No add-on effects of Unilateral and Bilateral Transcranial Direct Current Stimulation on Fine Motor Skill Training Outcome in Chronic Stroke. A Randomized Controlled Trial

Background: Transcranial direct current stimulation (tDCS) may improve motor recovery after stroke. This study investigated if uni- and bihemispheric tDCS of the motor cortex can enhances ne motor training outcome and transfer to clinical assessments of upper motor function. Methods: In a randomized, double-blinded, sham-controlled trial, forty chronic stroke patients underwent ve days of ne motor skill training of the paretic hand with either unilateral or bilateral (N=15/group) or placebo tDCS (N=10). Immediate and long-term (three months) effects on training outcome and motor recovery (Upper Extremity Fugl-Meyer, UE-FM, Wolf Motor Function Test, WMFT) were investigated. Results: Trained task performance signicantly improved independently of tDCS in a curvilinear fashion. Anodal, but not dual tDCS resulted in a steeper learning curve on the UE-FM. Neither training nor combined training-tDCS improved WMFT performance. Conclusions: Fine motor skill training can facilitate recovery of upper extremity function. Minimal add-on effects of tDCS were observed.


Background
Transcranial direct current stimulation, a non-invasive brain stimulation technique, can improve motor training outcome in stroke 1 . So far, few studies have investigated the effects of unilateral and bilateral tDCS on ne motor skill training outcome. This training may be particularly suited to improve motor function in severely impaired patients lacking preserved extensor muscle function for more complex training paradigms 2,3 . A previous study that administered unilateral M1-tDCS concurrently with four weeks of daily grip force training failed to show stimulation effects. However, tDCS was administered at a very low current intensity (0.5mA). The present study aimed to address three open questions: (1) Does ne motor skill training improve trained motor function, and/or generalize to clinical assessments of upper extremity function in chronic stroke?; (2) If immediate and long-term training gains occur, are they enhanced by concurrent unilateral or bilateral M1-tDCS?; (3) Does unilateral or bilateral tDCS result in differential transfer effects? We hypothesized that both anodal and dual tDCS would enhance motor training outcome and transfer to clinical assessments of upper motor function 4,5 .

Materials And Methods
Forty patients with chronic (> 6 months post-stroke; Table 1 for details) right or left hemispheric ischemic or hemorrhagic stroke participated in a randomized, double-blind, sham-controlled study. Inclusion criteria consisted of occurrence of ischemic or hemorrhagic stroke at least 6 month prior to enrollment; ability to complete the motor training; no previous or subsequent strokes; no additional neurologic, medical or psychiatric disorders; and no concurrent use of CNS-affecting drugs. Patients were strati ed to the stimulation groups by baseline UE-FM score to receive ve consecutive days of ne motor skill training with anodal (N = 15), dual (N = 15) or sham (N = 10) tDCS. Patients, care providers and investigators collecting the data were blinded to the stimulation conditions. The study was approved by the ethics committee of the Charité Universitätsmedizin Berlin (Protocol: EA1/026/11), where all data was collected.
Participants gave written informed consent prior to study inclusion. Sample size calculations were based on previous uni-and bihemispheric tDCS studies 6,7 . Due to the small expected effect of sham stimulation we proposed an unbalanced group size for the power analysis, which revealed the necessary group size for the stimulation group n = 15 and for the sham group n = 10 to achieve a statistical power of at least 95% (2-tailed, alpha = 0.05). Figure 1 displays the ow-chart of the study.

Primary and secondary research question
The primary research question was whether unilateral or bilateral tDCS in combination with a ne motor skill training improves UE-FM scores in stroke patients. Changes in training performance and WMFT score were secondary outcomes.

