Anodal tDCS over the left prefrontal cortex does not cause clinically significant changes in circulating metabolites

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY 4.0 https://creativecommons.org/licenses/by/4.0/ Anodal tDCS over the left prefrontal cortex does not cause clinically significant changes in circulating metabolites © 2020 Kortteenniemi, Ortega-Alonso, Javadi, Tolmunen, Ali-Sisto, Kotilainen, Wikgren, Karhunen, Velagapudi and Lehto. Published version Kortteenniemi, Aaron M.; Ortega-Alonso, Alfredo; Javadi, Amir-Homayoun; Tolmunen, Tommi; Ali-Sisto, Toni; Kotilainen, Tuukka; Wikgren, Jan; Karhunen, Leila; Velagapudi, Vidya; Lehto, Soili M. Kortteenniemi, A. M., Ortega-Alonso, A., Javadi, A.-H., Tolmunen, T., Ali-Sisto, T., Kotilainen, T., Wikgren, J., Karhunen, L., Velagapudi, V., & Lehto, S. M. (2020). Anodal tDCS over the left prefrontal cortex does not cause clinically significant changes in circulating metabolites. Frontiers in Psychiatry, 11, Article 403. https://doi.org/10.3389/fpsyt.2020.00403 2020 Anodal tDCS Over the Left Prefrontal Cortex Does Not Cause Clinically Significant Changes in Circulating Metabolites Aaron Kortteenniemi 1 , Alfredo Ortega-Alonso 2,3 , Amir-Homayoun Javadi 4,5,6 , Tommi Tolmunen 1,7 , Toni Ali-Sisto 1 , Tuukka Kotilainen 1 , Jan Wikgren 8 , Leila Karhunen 9 , Vidya Velagapudi 10 and Soili M. Lehto 11 * 1 Institute of Clinical Medicine, University of Eastern Finland, Kuopio, Finland, 2 Department of Psychology and Logopedics, Faculty of Medicine, University of Helsinki, University of Helsinki, Helsinki, Finland, 3 Department of Public Health Solutions, National Institute for Health and Welfare, Helsinki, Finland, 4 School of Psychology, University of Kent, Canterbury, United Kingdom, 5 Department of Experimental Psychology, Institute of Behavioural Neuroscience, University College London, London, United Kingdom, 6 School of Rehabilitation, Tehran University of Medical Sciences, Tehran, Iran, 7 Institute of Clinical Medicine and Clinical Nutrition, Kuopio University Hospital, Kuopio, Finland, 8 Centre for Interdisciplinary Brain Research, Department of Psychology, University of Jyväskylä, Jyväskylä, Finland, 9 Institute of Public Health and Clinical Nutrition, University of Eastern Finland, Kuopio, Finland, 10 Metabolomics Unit, Institute for Molecular Medicine Finland (FIMM), University of Helsinki, Helsinki, Finland, 11 Psychiatry, University of Helsinki and Helsinki University Hospital, Helsinki, Finland Background: Transcranial direct current stimulation (tDCS), a putative treatment for depression, has been proposed to affect peripheral metabolism. Metabolic products from brain tissue may also cross the blood–brain barrier, reflecting the conditions in the brain. However, there are no previous data regarding the effect of tDCS on circulating metabolites. Objective: To determine whether five daily sessions of tDCS modulate peripheral metabolites in healthy adult men. Methods: This double-blind, randomized controlled trial involved 79 healthy males (aged 20– 40 years) divided into two groups, one receiving tDCS (2 mA) and the other sham stimulated. The anode was placed over the left dorsolateral prefrontal cortex and the cathode over the corresponding contralateral area. Venous blood samples were obtained before and after the first stimulation session, and after the fifth stimulation session. Serum levels of 102 metabolites were determined by mass spectrometry. The results were analysed with generalised estimating equations corrected for the family-wise error rate. In addition, we performed power calculations estimating sample sizes necessary for future research. Results: TDCS-related variation in serum metabolite levels was extremely small and statistically non-significant. Power calculations indicated that for the observed variation to be deemed significant, samples sizes of up to 11,000 subjects per group would be required, depending on the metabolite of interest. Frontiers in Psychiatry | www.frontiersin.org May 2020 | Volume 11 | Article 4031 Edited by: Bartosz Zurowski, University Medical Center Schleswig-Holstein, Germany Reviewed by: Deniz Doruk, Mayo Clinic, United States Yosuke Morishima, University of Bern, Switzerland Ewelina Kinga Wardzinski, University of Lübeck, Germany *Correspondence: Soili M. Lehto soili.lehto@helsinki.fi Specialty section: This article was submitted to Neuroimaging and Stimulation, a section of the journal Frontiers in Psychiatry Received: 11 November 2019 Accepted: 21 April 2020 Published: 07 May 2020 Citation: Kortteenniemi A, Ortega-Alonso A, Javadi A-H, Tolmunen T, Ali-Sisto T, Kotilainen T, Wikgren J, Karhunen L, Velagapudi V and Lehto SM (2020) Anodal tDCS Over the Left Prefrontal Cortex Does Not Cause Clinically Significant Changes in Circulating Metabolites. Front. Psychiatry 11:403. doi: 10.3389/fpsyt.2020.00403 ORIGINAL RESEARCH published: 07 May 2020 doi: 10.3389/fpsyt.2020.00403 Conclusion: Our study found that five sessions of tDCS induced no major effects on peripheral metabolites among healthy men. These observations support the view of tDCS as a safe treatment that does not induce significant changes in the measured peripheral metabolites in healthy male subjects. Keywords: brain stimulation, transcranial direct current stimulation, tES, metabolism, mass spectrometry INTRODUCTION Transcranial direct current stimulation (tDCS) is a non-invasive method for modulating neuronal activity by introducing a small electric current into the brain via electrodes placed on the scalp. It has attracted increasing interest among both clinicians and researchers during the past decade. It has been speculated to be a promising candidate in the treatment of major depressive disorder (1), bipolar depression (2), dependence and craving (3), as well as neuropathic and idiopathic pain (3). The low cost, safety and simplicity of use make tDCS