High‐definition transcranial direct current stimulation—An open‐label pilot intervention in alleviating depressive symptoms and cognitive deficits in late‐life depression

Abstract The efficacy of high‐definition transcranial direct current stimulation (HD‐tDCS) in late‐life depression (LLD) remains unknown due to limited research on its therapeutic effects on the hallmarks of LLD—the depressive and cognitive symptoms. The present open‐label pilot study aimed to examine the effectiveness of HD‐tDCS as an augmentation therapy with antidepressants in improving the depressive and cognitive symptoms for LLD. Significant improvements were hypothesized in the depressive, cognitive, and daily functioning outcomes over time. A total of 15 subjects with LLD (13 females, mean age = 73.27 ± 6.25) received five consecutive daily sessions of 20‐minute active HD‐tDCS interventions weekly for 2 weeks, with a 2 mA anodal stimulation over F3 and cathodal stimulation over FC1, AF3, F7, and FC5. Depressive symptoms and cognitive and daily functioning were assessed across five assessment timepoints. The results revealed that the HD‐tDCS was effective in reducing the depressive severity and the remission rates, with a sustained effect at both the 1‐month and 3‐month follow‐up. Pre‐post improvements were seen in the overall cognitive functioning and in verbal fluency, but not in executive functioning. Our pilot study provides a preliminary result of HD‐tDCS in LLD, which was a safe and effective treatment in alleviating depressive symptoms, with mild cognitive improvements observed. Further larger scale randomized controlled trials are needed to confirm this result.

the side effects of the antidepressants. 7 Over 50% of people with LLD do not achieve symptomatic remission. 8 Between 30 and 50% do not respond to one antidepressant trial. 9 LLD is also closely correlated with cognitive impairments, with executive functioning and memory loss being the most predominant features in LLD. 10 As the aging population continues to surge, LLD will undeniably create further social and economic burdens on our society.

| Limitations of current treatment approaches
Although antidepressants have been well established as the first line of effective treatment for LLD, 11 up to one-third of patients with LLD show a suboptimal response or resistance to antidepressant therapy. 12 Furthermore, their efficacy in addressing depression-related cognitive deficits remains questionable. Cognitive dysfunctions are consistently coupled with depressive symptoms in LLD. 13 People with LLD may not return to normal levels of performance, particularly with respect to memory and executive functions, despite remission of depressive symptoms after antidepressant treatments. 14 Those with memory impairments were more susceptible to developing dementia relative to those without memory deficits. 15 This implies that antidepressants may not fully address the complex symptomology in LLD.
Hence, an alternative treatment approach targeting both depressive and cognitive symptoms is vital in improving the prognosis in LLD.
One such approach is transcranial direct current stimulation (tDCS).

| Descriptions and possible mechanisms of tDCS
tDCS is a non-invasive, neurostimulation technique in which a mild direct current (1-2 mA) is induced through the cerebral cortex via electrodes placed on the scalp, which in turn modifies cortical excitability, depending on the polarity directions. 16 No severe adverse events have been reported in over 40 previous studies involving the geriatric population. 17 It is a safe, easily administered, yet affordable, non-invasive neurostimulation technique, with persistent treatment effects that can last up to an hour. 18 While the exact mechanisms of tDCS are yet to be understood, 19 tDCS is said to exert its effects by modulating cortical excitability, which results in alterations in the corresponding cortical functioning and synaptic release probability uptake and sensitivity. 20 Anodal and cathodal stimulation triggers neuronal depolarization (ie, increased spontaneous firing) and hyperpolarization (ie, decreased neuronal firing), respectively. 21 Long-term plasticity is enhanced, with modulations in the rate of neurotransmitter release. 22

| Stimulation Target in Depression-DLPFC
Serotonin deficits and asymmetrical neural activities in the dorsal lateral prefrontal cortex (DLPFC) (ie, hypoactivity and hyperactivity in the left and right DLPFC 23 ) are two key neurological abnormalities in depression. tDCS and serotonin are known to enhance one another's functions. tDCS increases the release of serotonin, mediated by serotonin transporters, 24 while a continuous enhancement of serotonin by antidepressants strengthens the LTP-like glutamatergic plasticity induced by tDCS. 25 Moreover, tDCS has been shown to exert its antidepressant effects by modulating the hypoactivity in DLPFC in depression. Brunoni et al 26 have found a superior effect on treatment response, remission, and reduced depressive symptoms in intervention groups, relative to sham controls, across six randomized controlled trials that administered anodal tDCS at the left DLPFC in depressed adults. Its effect size was comparable to those receiving antidepressants or repetitive transcranial magnetic stimulation. 26 Similar treatment effects were seen in enhancing working memory 27 and executive functioning. 28 Furthermore, a reduction in executive deficits in patients with LLD may indirectly alleviate the depressive symptoms and enhance the treatment response. 29 Indeed, anodal stimulation over the left prefrontal cortex in schizophrenia patients showed an improvement in the functional capacity and depressive symptoms. 30 This lends support for tDCS's treatment potential for those with LLD as a monotherapy or augmentation with antidepressants. However, it should be noted that controversial findings observed no antidepressant differences between active and sham tDCS for depression. 31

