Abstract
Keywords
Introduction
Non-invasive brain stimulation (NIBS) is a safe and powerful method used to modulate human brain function. Repetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS) are two of the most promising NIBS modalities for the modulation of brain function (Nitsche et al., 2000; Pascual-Leone et al., 1994). Numerous studies have evaluated the therapeutic benefits of NIBS modalities with rTMS and tDCS in cognitive neuroscience, psychiatry, neurology, and neurorehabilitation (Lefaucheur et al., 2014; Brunoni et al., 2012). In addition, meta-analyses have demonstrated that NIBS has beneficial therapeutic effects on motor recovery after stroke (Hsu et al., 2012; Le et al., 2014). However, it remains unclear how NIBS protocols should be optimized to maximize their therapeutic effects. One method that may maximize the effects of NIBS is multi-modal and multi-site stimulation. Few studies, however, have investigated the effectiveness of bihemispheric stimulation using NIBS (Takeuchi et al., 2009; Sasaki et al., 2014; Koenigs et al., 2009; Sung et al., 2013; Wang et al., 2014; Meng et al., 2014). The connection of bilateral hemispheres might contribute to motor function recovery, so we suggest that stimulating bilateral hemispheres using different NIBS modalities could be an effective method for enhancing motorfunction.
Motor training-induced cortical excitability plays a major role in the motor recovery of chronic stroke patients (Nudo & Friel, 1999; Liepert et al., 2000; Murphy & Corbett, 2009). Recent studies have also highlighted that the combination of NIBS with goal-oriented motor training can produce long-term effects in motor function (Chang et al., 2010; Emara et al., 2010; Avenanti et al., 2012). Therefore, learning of complex and specific skills is a crucial component of higher cortical excitability and motor performance when applying NIBS in recovering strokepatients
In this study, we investigated the effects of dual-mode NIBS over the bilateral primary motor cortices (M1s) in conjunction with motor training, according to different combinations of NIBS modalities, and determined the most effective combination method to enhance neural plasticity and motor performance in chronic stroke patients. We hypothesized that the combination of NIBS modalities and sequential motor task training would induce consolidation of skill learning and recovery of motor performance.
Materials and methods
Participants
The two inclusion criteria were: (1) >3 months after the onset of the first-ever stroke, and (2) a unilateral upper limb motor deficits that allowed movement of the wrist greater than 20°, and movement of each finger individually. The exclusion criteria included patients who: (1) had a lesion of the primary motor cortex, (2) had a history of intractable seizure, (3) had an intracranial metallic implant, and (4) had a history of any neuropsychiatric comorbidity other than stroke. Twenty-three chronic stroke patients with hemiparesis were enrolled in this study. Two subjects who did not respond to motor-evoked potential (MEP) and one subject who was not able to perform sequential motor tasks were excluded. Thus, a total of 20 chronic stroke patients were ultimately included in the study. Thirteen of the patients were male. The mean age of all patients was 58.6±8.4 years. Demographic and clinical characteristics of the participants are shown in Table 1. Written informed consent was obtained from all participants prior to inclusion in the study, and the study protocol was approved by the institutional review board (IRB number: 2012-06-019).
Experimental design
The study was designed as a randomized, double-blind, crossover study. All participants completed five separate sessions wherein different dual-mode combinations of tDCS and rTMS were administered. Sessions were arranged in a counterbalanced order with at least a 24-hour interval between sessions. Each session had simultaneous, preconditioning, or sham bihemispheric dual-mode stimulation, which consisted of contralesional anodal or cathodal tDCS and lesional rTMS without any difference in total tDCS duration, rTMS pulses, or motor training duration. tDCS was delivered for 10 minutes at an intensity of 2 mA using saline-soaked 5×5-cm sponge electrodes and constant-current electrical stimulation using a Phoresor II Auto PM850 (IOMED, Salt Lake City, UT, USA). Active stimulation electrodes were placed over the contralesional M1, while reference electrodes were placed over the ipsilesional supraorbital area. The rTMS was applied in 100 trains of 10 Hz rTMS with 90% RMT over the lesion side motor cortex corresponding to the paretic hand and repeated 10 times with a 50-secondinter-train interval. A total of 1,000 pulses were delivered over the course of 10 minutes.
Conditions 1 and 2 included simultaneous application of cathodal or anodal tDCS, respectively, over the contralesional primary motor cortex (cM1) combined with 10 Hz rTMS over the ipsilesional primary motor cortex (iM1); conditions 3 and 4 included a preconditioning application of cathodal or anodal tDCS, respectively, followed by 10 Hz rTMS over the iM1; condition 5 included sham tDCS over the cM1 with 10 Hz rTMS over the iM1.
Each stimulation session was 20 minutes in duration. All participants were equipped with tDCS tools during the entire stimulation session in order to blind participants with respect to the actual treatment received. In the simultaneous sessions, sham-mode tDCS was applied during the first 10 minutes. tDCS was switched on for 10 seconds at the beginning of the stimulation session and then faded off to diminish the perception of sham. Anodal or cathodal tDCS and rTMS were applied simultaneously during the final 10 minutes of the stimulation session. In the preconditioning session, anodal or cathodal tDCS and rTMS were applied during the first 10 minutes followed by rTMS during the final 10 minutes. Active stimulation and cortical excitabilities were measured by the amplitude of MEP. Sequential hand motor task tests were performed to assess changes in functional hand movement. The measurements were performed prior to and immediately after each intervention (see Fig. 1).
