John Libbey Eurotext

Cortical silent period following transcranial magnetic stimulation in epileptic patients Volume 2, numéro 3, Septembre 2000

Transcranial magnetic stimulation (TMS) can be used for research in epileptic patients, as a safe and painless method of motor cortex stimulation. TMS has been used to determine the cortical TMS thresholds in different epileptic syndromes, the effects of anticonvulsant medications on cortical TMS thresholds, and the localization of seizure foci.

It has been showed that in different epileptic syndromes, the cortical TMS threshold is lower, the latencies of motor evoked potentials (MEP) are shorter and the amplitudes are higher than that of normal controls. These values are reversed in epileptic patients under anticonvulsant medication, that is, the TMS thresholds are higher, the latencies are longer and the amplitudes are lower than in normal groups [1-5]. Hufnagel et al. stated that the duration of epilepsy, localization of epileptic focus and seizure type did not affect the MEP and also noted that this conclusion is concordant with the concept of a global cortical disturbance in the excitation/inhibition balance in many forms of epilepsy [1, 2].

Silent period (SP) following TMS is mainly due to cortical inhibitory mechanisms [6-10]. It is generally agreed that the first 50 milliseconds of the cortical SP are due to peripheral factors, whereas the later part involves a cortical inhibitory mechanism [7]. There are a few studies of SP in epileptic patients [6, 11-13].

In this study, we aimed to differentiate the effect of antiepileptic drugs on cortical inhibitory mechanisms in different groups of epileptic patients. In order to evaluate this, cortical SP values in epileptic patients under medication but with uncontrolled seizures despite sufficient serum drug levels were compared with cortical SP in unmedicated epileptic patients, in epileptic patients with controlled seizures under medication and in normal subjects.

Subjects and method

Sixteen female and 9 male epileptic patients, aged between 11-42 years (mean 23 ± 10) were included in the study. All of the patients had been followed-up in Bakirköy State Hospital for Psychiatric and Neurological Diseases, Epilepsy Outpatient Clinic between February 1998 and July 1998. The control group consisted of 14 normal subjects (9 female and 5 male), aged between 22-38 years (mean 28 ± 5), who were normal on neurological examination and who had no epileptic seizure history or neurological disease. Epileptic patients were divided into 3 subgroups:

Epileptic patients with controlled seizures under medication (T+; S­). When the seizure frequency before the medication had decreased by at least 50% or more within 3 months of starting medication, seizures were regarded as controlled. Five female and 4 male (a total of 9 patients), aged between 14-42 years (mean 25 ± 11) were included in this group.

Epileptic patients with uncontrolled seizures under medication (T+S+). When the seizure frequency before the medication had decreased by less than 50%, remained unchanged or increased within the 3 months after starting medication, seizures were regarded as uncontrolled. There were 4 female, 2 male (a total of 6 patients), aged between 18-36 years (mean 28 ± 8) in this group.

Unmedicated epileptic patients with seizures (T­S+). Recently diagnosed epileptic patients having seizures and who had never received medicated or who had abandoned therapy for at least one year were regarded as unmedicated. There were 7 female, 3 male (a total of 10 patients), aged between 11-28 years (mean 19 ± 7) in this group.

The seizure types consisted of simple partial, complex partial, absence, myoclonic, generalized tonic-clonic seizures and their combinations. These seizure types were distributed randomly among the three subgroups of epileptic patients.

Carbamazepine was the antiepileptic medication given in 8 patients, valproic acid in 3, carbamazepine + lamotrigine in 1, carbamazepine + clonazepam in 1, carbamazepine + phenobarbital in 1 and phenytoin + phenobarbital in 1.

In medicated patients, antiepileptic drug (AED) plasma levels were detected before TMS. Having an AED level within therapeutic range was one of the inclusion criteria for medicated patients.

Cranial magnetic resonance imaging was performed in 14 patients and computerized brain tomography in 5. Imaging results were abnormal in 5 patients. These involved right mesial temporal sclerosis in 2 patients, right caudate nucleus ischemic lesion in 1, bilateral pachygyria-polymicrogyria in 1 and left parietal focal cortical atrophy in 1.