an attractive option for clinical applications, even though more research is needed to determine the optimal treatment protocols and patient characteristics. At the cellular level, tDCS is considered to alter neuronal resting membrane potentials (4,5). For example, anodal stimulation has been observed to lead to increased neuronal activity in the motor cortex, while cathodal stimulation of the same area has been considered to inhibit neuronal firing (5,6). In addition to local cortical effects in neuronal excitability, recent research has demonstrated that tDCS also appears to exert changes in neuronal activity in deeper brain structures, such as the midbrain nuclei (7). The brain acts as a control unit that regulates the functions of the entire organism, and is in turn modified by peripheral physiology. Nevertheless, only limited knowledge is available on possible peripheral changes induced by brain stimulation using tDCS. Previous research has indicated that tDCS induces changes in cerebral blood flow (8), brain neurotransmitter levels (9)and central and peripheral metabolic activity (10). Furthermore, anodal tDCS in the area corresponding to the left dorsolateral prefrontal cortex (DLPFC) has been postulated to lead to changes in autonomic nervous system activity, measured by highand lowfrequency heart rate variability and the secretion of cortisol under normal conditions (11), as well as in stressful situations (12,13). Thus, current evidence indicates that tDCS may induce very specific physiological responses in both the central nervous system and the periphery, although our understanding of possible systemic metabolic adaptations is very limited. Electrical fields created by tDCS exert force on polar and/or charged molecules, and many proteins and amino acids, which often serve as neurotransmitters of the brain, are charged. This could lead to alterations in the concentrations of polar molecules and trigger larger metabolic cascades. These consequences may explain some of the observed physiological effects of tDCS in both the central nervous system and the periphery. With regards to the peripheral effects of tDCS, a previous study conducted by Binkofski et al. (10) indicated that anodal tDCS led to a transient drop in high-energy phosphorus compounds (adenosine triphosphate (ATP), phosphocreatinine) both under the electrode and in the contralateral hemisphere, suggesting that there had been a widespread increase in neuronal glutaminergic activity. By using a hyperinsulinemic–euglycemic clamp, they also observed a simultaneous increase in glucose uptake in the peripheral circulation. The extent of the increase clearly correlated with the levels of the previously mentioned energetic compounds in the brain tissue, suggestive of elevated glucose uptake across the BBB. As many physiological processes in the central nervous system result in observable changes in metabolite profiles in the periphery, metabolomic analysis of possible tDCS effects could provide a valuable tool for investigating the safety profile and mechanisms of action of tDCS. As many circulating metabolites are biologically active compounds, alterations in the levels of such compounds could result in safety concerns in healthy or clinical populations. We investigated the acute effects of tDCS on peripheral metabolites in a sample of 79 healthy men. The working hypotheses were that the metabolic state of the participants would change as a result of the stimulation and that differences between the study groups could be reflected in altered metabolite profiles. MATERIALS AND METHODS Study Population This investigation formed part of the larger Optimizing Transcranial Electrical Stimulation for Clinical Applications (OptES) Study. The OptES Study was designed to generate novel information on the mechanisms of action of transcranial electrical stimulation, and to use this information to develop new, better clinical applications of transcranial electrical stimulation. The study protocol was approved by the Ethics Committee of the North Savo Hospital District (permit number 41/2015). All participants provided written informed consent after a full explanation of the study. The study conformed with the Declaration of Helsinki. In order to decrease random variance in our findings, we chose to include only male participants, since our metabolomics platform also included some compounds known to be affected by sex hormones either centrally (14) or in the periphery (15,16). A total of 80 male volunteers were recruited from the North Savo region of Finland. The participants received either tDCS or sham stimulation in a double-blind setting. The inclusion criteria were male gender (to avoid possible confounding metabolic effects of the menstrual cycle), aged between 18 and 40 years at the time of recruitment, right-handedness (i.e., belonging to the 1st to 10th right decile according to the Edinburgh Handedness Questionnaire (17)), and not having previously received tDCS. Kortteenniemi et al. Peripheral Metabolic Changes of tDCS Frontiers in Psychiatry | www.frontiersin.org May 2020 | Volume 11 | Article 4032 The exclusion criteria were metal implants inside the skull or eye, severe skin lesions in the electrode placement areas, a pacemaker, a history of epilepsy or previous seizures, a history of intracerebral bleeding during the previous six months, a selfreported history of substance dependence/abuse during the past six months and a history of any endocrinological condition [i.e., any physician-defined E00-E32 diagnosis according to the International Statistical Classification of Diseases and Related Health Problems version 10 (IDC-10) (18)]. We were unable to obtain the post-stimulation venous blood sample from one participant, and this individual was therefore excluded from all analyses. Thus, the final sample size was 40 in the tDCS group (mean age 28.3 years, mean