IJGG Journal

Introduction

Parkinson’s Disease (PD) is the second most common and fastest growing neurological condition. It affects roughly one million people in the United States and 10 million people worldwide [1,2]. PD is associated with cardinal motor symptoms — bradykinesia, tremor, rigidity, and postural instability — in addition to nonmotor symptoms [3]. These non-motor symptoms, such as cognitive impairment, fatigue, and depression, can significantly impact a patient’s quality of life but are overshadowed by the motor symptoms [4]. The lack of standard treatments for nonmotor symptoms, like cognitive impairment and fatigue, further complicates the management of non-motor symptoms. The motivation to advance knowledge in non-motor symptoms of PD is evident.

In the last decade, there has been exponential growth in examining the use of tDCS for the improvement of PD motor and non-motor symptoms. Transcranial direct current stimulation is a noninvasive brain stimulation technique that modulates cortical excitability. Anodal tDCS (atDCS, referenced as tDCS throughout the manuscript), often used in studies evaluating its effectiveness on cognitive function, increases the likelihood of an action potential occurring by depolarizing neurons. Recently, [5] identified that healthy adults showed an increase in extracellular dopamine following a single tDCS session. This effect of tDCS on dopamine levels emphasizes the use of tDCS as a potential therapy for the disruption of the dopaminergic system, which has been suggested to be involved in cognitive impairments [6] and is present in the pathology of PD. Further, an advantage of tDCS is that it has minimal adverse effects, with patients occasionally reporting skin irritation or headache at the site of stimulation [7,8]. Additionally, atDCS influences cortical excitability by facilitating long-term potentiation and reducing long-term depression. Thus, multisession tDCS has the potential of inducing more lasting changes in the brain [9].

Dopaminergic dysfunction of the prefrontal-basal ganglia-thalamocortical loops contribute to PD-related declines in attention, memory, and executive function [10-12]. The current study evaluated the effect of tDCS to improve attention and working memory. Attention is necessary for the filtering of information in one’s environment and can be automatic (simple) or effortful (complex; [11]. Memory, particularly working memory, drives our ability to hold information for a short period of time and to be able to manipulate that information [13].

Multiple studies have evaluated the effects of tDCS on attention and memory [14]. administered four 20-minutes sessions using 1.5 mA tDCS to the left dorsolateral prefrontal cortex (l-DLPFC) and showed improved complex attention (STROOP) and working memory (paragraph recall). It is important to note that Lawrence et al. tested tDCS alone and paired with cognitive training programs, in which the latter resulted in greater cognitive effects [15]. administered sixteen 20-minute sessions using 2 mA tDCS to the l-DLPFC four days a week for four weeks and found a trend for improved working memory at 16-weeks. However, short-term improvements in working memory (immediate memory index and story learning) were not found and decrements in executive attention (written coding test) declined at four weeks. [16] administered 2 mA at the l-DLPFC in a single session and found improved working memory (three-back letter task).

Currently, there are no treatments for fatigue in PD although roughly half of the PD population experience this non-motor symptom [17,18]. Fatigue is defined as “the subjective and general sensation of tiredness and difficulty initiating or engaging in activities” [19]. Fatigue can be divided into mental and physical domains. Mental fatigue refers to the extra effort required to pay attention. Physical fatigue refers to the extra effort required to initiate or complete an activity. Cognitive fatigability is “the deterioration in the performance of attention tasks over an extended period of time.” A key difference between fatigue and fatigability is the subjective versus objective nature of the two, respectively.

Studies in other neurological diseases, such as Multiple Sclerosis (MS), have found short- and long-term improvements in fatigue symptoms following five consecutive 20-minutes tDCS sessions at 2 mA to the l-DLPFC [20] and with 1.5 mA to the motor area for 15 minutes a day for five consecutive days [21]. One study in PD patients found reduced subjective fatigue following 2 mA stimulation for 20 minutes to the DLPFC for 8 total sessions [22].

The first aim of this study was to examine whether anodal tDCS to the left dorsolateral prefrontal cortex for five consecutive days improved cognitive function and reduced cognitive fatigue and fatigability in PD patients. We predicted that compared to the sham control group, the experimental group would show improved performance in complex attention, as measured by the STROOP task, and increased working memory, as measured by the Digit Span Test (Forward and Backward). We predicted that the treatment group would also show less fatigue, as measured by the Multidimensional Fatigue Inventory (MFI). We also predicted that the treatment group would show less fatigability compared to the sham group, as measured by performance across a challenging computer task (i.e., treatment group would show less worsening of reaction time or accuracy).

