The goal of the present study is to propose the use of global connectivity measures as quantitative indicators of long-term medication in pediatric patients with Attention-Deficit-Hyperactivity Disorder, combined type (ADHD-C). For this purpose, graph theoretical brain connectivity indices are computed from connectivity estimations across eyes-opened resting-state EEG recordings measured before and after the treatment with osmotic release oral system-methylphenidate for a month in 18 boys (aged between 7–12 years). In order to present the reliable results, neurofunctional correlations are firstly estimated in time (Pearson Correlation (PC), Spearman Correlation), frequency (Directed Transfer Function, Partial Directed Coherence) and phase (Phase Locking Value, Phase Lag Index) domains in between short segments of 2sec over single trials of 1min. Later, transitivity, clustering coefficients, assortativity, global efficiency and modularity are computed from EEG based connectivity matrices produced by each approach. Since the highest classification accuracy of 83.79% is provided by PC, statistical tests (one-way Anova, pair-wise multiple comparison) and step-wise logistic regression modelling are all examined to detect significant differences between pre- and post- treatment relevant connectivity measures. Statistical boxplots are also shown, as well. Overall results reveal that global brain connectivity can be increased by long-term medication in pediatric ADHD-C in terms of increased segregation & resilience. This is the first study to demonstrate that long-term medication can normalize the functional brain connectivity in ADHD, which is characterized by decreased connectivity compared to controls.
DISCISION AND CONCLUSION:
Discussion and Conclusion
In the present study, resting-state eyes-opened EEG series were recorded from 18 boys diagnosed with ADHD-C before medication phase. They had a specified treatment as OROS-MPH for a month. Then post-treatment resting-state eyes-opened EEG series were also recorded from them. Graph Theoretic brain network indices were estimated from both pre- and post-treatment related EEG recordings with respect to un-directed weighted adjacency matrixes that are extracted from brain dependency and connectivity estimations by PC, SC, PLV, PLI, DTF and PDC. In obtaining adjacency matrixes, threshold was 60% of maximum value in dependency matrix of interest in each short EEG segment of 2sec" role="presentation" > over trial of 1min" role="presentation" >.
The resulting estimations were investigated through both classifications with SVMs and statistical tests including one-way Anova and pair-wise multiple comparison. Finally, graphical box-plots of brain network indices were shown in pre- and post-treatment recordings in accordance with PC that provided superior results. Regarding step-wise logistic regression modelling, brain modularity (Q) is found to be less sensitive to treatment with OROS-MPH. The highest classification performance of 83.79%" role="presentation" > was obtained when three brain network indices (T,CC" role="presentation" > and r) were combined as feature set. The lowest statistical p-values were obtained for these three indices.
Combining quantitative results reveal that functional brain segregation (T and CC) and resilience (r) indices are mostly influenced by OROS-MPH, while integration and modularity indices are less affected by medication in boys with ADHD-C. However, the statistical box-plots on brain network indices showed the following brief results:
increased T indicates stronger connectivity of cortical regions to neighbors,
increased CC indicates increased efficiency of information flow,
increased r indicates increased number of interconnected neural populations thereby forming a resilient core to spread the info over the whole cortex,
decreased GE indicates decreased number of disconnected nodes,
decreased Q indicates decreased probability in determining cortical communication net into independent sub-nets.
From methodological point of view, statistical correlation method, PC provided the best results to compute brain dependency values that provide strong estimations in brain network measures from resting state eyes-opened surface EEG series. PC is found to be superior to both phase domain functional synchronization methods (PLV, PLI) and frequency domain effective connectivity methods (DTF, PDC).
In conclusion, our hypothesis is confirmed by the current results obtained by using time-domain dependency approach, PC in full-band eyes-opened resting state EEG segments. The past neuroimaging studies comparing patients and controls have a common emphasis on the decreased connectivity of the brain in ADHD. Therefore, the present findings are important contributions to the literature to show that the long-term medication increases functional connectivity in boys with ADHD-C. Moreover, graph theoretical analysis driven by default-mode EEG based statistical correlations have provided more insight to global connectivity in terms of integration, segregation, modularity and resilience in observing the impact of OROS-MPH in ADHD-C. Both segregation and resilience are increased by medication in accordance with the best results. From computational psychiatric point of view, quantitative visual results reveal that OROS-MPH clearly influence functional brain segregation and resilience in boys with ADHD-C. In detail, OROS-MPH provide the followings: (1) functional connectivity between cortical regions becomes more stronger, (2) neuronal undirected information flow becomes more efficient and (3) the number of interconnected regional neural populations that are required to form a resilient that provides information spread across the whole cortex. In future work, we are plan to record emotional EEG series from three groups of the patients (ADHD-I, ADHD-HI, ADHD-C) in response to affective static pictures (unpleasant vs pleasant) in both pre- and post- treatment periods in estimating brain network measures.
In the present study, pre- and post-treatment related resting-state EEG based global connectivity measures are estimated in several domains and then the best results are discussed with respect to specified network functions defined as segregation, integration and resilience assuming that functional connectivity is decreased in ADHD according to the past findings in ADHD research. Studies on how OROS-MPH changes regional FC in children with ADHD, supported by functional neuroimaging modalities, have shown that treatment normalizes the brain [56]. Unlike several structural studies, numerous fMRI studies have been conducted on MPH’s effects [57]. These studies can be divided into those investigating resting-state brain activation and those investigating task-related brain activation. Task-related studies have identified specific areas of activation with MPH treatment in the parietal areas during tasks involving attention, error monitoring, and interference inhibition [58], in the inferior frontal cortex during selective attention and response inhibition [59], and in the striatum during tasks involving reward and response inhibition [60]. Studies with tasks evaluating cognitive control have shown that MPH improves the ability to inhibit inappropriate responses in favour of more appropriate ones by increasing activation in frontostriatal circuits and ultimately normalizes brain activation patterns and executive functions during the task in children with ADHD with MPH treatment [61], [62]. One study performed with fMRI during the Stroop test showed that MPH treatment improved suppression of DMN activity in the ventral anterior cingulate and posterior cingulate cortices [63]. There is enough evidence to suggest that the poorly switched of mind-wandering functions of DMN at the time of cognitive task may be a cause of distractibility and impulsivity in children with ADHD and that increasing in DMN deactivation may play a role in the clinical improvement achieved with MPH treatment [64], [65], [56]. In the light of all these data, it can be concluded that MPH treatment has a clear protective and normalizing effect on the brain. Functional positive changes observed with MPH treatment, especially in the prefrontal cortex, and the improvement in DMN deactivation at the time of cognitive loading can be considered as evidence of the normalizing effect. The present study can minimize the relative factors that led to several inconsistency among the findings in the literature such that EEG based graph theoretical analysis of default-mode brain network is independent of both total number of electrodes and electrode placement on scalp surface.