BMI 26.0 kg/m 2 ) and 39 in the sham group (mean age 27.7 years, mean BMI 25.4 kg/m 2 ). Experimental Procedure The participants were randomly assigned to either 1) the active stimulation or 2) the sham stimulation group in a 1:1 fashion, utilizing a computer-generated scheme. Basic sociodemographic information, including age, height and weight used to calculate the body mass index (BMI, kg/m 2 ), was collected from questionnaires that the participants completed prior to the first stimulation. Three venous blood samples were collected via venipuncture by a trained nurse. Each participant received either tDCS or sham stimulation once per day for five consecutive days (Figure 1). Sociodemographic questionnaires were completed before the stimulation on day one. The baseline blood sample was collected immediately before the first stimulation session, with the second sample being drawn immediately after the first stimulation session (in order to investigate the acute effects of stimulation); the third sample was drawn immediately after the fifth stimulation session. A maximum of 5 min was allowed between stimulation and the collection of the blood samples. Before the experiment, the participants were instructed to abstain from alcohol for 12 h and to have consumed no more than two doses during the preceding 24 h, to abstain from products containing caffeine for 3 h, and to abstain from smoking and heavy exercise for 1 h prior to the experiment. In addition, all experiments were conducted in the morning, and the participants were instructed to fast for 10 h to allow fasting blood samples to be obtained. This procedure enabled minimization of the confounding metabolic effects of the consumption of dietary products. Before stimulation, participant compliance with the instructions was checked, and those not conforming to the instructions were re-scheduled to another testing day. Each participant received a 20-minute stimulation session with a current of 2 mA using a neuroConn DC-Stimulator (neuroConn GmbH, Ilmenau, Germany). The experimenter was unaware of the form of stimulation delivered (i.e., active stimulation vs. sham stimulation), and the participants were randomly divided between the study groups by a computer. The electrodes were made of conductive rubber placed inside sponge pads soaked with 12 ml of saline and held down with elastic straps. The electrode area was 25 cm 2 , resulting in a current density of 0.8 A/m 2 . The anode was placed at site F3 and the cathode at site F4 according to the international 10–20 electroencephalography system. In order to standardize the external stimuli, during the stimulation the participants were asked to refrain from talking and to watch a neutral landscape video with headphones on. The sham group received 15 seconds of ramping up and ramping down at the beginning, after which stimulation was discontinued. After the stimulation, both the participant and the experimenter filled in a form in which they were asked to provide their estimate (percentage) of the likelihood of the participant belonging to the sham group. Blood Sample Analysis Venous blood samples were collected into Vacuette 454078 4-ml serum gel tubes (Greiner Bio-One GmbH, Rainbach im Mühlkreis, Austria). They were left at room temperature for 30 min, followed by 10 min of centrifugation at 2,400×g, +20°C, to separate serum. The serum samples were frozen at −80°C until analysed. The metabolites were extracted from the serum samples using acetonitrile:formic acid (99:1 v/v) as a solvent (1:4, sample: solvent) and analysed using an ACQUITY UPLC-MS/MS system (Waters Corporation, Milford, MA, USA). A detailed protocol and instrument conditions have been published elsewhere (19). FIGURE 1 | Timeline of the study. Kortteenniemi et al. Peripheral Metabolic Changes of tDCS Frontiers in Psychiatry | www.frontiersin.org May 2020 | Volume 11 | Article 4033 Statistical Methods Preliminary inspections of the metabolomic data allowed the detection of a total of 2,124 missing observations (8.786% of the values in the dataset). Because an excessive number of missing values for some of the measured compounds could potentially introduce unanticipated biases, metabolites with ≥50% of missing values within either of the experimental groups (n = 9) were excluded from any further statistical testing (Supplementary Table 1,Supplementary Figure 1), and the remaining metabolites (n = 93) were included in the further analyses. After appropriate transformation and baseline standardization, each metabolite was subsequently entered into a generalized estimating equations model to investigate differences between groups in their respective values at baseline and after tDCS or sham stimulations were applied. The p-value threshold for statistical significancewassetatp≤5.376e−04, as the complexity of the analyses developed required accounting for both the family-wise error rate (Bonferroni method) and the intrinsic correlated nature of the metabolite data. Further details of the models implemented can be found in the Supplementary Methods and Supplementary Figure 2 These main statistical analyses were subsequently supplemented with a set of power-related computations utilizing statistical simulations within a sophisticated computer cluster environment. Firstly, we evaluated the sensitivity obtained with the current sample in its ability to detect significant longitudinal differences between groups with ≥80% power (for 0.05 and 5.376e−04 type-I error rates). This would allow an estimation of the minimum level of detection with the current experimental set-up. Subsequently, for each metabolite, we evaluated the sample size that would be required to detect the variation in metabolites caused by tDCS in this study as statistically significant with ≥80% power (type-I error rates 0.05 and 5.376e−04).Thiswasdoneinordertomakean estimate of the sample size needed