TMS was performed using a 9 cm diameter magnetic coil connected to Magstim 200 (Magstim Company LTD, Dyfed, UK). Maximum output was 1.5 Tesla. TMS threshold stimulation was defined as the stimulation which produces at least 3 MUPs of at least 100 µV in five, single, sequential stimulation. The initial output level for TMS was 30% of maximum output. Output was increased in steps of 5% until TMS threshold stimulation was reached. SP measurements were done at TMS threshold stimulation. Hand dominancy was determined by the Edinburgh Hand Preferency Test. If the left hemisphere was dominant then, it was stimulated by using face A of the magnetic coil (counter-clockwise) facing upwards, and if the right hemisphere was dominant then, it was stimulated by B facing upwards (clockwise).

EEG electrodes were used for recording the MEPs from the abductor digiti minimi (ADM) muscle, which is contralateral to the stimulated hemisphere. Recording was done during maximal contraction of the ADM muscle for measuring SP. During recording, display sweep time of electromyograph (Medelec Sapphire, Medelec Ltd, Surrey UK) was 100 millisecond, and display gain was 100 µV/division. Low cut and high cut filters were set to 3 Hz and 5 kHz, respectively.

The unpaired t-test, Mann-Whitney test, ANOVA and Tukey HSD tests were used for comparisons and statistical analysis. P values less than 0.05 were accepted for significance.


SP values of the patient and normal groups are shown in table 1. SP durations were longer in the T­S+ group compared to the normal control group. The difference between these two groups was statistically significant in the unpaired t-test (table 2). Comparing the T­S+ group to the T+S+ group, the difference between the SP values was not significant in the unpaired t-test. Similarly, there was no statistically significant difference between the SP values for the T­S+ and T+S­ groups. Neither was any significant difference found between the T+S+ and T+S­ groups. Comparing multiple groups to each other in the ANOVA with the Tukey HSD test, the difference between SP values for the normal control group and the T+S+ group was significant (table 2). Excluding the normal group, when SP values for patient subgroups were compared to each other in the Tukey HSD test, differences were not significant.


Cortical silent period (SP) following transcranial magnetic stimulation is mainly due to the cortical and peripheral inhibitory mechanisms [6, 7]. In a study by Cincotta et al., it was found that SP values for 8 patients with clonic seizures were longer than in the normal group and also longer than those of 10 patients with cryptogenic partial epilepsy without clonic seizures. In this study it was also reported that SP durations recorded from the healthy hemispheres were longer than those of epileptogenic hemispheres of these patients [6]. The authors claimed that interictal inhibitory mechanisms were probably overactive in epileptic patients with partial motor seizures originating from primary motor cortex. Furthermore, in order to balance the diseased hemisphere by a kind of compensatory mechanism, this inhibitory effect might be stronger in the healthy hemisphere than in the epileptic one [6]. Similarly, in our study, mean SP durations in epileptic patients were significantly longer than that in the normal group. In addition, SP differed among the patient subgroups. However, probably due to the small number of patients included in the subgroups, this difference was not statistically significant. The difference among the SP of patient subgroups in our study can be interpreted from the same point of view as in the study of Cincotta et al. [6]. SP of the medicated patients with controlled seizures were longer than those in the normal group but shorter than those in the unmedicated patients with uncontrolled seizures. Therefore, it can be proposed that the cortical inhibitory mechanisms in unmedicated patients could be overactive in order to suppress epileptogenic activity, whereas the condition which forces inhibitory mechanisms is no longer present in medicated patients with controlled seizures. In medicated patients with uncontrolled seizures, SP was the longest, probably due to the overactive cortical inhibitory mechanisms.


In conclusion, our findings probably indicate the enhanced interictal inhibitory mechanisms in epilepsy which is resistant to antiepileptic medication. However, the differences in SP values among the patient subgroups were not statistically significant. The reason for this could be that the epileptic seizure types were not homogeneously distributed among the patient subgroups.