| HD-tDCS
As evidence has shown that the highest cortical current density in tDCS might not be induced directly under the target electrode, 32 the spatial focality of conventional tDCS thus remains questionable. This implies that the treatment efficacy of tDCS could be adversely affected, which might also explain the discrepancy in previous findings. 26,31 Unlike conventional tDCS, high-definition tDCS (HD-tDCS) is typically administered with two or more smaller electrodes. A 4 × 1 ring set-up would be the most typical design, whereby a central anodal electrode is surrounded by four return cathodal electrodes. The density of the cortical field and spatial focality can be adjusted by altering the diameter of the ring set-up. 31 Other strengths of HD-tDCS over tDCS include longer lasting treatment effects due to a more precise cortical field 25 and better tolerability. 33 To our knowledge, no HD-tDCS study has been performed on patients with LLD.
Although some promising results of tDCS were seen, including a reduction of working memory deficits in LLD, 34 these results did not include measures for both the depressive and cognitive symptoms, nor was HD-tDCS administered; only tDCS was administered.
Moreover, with little tDCS research on LLD, it is vital to explore the efficacy and tolerability of HD-tDCS on LLD, a treatment approach that is safe and easy to administer, with proven efficacy in ameliorating depressive symptoms.

| Aims
Therefore, we aimed to perform an open-label pilot study to examine the effectiveness of HD-tDCS as an augmentation therapy with antidepressants in improving depressive and cognitive symptoms in patients with LLD. Significant improvements were hypothesized in the depressive, cognitive, and daily functioning outcomes across various assessment timepoints.

| Study design
This was a 2-week open-label study whereby all participants would receive ten sessions of HD-tDCS (5 consecutive daily sessions of 30 minutes weekly, for a total of 10 sessions) in a psychiatry outpatient clinic in Hong Kong.

| Ethical approval
Written informed consent was obtained from all participants. The study was approved by the Institutional Review Board and was conducted in accordance with the Good Clinical Practice and the Declaration of Helsinki. This study was registered on the HKU Clinical Trial Registry (HKUCTR-2357).

| Participants
The participants were recruited between July 2018 and Mar 2019 from a local public psychiatry outpatient clinic. A total of fifteen pa- Exclusion criteria were (a) a DSM-V diagnosis of other than MDD or anxiety disorders (eg, bipolar affective disorder and schizophrenia); (b) a HK-MoCA score that is below the second percentile according to the subject's age and education level; and (c) any concomitant major medical/neurological conditions or evidence of active infections or significant communicative impairments.

| HD-tDCS procedures
The HD-tDCS intervention was administered at the psychiatric outpatient clinic. Nursing and supporting staff were available in case of emergencies. The intervention was administered using Starstim® produced by Neuroelectrics. The HD-tDCS device was controlled wirelessly via the computer, using the Starstim® software. The montages were the 4 × 1 ring set-up, which is a typical HD-tDCS stimulation protocol. There was a central anodal electrode surrounded by four return cathodal electrodes. The anode was placed over the left DLPFC, which was located at F3, based on the 10/20 electroencephalogram system. The four cathodal electrodes were placed at FC1, AF3, F7, and FC5, forming a circle with a radius of 4.5 cm (Figure 1). Conductive electrode gel was applied on the scalp at all the designated stimulation areas. To ensure the electrodes were secured in place, a different cap size was used depending on the subject's head size. Prior to each session, impedance checks were performed using the Starstim® software. The participants were instructed to relax for the first 5 minutes of each session during the stimulation set-up. A 2 mA stimulation was then delivered for 20 minutes, with a gradual increase and decrease of the current over the first 30 seconds. Each patient was asked to relax and do nothing during the intervention. The administrator closely monitored the impedances throughout each session and recorded any side effects experienced by the participants. They were allowed to rest for 5 minutes after the intervention and were systematically asked if they experienced any discomfort. Each session lasted for approximately 30 minutes, and the sessions took place for five consecutive days each week, for two consecutive weeks.

