Home-based tDCS: are we there yet?

This editorial addresses the paper by Ko and colleagues (in press)¹, published in Stroke, on the effectiveness of home transcranial direct current stimulation (tDCS) as an adjunct for previously shown effective in-home computerized cognition treatment for post-stroke cognitive rehabilitation². Previous studies, amongst which the largest is Julius Fridriksson’s study of non-futility in the use of tDCS in post-stroke language rehabilitation (2018)³, have advocated for further investigation of tDCS applications in post-stroke rehabilitation, and the present study is a worthy sequel. This is the first effectiveness study that adds to previous feasibility studies of home-based, remotely supervised tDCS⁴. Several aspects of the paper are of special interest as they are the themes of current research: (1) the focus on cognitive, non-language, deficits post-stroke, something that is a new and exciting field of investigation⁵, (2) the possibility of transferring neuromodulation (tDCS) from the clinical setting to the home for continuous rehabilitation for chronic conditions, using home-based devices and computerized cognitive training⁶, and (3) the application of neuromodulation in long-term and individualized schemes (6 months). The current study addresses all these three prerequisites that would advocate broader application of tDCS for at-home use in post-stroke rehabilitation.

I would like to discuss the design and results of the study, and then critically address concerns about neuromodulation in post-stroke rehabilitation.

Although the present study is not large, (comprising 26 post-stroke patients, with either left or right hemisphere damage who were randomized in two arms), it is the largest to-date, double-blind, randomized sham-controlled trial in the field for home-based tDCS, and has the longest use of tDCS (4 weeks, 5 sessions per week) along with computerized cognitive rehabilitation. Neuromodulation and language/cognitive rehabilitation studies, in general, frequently face the criticism of not being large, on the scale of pharmacological studies. This limitation is understandable as these patients usually face a combination of motor, cognitive and emotional challenges on a daily basis. The high compliance (87%) in the study attests to the feasibility of home-based tDCS devices. So, this new study is an important proof-of-concept, albeit preliminary, study, as the authors admit. But is feasibility accompanied by effectiveness for non-language cognitive rehabilitation? The preliminary study by Ko and colleagues (in press) provides an answer: “yes, under certain conditions”.

Probably the most important findings of the study are that tDCS is more beneficial on a global cognitive measure such as the Montreal Cognitive Assessment (MoCA) for patients with moderate than mild impairments, and for those with left-hemisphere (LH) rather than right-hemisphere (RH) stroke. Are these differences important? The question of who will improve is an important one, not only in post-stroke rehabilitation, but also in rehabilitation of other neurological diseases. Unfocused populations and confounding patient conditions are among the main reasons that large trials may fail. Therefore, there is imperative to fine tune the targeted population in clinical trials, and the findings of the present study provide key insights in this direction.

With regard to the finding that patients with moderate impairment benefited more from tDCS than those with mild improvement, one could claim that this would be expected. The relative merits and limitations of methods to measure and quantify improvement continue to challenge the science of rehabilitation. Patients with moderate impairment have more capacity to improve, and it might be easier, for example, to improve 30 degrees in absolute percentage scale from 40 to 70 than from 70 to 100 (a perfect score indicating full recovery). Several methods to quantify individual patient’s improvement have been suggested, such as potential maximal gain⁷, which quantifies improvement in percent of remaining potential gain thereby homogenizing the effect of initial performance. Future studies of home-based tDCS should employ several methods to better characterize individual improvements.

The novel finding that patients with LH stroke improved more than those with RH stroke after LH tDCS, in combination with the impact of baseline performance on the effect of tDCS, (the latter finding replicating ⁸), has two implications. The first is that LH stroke patients are more impaired in executive functions (targeted by the treatment) or the MoCA (measured global cognitive outcome) than are RH stroke patients, and therefore they showed more improvement. The second account is that it is more beneficial to target the compromised (but probably still functioning) areas than non-compromised areas. That is, the LH dorsolateral prefrontal cortex that was targeted by the tDCS in both groups was in the compromised hemisphere in LH-stroke patients but in the noncompromised hemisphere in the RH-stroke patients. To resolve this issue, a study would be needed in which LH- and RH- stroke patients are matched for the targeted executive functions and MoCA at baseline, and the right DLPFC is targeted for RH-stroke patients. The present results nevertheless confirm the principle of ‘functional targeting’ (at least for the LH) as advocated by Bikson and Rahman⁹, in which only active cells and networks are changed by weak stimulation, namely those that are engaged by the behavioral task.

The most important contribution of this recent study is that it allows us to form hypotheses to move forward. To answer the question whether we are ready to implement home-based tDCS, we need larger randomized clinical trials that: (a) target LH-stroke patients with executive functions rehabilitation, and (b) target RH-stroke patients with executive functions rehabilitation and RH stimulation. Such trials should be stratified by symptom severity for both lesion sites, and implement hemisphere-specific language outcomes. Despite the preliminary nature of this study on the efficacy of home-based tDCS, the present study provides new hope for accessible and continuous rehabilitation of cognitive symptoms in patients with stroke and their families.

Non-standard Abbreviations and Acronyms.

tDCS

transcranial Direct Current Stimulation

LH

left-hemipshere

RH

right hemisphere