016.