in future experimental settings aiming to detect similar changes in metabolites resulting from tDCS (assuming that the detected values in this experiment would apply to other healthy populations with similar characteristics). Further details of these computations are offered in the Supplementary Methods. Lastly, the success of the blinding procedure was investigated by running a set of Mann–Whitney U-tests for both the participants and the experimenters on days one and five. They were asked how likely they thought it was, as a percentage, that they were part of the sham group, and the answers were used in the analysis. RESULTS The current double-blind, randomized controlled trial failed to detect any metabolic changes due to tDCS after five treatment sessions; none of the models implemented for any of the investigated metabolites displayed statistically significant coefficients. Figure 2 presents the model coefficients obtained for each metabolite, while the specific values provided by the statistical tests can be found in Table 1. The computations related to statistical power made it possible to draw relevant conclusions concerning the usability of the current data. Firstly, the conducted analyses suggested that our study sample conferred enough power (80% level) to detect relatively small differences between groups due to tDCS (an absolute value for the “Time x Group”model coefficient of ≥0.1493 would be detectable with a type-I error rate of 5.376 · 10 −4 ). However, the majority of the tDCS effects observed here were substantially smaller than that threshold, and only a single metabolite, 2-aminoisobutyric acid, surpassed it. However, this metabolite was not ultimately considered as significant, because the model estimates displayed high uncertainties (for instance, see the extremely large standard error for the corresponding coefficients in Table 1). Secondly, future replication studies aiming to detect significant group differences due to tDCS will need to increase their sample size considerably, as the treatment effects detected in our data were extremely small. For example, increasing the sample size to n = 150 participants per group (i.e., n = 300 in total) would have made it possible to identify six metabolites as displaying statistically significant differences between the study groups: 2-aminoisobutyric acid and propionylcarnitine being upregulated and hippuric acid, L-glutamine, xanthine and gamma-aminobutyric acid being downregulated in the tDCS group. Figures 3 and 4provide more details on this sample size estimation. Detailed results from the calculations are available in Supplementary Table 2. Finally, it should be acknowledged here that the blinding protocol was not totally successful, as a larger number of participants and experimenters on day 1 compared to day 5 successfully guessed that they belonged to the sham group (p = 0.041 for participants, 0.012 for experimenters). DISCUSSION This investigation revealed that five daily sessions of tDCS applied to healthy male adults exerted no significant impact on serum metabolite levels when compared to sham stimulation. Although our study was adequately powered to detect relatively small effect sizes (more specifically, ≥0.1493), tDCS produced either no or extremely small effects for the majority of the metabolites evaluated. This study further demonstrated that in order to detect the observed minuscule effects as statistically significant, the sample sizes would need to be extremely large, making such a hypothetical study unfeasible. In principle, tDCS could affect peripheral metabolite levels via two mechanisms: either indirectly through changes in the central nervous system that would modify peripheral neural responses and consequently peripheral metabolism, or directly through central metabolites passing through the BBB. While tDCS does not appear to lead to long-lasting changes in BBB permeability (20), short-lasting functional changes in its permeability could evoke changes in metabolite levels in the peripheral circulation. Our findings did not support a clinically significant presence of such changes, and the above hypotheses remain to be further tested in animal or human models. Peripheral metabolic effects and serum effects caused by tDCS have previously been investigated in a limited number of Kortteenniemi et al. Peripheral Metabolic Changes of tDCS Frontiers in Psychiatry | www.frontiersin.org May 2020 | Volume 11 | Article 4034 studies.(10–13,21–23) In these studies, one session of tDCS was found to cause changes in glucose metabolism (10,21) and cortisol secretion (11–13,23) when investigating healthy individuals. The sample sizes of the glucose studies were 15 (sham-controlled cross-over design), and 9 (within-subjects design, no sham, measurements before and after stimulation), respectively, while the sample sizes of the cortisol studies ranged from 20 to 60. Interestingly, one of the studies indicated that the hormone-related effect of tDCS varies based on pre-existing conditions. Subjects with high math anxiety experienced reduced serum cortisol with tDCS compared to sham stimulation, while receiving tDCS was associated with no change in serum cortisol in subjects with low math anxiety (13). In addition, Khedr et al. (22), in a study with 40 participants, observed that 10 consecutive sessions of tDCS lowered pain, a finding that correlated with an increase in serum endorphin levels FIGURE 2 | Coefficient values with 95% confidence intervals from each metabolite model analysis. Kortteenniemi et al. Peripheral Metabolic Changes of tDCS Frontiers in Psychiatry | www.frontiersin.org May 2020 | Volume 11 | Article 4035 TABLE 1 | Descriptive statistics from the pre-processed data, with main results from the models evaluated for each metabolite. Metabolite Name Baseline Post session 1 Post session 5 Coefficient for “Time”** Coefficient for “Group”** ^ Coefficient for “Time × Group”** tDCS Group * Sham Group * tDCS Group * Sham Group * tDCS Group * Sham Group * 2-Aminoadipic acid 39; 0.012 (0.166) 39; −0.012 (0.156) 40; 0.092 (0.153) 39; 0.062 (0.177) 39; 0.228 (0.136) 39; −0.106 (0.151) −0.018 (0.027); 0.434; 0.51 0.001 (0.193); 0; 0.994 0.073 (0.041); 3.113; 0.078 2-Aminoisobutyric acid 40; 0.127 (0.129) 39; −0.13 (0.185) 40; 0.137 (0.131) 39; −0.065 (0.18) 40; 0.283 (0.141) 39; 0.217 (0.156) 0.092 (0.726); 0.016; 0.899 0.629 (3.6); 0.031; 0.861 −0.172 (1.137); 0.023; 0.88 3-Hydroxyanthranilic acid 40; 0.055 (0.151) 39; −0.056 (0.169) 40; −0.051 (0.128) 39; −0.139 (0.168) 40; −0.082 (0.153) 39; −0.092 (0.143) −0.011 (0.025); 0.206; 0.65 0.082 (0.206); 0.158; 0.691 −0.012 (0.036); 0.11; 0.74 4-Pyridoxic acid 40; −0.044 (0.121) 39; 0.045 (0.192) 40; -0.067 (0.118) 39; 0.034 (0.186) 40; −0.05 (0.16) 39; 0.214 (0.19) 0.037 (0.019); 3.726; 0.054 −0.091 (0.229); 0.158; 0.691 −0.03 (0.029); 1.032; 0.31 Acetoacetic acid 40; 0.005 (0.154) 39; −0.005 (0.167) 40; −0.044 (0.158) 39; −0.193 (0.174) 40; −0.223 (0.18) 39; −0.148 (0.163) −0.012 (0.018); 0.413; 0.52 0.075 (0.228); 0.107; 0.744 −0.031 (0.028); 1.207; 0.272 Adenine 40; −0.021 (0.176) 39; 0.022 (0.142) 40; −0.174 (0.132) 39; 0.007 (0.195) 40; 0.04 (0.135) 39; −0.044 (0.134) −0.012 (0.027); 0.205; 0.651 −0.103 (0.223); 0.215; 0.643 0.029 (0.042); 0.46; 0.498 Adenosine 40; −0.066 (0.164) 39; 0.067 (0.155) 40; 0.252 (0.163) 39; 0.225 (0.177) 40; 0.057 (0.186) 39; 0.378 (0.167) 0.054 (0.024); 4.931; 0.026 −0.03 (0.223); 0.018; 0.893 −0.061 (0.044); 1.963; 0.161 Adenosine monophosphate 40; 0.04 (0.142) 39; −0.041 (0.177) 40; 0.101 (0.138) 39; 0.349 (0.145) 40; 0.251 (0.123) 39; 0.211 (0.14) 0.044 (0.027); 2.721; 0.099 −0.082 (0.205); 0.161; 0.688 0.005 (0.038); 0.017; 0.895 Allantoin 40; 0.038 (0.173) 39; −0.039 (0.144) 40; −0.125 (0.174) 39; −0.103 (0.154) 40; −0.312 (0.177) 39; −0.177 (0.154) −0.017 (0.025); 0.439; 0.507 0.067 (0.228); 0.086; 0.769 −0.049 (0.042); 1.348; 0.246 alpha-D-glucuronic acid 40; −0.016 (0.165) 39; 0.016 (0.155) 40; −0.176 (0.158) 39; −0.071 (0.172) 40; −0.054 (0.15) 39; −0.132 (0.142) -0.028 (0.021); 1.87; 0.172 −0.073 (0.226); 0.103; 0.748 0.032 (0.034); 0.879; 0.348 Asymmetric dimethylarginine 40; −0.145 (0.16) 39; 0.149 (0.157) 40; 0.171 (0.135) 39; 0.062 (0.134) 40; 0.034 (0.132) 39; 0.236 (0.154) 0.022 (0.041); 0.282; 0.595 −0.13 (0.176); 0.547; 0.459 0.012 (0.051); 0.055; 0.815 Betaine 40; 0.008 (0.16) 39; −0.008 (0.16) 40; −0.083 (0.153) 39; 0.048 (0.133) 40; 0.01 (0.176) 39; 0.003 (0.119) 0.005 (0.027); 0.036; 0.85 −0.04 (0.212); 0.036; 0.849 0.003 (0.044); 0.004; 0.95 Carnosine 40; 0.114 (0.153) 39; −0.116 (0.165) 40; 0.05 (0.143) 39; 0.061 (0.134) 40; 0.064 (0.129) 39; −0.004 (0.115) 0.021 (0.025); 0.752; 0.386 0.177 (0.217); 0.662; 0.416 −0.028 (0.033); 0.711; 0.399 Chenodeoxycholic acid 40; 0.046 (0.143) 39; −0.047 (0.176) 40; 0.026 (0.138) 39; −0.14 (0.167) 40; −0.286 (0.123) 39; −0.133 (0.186) −0.013 (0.023); 0.321; 0.571 0.159 (0.224); 0.501; 0.479 −0.061 (0.034); 3.145; 0.076 Cholic acid 30; 0.08 (0.186) 28; −0.086 (0.188) 26; −0.159 (0.209) 25; −0.175 (0.161) 27; 0.002 (0.194) 23; −0.387 (0.202) −0.018 (0.047); 0.151; 0.697 0.252 (0.27); 0.867; 0.352 −0.004 (0.072); 0.003; 0.955 Choline 40; 0.03 (0.148) 39; −0.031 (0.171) 40; −0.024 (0.175) 39; −0.044 (0.157) 40; −0.121 (0.158) 39; 0.05 (0.197) 0.019 (0.036); 0.287; 0.592 0.006 (0.223); 0.001; 0.979 −0.038 (0.05); 0.578; 0.447 Citrulline 40; 0.035 (0.166) 39; −0.036 (0.154) 40; −0.127 (0.152) 39; −0.163 (0.151) 40; −0.229 (0.182) 39; −0.164 (0.18) −0.014 (0.026); 0.284; 0.594 0.081 (0.217); 0.138; 0.71 −0.036 (0.034); 1.085; 0.298 Creatine 40; 0.302 (0.151) 39; −0.309 (0.154) 40; 0.308 (0.146) 39; −0.074 (0.133) 40; 0.229 (0.141) 39; −0.051 (0.137) 0.057 (0.019); 8.719; 0.003 0.516 (0.21); 6.014; 0.014 −0.072 (0.028); 6.578; 0.01 Creatinine 40; −0.087 (0.154) 39; 0.089 (0.165) 40; −0.098 (0.149) 39; −0.024 (0.157) 40; −0.026 (0.164) 39; 0.076 (0.176) 0.005 (0.015); 0.098; 0.754 −0.118 (0.218); 0.293; 0.588 0.008 (0.022); 0.129; 0.72 Cyclic AMP 40; 0.129 (0.174) 39; −0.133 (0.141) 40; 0.086 (0.141) 39; −0.054 (0.14) 40; 0.043 (0.135) 39; 0.021 (0.141) 0.02 (0.023); 0.77; 0.38 0.184 (0.188); 0.956; 0.328 −0.044 (0.045); 0.967; 0.326 Cytidine 40; 0.038 (0.14) 39; −0.039 (0.179) 40; −0.013 (0.133) 39; −0.054 (0.177) 40; −0.045 (0.167) 39; 0.066 (0.195) 0.016 (0.035); 0.207; 0.649 0.089 (0.228); 0.153; 0.696 −0.04 (0.046); 0.762; 0.383 Cytosine 40; −0.013 (0.18) 39; 0.013 (0.137) 40; −0.079 (0.16) 39; 0.015 (0.139) 40; −0.304 (0.16) 39; −0.232 (0.14) −0.04 (0.026); 2.348; 0.125 −0.033 (0.232); 0.021; 0.886 −0.012 (0.042); 0.084; 0.772 D-Ribose-5-phosphate 40; 0.053 (0.162) 39; −0.054 (0.157) 40; 0.038 (0.147) 39; 0.02 (0.144) 40; 0.129 (0.128) 39; −0.023 (0.14) 0.02 (0.027); 0.555; 0.456 0.09 (0.22); 0.17; 0.68 −0.002 (0.042); 0.002; 0.969 Decanoylcarnitine 40; 0.04 (0.156) 39; −0.041 (0.165) 40; −0.105 (0.18) 39; −0.401 (0.18) 40; −0.36 (0.173) 39; −0.287 (0.112) 0.005 (0.039); 0.015; 0.902 0.061 (0.226); 0.072; 0.788 −0.069 (0.051); 1.878; 0.171 Deoxycytidine 40; 0.018 (0.128) 39; −0.018 (0.188) 40; −0.036 (0.123) 39; −0.127 (0.184) 40; −0.267 (0.136) 39; −0.044 (0.177) −0.01 (0.014); 0.528; 0.468 0.057 (0.227); 0.064; 0.8 −0.051 (0.025); 4.151; 0.042 Deoxyuridine 40; −0.02 (0.151) 39; 0.02 (0.169) 40; 0.058 (0.153) 39; −0.025 (0.17) 40; 0.031 (0.139) 39; 0.121 (0.189) 0.026 (0.02); 1.6; 0.206 −0.026 (0.225); 0.013; 0.908 −0.018 (0.029); 0.372; 0.542 Dimethylglycine 40; −0.106 (0.145) 39; 0.109 (0.173) 40; −0.129 (0.145) 39; -0.007 (0.162) 40; −0.089 (0.13) 39; 0.112 (0.169) −0.061 (0.031); 3.848; 0.05 −0.311 (0.24); 1.673; 0.196 0.053 (0.037); 2.086; 0.149 Folic acid 40; 0.063 (0.175) 39; −0.064 (0.142) 40; 0.181 (0.16) 39; 0.016 (0.134) 40; −0.053 (0.184) 39; −0.169 (0.148) −0.034 (0.026); 1.706; 0.191 0.083 (0.21); 0.156; 0.693 −0.006 (0.037); 0.025; 0.875 Gamma-Aminobutyric acid 40; −0.116 (0.163) 39; 0.119 (0.155) 40; 0.004 (0.155) 39; −0.102 (0.149) 40; 0.184 (0.148) 39; −0.18 (0.129) −0.05 (0.032); 2.479; 0.115 −0.112 (0.183); 0.372; 0.542 0.105 (0.051); 4.207; 0.04 Gamma-Glutamylcysteine 38; −0.05 (0.169) 34; 0.055 (0.165) 37; −0.205 (0.133) 38; 0.035 (0.175) 38; 0.014 (0.159) 34; −0.005 (0.145) −0.014 (0.019); 0.505; 0.477 −0.188 (0.218); 0.743; 0.389 0.042 (0.029); 2.127; 0.145 Glyceraldehyde 40; 0.167 (0.143) 39; −0.171 (0.172) 40; −0.089 (0.135) 39; −0.271 (0.163) 40; 0.089 (0.139) 39; −0.243 (0.154) −0.007 (0.029); 0.064; 0.8 0.298 (0.209); 2.026; 0.155 0.001 (0.045); 0; 0.991 Glycine 40; −0.296 (0.169) 39; 0.304 (0.133) 40; −0.319 (0.177) 39; 0.36 (0.143) 40; −0.304 (0.183) 39; 0.094 (0.125) −0.039 (0.018); 4.72; 0.03 −0.685 (0.216); 10.095; 0.001 0.033 (0.029); 1.291; 0.256 Glycocholic acid 40; 0.011 (0.165) 39; −0.012 (0.155) 40; −0.31 (0.172) 39; −0.465 (0.164) 40; −0.61 (0.163) 39; −0.467 (0.153) −0.092 (0.036); 6.521; 0.011 0.067 (0.233); 0.082; 0.775 −0.039 (0.048); 0.681; 0.409 Guanidoacetic acid 40; 0.028 (0.149) 39; −0.029 (0.171) 40; −0.159 (0.144) 39; −0.209 (0.169) 40; 0.005 (0.2) 39; −0.216 (0.187) −0.021 (0.023); 0.838; 0.36 0.071 (0.223); 0.102; 0.749 0.025 (0.037); 0.439; 0.508 Guanosine 40; −0.068 (0.159) 39; 0.07 (0.16) 40; 0.179 (0.133) 39; 0.023 (0.161) 40; −0.064 (0.137) 39; 0.06 (0.161) 0 (0.022); 0; 0.994 −0.014 (0.198); 0.005; 0.943 −0.026 (0.031); 0.681; 0.409 Hexanoylcarnitine 40; 0.133 (0.162) 39; −0.136 (0.155) 40; −0.165 (0.161) 39; −0.496 (0.165) 40; −0.251 (0.177) 39; −0.45 (0.127) −0.067 (0.031); 4.866; 0.027 0.253 (0.227); 1.235; 0.266 −0.002 (0.044); 0.002; 0.961 Hippuric acid 40; −0.254 (0.145) 39; 0.261 (0.164) 40; −0.451 (0.135) 39; 0.039 (0.159) 40; 0.003 (0.192) 39; 0.116 (0.167) −0.029 (0.034); 0.718; 0.397 −0.515 (0.224); 5.288; 0.021 0.083 (0.045); 3.437; 0.064 Homocysteine 40; -0.08 (0.15) 39; 0.082 (0.169) 40; −0.094 (0.156) 39; 0.088 (0.168) 40; −0.118 (0.146) 39; 0.125 (0.158) 0.017 (0.035); 0.235; 0.628 −0.171 (0.25); 0.468; 0.494 −0.029 (0.047); 0.387; 0.534 Homogentisic acid 40; 0.144 (0.165) 39; −0.147 (0.151) 40; −0.06 (0.135) 38; −0.261 (0.183) 40; −0.131 (0.131) 39; −0.102 (0.137) 0.02 (0.022); 0.865; 0.352 0.281 (0.215); 1.705; 0.192 −0.066 (0.034); 3.668; 0.055 Hydroxykynurenine 40; 0.053 (0.158) 39; −0.055 (0.161) 40; −0.004 (0.128) 39; −0.043 (0.171) 40; −0.092 (0.154) 39; 0.143 (0.156) 0.038 (0.025); 2.291; 0.13 0.058 (0.213); 0.074; 0.785 −0.069 (0.037); 3.59; 0.058 Hypoxanthine 40; 0.135 (0.155) 39; −0.138 (0.163) 40; −0.129 (0.117) 39; −0.02 (0.141) 40; 0.041 (0.131) 39; −0.181 (0.119) −0.005 (0.033); 0.022; 0.883 0.136 (0.214); 0.404; 0.525 −0.02 (0.046); 0.19; 0.663 Inosine 40; −0.109 (0.159) 39; 0.112 (0.159) 40; 0.154 (0.12) 39; 0.083 (0.155) 40; 0.062 (0.148) 39; 0.055 (0.16) −0.011 (0.028); 0.155; 0.694 −0.133 (0.198); 0.451; 0.502 0.028 (0.039); 0.526; 0.468 Inosinic acid 40; 0.008 (0.17) 39; −0.008 (0.149) 40; −0.085 (0.171) 39; 0.136 (0.161) 40; −0.113 (0.174) 39; −0.159 (0.159) −0.046 (0.022); 4.373; 0.037 −0.128 (0.218); 0.344; 0.558 0.032 (0.04); 0.641; 0.423 Isobutyryl-L-carnitine 40; −0.09 (0.145) 39; 0.092 (0.174) 40; −0.285 (0.134) 39; −0.131 (0.165) 40; −0.555 (0.139) 39; −0.192 (0.175) −0.067 (0.022); 9.072; 0.003 −0.102 (0.225); 0.206; 0.65 −0.042 (0.031); 1.906; 0.167 Isovalerylcarnitine 40; 0.035 (0.134) 39; −0.035 (0.183) 40; −0.001 (0.128) 39; −0.214 (0.176) 40; −0.385 (0.145) 39; −0.147 (0.154) −0.025 (0.03); 0.698; 0.403 0.073 (0.219); 0.112; 0.738 −0.059 (0.041); 2.097; 0.148 Kynurenic acid 40; 0.057 (0.137) 39; −0.059 (0.181) 40; −0.065 (0.133) 39; −0.23 (0.167) 40; 0.118 (0.155) 39; −0.066 (0.175) 0.013 (0.027); 0.236; 0.627 0.133 (0.225); 0.351; 0.554 0.011 (0.038); 0.079; 0.779 L-Acetylcarnitine 40; 0.049 (0.154) 39; −0.05 (0.166) 40; −0.027 (0.171) 39; −0.111 (0.168) 40; 0.044 (0.2) 39; −0.009 (0.16) 0.004 (0.044); 0.007; 0.931 0.129 (0.227); 0.322; 0.57 −0.009 (0.058); 0.023; 0.88 (Continued) Kortteenniemi et al. Peripheral Metabolic Changes of tDCS Frontiers in Psychiatry | www.frontiersin.org May 2020 | Volume 11 | Article 4036 TABLE 1 | Continued Metabolite Name Baseline Post session 1 Post session 5 Coefficient for “Time”** Coefficient for “Group”** ^ Coefficient for “Time × Group”** tDCS Group * Sham Group * tDCS Group * Sham Group * tDCS Group * Sham Group * L-Alanine 40; −0.119 (0.166) 39; 0.122 (0.151) 40; −0.152 (0.184) 39; 0.02 (0.142) 40; −0.485 (0.168) 39; −0.373 (0.149) −0.093 (0.028); 10.741; 0.001 −0.262 (0.23); 1.298; 0.255 0.029 (0.039); 0.569; 0.451 L-Arginine 40; 0.023 (0.153) 39; −0.024 (0.167) 40; 0.054 (0.149) 39; −0.008 (0.151) 40; 0.225 (0.148) 39; 0.097 (0.172) 0.022 (0.019); 1.408; 0.235 0.025 (0.226); 0.012; 0.912 0.016 (0.03); 0.277; 0.599 L-Asparagine 40; −0.08 (0.166) 39; 0.082 (0.152) 40; −0.129 (0.181) 39; 0.023 (0.155) 40; 0.13 (0.167) 39; 0.092 (0.161) 0.013 (0.025); 0.261; 0.61 −0.17 (0.226); 0.564; 0.453 0.046 (0.032); 2.072; 0.15 L-Aspartic acid 40; 0.107 (0.165) 39; −0.109 (0.153) 40; −0.375 (0.143) 39; −0.205 (0.163) 40; −0.091 (0.175) 39; −0.333 (0.157) −0.038 (0.031); 1.457; 0.227 0.065 (0.219); 0.087; 0.767 0.025 (0.051); 0.241; 0.624 L-Carnitine 40; −0.027 (0.16) 39; 0.028 (0.16) 40; −0.065 (0.163) 39; 0.059 (0.154) 40; −0.128 (0.17) 39; 0.159 (0.154) 0.026 (0.014); 3.701; 0.054 −0.113 (0.227); 0.248; 0.618 −0.049 (0.02); 5.801; 0.016 L-Cystathionine 40; -0.184 (0.13) 39; 0.189 (0.181) 40; −0.378 (0.122) 39; 0.058 (0.187) 40; −0.457 (0.117) 39; 0.09 (0.175) −0.027 (0.023); 1.368; 0.242 −0.384 (0.22); 3.037; 0.081 −0.039 (0.035); 1.229; 0.268 L-Glutamic acid 40; 0.159 (0.171) 39; −0.163 (0.143) 40; 0.062 (0.136) 39; −0.033 (0.152) 40; 0.053 (0.126) 39; −0.198 (0.131) 0.017 (0.022); 0.564; 0.453 0.292 (0.218); 1.788; 0.181 −0.047 (0.036); 1.776; 0.183 L-Glutamine 40; −0.089 (0.15) 39; 0.091 (0.169) 40; −0.04 (0.148) 39; 0.136 (0.17) 40; 0.064 (0.136) 39; 0.095 (0.153) −0.083 (0.088); 0.896; 0.344 −0.353 (0.473); 0.558; 0.455 0.088 (0.121); 0.53; 0.467 L-Histidine 40; 0.006 (0.161) 39; −0.006 (0.159) 40; 0.013 (0.144) 39; −0.036 (0.16) 40; 0.155 (0.164) 39; 0.104 (0.176) 0.026 (0.025); 1.094; 0.296 −0.009 (0.221); 0.002; 0.968 0.004 (0.035); 0.015; 0.903 L-Homoserine 40; −0.017 (0.167) 39; 0.017 (0.152) 40; −0.243 (0.156) 39; −0.11 (0.145) 40; −0.319 (0.155) 39; −0.604 (0.148) −0.116 (0.028); 17.574; 0 −0.081 (0.201); 0.161; 0.688 0.072 (0.047); 2.341; 0.126 L-Isoleucine 40; 0.043 (0.161) 39; −0.044 (0.158) 40; −0.162 (0.133) 39; −0.154 (0.171) 40; −0.13 (0.144) 39; −0.122 (0.155) −0.008 (0.014); 0.29; 0.59 0.048 (0.223); 0.047; 0.828 −0.02 (0.027); 0.516; 0.473 L-Kynurenine 40; −0.082 (0.148) 39; 0.084 (0.171) 40; −0.128 (0.138) 39; 0.101 (0.163) 40; 0.159 (0.159) 39; 0.252 (0.191) 0.035 (0.032); 1.198; 0.274 −0.255 (0.221); 1.33; 0.249 0.024 (0.043); 0.307; 0.58 L-Leucine 40; 0.051 (0.151) 39; −0.052 (0.169) 40; −0.102 (0.167) 39; −0.24 (0.161) 40; 0.05 (0.169) 39; 0.049 (0.188) 0.042 (0.016); 6.525; 0.011 0.117 (0.226); 0.266; 0.606 −0.03 (0.028); 1.136; 0.286 L-Lysine 40; 0.184 (0.152) 39; −0.188 (0.163) 40; 0.266 (0.162) 39; −0.151 (0.159) 40; 0.236 (0.159) 39; 0.061 (0.202) 0.062 (0.029); 4.611; 0.032 0.349 (0.226); 2.375; 0.123 −0.053 (0.042); 1.572; 0.21 L-Methionine 40; 0.015 (0.154) 39; −0.016 (0.166) 40; −0.182 (0.151) 39; −0.086 (0.169) 40; −0.103 (0.16) 39; −0.224 (0.148) −0.039 (0.022); 3.154; 0.076 −0.037 (0.227); 0.027; 0.87 0.012 (0.035); 0.118; 0.731 L-Octanoylcarnitine 40; 0.049 (0.162) 39; −0.05 (0.158) 40; −0.032 (0.187) 39; −0.292 (0.163) 40; −0.236 (0.179) 39; −0.237 (0.115) −0.012 (0.033); 0.131; 0.717 0.095 (0.226); 0.176; 0.675 −0.037 (0.043); 0.768; 0.381 L-Ornithine 40; 0.051 (0.161) 39; −0.052 (0.158) 40; −0.152 (0.167) 39; −0.094 (0.168) 40; −0.161 (0.169) 39; −0.086 (0.155) −0.005 (0.019); 0.059; 0.809 0.032 (0.23); 0.019; 0.889 −0.019 (0.031); 0.373; 0.541 L-Phenylalanine 40; 0.061 (0.167) 39; −0.062 (0.151) 40; −0.129 (0.145) 39; −0.225 (0.145) 40; 0.011 (0.153) 39; −0.201 (0.14) −0.046 (0.014); 10.562; 0.001 0.088 (0.228); 0.15; 0.698 0.013 (0.03); 0.194; 0.66 L-Proline 40; 0.162 (0.171) 39; −0.166 (0.143) 40; 0.06 (0.169) 39; −0.163 (0.132) 40; −0.212 (0.186) 39; −0.355 (0.153) −0.03 (0.031); 0.936; 0.333 0.33 (0.228); 2.091; 0.148 −0.05 (0.041); 1.504; 0.22 L-Serine 40; −0.006 (0.129) 39; 0.006 (0.188) 40; −0.056 (0.127) 39; 0.168 (0.194) 40; 0.093 (0.145) 39; 0.233 (0.18) 0.037 (0.026); 2.04; 0.153 −0.136 (0.226); 0.363; 0.547 −0.016 (0.033); 0.23; 0.632 L-Threonine 40; 0.004 (0.149) 39; −0.004 (0.171) 40; −0.098 (0.143) 39; −0.047 (0.169) 40; −0.241 (0.189) 39; −0.443 (0.137) −0.081 (0.026); 10.062; 0.002 −0.022 (0.221); 0.01; 0.922 0.026 (0.042); 0.375; 0.54 L-Tryptophan 40; −0.009 (0.138) 39; 0.009 (0.18) 40; −0.272 (0.145) 39; −0.262 (0.174) 40; −0.179 (0.154) 39; −0.298 (0.151) −0.078 (0.029); 7.173; 0.007 −0.083 (0.228); 0.134; 0.714 0.036 (0.038); 0.871; 0.351 L-Tyrosine 40; 0.035 (0.153) 39; −0.036 (0.167) 40; −0.225 (0.16) 39; −0.226 (0.162) 40; −0.323 (0.155) 39; −0.216 (0.187) −0.032 (0.018); 3.203; 0.073 0.035 (0.23); 0.023; 0.879 −0.045 (0.028); 2.544; 0.111 L-Valine 40; 0.148 (0.138) 39; −0.152 (0.177) 40; −0.008 (0.145) 39; −0.282 (0.164) 40; 0.025 (0.168) 39; −0.096 (0.169) 0.022 (0.025); 0.745; 0.388 0.312 (0.225); 1.926; 0.165 −0.04 (0.037); 1.185; 0.276 Myo-inositol 40; −0.009 (0.156) 39; 0.01 (0.164) 40; −0.171 (0.142) 39; 0.024 (0.196) 40; 0.181 (0.16) 39; 0.09 (0.163) 0.024 (0.027); 0.833; 0.361 −0.133 (0.227); 0.345; 0.557 0.031 (0.043); 0.517; 0.472 N(tele)-methylhistamine 40; −0.108 (0.158) 39; 0.111 (0.161) 40; −0.047 (0.148) 39; 0.175 (0.15) 40; 0.016 (0.152) 39; 0.094 (0.162) −0.004 (0.013); 0.072; 0.788 −0.349 (0.214); 2.66; 0.103 0.032 (0.023); 1.863; 0.172 NAD 40; −0.08 (0.168) 39; 0.082 (0.15) 40; −0.056 (0.135) 39; 0.135 (0.177) 40; −0.08 (0.133) 39; 0.022 (0.156) −0.019 (0.025); 0.593; 0.441 −0.222 (0.205); 1.178; 0.278 0.015 (0.042); 0.134; 0.714 Niacinamide 40; 0.23 (0.143) 39; −0.236 (0.168) 40; 0.929 (0.189) 39; 0.394 (0.147) 40; 0.377 (0.162) 39; −0.08 (0.175) 0.045 (0.037); 1.425; 0.233 0.463 (0.226); 4.194; 0.041 −0.003 (0.052); 0.003; 0.958 Normetanephrine 40; 0.053 (0.176) 39; −0.054 (0.142) 40; 0.287 (0.145) 39; 0.376 (0.177) 40; 0.22 (0.144) 39; 0.301 (0.148) 0.065 (0.023); 7.89; 0.005 0.026 (0.221); 0.013; 0.908 −0.032 (0.039); 0.672; 0.412 O-Phosphoethanolamine 40; 0.002 (0.136) 39; −0.002 (0.182) 40; −0.053 (0.138) 39; 0.079 (0.122) 40; −0.049 (0.143) 39; 0.075 (0.15) 0.018 (0.036); 0.242; 0.623 −0.072 (0.198); 0.133; 0.716 −0.029 (0.041); 0.508; 0.476 Orotic acid 40; 0.087 (0.155) 39; −0.089 (0.164) 40; −0.139 (0.187) 39; 0.256 (0.219) 40; 0.026 (0.158) 39; 0.045 (0.161) −0.007 (0.029); 0.058; 0.809 −0.117 (0.228); 0.264; 0.607 0.019 (0.039); 0.237; 0.626 Pantothenic acid 40; 0.058 (0.153) 39; −0.059 (0.166) 40; −0.074 (0.151) 39; −0.054 (0.191) 40; 0.011 (0.157) 39; −0.123 (0.193) −0.018 (0.023); 0.652; 0.419 0.029 (0.227); 0.016; 0.899 0.023 (0.033); 0.499; 0.48 Propionylcarnitine 40; 0.051 (0.138) 39; −0.052 (0.18) 40; −0.168 (0.141) 39; −0.235 (0.174) 40; −0.404 (0.142) 39; −0.103 (0.18) −0.027 (0.017); 2.461; 0.117 0.09 (0.231); 0.153; 0.696 −0.085 (0.026); 10.381; 0.001 Sorbitol 40; −0.02 (0.184) 39; 0.02 (0.13) 40; −0.199 (0.187) 39; −0.012 (0.133) 40; 0.028 (0.188) 39; −0.174 (0.142) −0.032 (0.017); 3.476; 0.062 −0.095 (0.228); 0.173; 0.677 0.036 (0.037); 0.962; 0.327 Spermidine 40; −0.156 (0.173) 39; 0.16 (0.14) 40; −0.033 (0.171) 39; 0.47 (0.181) 40; 0.378 (0.191) 39; 0.562 (0.162) 0.069 (0.024); 8.464; 0.004 −0.441 (0.229); 3.722; 0.054 0.045 (0.035); 1.636; 0.201 Succinic acid 40; −0.057 (0.157) 39; 0.058 (0.163) 40; −0.282 (0.158) 39; −0.253 (0.152) 40; −0.27 (0.141) 39; −0.121 (0.171) −0.015 (0.022); 0.47; 0.493 −0.038 (0.218); 