844
Letters to the Editor / Brain Stimulation 8 (2015) 838e849
Bifrontal Anodal Transcranial Direct Current Stimulation (tDCS) Improves Daytime Vigilance and Sleepiness in a Patient With Organic Hypersomnia Following Reanimation Dear Editor: In this case report, we provide first and preliminary evidence that bifrontal anodal transcranial direct current stimulation (tDCS) can improve vigilance and reduce daytime sleepiness in a patient with organic hypersomnia following reanimation. Topdown models of sleep regulation propose that local activity changes of the frontal cortex modulate arousal and sleep via corticothalamo-cortical feedback loops [1]. Here, we used anodal tDCS to enhance the excitability of the frontal cortex [2] and to improve the clinical symptoms of organic hypersomnia, an adverse health condition resulting in a substantially reduced quality of life with limited treatment options [3]. Methods Patient We report a 52-year-old construction worker who sustained a severe allergic reaction with cardiac arrest and cardio-pulmonal reanimation following a bee sting 10 years prior to the study. In good mental and somatic health until then, he suffered from the de novo onset of excessive daytime sleepiness of up to several
hours per day in the absence of any other neurological deficits, establishing the diagnosis of organic hypersomnia (ICD-10: G47.1). He remained incapable of work. With a time lag of one year, he suffered from a first depressive episode with reduced mood and a loss of pleasure. Under antidepressant treatment, these symptoms returned to normal; the excessive daytime sleepiness persisted. In addition, a comorbid obstructive sleep apnea syndrome was diagnosed, which was continuously and sufficiently treated with automatic continuous positive airway pressure (auto-CPAP; apnoea hypopnea index, AHI 2.2/h, normal level). Prior to the study, extensive pharmacological treatments, including SSRI, venlafaxine, modafinil, and amantadine, as well as cognitive-behavioral therapy had not substantially improved reduced vigilance and excessive daytime sleepiness. During the study, the medication remained unchanged (venlafaxine 225 mg, modafinil 200 mg, amantadine 200 mg). The study has been approved by the local ethics committee and the patient has given written informed consent.
Study design The study comprised two phases (Fig. 1). In experimental phase I, the patient underwent three sessions of anodal stimulation and three sessions of sham stimulation in an alternating order on six consecutive days (single blind). Vigilance was investigated prior to and after the interventions using the psychomotor vigilance task (PVT, [4]). Means of the PVT outcome parameters response speed and response errors were calculated across the experimental days for each point in time of the assessment (pre- and post-stimulation) and each condition (anodal and sham stimulation) separately. Analyses of variance with the repeated measures factors Time (pre- vs. post-stimulation) and Condition (anodal vs. sham stimulation) were calculated for both outcome parameters. In observational phase II, two blocks comprising three sessions of anodal stimulation on three consecutive days were followed by one month of clinical self-monitoring after each tDCS block.
Figure 1. A1) Study protocol for phase I. Three sessions of anodal stimulation and three sessions of sham stimulation were applied in an alternating order on six consecutive days, with vigilance measured pre- and post-stimulation using the psychomotor vigilance task (PVT). A2) Mean reaction speed deteriorated after sham but improved after anodal tDCS. The number of reaction errors remained unchanged. ANOVA with the repeated measures factors Time (pre- vs. post-stimulation) and Condition (anodal vs. sham stimulation). Bars indicate the standard error of the mean. * P < .05. B1) Study protocol for phase II. Anodal stimulation was applied on two times three consecutive days followed by one month of clinical self-monitoring at each case. B2) Regression analysis showed a significant time-dependent improvement of subjective vigilance (regression line and linear trend of residuals given). B3) The duration of daytime sleep decreased over the reported time (descriptive data).
