Measurement of transcallosal inhibition in traumatic brain injury by transcranial magnetic stimulation.

Primary objective: To study whether transcallosal inhibition (TCI) can evaluate the severity of traumatic brain injury (TBI). Research design: Case-control study. Methods and procedures: Twenty patients with a chronic TBI and 20 control subjects were studied. The following transcranial magnetic stimulation parameters were checked; resting motor threshold, central motor latency times, onset latency of TCI, duration of TCI, transcallosal conduction times and amount of TCI. The severity of TBI was evaluated using the Glasgow Coma Scale (GCS). Main outcome and results: The amount of TCI was significantly lower in the patients than the control subjects (p < 0.001). The amount of TCI was highly correlated with the GCS (r = 0.787, p < 0.001). Conclusion: An assessment of TCI was found to be a more sensitive and useful method for an evaluation of the severity of TBI.


Introduction
The functional integrity of the corpus callosum (CC) that connects homologous motor cortices can be tested electrophysiologically in humans by transcranial magnetic stimulation (TMS) [1]. When TMS is performed during an ongoing tonic voluntary contraction, the activity of the muscles ipsilateral to the site of the cortical stimulation is temporarily suppressed. This transcallosal inhibition (TCI) is not detected in patients with CC lesions [2]; hence, a transcallosal route is presumed to account for it. In recent years, TCI has been found to be a useful diagnostic tool in patients with multiple sclerosis, corticobasal ganglionic degeneration, and progressive supranuclear palsy [3][4][5].
CC lesions are commonly detected in patients with traumatic brain injury (TBI) [6][7][8][9][10][11][12]. The diffuse axonal injury (DAI) due to acceleration-deceleration and rotational forces is considered to be an important factor in the formation of a CC lesion [6-8, 10, 11]. A majority of the TBI survivors recover from coma and show remarkable progress toward regaining their preinjury functional abilities. However, patients with TBI often have cognitive impairments, including attention, memory, and executive function deficits [8,13]. When attempting to correctly evaluate the severity of TBI, a method that requires little cooperation from the patient is desirable so that any cognitive impairment is excluded.
Our hypothesis was that the TCI method could detect an abnormality of the CC in patients with TBI. The TCI method is simple and objective and requires little cooperation from the patient; hence, this method has the advantage of excluding any cognitive impairment when the severity of the TBI was evaluated. To our knowledge, imaging was the only available technique that could evaluate the severity of DAI while the patient was alive [9,12,14]. Therefore, TCI may be a unique diagnostic tool for electrophysiologically evaluating the DAI by monitoring the CC lesions. In the present investigation, we examined whether TCI could evaluate the severity of TBI that was indicated by the Glasgow Coma Scale (GCS) [15].

Insert table 1 about here
The mean period after their TBI was 25.3 ± 10.7 months. We used the GCS scores that were obtained at the time of their first admission after accident to evaluate the severity of TBI (scale mean, 8.7 ± 2.5; range, 6 to 14). All the patients who were enrolled for the study fulfilled the following criteria: (1) age from 18 to 60 years, (2) drugs that are known to influence the excitability of the central nervous system (for example, an anti-epileptic or psychoactive drug) had not been administered in the past month, and (3) no concomitant cervical or upper limb injury that could affect conduction along the peripheral nerves and spinal cord. The exclusion criteria for patients in this study were as follows: (1) the subjects, their guardians, or legal representatives were unwilling to give consent for participating in the study and (2) the patient had a history of neurological disease.
Twenty normal subjects, who were matched for age and sex (15 men and 5 women, aged 20-56 years; mean age, 35.3 ± 11.2 years), participated as control subjects in the study. They did not have any history or clinical evidence of any neurological disease.
Informed consent for the study was obtained from all the patients and control subjects. because this direction is the most effective for eliciting TCI [16]. The electromyographic (EMG) responses were recorded bilaterally from the first dorsal interosseous muscle (FDI) using Ag-AgCl surface electrodes. The EMG signal was amplified, filtered, and stored in a personal computer for off-line analysis (Neuropack; Nihon Koden, Tokyo, Japan).
The resting motor threshold (rMT) was defined as the lowest stimulator output that could produce motor evoked potentials (MEPs) with a peak-to-peak amplitude that was greater than 50 µV in at least 5 of 10 trials [17]. In a TCI session, each motor cortex was stimulated 20 times, with an intensity of 150% rMT, during unilateral maximal tonic contraction of the ipsilateral FDI. During each stimulation, the subjects maintained a sustained maximal tonic contraction of the FDI muscle, with visual and auditory feedback, for approximately 2 s. To avoid central or peripheral fatigue during maximal tonic muscle contraction, the subjects rested for 3 min after a series of 10 stimuli. The stimuli at a frequency of 0.1 Hz were applied over the cortex. The peripheral latencies were obtained by magnetic stimulation of the cervical nerve roots using a 90-mm circular coil.