Determination of motor hot spots and Resting Motor Thresholds (RMTs)
Patients were comfortably seated with both hands pronated on a pillow. Electromyography (EMG) recordings from the contralateral first dorsal interosseous (FDI) muscle were acquired with silver-silver chloride surface electrodes using a muscle belly-tendon montage. EMG activity was amplified using the Synergy electromyography EP system (Medelec, Oxford, UK), and the data were band-pass filtered at 10 to 2,000 kHz. The optimal scalp location (“hot spot”) was determined using a Rapid2® stimulator TMS system (Magstim Company Ltd, Wales, UK) and a 70-mm figure-of-eight coil. The handle of the coil was oriented to a direction posterior to the midline at a 45° angle so as to cause electromagnetic current to flow perpendicularly to the central sulcus, and the stimulator was moved over the scalp in 1-cm increments. Once a hot spot was identified, a single-pulse TMS was delivered to the location to determine the RMT, defined as the lowest stimulus intensity necessary to produce MEP with ≥50-mV peak-to-peak amplitude in five of 10subsequent trials. Real-time EMG confirmed that muscles were relaxed prior to stimulation.
Measurement of cortical excitability and rTMS application
To observe motor cortical excitability, single magnetic stimulations at 120% of the RMT before each session were administered over the motor hot spot of the affected hemisphere using a 70-mm figure-of-eight coil. The MEP was recorded on the contralateral FDI muscle. The stimuli were delivered at 5-second intervals. Ten sweeps of data were collected, and the mean peak-to-peak amplitude of the MEP was calculated and designated as the baseline MEP magnitude.
rTMS was delivered to the scalp over the hot spot of the affected hemisphere at an intensity of 90% RMT and frequency of 10 Hz using a Rapid® II stimulator with two Booster Modules (Magstim Company Ltd, Wales, UK). The motor hot spot was stimulated by holding the figure-of-eight coil tangentially to the skull.
Motor training
The sequential finger motor task paradigm was a repetitive push-button task in response to a number displayed on a computer. Each patient was seated 50 cm from the front of a 15-inch monitor. A random single-digit number (1, 2, 3, 4, or 5) was displayed at the center of the monitor screen. Each button was labeled with a number representing the finger to be used; 1, 2, 3, 4, and 5 represented the thumb, index, middle, ring, and little finger, respectively. The patients were instructed to push the numbered response buttons as accurately and quickly as possible using the paretic fingers (see Fig. 1.B).
Behavioral measures
Performance in the sequential finger motor task
Patient performance in the sequential finger motor task was assessed by movement accuracy (MA; the ratio of correct button presses out of the maximum potential score) and movement time (MT; the time between the presentation of each number on the monitor and the subsequent button press, expressed in milliseconds) using SuperLabPro® 2.0 software. Performance in the motor task blocks was assessed before and immediately after each intervention.
Statistical analyses
All analyses were performed with the use of SPSS 23.0 (SPSS, Inc., Chicago, IL, USA). The Shapiro-Wilk test revealed that the data were not normally distributed in our study. Therefore, Wilcoxon signed rank tests were used to evaluate differences in motor performance and cortical excitability between post-interventional and baseline values among the various conditions. The Kruskal-Wallis test was performed to evaluate differences among baseline values and post-interventional rates of change. The Mann-Whitney U test was used to compare differences between specific paired conditions in post hoc analysis for multiple comparisons. P-values <0.05 were considered statistically significant.
Results
None of the subjects reported severe adverse effects during or after the experiments, and all subjects tolerated dual-mode stimulation without inte-rrupting the experiment.
Changes in hand motor performance
In Fig. 3, Patient motor performance was significantly improved in MA under conditions 1 and 5 (P = 0.022 and 0.030 respectively) and in MT under conditions 1, 3, and 5 (P = 0.000, P = 0.014, and 0.047 respectively). Baseline values for MA and MT were not significantly different among the intervention methods (P = 0.850 and 0.983, respectively). On the other hand, post-interventional rates of changes for MT were significantly different between the five conditions (χ2 = 21.878, P = 0.000). Post hoc analysis for multiple comparisons showed the rate of change of condition 1 was significantly higher than those of other conditions (P = 0.000, 0.003, 0.000, 0.002, respectively).
Changes in cortical excitability
No statistically significant differences were observed in the baseline values of cortical excitability among the interventions (P = 0.917). Cortical excitability measured by amplitudes of MEP was significantly increased after intervention under conditions 1 and 5 (P = 0.001, P = 0.040). There were no significant differences in MEP amplitude change among the five conditions (χ2 = 7.285 P = 0.122). However, the degree of MEP amplitude change was significantly higher under condition 1 than under conditions 2 or 4 (P = 0.013, 0.043 respectively) (see Fig. 2).