Letters to the Editor / Brain Stimulation 8 (2015) 838e849
Subjective vigilance (visual analog scale; 0e100, low to high) and the duration of daytime sleep (self-report) were assessed. tDCS protocol tDCS was delivered between 12.30 PM and 1.15 PM by a constant current stimulator (neuroConn, Germany) via two frontal stimulation electrodes (5 7 cm, FP1/FP2, 10e20 system) and two parietal reference electrodes (10 10 cm, P3/P4). To induce robust after-effects for the modulation of tonic arousal processes, an optimized repetitive stimulation protocol was selected with anodal stimulation in two blocks of 13 min with a 20 min inter-stimulation interval [5]. A constant current of 1 mA over each stimulation electrode was applied (2 mA stimulator output, Y-cable split for stimulation and reference electrodes) in a 30-s fade-in/fade-out design to decrease potential skin sensations [6]. For sham stimulation, the current turned off automatically after 15 s. Results Figure 1 visualizes the main findings. In phase I (Fig. 1A), analysis of variance of response speed revealed a significant effect for Time (pre- vs. post-stimulation, F ¼ 4.2, P < .05) and Condition (anodal vs. sham stimulation, F ¼ 42.2, P < .001) and a significant interaction effect (F ¼ 24.5, P < .01). Post-hoc analyses indicated that the response speed significantly deteriorated following sham (t ¼ 2.1, P < .05) and improved following anodal stimulation (t ¼ 4.9, P < .001) compared to pre-stimulation measurements. In addition, the response speed was significantly faster after anodal than after sham stimulation (t ¼ 8.0, P < .001), with similar pre-stimulation levels (t ¼ 1.1, P ¼ .29). The number of response errors did not differ between the conditions and time points of measurement (X2 ¼ .27, P ¼ .78). In phase II (Fig. 1B), a regression analysis of subjective ratings showed a significant increase in subjective vigilance over time (b ¼ .42, R2 ¼ .28, P < .001). The reported duration of daytime sleep decreased over time. During the first month, the patient slept for up to 3.5 h during daytime on 4 days a week, while during the second month, he slept for up to 2.5 h on 1.8 days a week. Discussion and conclusions Bifrontal anodal tDCS resulted in a significant improvement of objective and subjective vigilance and a decrease in daytime sleep. This provides first preliminary evidence that major symptoms of chronic organic hypersomnia can be improved with tDCS. The first experimental phase of the study demonstrated an immediate improvement of vigilance (PVT) after anodal vs. sham stimulation. This improvement was not compromised by a speed-accuracy trade-off, but rather reflected an overall improvement in performance. Similar improvements had already been demonstrated in healthy individuals, where tDCS appeared to be a strong enhancer of sustained attention in the PVT comparable and partly superior to caffeine [7]. Even short-term enhancements of vigilance could be of great clinical importance for patients with hypersomnia and could be used prior to specific cognitive demands. The second observational phase of the study suggested an improvement of subjective vigilance and a decrease in daytime sleep after two blocks of three consecutive days with tDCS over two months. The patient and his wife reported no relevant side
845
effects and an improvement in quality of life. The level of evidence of this uncontrolled observation is low, but the observed improvement contrasted with the non-response to numerous pharmacological and behavioral interventions over the previous 10 years. Our observations are preliminary and future studies are needed to test whether tDCS protocols can be considered to represent a short- and long-term treatment for patients with different types of hypersomnia, a disabling medical condition with currently limited treatment options. Conflict of interest statement: DR has received a consulting fee from Abbvie Germany. MAN is in the Advisory Board of Neuroelectrics. ChN has received speaker honoraria from Servier. All other authors declare no conflicts of interest.