Clinical assessment
Subjects underwent standardized motor function and impairment assessments using UE-FM and WMFT prior to and immediately after the intervention. The UE-FM examines multi-joint movements of the upper limb (max. score = 66, lower scores = greater impairment) 4 . The WMFT comprises 15 time-based items ranging from whole arm movements to ne nger control. WMFT completion times were logarithmized to account for skewed data distribution 5 . This score has a maximum value of 2.08s [log] with lower values re ecting better arm function. All tests (including the training task without-tDCS) were repeated three months later to investigate potential long-term effects of tDCS on motor function. Assessments were videotaped and analyzed by two independent raters.
Motor Training Similar to a previous study 8 , the training consisted of isometric abductions with the paretic thumb, which was placed in a sling attached to a Grass® Force-Displacement-Transducer FT10 (Grass Instruments).
Velcro straps xated the forearm to minimize unwanted movement. Signal software (Cambridge Electronic Design Ltd.) was used for data acquisition and task presentation. Force displacement was ampli ed and digitized using a CPT22 AC/DC Straining Gage Ampli er (Grass® Technologies; ampli cation: 2000Hz, high lter: 3Hz). Each training day consisted of eight blocks of 30 trials, four seconds/trial. A target force window was de ned between 30-40% of the individual maximum force output on the y-axis and 2800-3200ms on the x-axis. Abductions in the target window were considered "hits".
tDCS Stimulation tDCS was administered using a battery driven direct current stimulator (DC-Stimulator PLUS, NeuroConn).
The anode (5x7cm²) was placed above the ipsilesional M1 (C3/C4 of the international 10-20 EEGsystem, depending on the lesion). The cathode was placed above the contralesional supraorbital ridge (anodal, size 10x10cm²) or the contralesional M1 (dual, size 5x7cm²). The current was increased to 1mA over 10 seconds and lasted for 25 minutes. The sham group was pseudo-randomly assigned to a montage (50% anodal/dual) and received 30 seconds of tDCS to ensure the typical initial tingling sensation. A second investigator con gured DC-Stimulator to ensure investigator blinding.
Performance declines during the follow-up were comparable between groups ( Fig. 2A).

Discussion
Previous studies with comparable isometric pinch tasks demonstrated better trained task performance and UE-FM improvements, but not WMFT 2 , and improvements in the Jebsen-Taylor hand function and Grooved-Pegboard tests 10 . In line with these ndings, our study demonstrated long-term improvements of trained task performance and functionally relevant recovery of upper extremity function.
However, the present study did not con rm substantial add-on effects of unilateral or bilateral tDCS. Previous studies had only investigated effects of unilateral tDCS on ne motor skill training and reported mixed results. While bene cial effects of anodal tDCS on the shoulder-elbow-subscale of the UE-FM have been demonstrated, no effects on trained task performance were reported 2 . In contrast another study reported improved trained task performance with anodal tDCS, but not enhanced functional recovery 10 .
Both studies report no long-term effect of anodal tDCS on task performance, but bene cial effects on the shoulder-elbow-subscale lasting up to two months 2 . Hence, the mainly weak effects of tDCS reported previously, and the results of our study, question the utility of tDCS to enhance the outcome of this particular type of training.
Due to the small sample size, low statistical power, heterogeneity of the sample and drop-outs in the training task the results should be interpreted with caution. However, smaller proof-of-principle studies are imperative prior to investing limited resources into larger randomized control trials and our results in combination with previous studies do not encourage large follow-up trials. The sample comprised patient with left or right hemispheric ischemic or hemorrhagic lesions, which may have been suboptimal. Yet the sample accurately represents the general patient population and bene cial training effects were found despite variable lesion types, sizes and locations. Nonetheless a more homogeneous sample would be preferable with regard to tDCS. Furthermore, missing data was balanced across groups and the mixed effects models used are robust regarding missing data. Nonetheless, while there were no substantial effects at the group level, individual participants may have bene ted from tDCS and future studies are needed to include individualized modeling of current ow and functional imaging to investigate characteristics of potential responders.

Conclusion
We con rm signi cant performance improvements due to the ne motor skill training in patients with chronic stroke. Only limited add-on effects were induced by tDCS.