The second aim was to assess the short-term and long-term efficacy of the tDCS intervention. We assessed short-term effects by comparing the baseline measurements to those taken immediately after the final tDCS session on the 5th day (i.e., post-test). We assessed long-term effects by comparing baseline measurements to those taken at 1-week and 2-week follow-ups (i.e., 7 and 14 days after last tDCS session, respectively). We predicted that 5 consecutive sessions of atDCS targeting the l-DLPFC would induce both immediate and sustained improvements in cognitive function, as well as reductions in cognitive fatigue and fatigability in PD patients.

Materials and Methods Participants

We recruited PD patients from the Movement Disorders Clinic at the Sanford Brain and Spine Center in Fargo, North Dakota. All patients spoke fluent English and met clinical diagnosis of PD with at least two of the four diagnostic criteria for PD: tremor, rigidity, bradykinesia, and postural instability. All participants were responsive to levodopa and maintained their treatment regimen for the duration of the study. The following exclusion criteria were applied: (1) patients with dementia (MOCA score < 21), (2) psychosis, (3) presence of other neurological conditions such as multiple sclerosis, stroke, epilepsy, (4) internal electric stimulators such as deep brain stimulator, pacemaker, spinal cord stimulator, or (5) severe medical illnesses such as COPD, congestive heart failure, or renal failure. The study initially required participants to meet criteria for mild cognitive impairment (21 < MOCA < 26). The eligibility criterion pertaining to cognitive status was modified to include individuals with normal cognition in order to evaluate the effectiveness of tDCS on individuals who may be experiencing minimal or early cognitive changes.

This study was performed in accordance with the Declaration of Helsinki. This study was approved by the Sanford Health IRB, and we obtained informed consent from all participants. Participants were paid $300 for their participation in the study. The study is registered at the United States Registry of Clinical Trials with the reference ID NCT03191916.

Procedure

The study was a randomized, single-blind, placebo-controlled experiment to compare the two groups of PD participants. Research personnel randomly assigned participants to either the stimulation (“stim”) group or control (“sham”) group using a virtual coin flip simulator. The research team only was aware of the intervention assignment. Following a screening visit to determine eligibility, all consented participants completed the primary and secondary outcome measures and the first tDCS session on a Monday. Stimulation was administered for five consecutive days (MondayFriday). On the fifth day (Friday), and at the 1-week and 2-week follow-up visits, participants again completed all primary and secondary outcome measures. See Figure 1 for a flowchart of the study.

Figure 1: Study flow diagram

Transcranial Direct Current Stimulation

Transcranial direct current stimulation (MagStim HDCkit tDCS) was administered daily for five consecutive days. The current intensity was 2 mA and was applied for 20 minutes with a 30-second ramp up and ramp down period. The anode electrode was 25 cm² and the cathode electrode was 51 cm² and were placed at the left dorsolateral prefrontal cortex (F5 according to the 1020 international system for EEG electrode placement) and the right supraorbital area, respectively. The pads that surrounded the electrodes were soaked in a saline solution before each session. The sham group did not receive stimulation during the 20-minute session but had the same electrode placement and ramp-up and ramp-down period.

Risks of tDCS

Transcranial direct current stimulation is a safe procedure. It has been tested in thousands of subjects worldwide with no evidence of adverse effects to date [8]. Hundreds of studies have used tDCS in different clinical conditions without causing any adverse effects. To further examine the safety of tDCS specifically, [23] retrospectively reviewed adverse effects in 77 healthy subjects and 25 patients who underwent a total of 567 tDCS sessions at 1mA. They found that adverse effects caused by tDCS are not different from those of placebo stimulation, which include mild tingling sensations (75%), light itching sensations (30%), moderate fatigue (35%), and headache (11.8%). A recent article by [24] investigated the adverse events associated with repeated sessions of tDCS and found adverse events did not increase with higher levels of tDCS exposure. A comprehensive review of over 33,000 tDCS sessions found no evidence of serious adverse events [7]. Of the 20 participants in our study that received the tDCS intervention, 2 (10%) experienced a “slightly painful poking sensation” at the location of the anode electrode placement during the tDCS stimulation, 1 (5%) experienced a mild headache after the tDCS, and 15 (75%) felt mild tingling sensations during the tDCS. No noted effects were found in the sham group. The lab director and medical expert reviewed all events of reported adverse reactions.