0.03; 0.863 −0.014 (0.032); 0.198; 0.656 Sucrose 40; 0.054 (0.156) 39; −0.056 (0.164) 40; −0.049 (0.16) 39; 0.088 (0.134) 40; 0.068 (0.147) 39; −0.039 (0.144) 0.003 (0.02); 0.022; 0.882 0.029 (0.206); 0.02; 0.889 0.004 (0.03); 0.021; 0.885 Symmetric dimethylarginine 40; −0.224 (0.161) 39; 0.23 (0.15) 40; −0.169 (0.154) 39; 0.113 (0.148) 40; 0.201 (0.154) 39; 0.378 (0.134) 0.037 (0.026); 2.134; 0.144 −0.481 (0.22); 4.752; 0.029 0.06 (0.04); 2.222; 0.136 Taurine 40; 0.102 (0.158) 39; −0.104 (0.16) 40; −0.285 (0.138) 39; −0.114 (0.155) 40; −0.057 (0.161) 39; −0.079 (0.159) 0.014 (0.04); 0.129; 0.72 0.126 (0.218); 0.334; 0.563 −0.039 (0.052); 0.575; 0.448 Taurochenodesoxycholic acid 40; 0.057 (0.171) 39; −0.059 (0.147) 40; −0.222 (0.158) 39; −0.401 (0.129) 40; −0.487 (0.131) 39; −0.332 (0.145) −0.078 (0.03); 6.61; 0.01 0.187 (0.223); 0.705; 0.401 −0.059 (0.042); 1.97; 0.16 Taurocholic acid 40; 0.02 (0.138) 39; −0.021 (0.18) 40; −0.064 (0.138) 39; −0.152 (0.17) 40; 0.011 (0.16) 39; −0.047 (0.186) −0.014 (0.035); 0.155; 0.694 0.064 (0.224); 0.083; 0.773 0.005 (0.047); 0.01; 0.919 trans-4-Hydroxy-L-proline 40; 0.012 (0.165) 39; −0.012 (0.154) 40; 0.01 (0.163) 39; 0.007 (0.152) 40; −0.301 (0.159) 39; −0.166 (0.165) 0.008 (0.033); 0.062; 0.803 0.006 (0.233); 0.001; 0.979 −0.032 (0.043); 0.555; 0.456 Trimethylamine N-oxide 40; −0.046 (0.144) 39; 0.047 (0.175) 40; −0.058 (0.139) 39; 0.099 (0.181) 40; −0.131 (0.147) 39; 0.043 (0.178) 0.018 (0.03); 0.347; 0.556 −0.019 (0.223); 0.007; 0.931 −0.037 (0.038); 0.934; 0.334 Uracil 40; 0.038 (0.177) 39; −0.039 (0.141) 40; −0.165 (0.176) 39; −0.158 (0.154) 40; −0.011 (0.134) 39; −0.235 (0.155) −0.031 (0.031); 0.986; 0.321 0.058 (0.202); 0.083; 0.773 0.03 (0.044); 0.482; 0.488 Xanthine 40; 0.107 (0.155) 39; −0.11 (0.163) 40; −0.092 (0.145) 39; −0.092 (0.169) 40; 0.278 (0.158) 39; −0.296 (0.135) −0.039 (0.03); 1.726; 0.189 0.065 (0.212); 0.095; 0.758 0.103 (0.042); 5.941; 0.015 Xanthosine 40; 0.138 (0.153) 39; −0.141 (0.164) 40; 0.086 (0.166) 39; 0.023 (0.144) 40; 0.064 (0.177) 39; −0.155 (0.149) −0.003 (0.024); 0.018; 0.893 0.227 (0.209); 1.183; 0.277 −0.001 (0.044); 0; 0.989 *values represent: “N; Mean (SE)”;**values represent: “Estimate (SE); Wald statistic; P-value”;^“Group”coefficient specified in the model as variation in the tDCS group respect to the Sham. Kortteenniemi et al. Peripheral Metabolic Changes of tDCS Frontiers in Psychiatry | www.frontiersin.org May 2020 | Volume 11 | Article 4037 during the 10-session intervention. Brunoni et al. (24–26) investigated the effect of tDCS on circulating neurotrophins and their receptors (brain-derived neurotrophic factor (BDNF), neurotrophins 3 and 4 (NT-3 and NT-4), nerve growth factor, glial cell line derived neurotrophic factor (GDNF), and soluble tumour necrosis factor receptors 1 and 2), but found no difference between individuals receiving tDCS vs. sham stimulation. A few year later (27), the same group expanded their research to interleukins and tumour necrosis factor alpha, and again observed no differences observed between participants receiving tDCS vs. sham stimulation. In contrast, in a study by Hadoush et al. (28), belonging to the tDCS group vs. sham stimulation was associated with an increase in serum BDNF levels in Parkinson’s disease patients undergoing a 10-session tDCS intervention. The effects of tDCS on the concentrations of soluble neuronal cell adhesion molecules in minimally conscious subjects have also been studied; the authors observed no significant tDCS-related changes (29). Unfortunately, our metabolomics panel did not include any of the aforementioned metabolites, which limits comparisons with these previous observations. However, these studies (with the exception of glucose studies) focused more on larger molecules in peripheral blood in contrast to the small metabolic products we measured. Some earlier reports have suggested that tDCS may cause intracerebral changes in the concentrations of certain metabolites. For example, Dickler et al. (30) found an increase indorsolateralprefrontalcorticalGABAlevelsunderthe stimulated area in patients with a gambling disorder, while Hone-Blanchet et al. (31) recorded elevated striatal levels of Nacetylaspartate and Glx (a combined measurement of glutamate and glutamine), but no differences in the levels of GABA in healthy subjects. Although a previous study (10) did claim that tDCS-induced alterations in central metabolism are reflected in the peripheral circulation, we observed no tDCS-related alterations in the serum levels of N-acetylaspartate or GABA among healthy individuals. Nevertheless, the possibility of such alterations at the cerebral level cannot be ruled out; tDCS may have induced central nervous system alterations that are not reflected as meaningful alterations at the peripheral level. The main strengths of our study are the relatively large sample size (previous studies focusing on tDCS-induced possible peripheral metabolic changes have utilised samples ranging from 14 to 60 individuals) (13,32) and the implementation of a randomized, double-blind, controlled study design. Furthermore, in order to reduce potential confounding due to individual lifestyle factors, we provided the study participants with detailed instructions regarding lifestyle behaviours potentially modifying the effects of tDCS before each tDCS/sham stimulation session. All participants were instructed to fast before providing blood samples. Controlling for these factors most likely contributed to FIGURE 3 | Power curves for the detection of “Time × Group”coefficient values. The current sample size per experimental group confers ≥80% power to detect “Time Group”coefficients of ≥0.0976 (type-I error rate: 0.05) and ≥0.1493 (type-I error rate: 5.376 · 10 −4 ). Kortteenniemi et al. Peripheral Metabolic Changes of tDCS Frontiers in Psychiatry | www.frontiersin.org May 2020 | Volume 11 | Article 4038