Abstract
We studied the excitability of the visual and motor cortex in 36 patients with frequent migraine without aura (30 women, mean age 38.6 ± 10.0 years) before and after treatment with topiramate (100 mg/day) using transcranial magnetic stimulation. Treatment with topiramate resulted in reduction of both headache frequency (12.0 ± 1.3 to 5.8 ± 3.2 migraine days per month; P = 0.004) and cortical excitability: motor cortex thresholds increased on the right side from 43.8 ± 7.5% to 47.7 ± 9.2% (P = 0.049) and on the left side from 43.4 ± 7.0% to 47.2 ± 9.6% (P = 0.047), and phosphene thresholds increased from 58.9 ± 11.1% to 71.2 ± 11.2% (P = 0.0001). Reduction of headache frequency correlated inversely with an increase of visual thresholds and did not correlate with motor thresholds. The effect of topiramate in migraine prevention is complex and can not be explained simply by inhibition of cortical excitability.
Introduction
Migraine is a common and disabling neurological disorder characterized by attacks of pulsating headache accompanied by nausea, vomiting, photo- and phonophobia (1). The pathophysiology of migraine is complex and the aetiology remains elusive. There is a growing body of evidence demonstrating that neuronal excitability and responsiveness to sensory stimulation is increased in migraine, both at a cortical (2–5) and at a brainstem level (6). Studies using visual and auditory cortical evoked potentials have demonstrated that the deficit of cortical habituation in migraine normalizes following successful preventive treatment (7, 8).
Transcranial magnetic stimulation (TMS) is a powerful tool to assess cortical excitability (for review see (9)). TMS allows the analysis of the excitability of the motor and visual cortex during the migraine attack and in the interictal period (10–17). It has also been suggested as a useful tool to assess the effects of migraine-preventive drugs (18).
Topiramate has been introduced as a potent migraine-preventive drug (19–21). Pharmacological mechanisms of its antimigraine activity may include the blockade of voltage-gated sodium (Na+) channels, enhancement of γ-aminobutyric acid (GABA)-mediated neurotransmission, block of
The aim of this study was to investigate the effects of topiramate on cortical excitability in patients with frequent migraine using TMS.
Methods
The study was approved by the local ethics committee ‘Vein clinic ethics committee’. Patients' written consent was obtained prior to inclusion.
Study population
Thirty-six out-patients of the ‘Alexander Vein Foundation’ Headache Clinic Moscow suffering from frequent migraine without aura (MoA) were studied. Inclusion criteria were: (i) age 18–65 years; (ii) duration of illness > 6 months; and (iii) current migraine frequency 8–15 days/month. Exclusion criteria were: (i) symptomatic headache; (ii) chronic migraine (migraine frequency > 15 days/month); (iii) coexistence of tension-type headache; (iv) psychiatric comorbid disorders; (v) overuse of acute headache medication; and (vi) current intake of migraine preventive medication.
Study design
The study was designed as a prospective, open, non-randomized, exploratory investigation.
None of the participants was taking preventive drugs at least 4 weeks prior to the study or was taking drugs known to alter central nervous system excitability (sedatives, hypnotics, anticonvulsants, etc.).
Topiramate treatment was started with 25 mg/day, increased by 25 mg every week up to a target dose of 100 mg/day and continued so for the next 4 weeks. Hence, the titration period was 3 weeks and the treatment period was 5 weeks.
Clinical evaluations were performed three times: at baseline and after 4 and 8 weeks' treatment. Data were obtained from headache diaries: (i) number of headache days per month; (ii) number of attacks per month; (iii) number of days with acute migraine treatment; (iv) number of analgesic tablets per month; (v) number of severe attacks per month; and (vi) documentation of adverse events.
Transcranial magnetic stimulation
TMS was performed twice, at baseline and after 8 weeks by one of the authors (A.L.K.). Motor and visual thresholds were assessed interictally at least 3 days before and after a migraine attack. All female patients were studied between days 2 and 6 of their menstrual cycle.