Response indices
The peak-to-peak amplitude of 10 averaged MEPs of the contralateral EMG obtained with an intensity of 120% rMT was determined. The central motor latency times (CMLTs) were calculated by subtracting the longest peripheral conduction time following magnetic stimulation of the cervical nerve roots from the onset latency of the cortically elicited contralateral EMG response.
The TCI parameters were evaluated by rectifying and averaging 20 EMG signals of the active FDI ipsilateral to the site of the cortex stimulation. The TCI was quantified by the period of relative EMG suppression after the stimulus, i.e., when the EMG activity dropped below the background activity. The onset latency of TCI was measured from the stimulus to where the EMG activity clearly fell below the mean amplitude of the EMG activity in the 100 ms before the stimulus. The duration of TCI was measured from its onset to where the EMG activity again reached the mean amplitude of the EMG activity before the stimulus. The transcallosal conduction times (TCTs) were determined by subtracting the onset latency of the corticospinally mediated contralateral responses from the onset latency of TCI in the same FDI. The area of suppressed EMG activity below the mean amplitude of the EMG activity before the stimulus was also averaged.
The amount of TCI was then defined as the percentage of this mean suppressed activity in the mean amplitude of the EMG activity before the stimulus [18]. That is to say, the more the EMG activity was suppressed the greater was the amount of TCI.

Statistical analysis
Any parameters of the MEPs (rMT, CMLTs, amplitude) and TCI (amount of TCI, duration of TCI, TCTs) of the patients that exceeded mean value ± 2.5 SD (the pooled data across the left and right sides) of the control subjects were considered to be abnormal. The observed data (MEPs and TCI parameters) from the patients and the control subjects were compared by using a Mann-Whitney U test. Any possible correlation between the amount of TCI and GCS was determined by using the Spearman rank correlation test.

Results
The data for the MEPs and the TCI parameters that were obtained for the control subjects and the patients are listed in Table 2.

MEPs parameters
We did not observe any difference between the mean data of the control subjects and the patients for the rMT, CMLTs, or amplitude ( Table 2). In addition, we did not observe a significant difference in laterality in the MEPs parameters (rMT, CMLTs, and amplitude) for the control subjects and patients. An increased rMT was found in 15% (3 of 20), a decreased amplitude in 15% (3 of 20), and prolonged CMLTs in 20% (4 of 20) of the patients with TBI on either or both sides.

TCI parameters
The amount of TCI was significantly lower in the patients than in the control subjects (p < 0.001). One patient (patient no. 13) did not display TCI on either side. Therefore, we excluded her from the TCI correlation study. In 70% of the patients (14 of 20), the amount of TCI was abnormal. However, when compared with the control subjects the latency onset of TCI, the duration of TCI, and the TCTs of the patients with TBI were not significantly different. The typical TCI that were observed for the control subjects and patients are shown in Figure 1. The amount of TCI was significantly correlated with the GCS (Figure 2; r = 0.787, p < 0.001).