Lukas Frase Jonathan G. Maier Sulamith Zittel Department of Psychiatry and Psychotherapy University Medical Center Freiburg Germany Department of Clinical Psychology and Psychophysiology University Medical Center Freiburg Germany Tobias Freyer Department of Psychiatry and Psychotherapy University Medical Center Freiburg Germany Dieter Riemann Department of Clinical Psychology and Psychophysiology University Medical Center Freiburg Germany Claus Normann Department of Psychiatry and Psychotherapy University Medical Center Freiburg Germany Bernd Feige Department of Psychiatry and Psychotherapy University Medical Center Freiburg Germany Department of Clinical Psychology and Psychophysiology University Medical Center Freiburg Germany Michael A. Nitsche Department of Clinical Neurophysiology University Medical Center Göttingen Germany Leibniz Research Centre for Working Environment and Human Factors, Dortmund, Germany Department of Neurology University Medical Hospital Bergmannsheil Bochum, Germany Christoph Nissen* Department of Psychiatry and Psychotherapy University Medical Center Freiburg Germany Department of Clinical Psychology and Psychophysiology University Medical Center Freiburg Germany
846
Letters to the Editor / Brain Stimulation 8 (2015) 838e849
* Corresponding author. Department of Psychiatry and Psychotherapy, University Medical Center Freiburg, Hauptstr. 5, 79104 Freiburg, Germany. Tel.: þ49 761 270 65010; fax: þ49 761 270 66190. E-mail address: [email protected] (C. Nissen)
Received 5 May 2015
http://dx.doi.org/10.1016/j.brs.2015.05.009
References [1] Steriade M, Contreras D, Curró Dossi R, Nuñez A. The slow (1 Hz) oscillation in reticular thalamic and thalamocortical neurons: scenario of sleep rhythm generation in interacting thalamic and neocortical networks. J Neurosci 1993;8: 3284e99. [2] Nitsche MA, Paulus W. Sustained excitability elevations induced by transcranial DC motor cortex stimulation in humans. Neurology 2001;10:1899e901. [3] Billiard M, Dauvilliers Y. Idiopathic hypersomnia. Sleep Med Rev 2001;5: 349e58. [4] Dinges DF, Powell JW. Microcomputer analysis of performance on a portable, simple visual RT task sustained operations. Behav Res Methods Instrum Comput 1985;17:652e5. [5] Monte-Silva K, Kuo M, Hessenthaler S, Fresnoza S, Liebetanz D, Paulus W, et al. Induction of late LTP-like plasticity in the human motor cortex by repeated noninvasive brain stimulation. Brain Stimul 2013;3:424e32. [6] Nitsche MA, Cohen LG, Wassermann EM, Priori A, Lang N, Antal A, et al. Transcranial direct current stimulation: state of the art 2008. Brain Stimul 2008;3: 206e23. [7] McIntire LK, McKinley RA, Goodyear C, Nelson J. A comparison of the effects of transcranial direct current stimulation and caffeine on vigilance and cognitive performance during extended wakefulness. Brain Stimul 2014;4: 499e507.
Conceptual and Procedural Shortcomings of the Systematic Review “Evidence That Transcranial Direct Current Stimulation (tDCS) Generates Littleto-no Reliable Neurophysiologic Effect Beyond MEP Amplitude Modulation in Healthy Human Subjects: A Systematic Review” by Horvath and Co-workers Dear Sir, We are writing in reply to the above-mentioned paper by Horvath and colleagues, published in Neuropsychologia recently [1]. In this article, the authors conclude, based on a systematic review of research data exploring the physiological effects of tDCS, a non-invasive brain stimulation technique that beyond an effect on motor evoked potentials, tDCS has no impact on physiological
parameters. We are concerned about the validity of the conclusions for various reasons. Since this paper reviews a whole field of research and comes to debatable assumptions, it is especially important that basic quality requirements are fulfilled, which is unfortunately not the case. First, this review suffers from numerous conceptual flaws and misunderstandings. Second, the work contains relevant design problems, several errors and many incompletely or incorrectly cited data. A complete list of the factual errors is beyond the scope of the present letter. It would require a complete re-review of all original studies, which is not possible in a reasonable time frame, also because even parts of the data supplied in the tables referring to the original studies are wrong. We will focus our reply primarily on studies from our groups. The introduction contains some relevant shortcomings about basic concepts of tDCS. The statement that the primary effect of tDCS is a modulation of the “resting membrane potential of neuronal populations via ionic adjustment of extracellular space” is a misconception and not supported by the respective reference [2], nor by animal slice and human in vivo experiments (e.g. Refs. [3e6]). In these studies, different return electrode positions, and neuronal orientations, result in antagonistic effects of tDCS delivered via identically positioned target electrodes, which would not be the case if ionic adjustment of