A Maglite (Medtronic, Copenhagen, Denmark) magnetic stimulator and an MC125 circular coil with an outside diameter of 125 mm were used. For stimulation of the visual cortex the coil was centred over the occipital skull with the handgrip pointing horizontally in a lateral direction. Subjects were blindfolded and wore a swimmer cap with a grid system parallel to the medio-sagittal and interaural lines. The skull surface grid had 1 × 1 cm intersections to allow for exact repositioning of the coil in each subject. To avoid changes of visual cortex excitability due to prolonged sensory deprivation by darkness, the examination was interrupted every 10–15 min by exposure to daylight. Patients were asked to report every bright or coloured visual perception during TMS. The coil was moved in steps of 1 cm in medio-lateral and cranio-caudal directions to identify the optimal position in which brief flashes or white patches of light were consistently reported foveally or within the left or right visual hemifield. To assess the phosphene thresholds TMS was initially applied with a stimulus intensity of 30% of maximal stimulator output and further increased in steps of 5% until phosphenes were reported. Then the threshold was fine-tuned by varying the stimulus intensity in steps of 2.5%. The intertrial intervals were ≥ 10 s apart. To avoid order effects, the direction of increasing and decreasing sequences was randomized. Phosphene threshold was defined as the minimal stimulus intensity at which subjects reported phosphenes in at least five out of 10 stimulations at a given coil position.
Motor evoked potential (MEP) thresholds were assessed according to guidelines of the International Federation of Clinical Neurophysiology. MEPs were registered at rest from the abductor pollicis brevis muscle. The coil was applied to the cull vertex. Stimulation started from 30% of maximal stimulator output and was further increased in steps of 5% until MEPs were recorded. Stimulation frequency did not exceed 0.2 Hz. The coil was moved in steps of 1 cm in medio-lateral and cranio-caudal directions to identify the optimal position in which MEPs with maximal amplitude were recorded. The MEP was then fine-tuned by varying the stimulus intensity in steps of 2.5%. The motor threshold was defined as the minimal stimulus intensity at which MEPs were recorded with minimal amplitude of 30 µV in at least five out of 10 stimulations at a given coil position.
Statistical analysis
Statistics was processed using SPSS software (Version 14.1; SPSS Inc., Chicago, IL, USA).
We compared clinical outcome (headache days per months) at three time points using
Correlations between migraine days per month and motor and visual threshold values at baseline and after treatment were calculated by using Spearman's correlation coefficient.
We furthermore calculated changes in migraine frequency and changes in motor and visual thresholds following treatment. Change in migraine frequency was defined as a difference between migraine days per month at baseline minus migraine days per month after 8 weeks. Changes in visual and motor thresholds were calculated as a difference between threshold values at 8 weeks minus baseline. Correlations between changes in visual and motor thresholds and changes in migraine frequencies were calculated using Spearman's correlation coefficient. The level of significance was set at P < 0.05.
Results
Thirty-six patients participated. Four patients dropped out: two made the decision to discontinue treatment and two were lost to follow-up; 32 patients reached the target dose of 100 mg and successfully completed the planned treatment course. Thirty of them were women, mean age 38.6 ± 10.0 years.
The clinical effects of topiramate are shown in Table 1. Treatment with topiramate resulted in a significant decrease of headache frequency, decrease in days with acute migraine treatment and amount of analgesic tablets used per month. The mean number of migraine days per month decreased from 12.0 ± 1.3 per month to 8.9 ± 3.2 per month (P < 0.001) during the first month and to 5.8 ± 3.2 (P < 0.001) during the second month of treatment (
Clinical effects of topiramate
Adverse events of topiramate
Data on motor and phosphene thresholds before and after treatment are shown in Table 3. Overall treatment with topiramate resulted in an increase of both thresholds for motor-evoked responses and eliciting phosphenes. Right-sided motor thresholds increased by 3.9 ± 5.1% (P < 0.001), left-sided motor thresholds increased by 3.8 ± 6.9% (P = 0.005), and phosphene thresholds increased by 12.3 ± 5.9% (P < 0.001).
Changes of cortical excitability after treatment with topiramate
P < 0.05.
No significant correlations were found between headache days and motor and visual thresholds either at baseline or after withdrawal. Comparing changes in headache days and changes in thresholds, we found a significant inverse correlation between changes in headache frequency and visual thresholds (difference in headache days vs. visual thresholds Spearman's ρ =−0.553, P = 0.002), but no correlation between changes in headache frequency and changes in motor thresholds (change in headache days vs. change in right-sided motor thresholds Spearman's ρ = 0.115, P = 0.53; change in headache days vs. change in left-sided motor thresholds Spearman's ρ = 0.03, P = 0.83; Fig. 1).