All participants had the option to withdraw from the study.

Outcome Measurements Primary Outcome Measure: Computer Task

To investigate cognitive fatigability, we used a visual search computer task. The primary outcome measures were Reaction Time (RT) and Accuracy (hits minus false alarms; HFA). Stimuli were presented on a 15-inch monitor set to a refresh rate of 60 Hz. Stimuli were presented and performance was recorded using Presentation Software (Version 18.1, Neurobehavioral Systems, Inc., Berkley, CA). All stimuli were presented in black on a white background. Participants were seated 34 cm from the computer monitor, with the distance held constant with the use of a chin rest. Participants were instructed to respond by pressing designated buttons on a button box.

Figure 2 illustrates the basic trial sequence. A trial began with a black central fixation dot. After 500 ms, the dot is replaced by a search array consisting of 11 L’s in various orientations and 1 T. Participants located the “T” in the array of L’s. They then indicated the target orientation by pressing a key. If the vertical line of the “T” was on the left, participants were instructed to press the left button. Conversely, if the vertical line of the “T” was on the right, participants were instructed to press the right button. Participants were instructed to answer as quickly as possible, while limiting mistakes. The trial ends after the participant responds, or, if the participant fails to respond within 8,000 ms, the trial is terminated. Before the next trial begins, a white screen is presented for 1,000 ms.

Figure 2: Schematic of the visual search task. A blank screen is presented for 500 ms. A search array is presented in which the participant must discriminate the horizontal orientation of one “T’ amongst 11 “L’s” at different orientations. The trial ends upon participant response or after 8,000 ms and is followed by a 1,000 ms intertrial interval.

Each participant completed a practice block with 16 trials, and feedback was provided following each trial. During practice, the researcher monitored and provided additional instructions, if needed. Following the practice block, there were 40 test blocks with 24 trials per block. Each block was followed by a screen stating that participants could take an optional break; pressing the central button would initiate the start of the next block. Reaction time and error rate for each block were recorded, with cognitive fatigue measured as deterioration of reaction time (RT) or Accuracy (HFA) over the completion of the task (evaluated over 5 epochs of 8 blocks per epoch).

Secondary Outcome Measures

The secondary outcome measures included the Multidimensional Fatigue Inventory (MFI), Center for Epidemiological Studies Depression Scale (CES-D), Single Item McGill Quality of Life (QOL) Scale, Stroop Test A, B & C, and the Digit Span Test (Forward and Backward).

The Multidimensional Fatigue Inventory (MFI) is a 20-item self–report instrument that measures five dimensions of fatigue independently: general fatigue, physical fatigue, mental fatigue, reduced motivation, and reduced activity [25]. Each dimension score ranges from 4 to 20 (most severe). Our study examined the mental and physical fatigue domains.

The Center for Epidemiological Studies Depression Scale (CES-D) is a 20-question self-assessment that reports a general score of depression [26]. A score greater than 16 is indicative of a diagnosis of depression. Depression and fatigue have similar symptomology—lack of energy, tiredness, difficulty initiating and completing tasks—which highlights the necessity to evaluate possible comparative variables.

The Single Item McGill Quality of Life (QOL) Scale is a questionnaire that assesses a person’s subjective quality of life in consideration to the following areas of their life: physical, emotional, social, spiritual, and financial [27]. The scale ranges 1-10 with 10 indicating the highest QOL.

The Stroop Test A, B & C, is a measure of processing speed, attention, and inhibition [28]. For each sub-test, the participant reads the words aloud as quickly and accurately as possible for a 45-second duration. In Stroop A, the participant is required to read color words (Red, Green, or Blue) printed in black ink. In Stroop B, the participant is required to name ink colors out loud (text consists of the letter “x”). In Stroop C, the participants are required to name the color ink of opposing color words (e.g., the word Blue written in Red ink, with the correct response being “Red”). The scores are then standardized based on the participant’s age and education level. Ultimately, an Interference score is calculated to reflect cognitive flexibility and inhibitory control by effectively ignoring distracting information. A score of 40 or greater would demonstrate statistically normal cognitive flexibility and inhibitory control.