extracellular space would be the main driver for the effects. It is also a misconception that the alteration of resting membrane potentials lasts for some minutes after stimulation termination, and also this statement is not supported by appropriate references. The abolishment of also short-lasting after-effects via NMDA-receptor block is in accordance with a synaptic effect [7,8]. The methodological approach chosen by the authors is not well suited to explore the data base from the original studies. For example, in case of tDCS, but similar to other neuromodulatory interventions, like repetitive transcranial magnetic stimulation (rTMS), or pharmacological interventions, one main intrinsic aspect is physiologically based state-dependency and non-linearity of effects. In this connection, the diversity of intervention protocols, tasks, and subject groups, with rare availability of studies in which strict replication was performed, makes it difficult to compare studies retrospectively by pooled data. This problem could be solved if for a specific task/measure numerous studies would be available, which differ in a limited set of factors, enabling the systematic exploration of the impact of protocol variants on the results. Unfortunately, this is not the case for tDCS in many instances. Therefore results of a quantification, which lump studies together without taking critical protocol components (stimulation duration, intensity, target and return electrode positions) and anatomical/physiological (e.g. dominant-non-dominant sides) differences into account, can be misleading. To give some examples, antagonistic effects due to different return electrode positions in different studies will result in a null outcome, if put together in a single analysis, as shown for the impact of tDCS on visual evoked potentials [3,4]. Pooling results of 13 min and 26 min anodal tDCS on motor cortex excitability, where anodal tDCS enhances excitability in the 13 min stimulation condition, but results in an excitability diminution in case of 26 min stimulation, probably due to calcium overflow mechanisms [9], will also result in an overall zero effect. This, however, does not mean that these stimulation protocols are inefficient, but that protocol differences result in physiologically based discernable effects. Statistical heterogeneity might also apply, and can and should be tested before conducting such an analysis (Cochrane Handbook for Systematic Reviews of Interventions, 2015). Unfortunately, in this review these aspects have not constantly been taken into account. These would have likely resulted in the conclusion that the type of analysis conducted in
Letters to the Editor / Brain Stimulation 8 (2015) 838e849
Bifrontal Anodal Transcranial Direct Current Stimulation (tDCS) Improves Daytime Vigilance and Sleepiness in a Patient With Organic Hypersomnia Following Reanimation Dear Editor: In this case report, we provide first and preliminary evidence that bifrontal anodal transcranial direct current stimulation (tDCS) can improve vigilance and reduce daytime sleepiness in a patient with organic hypersomnia following reanimation. Topdown models of sleep regulation propose that local activity changes of the frontal cortex modulate arousal and sleep via corticothalamo-cortical feedback loops [1]. Here, we used anodal tDCS to enhance the excitability of the frontal cortex [2] and to improve the clinical symptoms of organic hypersomnia, an adverse health condition resulting in a substantially reduced quality of life with limited treatment options [3]. Methods Patient We report a 52-year-old construction worker who sustained a severe allergic reaction with cardiac arrest and cardio-pulmonal reanimation following a bee sting 10 years prior to the study. In good mental and somatic health until then, he suffered from the de novo onset of excessive daytime sleepiness of up to several
hours per day in the absence of any other neurological deficits, establishing the diagnosis of organic hypersomnia (ICD-10: G47.1). He remained incapable of work. With a time lag of one year, he suffered from a first depressive episode with reduced mood and a loss of pleasure. Under antidepressant treatment, these symptoms returned to normal; the excessive daytime sleepiness persisted. In addition, a comorbid obstructive sleep apnea syndrome was diagnosed, which was continuously and sufficiently treated with automatic continuous positive airway pressure (auto-CPAP; apnoea hypopnea index, AHI 2.2/h, normal level). Prior to the study, extensive pharmacological treatments, including SSRI, venlafaxine, modafinil, and amantadine, as well as cognitive-behavioral therapy had not substantially improved reduced vigilance and excessive daytime sleepiness. During the study, the medication remained unchanged (venlafaxine 225 mg, modafinil 200 mg, amantadine 200 mg). The study has been approved by the local ethics committee and the patient has given written informed consent.
