This is a single-center, parallel, randomized, double-blinded, sham-controlled trial, conducted in the Laboratory of Functional Electrostimulation at the Federal University of Bahia (Salvador, Brazil). This study will include descriptive and analytical steps designed to investigate the neurobiological aspects of patients with ADHD, and the response of inhibitory control parameters to tDCS.

Patients with ADHD will be recruited by sending letters and emails to neuropsychiatric societies and associations. Professional experts in ADHD, contacted by mail and telephone, will be invited to refer potential patients for the study. Advertisements will be published on the Internet and social networking sites, according to the inclusion criteria.

The inclusion criteria for the study are (1) the ability to understand and sign the informed consent form, (2) a diagnosis of ADHD according to DSM-IV-TR and the Adult Self-Report Scale (ASRS), (3) resident of Bahia, and (4) ≥18 years old.

Patients who meet the following criteria will be excluded (1) major psychiatric disorders such as schizophrenia and bipolar disorder, (2) inability to understand the questionnaires applied (cognitive impairment score of ≤24 on the Mini-Mental State Examination) or illiterate, and (3) abuse of psychoactive substances or alcohol, except nicotine and caffeine, during the last 12 months.

A sample size of 50 subjects was calculated assuming a difference in proportion between the active tDCS and sham control groups of 40% on the go/no go performance, before and after interventions. This calculation was performed using the statistical program STATA 12.0, considering an alpha of .05 and a power of 0.80, resulting in a sample of 25 subjects in each group (intervention and control). To address unexpected factors, we applied a dropout rate of 20%, reaching a total sample size of 60 individuals.

Individuals will be divided into two groups by block randomization aiming for a balanced distribution between the groups (30 subjects each), and considering gender and age as variables. Blocks will be composed of 10 subjects, totaling 6 blocks. Each intervention group will comprise 5 individuals from each block.

An external researcher will perform the randomization and will generate a list to allocate patients and ensure a concealed allocation.

Excluding the external researcher, investigators applying the go/no go tasks and registering the EEG, and the subjects will be blind to the intervention. Raters and researchers responsible for the statistical analysis will not be aware of the treatment group that the patient will be enrolled. For analysis purposes, the intervention groups will be identified as “0” and “1”. Blinding code will be known only by the principal investigator and by the person responsible for the stimulation.

In accordance with the Declaration of Helsinki [18], this study strictly follows the ethical principles in research involving human subjects. All participants will be informed about the nature of the study and all procedures prior to enrollment. Following the resolution 196/96 of the National Health Council (Brazil), only those that will sign the informed consent form will be included. The signature of a witness will be required for the patients unable to sign the free and informed consent form.

This trial was approved on October 2012 (IRB approval number 19311) by the institutional review board of the Maternidade Climério de Oliveira-Federal University of Bahia.

Signed consent will be obtained from each participant. Only subjects capable of understanding and agreeing to the consent will be enrolled. In addition to a signed consent form and prior to any intervention, participants will receive a detailed explanation about the trial aims and procedures that will be performed. Two experienced researchers will explain that participation is voluntary and they may withdraw at anytime without losses. The trial staff also will be available to clarify any question; at the same time they will assure that subjects understood all of the steps of the study. Participants will have an indefinite amount of time to make a decision, and those that agree, will be asked to sign two copies of the informed consent form.

Following the consent procedures, subjects will undergo a cognitive screening using the MiniMental State Examination (MMSE) [19,20], the Mini International Neuropsychiatric Interview Brazilian (MINI PLUS) [21], the Adult Self-Report Scale (ASRS) [22], the Adult ADHD Quality of Life Questionnaire (AAQoL), and an interview with a questionnaire. This evaluation script is an instrument based on the ADHD diagnostic criteria from DSM-IV-TR [7,8], and consists of epidemiological and clinical questions.

Posterior electrical brain activity will be recorded by EEG with 32 channels placed on the scalp, according to the 10-20 international EEG system. Each recording will last 5 minutes and will be performed in a resting state, with the patient's eyes open and looking for a fixed point during the first minute, and with closed eyes for the remaining four. Participants will then be asked to perform tasks on the computer: 2 go/no go tasks adapted from the original version using a fruit and a letter as the first and second target, respectively. Following the cognitive tasks, subjects will undergo the active or sham stimulation, according to the previous randomization. At the end of the intervention, subjects will perform the 2 go/no go tasks again (randomly selected to avoid learning effect), and they will undergo another EEG recording.

There will be no follow-up visits; participants will undergo a single visit of approximately two hours in duration. All subjects will be asked to abstain from caffeine, alcohol, and nicotine one day before the experiments.

Participants of the intervention group will have the tDCS applied at 1 mA, with the anode electrode over the left dorsolateral prefrontal cortex and cathode electrode over the equivalent area on the right side. For the control group subjects, electrodes will be placed at identical positions and sham stimulation will be applied. The stimulation device will be turned on for 30 seconds so that the patient will feel the initial sensation, then will be shut down. The intervention and sham procedure will take place in one single 20-minute session to evaluate the immediate effect.

The application of tDCS and the assessment of its effects will be performed by clinical observation using a reduction of symptomatology as evaluated by the neuropsychological task, go/no go, and by neurophysiological assessment, measured by quantitative electroencephalogram (qEEG) and the reconstruction of brain networks, as parameters.

The primary outcome of this trial is the inhibitory response measured as behavioral performance by two go/no go tasks. These tests are adapted from a previous version [23], and will present as targets a fruit (version 1) and a letter (version 2). Participants will be asked to react when a target ("go" stimulus), and not when a non-target ("no go" stimulus) is presented. All stimuli will be presented in black and white ink with equal dimensions. In both versions of the go/no go tasks, one selected letter or fruit will be previously defined as the "go" stimulus and subjects will be instructed to press the left button of a computer mouse with the right finger as soon as this target is presented on the screen. Stimuli will be presented one-by-one on the computer screen during 650 milliseconds with 1000 millisecond interstimulus intervals (ISI). In each version of the task, 150 trials consisting of 80% "go" and 20% "no go" will be presented. Correct responses, impulsivity, and omissions errors will be computed separately for each version of the go/no go task, and will be the primary outcome variables.

The secondary outcome variables are power analysis of frequency bands through qEEG and the brain networks. For the EEG, data will be recorded from the 32 channels and a reference electrode (Cz) placed according to the 10-20 international EEG system with additional electrodes (FC3, FC4, CP3, CP4, FT7, FT8, TP7, TP8, and Oz). The EEG data will be filtered between 0.5-50 Hz, and analyzed through EEGLAB (The Mathworks, Inc.). Brain network reconstructions will be performed based on the EEG data.

A 5-step statistical analysis will be conducted (see Textbox 1). These analyses will be performed using Stata, version 13.0 for Windows. Results will be considered statistically significant if P<.05.