| Assessments
All assessments and treatment sessions were administered by a trained research assistant (HLW) or a psychiatrist (PWC). All participants were assessed in terms of their depressive symptoms and cognitive and daily functioning across five timepoints, including the baseline (t0), the 5th day of intervention (t1) and the 10th day of intervention (t2), as well as 1 month (t3) and 3 months (t4) after the treatment's completion. Rating Scale (HAM-D17 36 ), which is a widely used and reliable measure of depressive symptoms. 37 The total scores range from 0 to 52, with a higher score suggesting a greater severity in depression. A score of less than seven was defined as remission. A clinical response was defined as a 50% or greater reduction in the HAM-D total scores from the baseline.

Adverse effects
A checklist of potential adverse effects associated with the HD-tDCS administration was generated from available literature reports 46 (see Appendix S1). To monitor tolerability and any adverse events during the intervention, all participants were asked systematically, at the end of each session, if they had experienced any side effects.  Further demographic details can be found in Table 1.

| Depressive severity
A significant effect of time was found in the overall severity of the depressive symptoms, as indicated by the HAM-D-17 total scores (P < .001; see Table 2). When compared with the severity at the

| Apathy and anhedonia
There was a significant effect of time on the levels of apathy   Table 3).

| Neuropsychological & daily functioning
No time effects were found in overall cognitive functioning, as indicated by HK-MoCA total scores (P = .353). For daily functioning as indicated by the Chinese Lawton IADL total scores, no significant effect of time was seen (P > .05).

| Adverse outcomes and side effects
Nine subjects reported mild side effects, such as tingling, itchiness, and mild skin redness at the stimulation site, with no adverse outcomes being reported. Treatment compliance was excellent, with no dropouts being seen in the 2-week stimulation phase, suggesting good tolerability of the intervention. In particular, the improvements in the overall depressive severity and the remission rates were not only seen at the pre-post phase, but also were maintained at the 1-month and 3-month follow-up, implying sustained treatment effects. Although the proportion of participants with a clinical response did not significantly increase over time, a growing trend was found. The levels      Previous research has also suggested that psychiatric patients with greater cognitive deficits (eg, those with schizophrenia) would exhibit more pronounced cognitive improvements in working memory upon receiving tDCS treatment. 51 As our present sample involved LLD patients with a relatively mild depressive profile, it is possible that any subtle cognitive enhancements might not be observed.

| Strengths of the present study
To the best of our knowledge, the present study is the first to examine both the short-and long-term therapeutic efficacy of HD-tDCS as an augmentation therapy with antidepressants in ameliorating both depressive and cognitive symptoms in patients with LLD.
Furthermore, we have attempted to operationalize the primary outcome of our present study-that is, defining depressive symptoms by three levels, namely, the overall severity, the clinical response, and the rates of remission. This is to allow comparisons to be made with previous literature regarding the treatment efficacy of tDCS on reducing depressive symptoms. As research has suggested, the conflicting findings could be partially explained in terms of the diverse definitions used in operationalizing the levels of depressive symptoms. 52 For instance, tDCS was only found to be effective in improving depressive symptoms than sham controls when standard depression scales were used, 53 but not when remission rates or clinical responses were adopted as outcome measures. 54 It is also hoped that our initial findings will provide some insight and framework for designing future studies on HD-tDCS to reduce both depressive and cognitive symptoms in patients with LLD, whether it be the stimulation protocol (ie, the stimulation sites, the frequency or duration of the sessions, or the current strengths) or the types of outcome measures being used.

| Limitations of the current study
Despite the promising results, our study had several methodo-

| Implications for future research
Thus, to counteract the aforementioned limitations, future research should adopt a randomized controlled trial design with a larger sample size, with control arms that involve treatment-as-usual (or on antidepressants only) or other effective interventions that target cognitive dysfunctions in LLD, such as cognitive training, 56 in order to determine if HD-tDCS alone or a combined treatment will maximize the therapeutic benefits for patients with LLD.

| CON CLUS ION
In conclusion, the current study has shown that HD-tDCS was effective in eliciting improvements in the depressive symptoms with mild cognitive enhancements. Future studies should aim for a larger scale, randomized controlled trial in determining the optimal stimulation protocol and the clinical profiles, which could best benefit from HD-tDCS in reducing the hallmarks of LLD-the depressive symptoms and cognitive dysfunctions that play a contributing role in the prognosis and the quality of life in these patients.