Study design The study comprised two phases (Fig. 1). In experimental phase I, the patient underwent three sessions of anodal stimulation and three sessions of sham stimulation in an alternating order on six consecutive days (single blind). Vigilance was investigated prior to and after the interventions using the psychomotor vigilance task (PVT, [4]). Means of the PVT outcome parameters response speed and response errors were calculated across the experimental days for each point in time of the assessment (pre- and post-stimulation) and each condition (anodal and sham stimulation) separately. Analyses of variance with the repeated measures factors Time (pre- vs. post-stimulation) and Condition (anodal vs. sham stimulation) were calculated for both outcome parameters. In observational phase II, two blocks comprising three sessions of anodal stimulation on three consecutive days were followed by one month of clinical self-monitoring after each tDCS block.
Figure 1. A1) Study protocol for phase I. Three sessions of anodal stimulation and three sessions of sham stimulation were applied in an alternating order on six consecutive days, with vigilance measured pre- and post-stimulation using the psychomotor vigilance task (PVT). A2) Mean reaction speed deteriorated after sham but improved after anodal tDCS. The number of reaction errors remained unchanged. ANOVA with the repeated measures factors Time (pre- vs. post-stimulation) and Condition (anodal vs. sham stimulation). Bars indicate the standard error of the mean. * P < .05. B1) Study protocol for phase II. Anodal stimulation was applied on two times three consecutive days followed by one month of clinical self-monitoring at each case. B2) Regression analysis showed a significant time-dependent improvement of subjective vigilance (regression line and linear trend of residuals given). B3) The duration of daytime sleep decreased over the reported time (descriptive data).
Letters to the Editor / Brain Stimulation 8 (2015) 838e849
Subjective vigilance (visual analog scale; 0e100, low to high) and the duration of daytime sleep (self-report) were assessed. tDCS protocol tDCS was delivered between 12.30 PM and 1.15 PM by a constant current stimulator (neuroConn, Germany) via two frontal stimulation electrodes (5 7 cm, FP1/FP2, 10e20 system) and two parietal reference electrodes (10 10 cm, P3/P4). To induce robust after-effects for the modulation of tonic arousal processes, an optimized repetitive stimulation protocol was selected with anodal stimulation in two blocks of 13 min with a 20 min inter-stimulation interval [5]. A constant current of 1 mA over each stimulation electrode was applied (2 mA stimulator output, Y-cable split for stimulation and reference electrodes) in a 30-s fade-in/fade-out design to decrease potential skin sensations [6]. For sham stimulation, the current turned off automatically after 15 s. Results Figure 1 visualizes the main findings. In phase I (Fig. 1A), analysis of variance of response speed revealed a significant effect for Time (pre- vs. post-stimulation, F ¼ 4.2, P < .05) and Condition (anodal vs. sham stimulation, F ¼ 42.2, P < .001) and a significant interaction effect (F ¼ 24.5, P < .01). Post-hoc analyses indicated that the response speed significantly deteriorated following sham (t ¼ 2.1, P < .05) and improved following anodal stimulation (t ¼ 4.9, P < .001) compared to pre-stimulation measurements. In addition, the response speed was significantly faster after anodal than after sham stimulation (t ¼ 8.0, P < .001), with similar pre-stimulation levels (t ¼ 1.1, P ¼ .29). The number of response errors did not differ between the conditions and time points of measurement (X2 ¼ .27, P ¼ .78). In phase II (Fig. 1B), a regression analysis of subjective ratings showed a significant increase in subjective vigilance over time (b ¼ .42, R2 ¼ .28, P < .001). The reported duration of daytime sleep decreased over time. During the first month, the patient slept for up to 3.5 h during daytime on 4 days a week, while during the second month, he slept for up to 2.5 h on 1.8 days a week. Discussion and conclusions Bifrontal anodal tDCS resulted in a significant improvement of objective and subjective vigilance and a decrease in daytime sleep. This provides first preliminary evidence that major symptoms of chronic organic hypersomnia can be improved with tDCS. The first experimental phase of the study demonstrated