Transcranial Direct Current Stimulation: Theory, Treatment of Major Depressive Disorder, and Other Neuropsychiatric Applications
The use of non-invasive brain stimulation for the treatment of various neuropsychiatric disorders, including major depressive disorder (MDD), has rapidly expanded recently. Transcranial direct current electrical stimulation (tDCS), variants of which have been used experimentally for psychiatric (Boggio 2008, Loo 2012, Brunoni 2014, McIntire 2014, Brunoni 2013), neurologic (de 2015, Khedr 2014), and physical rehabilitation (Fregni 2006, Marlow 2013) applications, has garnered a great deal of attention. While it is not yet FDA-approved for any indication, its promise is related to its low cost and wide range of applications; although the breadth of its applicability has been questioned due to heterogeneous data (Horvath 2015), this heterogeneity has been attributed to methodological variability (Antal 2015).
The safety and tolerability of tDCS were outlined by an early study including 567 sessions in 102 patients. The most common adverse effects were mild tingling/itching at the stimulation site and moderate fatigue. Less frequent effects included headaches (11.8%), nausea (2.9%), and insomnia (0.98%), all of which were mild and transient (Poreisz 2007).
The underlying theory is that tDCS modulates the excitability of certain cortical regions (Merzagora 2010, Hampstead 2014, Medeiros 2012) by passage of a small electrical current through conducting pads applied to the scalp in a minimally painful manner. While the precise mechanism is not fully understood, it likely enhances cortical excitability at the anode and depresses it at the cathode (Tremblay 2014, Merzagora 2010, Hampstead 2014).
Proposed mechanisms have been based on data demonstrating relationships between tDCS stimulation and neuropharmacologic effects, cortical electrophysiology, and functional neuroimaging changes. Effects of tDCS on neuroplasticity and cortical excitability have been shown to be differentially modulated by agents affecting neurotransmission via serotonin (citalopram), dopamine (L-dopa), NMDA (dextromethorphan and d-cycloserine), and GABA (lorazepam). Electrophysiologic changes include differential modulation in the presence of agents that modulate sodium channels (carbamazepine) and calcium channels (flunarizine) (Medeiros 2012). Active tDCS shows significant increases in prefrontal cortex activity as measured by functional near infrared spectroscopy (fNIRS), a technique used to measure cortical oxygenation, during and after stimulation – notably, fNIRS measurements may be limited by interference due extracranial blood flow and inability to assess deeper structures, so they merely approximate the functional magnetic resonance imaging (fMRI) signal in superficial structures (Merzagora 2010). Stimulation also increases fMRI activation and connectivity of the underlying cortical regions and hippocampi, though the clinical significance of this is uncertain given that this same study found no behavioral changes (Hampstead 2014).
Since the FDA approval of rTMS of the dorsolateral prefrontal cortex (DLPFC) for treatment of MDD, there has been extensive research regarding non-invasive stimulation of the DLPFC. This has led to some literature showing various types of tDCS to have efficacy as standalone treatment for MDD (Boggio 2008, Loo 2012), as an adjunct to antidepressants (Brunoni 2013), and as a means to enhance psychotherapy response via its cognitive effects (Brunoni 2014). Some conflicting results have called these findings into question, suggesting that further research is needed to establish the most appropriate methodology (Tremblay 2014).
The most widely studied use for tDCS has focused on stimulation of the dominant DLPFC for cognitive enhancement. A recent systematic review of 61 studies found variability regarding tDCS effects on attention, executive function, working memory, and learning – some of which showed improvement, while others did not – and attributed this inconsistency to varying experimental methods (Tremblay 2014). The most pronounced effect has been for attention and stimulatory effects – anodal tDCS of the DLPFC was found to be more effective than 200 mg of caffeine for improving performance on sustained attention tasks for several hours after stimulation, and was comparable to caffeine after 8 hours of sleep deprivation. Furthermore, the effects of the two were comparable when measuring performance on tasks of reaction time and short-term memory in sleep-deprived patients. Subjectively, tDCS outperformed both caffeine and sham in self-ratings of mood, energy, drowsiness, fatigue, and sustained vigilance (McIntire 2014).
An early study on the use of tDCS monotherapy for MDD was by Boggio et al., who demonstrated significantly better improvement in depressive symptoms immediately following stimulation of the dominant DLPFC when compared occipital cortex stimulation and sham (Boggio 2008). Loo et al. subsequently demonstrated significant improvement in mood after active tDCS compared to sham in patients already taking antidepressants, but there was no difference in overall response rates between groups after 3 weeks (Loo 2012).