Discussion
We investigated the effects of dual-mode bihemispheric stimulation by applying high-frequency rTMS over the ipsilesional M1 combined with simultaneous and preconditioning tDCS over the contralesional M1 with motor training. Our results demonstrated that movement time in sequential motor tasks was significantly improved following high-frequency rTMS over the ipsilesional M1 combined with simultaneous cathodal tDCS over the contralesional M1, compared to other combinations of stimulation, including sham tDCS with high-frequency rTMS, suggesting that this dual-mode stimulation method may be more effective than other combinations of dual-mode stimulation and unimodal stimulation.
Bihemispheric brain stimulation is based on the interhemispheric competition theory, which hypothesizes that unilateral lesions in the brain cause an imbalance between dominant and non-dominant hemispheres, and that this may interfere with the natural recovery process, especially in the injured brain (Nowak et al., 2009; Takeuchi et al., 2012). Theoretically, increasing the excitability of the ipsilesional cortical region and decreasing the excitability of the contralesional cortical region may help to restore the balance between both hemispheres and thus promote the recovery process (Gorsler et al., 2003; Takeuchi et al., 2012). Several studies have been based on this theory. They showed improvements in motor function after applying ipsilesional high-frequency orcontralesional low-frequency rTMS, and after applying ipsilesional anodal or contralesional cathodal tDCS, in stroke patients (Fregni et al., 2005; Kim et al., 2006; Takeuchi et al., 2008; Hummel et al., 2010). To date, few studies have compared the effects of bilateral stimulation with those of unilateral simulation. One study found that bihemispheric tDCS facilitates greater improvement than does unilateral tDCS (Lindenberg et al., 2010); however, another study did not conclude that bilateral tDCS was superior to unilateral tDCS (Kidgell et al., 2013). Although an rTMS study using bilateral stimulation with low-frequency, high-frequency, and intermittent theta burst found a greater beneficial effect than with single rTMS (Takeuchi et al., 2009; Sung et al., 2013; Yamada et al., 2013), methodological discrepancies in the stimulation protocols affected the consistency of these results. Moreover, none of these studies examined dual-mode NIBS (tDCS with rTMS), nor have any directly compared simultaneous simulation to preconditioning stimulation.
The appropriate timing of stimulation should be considered when combining modalities. In this study, we applied preconditioning tDCS, which did not result in a statistically significant improvement in motor function or cortical excitability after intervention. However, our results did show that both corticomotor excitability and hand function tended to increase after intervention under simultaneous tDCS. These conflicting results may be due to the timing of tDCS application. In the preconditioning method, we performed rTMS after completion of tDCS with a 10-minute interval between stimulation and hand motor training.
According to the model of homeostatic plasticity, training-induced plasticity is also affected by preceding postsynaptic neural activity. A recent study showed that voluntary contraction after tDCS tended to reverse tDCS-induced cortical motor excitability (Thirugnanasambandam et al., 2011). However, another study demonstrated that concurrent tDCS with the patient pushing a ball or doing a voluntary grip exercise facilitated greater cortical excitability (Antal et al., 2007; Kim & Ko, 2013). It seems that there are subtle considerations that affect the outcomes of combined stimulation methods and motor task training with respect to synaptic plasticity. However, several studies described the impact of these neuromodulation interventions on procedural learning consolidation, suggesting that it may be worthwhile to test the effects of intervention with both rTMS (Muellbacher et al., 2002) and tDCS (Reis et al., 2009; Tecchio et al., 2010) performed just after training. Combining NIBS without motor training would likely be a suboptimal experimental approach with limited functional benefits in stroke patients.
The current study had several limitations. First, 24-hour interval between the stimulations is a short period of time. Second, the high variability of the baseline MEP and some MEP results did not show significant differences among conditions. The large standard deviations of stroke duration of participants may affect the post-interventional MEP results. Third, this study did not include the group of both sham rTMS and sham tDCS. Fourth, there was no measurement of corticospinal excitability after NBS in this study. This is one of limitations. The measurement of bilateral MEP is needed to confirm the effect of interhemispheric interaction on cortical excitability after dual-mode stimulation. The further study with the measurement of bilateral MEP will be needed to confirm the effect of interhemispheric interaction on cortical excitability after dual-mode stimulation. Finally, this study was designed to include only a single session of measuring reaction times and motor thresholds in high-functioning stroke patients. Future studies using multisession interventions need to be conducted to clarify the effects of dual-mode combination methods for motor recovery after stroke.
In summary, bihemispheric dual-mode stimulation using rTMS combined with tDCS enhanced cortical excitability and hand motor function in patients suffering from stroke. The present findings expand our understanding of the combined use of non-invasive brain stimulation and provide evidence for an effective combination therapy for the motor rehabilitation of stroke patients.
Declaration of conflicting interests
None of the authors have potential conflicts of interest to declare with respect to the research, authorship, and/or publication of this article.
Footnotes
Acknowledgments
This study was supported by grants from the National Research Foundation of Korea (NRF) funded by the Korean government (MSIP) (NRF-2014R1A2A1A01005128).