an immediate improvement of vigilance (PVT) after anodal vs. sham stimulation. This improvement was not compromised by a speed-accuracy trade-off, but rather reflected an overall improvement in performance. Similar improvements had already been demonstrated in healthy individuals, where tDCS appeared to be a strong enhancer of sustained attention in the PVT comparable and partly superior to caffeine [7]. Even short-term enhancements of vigilance could be of great clinical importance for patients with hypersomnia and could be used prior to specific cognitive demands. The second observational phase of the study suggested an improvement of subjective vigilance and a decrease in daytime sleep after two blocks of three consecutive days with tDCS over two months. The patient and his wife reported no relevant side
845
effects and an improvement in quality of life. The level of evidence of this uncontrolled observation is low, but the observed improvement contrasted with the non-response to numerous pharmacological and behavioral interventions over the previous 10 years. Our observations are preliminary and future studies are needed to test whether tDCS protocols can be considered to represent a short- and long-term treatment for patients with different types of hypersomnia, a disabling medical condition with currently limited treatment options. Conflict of interest statement: DR has received a consulting fee from Abbvie Germany. MAN is in the Advisory Board of Neuroelectrics. ChN has received speaker honoraria from Servier. All other authors declare no conflicts of interest.
Lukas Frase Jonathan G. Maier Sulamith Zittel Department of Psychiatry and Psychotherapy University Medical Center Freiburg Germany Department of Clinical Psychology and Psychophysiology University Medical Center Freiburg Germany Tobias Freyer Department of Psychiatry and Psychotherapy University Medical Center Freiburg Germany Dieter Riemann Department of Clinical Psychology and Psychophysiology University Medical Center Freiburg Germany Claus Normann Department of Psychiatry and Psychotherapy University Medical Center Freiburg Germany Bernd Feige Department of Psychiatry and Psychotherapy University Medical Center Freiburg Germany Department of Clinical Psychology and Psychophysiology University Medical Center Freiburg Germany Michael A. Nitsche Department of Clinical Neurophysiology University Medical Center Göttingen Germany Leibniz Research Centre for Working Environment and Human Factors, Dortmund, Germany Department of Neurology University Medical Hospital Bergmannsheil Bochum, Germany Christoph Nissen* Department of Psychiatry and Psychotherapy University Medical Center Freiburg Germany Department of Clinical Psychology and Psychophysiology University Medical Center Freiburg Germany
846
Letters to the Editor / Brain Stimulation 8 (2015) 838e849
* Corresponding author. Department of Psychiatry and Psychotherapy, University Medical Center Freiburg, Hauptstr. 5, 79104 Freiburg, Germany. Tel.: þ49 761 270 65010; fax: þ49 761 270 66190. E-mail address: [email protected] (C. Nissen)
Received 5 May 2015
http://dx.doi.org/10.1016/j.brs.2015.05.009
References [1] Steriade M, Contreras D, Curró Dossi R, Nuñez A. The slow (1 Hz) oscillation in reticular thalamic and thalamocortical neurons: scenario of sleep rhythm generation in interacting thalamic and neocortical networks. J Neurosci 1993;8: 3284e99. [2] Nitsche MA, Paulus W. Sustained excitability elevations induced by transcranial DC motor cortex stimulation in humans. Neurology 2001;10:1899e901. [3] Billiard M, Dauvilliers Y. Idiopathic hypersomnia. Sleep Med Rev 2001;5: 349e58. [4] Dinges DF, Powell JW. Microcomputer analysis of performance on a portable, simple visual RT task sustained operations. Behav Res Methods Instrum Comput 1985;17:652e5. [5] Monte-Silva K, Kuo M, Hessenthaler S, Fresnoza S, Liebetanz D, Paulus W, et al. Induction of late LTP-like plasticity in the human motor cortex by repeated noninvasive brain stimulation. Brain Stimul 2013;3:424e32. [6] Nitsche MA, Cohen LG, Wassermann EM, Priori A, Lang N, Antal A, et al. Transcranial direct current stimulation: state of the art 2008. Brain Stimul 2008;3: 206e23. [7] McIntire LK, McKinley RA, Goodyear C, Nelson J. A comparison of the effects of transcranial direct current stimulation and caffeine on vigilance and cognitive performance during extended wakefulness. Brain Stimul 2014;4: 499e507.