Discussion
There was a significant difference between the control subjects and the patients with TBI for the amount of TCI, but not rMT, CMLTs, or amplitude. This result suggested that the TCI method of evaluating CC was more sensitive for detecting abnormal findings in patients with TBI than the method of evaluating the excitability of the cortex and corticospinal tract function by TMS.
The GCS score has been reported to be strongly correlated with the severity of the TBI [19][20][21]. Therefore, the severity of TBI is usually described by the amount of impaired consciousness as defined by the GCS. Moreover, the GCS has a significant correlation with a CC lesion [6,10]. TCI methods by TMS can evaluate the function of integrity of the CC connecting homologous motor cortices [1][2][3][4][5]. In our study, the amount of TCI was significantly correlated with the GCS. Based on these reports, it may be considered that the TCI method by TMS is useful for evaluating the severity of TBI by monitoring the CC function.
No significant difference in the MEPs parameters (rMT, CMLTs, and amplitude) of the control subjects and patients was observed. The normal rMT and CMLTs in our patients with TBI may have indicated that they had only minimal or no loss of motor cortex excitability and conduction in the pyramidal tract. These results indicated that the abnormal findings in patients with TBI could not be easily detected by the well established method of determining the corticospinal tract function by using TMS. In contrast to our results, Chistyakov et al. have reported that patients with TBI had a higher rMT than the control subjects [22,23]. However, this report had compared the more damaged side in TBI patients with that in control subjects [23]. In our investigation, the patients with TBI exhibited normal rMT and did not show significant laterality. Chistyakov et al., in a follow-up study, have also reported some improvement in the observed high rMT [22]. In fact, in our study, the period after the TBI was longer than the postaccident period of the Chistyakov's study that reported the high rMT of patients with TBI. In addition, a severe brain injury that might preclude consciousness and voluntary movement did not invariably predicate an abnormal rMT and CMCT [24].
Based on these reports, the normal rMT observed in our study was not thought to be specific with regard to the studies that reported high rMT in patients with TBI. In addition, the TCI route is thought to be longer than rMT, CMLTs, and amplitude because in addition to the corticospinal tract, the TCI route included the commissural fibers via the CC. Therefore, the TCI method might detect the abnormal findings in patients with TBI more than the well established method of evaluating corticospinal tract function by using TMS.
The patients with TBI often have cognitive impairments, including attention, memory, and executive function deficits [8,13]. A recent study reported that the area of the CC in patients with TBI correlated with the memory function [11]. Therefore, the TCI that can monitor the CC function might also be correlated with cognitive impairment in patients with TBI. In addition, the TCI as well as imaging is useful for accurately evaluating the patients with TBI because TCI is a simple method that requires little cooperation from the patient. Moreover, the CC lesion is considered to reflect the DAI [6-8, 10, 11]; therefore, TCI may be a diagnostic tool for electrophysiologically evaluating the DAI in vivo by monitoring the CC lesions. It has been recently considered that the diffusion MR tensor imaging may be able to detect the DAI [9,14]; hence, some in vivo studies used the diffusion MR tensor imaging to detect the DAI [9,12,14]. Therefore, we must study the correlation between diffusion MR tensor imaging and the TCI to suggest the utility of TCI for evaluating the DAI in patients with TBI.
In conclusion, we confirmed that the TCI was positively correlated with the severity of the TBI. Moreover, an assessment by TCI was found to be a more sensitive method for detecting abnormal findings in patients with TBI than the well established method of evaluating corticospinal tract function by using TMS. This study is the first to report the evaluation of the severity of TBI by using a TCI method that tests the functional integrity of CC by using TMS. The amount of TCI had a significant correlation with GCS in 19 patients with TBI (r = 0.787 p < 0.001).

Legends
TBI: traumatic brain injury; TCI: transcallosal inhibition; GCS: Glasgow coma scale Table 1 Clinical characteristics of patients with traumatic brain injury.