The Forward and Backward Digit Span Test assesses multiple areas of cognitive function. The Forward Digit Span is a measure of simple attention [29]. Participants repeat progressively longer series of numbers until the participant can no longer complete the task. The Backward Digit Span test is a measure of complex attention and working memory. The participant again repeats a series of numbers, except in reverse order from what the researcher says. Scores reflect the total points earned (one point for each correct line; ranges from 0-14) and the longest string of digits (LDS) completed successfully.

Data Analysis

At the time of the protocol design, there were few studies that had studied the effect of tDCS in PD. Therefore, sample size was determined based on the number of participants in the background studies. Previous literature had single samples with 8-16 participants [15,16]. In this comparative study, we aimed for twenty participants in each test group.

All statistical tests were performed using SPSS version 29 (IBM SPSS Statistics). The test results were considered significant at α = .05. Tests for qualitative variables were compared using the chisquare test, and tests for quantitative variables used a t- or 2-way ANOVA test (first testing for normality and equality of variance). Raw scores for each outcome were submitted to a 2 (Group: tDCS, Sham) x 2 (Visit) mixed ANOVA, with Group as the betweensubjects variable and Visit as the within-subjects variable. Based on the hypotheses of this study (short and long-term effects of tDCS), three two-way ANOVA tests were performed to make specific comparisons. Visit compared 1) baseline performance to post-visit (immediately after treatment on day five), 2) baseline to 1-week follow-up, and 3) baseline to 2-week follow-up. For the purposes of this study, we did not explore changes in participant performance between the three post-visits.

Results

A total of 43 PD participants were enrolled in this study from 2015 until 2025 (interruptions in recruitment and data collection due to personnel changes). The study was registered in 2017 due to initiators of the study being unaware of the requirement for prospective registration of all interventional trials. Six participants were enrolled during this period. We registered the trial as soon as we became aware of this policy. Four participants were excluded, three prior to administration of stimulation: two were not eligible due to a MOCA < 21, indicating potential dementia, and one voluntarily withdrew participation prior to the first day of testing and stimulation.  One participant was withdrawn due to being unable to complete all tDCS sessions due to adverse weather. Enrollment was ended upon obtaining the proposed sample size, based on those who initiated the intervention. A total of 39 PD participants (tDCS: 20, sham: 19) completed the study and were included in the final analysis. Participant characteristics for the study are shown in Table 1.

Table 1: Characteristics of tDCS and Sham Parkinson’s disease participants

Computer Task

Trials with RTs less than 200 ms were excluded from the analysis. Participant responses under 200 ms on the discrimination task were considered anticipatory responses or inattention errors. Error trials were also removed from the analysis. Data for two participants were removed due to HFA rates below 0.5 for a visit, suggesting that they were performing at the same rate as guessing for the discrimination task. An additional participant’s data was removed for the 1-week follow-up only, as the error rate was below 0.5 for the visit. The slope of median RTs across each epoch (8 blocks per epoch) was calculated and submitted to a two-way mixed ANOVA. The slope was used to evaluate the change in fatigue across the task. A negative slope reflected hastening of speed and a positive slope indicated slowing in responses in a single computer task session. A main effect of Visit was found for the baseline to post-test comparison (p = .003), and baseline to 2-week follow-up comparison (p = .002). Both comparisons identified that compared to the baseline visit, at the follow-up visit all participants responded more quickly at the end of the computer task compared to the start. None of the Group x Visit interactions were significant (p > .05).

Accuracy was analyzed as hits (correct trials) minus false alarms (correct trials within the 200 ms window). The slope of mean accuracy rates were submitted to the two-way mixed ANOVA. A positive slope indicated participants were making more errors at the end of the computer task compared to the start. A negative slope indicated that participants were making fewer errors as they completed the computer task. A main effect of Visit was found for the baseline to 2-week follow-up comparison (p = .046), indicating that all participants showed improved accuracy from start-to-finish at the follow-up compared to baseline visit. No other main effects were found nor were any interactions significant (p > .05). Reaction time and Accuracy data are presented in Table 2.

Computer Task

Table 2: Primary measures at baseline, post-visit, 1-week, and 2- week follow-up visits