Conceptual and Procedural Shortcomings of the Systematic Review “Evidence That Transcranial Direct Current Stimulation (tDCS) Generates Littleto-no Reliable Neurophysiologic Effect Beyond MEP Amplitude Modulation in Healthy Human Subjects: A Systematic Review” by Horvath and Co-workers Dear Sir, We are writing in reply to the above-mentioned paper by Horvath and colleagues, published in Neuropsychologia recently [1]. In this article, the authors conclude, based on a systematic review of research data exploring the physiological effects of tDCS, a non-invasive brain stimulation technique that beyond an effect on motor evoked potentials, tDCS has no impact on physiological
parameters. We are concerned about the validity of the conclusions for various reasons. Since this paper reviews a whole field of research and comes to debatable assumptions, it is especially important that basic quality requirements are fulfilled, which is unfortunately not the case. First, this review suffers from numerous conceptual flaws and misunderstandings. Second, the work contains relevant design problems, several errors and many incompletely or incorrectly cited data. A complete list of the factual errors is beyond the scope of the present letter. It would require a complete re-review of all original studies, which is not possible in a reasonable time frame, also because even parts of the data supplied in the tables referring to the original studies are wrong. We will focus our reply primarily on studies from our groups. The introduction contains some relevant shortcomings about basic concepts of tDCS. The statement that the primary effect of tDCS is a modulation of the “resting membrane potential of neuronal populations via ionic adjustment of extracellular space” is a misconception and not supported by the respective reference [2], nor by animal slice and human in vivo experiments (e.g. Refs. [3e6]). In these studies, different return electrode positions, and neuronal orientations, result in antagonistic effects of tDCS delivered via identically positioned target electrodes, which would not be the case if ionic adjustment of extracellular space would be the main driver for the effects. It is also a misconception that the alteration of resting membrane potentials lasts for some minutes after stimulation termination, and also this statement is not supported by appropriate references. The abolishment of also short-lasting after-effects via NMDA-receptor block is in accordance with a synaptic effect [7,8]. The methodological approach chosen by the authors is not well suited to explore the data base from the original studies. For example, in case of tDCS, but similar to other neuromodulatory interventions, like repetitive transcranial magnetic stimulation (rTMS), or pharmacological interventions, one main intrinsic aspect is physiologically based state-dependency and non-linearity of effects. In this connection, the diversity of intervention protocols, tasks, and subject groups, with rare availability of studies in which strict replication was performed, makes it difficult to compare studies retrospectively by pooled data. This problem could be solved if for a specific task/measure numerous studies would be available, which differ in a limited set of factors, enabling the systematic exploration of the impact of protocol variants on the results. Unfortunately, this is not the case for tDCS in many instances. Therefore results of a quantification, which lump studies together without taking critical protocol components (stimulation duration, intensity, target and return electrode positions) and anatomical/physiological (e.g. dominant-non-dominant sides) differences into account, can be misleading. To give some examples, antagonistic effects due to different return electrode positions in different studies will result in a null outcome, if put together in a single analysis, as shown for the impact of tDCS on visual evoked potentials [3,4]. Pooling results of 13 min and 26 min anodal tDCS on motor cortex excitability, where anodal tDCS enhances excitability in the 13 min stimulation condition, but results in an excitability diminution in case of 26 min stimulation, probably due to calcium overflow mechanisms [9], will also result in an overall zero effect. This, however, does not mean that these stimulation protocols are inefficient, but that protocol differences result in physiologically based discernable effects. Statistical heterogeneity might also apply, and can and should be tested before conducting such an analysis (Cochrane Handbook for Systematic Reviews of Interventions, 2015). Unfortunately, in this review these aspects have not constantly been taken into account. These would have likely resulted in the conclusion that the type of analysis conducted in