Comparison with SSRIs was first attempted recently in a 2x2 factorial study comparing tDCS/sertraline, tDCS/placebo, sham/sertraline, and sham/placebo over 6 weeks. The combined treatment was superior to all other groups, while the difference between tDCS and sertraline was insignificant. The difference between sertraline and placebo did not reach statistical significance, although tDCS alone was superior to placebo. Notably, patients correctly guessed which treatment they received, compromising the integrity of the blind; there was no difference between sertraline and tDCS in this regard, leading the authors to suspect that correct guesses were driven by clinical improvement (Brunoni 2013).
Based on the potential efficacy of DLPFC stimulation for both attention/cognition and depression, it has been speculated that this electrode placement would be a useful adjuvant to cognitive psychotherapy, which is otherwise limited by the concentration impairments inherent to MDD. Brunoni et al investigated this in a study of tDCS in cognitive control therapy (CCT), a form of self-directed cognitive therapy for depression. Interestingly, this study found that patients over age 50 had significantly better response to CCT when stimulated with tDCS, while improvement in younger patients did not reach statistical significance; the authors’ speculative explanation for this finding is that tDCS-induced neuroplasticity may mitigate the effects of subclinical prefrontal atrophy in older patients (Brunoni 2014). Conversely, concurrent CCT was also independently shown to enhance response to tDCS (Segrave 2014).
Notably, results of studies evaluating the cognitive effects of tDCS are heterogeneous (Tremblay 2014, Horvath 2015). This heterogeneity is associated not only with variable experimental methods, but also with inter-subject variability in size, shape, and fat tissue content of the patient’s scalp. A recent meta-analysis by Horvath et al. challenged the idea that tDCS has any reliable cognitive effects (Horvath 2015), although a subsequent critique of this review demonstrated that it failed to account for significant differences in effects with different treatment parameters - for instance, the pooled data sets included studies with both short treatment durations and long treatment durations, which are known to produce conflicting results due to calcium overflow mechanisms. Furthermore, the review was found to contain several errors, incorrectly/incompletely cited data, and other conceptual flaws in the data pooling methods (Antal 2015).
As a result of these findings, multi-site trials are in progress investigate larger sample sizes (ClinicalTrials.gov identifiers: NCT01562184, NCT01644747, NCT01346306).
tDCS has also been implicated as a potential treatment for several other neuropsychiatric disorders. Early studies suggest efficacy of tDCS as an adjunct in stroke rehabilitation (de 2015) and fibromyalgia (Fregni 2006, Marlow 2013), while more recent studies have suggested possible uses in Alzheimer’s dementia (Hsu 2015).
The most prominent evidence in stroke rehabilitation has been found in post-stroke aphasia; anodal excitation of the damaged Broca’s area and cathodal inhibition of the contralesional area led to faster overall recovery and improved speech fluency. Furthermore, excitation of the contralesional area may accelerate cortical relocation to the nondominant hemisphere in patients with large lesions (de 2015). While a recent Cochrane review found that changes in post-stroke aphasia did not reach statistical significance, its sample size was limited by the fact that it only included studies that reported picture naming as an outcome measure rather than speech fluency and recovery speed (Elsner 2015).
Motor cortex stimulation has also been used for fibromyalgia. Fregni et al demonstrated at least 21 days of improvement in pain scores for fibromyalgia patients with five days of anodal excitation over the primary motor cortex when compared to left DLPFC and sham stimulation (Fregni 2006). Several other studies have replicated this phenomenon (Marlow 2013). A recent Cochrane review did not find a significant effect for chronic pain syndromes overall, but this review did not make a distinction between fibromyalgia and other causes of chronic pain in its pooled analysis (O’Connell 2014).
Promising findings for improving attention, memory, and executive function have led to recent research aimed at evaluating DLPFC stimulation for alleviation of cognitive deficits seen in Alzheimer’s dementia. A recent randomized controlled double-blind trial in patients taking memantine found that a series of 10 daily sessions led to statistically significant improvements in mini-mental status examination score, digit span, and performance IQ; these effects were noted immediately and sustained for at least two months after treatment (Khedr 2014). While this study was limited by small sample size, a subsequent meta-analysis fo