Correlations between differences in migraine frequencies and differences in right-sided motor thresholds (A), left-sided motor thresholds (B) and phosphene thresholds (C).
Discussion
We have evaluated the influence of topiramate on phosphene and motor thresholds as well-established markers of cortical excitability. We have been able to demonstrate an inhibitory effect of topiramate on excitability of both motor and visual cortex. These findings were in parallel with clinical effects of topiramate treatment reducing migraine frequency by half.
Our data confirm the well-known efficacy of topiramate in prevention of migraine which has been observed already within the first month of treatment. The 2 months' treatment with topiramate resulted in a significant reduction of migraine frequency and of the use of acute headache medications. Treatment was well tolerated.
The main aim of this study, however, was to evaluate the modulatory effect of topiramate on the cortical excitability as a contributory pathophysiological mechanism in migraine prevention. We found a significant increase in motor response thresholds on average of 4%, which was consistent following the right- and the left-sided stimulation. Phosphene thresholds increased by 12%. Hence, the data clearly demonstrate the inhibitory effect of topiramate on excitability of motor and visual cortical neurons. These findings are in line with previous studies demonstrating the inhibitory/modulatory effect of topiramate on cortical excitability in healthy subjects probably via its GABA-ergic and/or glutamatergic mechanisms (23–25). It is interesting that administration of a single dose of topiramate (50–200 mg) did not influence motor thresholds (25), whereas in our study chronic treatment over 6 weeks with a similar dosage resulted in a significant increase of both motor and visual thresholds.
The main hypothesis of the study was that the clinical efficacy of topiramate could be explained by inhibition of cortical excitability. The fact that treatment with topiramate resulted in a decrease of migraine frequency and an increase of motor and visual thresholds supports the inhibitory role of topiramate on cortical excitability However, at baseline we could not identify a correlation of migraine frequency with absolute values of motor and visual thresholds. Also, no significant correlations were found between changes in migraine frequency and changes in motor thresholds. A further unexpected finding was the inverse correlation between decrease of migraine frequency and increase of phosphene thresholds. This seems counterintuitive at first glance and indicates that the preventive effects of topiramate in migraine are complex and cannot be sufficiently explained by inhibition of cortical excitability.
A similar recent study has evaluated the effect valproic acid on excitability of visual cortex in patients with migraine with aura (MA) and MoA (18). The authors found an increase of phosphene thresholds in patients with MA only, but not in patients with MoA. This is in contrast to our results. A possible explanation could be differences in methodology. Another explanation could be the sample size. Mulleners et al. analysed seven patients with MoA and eight patients with MA. It is possible that the relatively small number was not enough to reach statistical significance. The authors found a week correlation between clinical and electrophysiological effects of valproic acid only in patients with MA.
The most direct explanation of our findings is that the effect of topiramate can be substantial either by increasing TMS measured thresholds (linked to cortical excitability) or by modulating neuronal networks (which are involved in regulating migraine activity), but not on both simultaneously. It is also possible that a postulated relationship between migraine disease activity and cortical excitability is non-linear, with an initial slope and following saturation plateau at higher attack frequencies as in our sample. Our findings could be influenced by the fact that we studied a selected patient population with frequent migraine attacks with small spread of migraine days per month at baseline, which resulted in large spread of migraine frequencies at follow-up. Finally, it is possible that long-term stability of cortical excitability could predict good clinical outcome. A recent study has demonstrated higher intra- and interindividual variability of phosphene thresholds in migraine patients compared with healthy controls. Moreover, the highest and lowest values were found to be possible predictors for subsequent migraine attack (26).
A limitation of this study is the fact that we did not use a placebo group which would allow differentiating between drug-specific and non-specific effects. The significant overall reduction in thresholds and a significant negative correlation between clinical benefit and phosphene threshold effects should be less pronounced in placebo responders and cannot explain our results as placebo responses.
Our findings suggest that, at least in patients with high attack frequency, cortical excitability as measured by our TMS method is not a sufficient surrogate marker for disease activity. For a more generalizable understanding of the role of cortical excitability in migraine prevention it will be necessary to study more prophylactic drugs and the relation with TMS, comparing cortical neurophysiology and clinical effects.
Footnotes
Acknowledgements
The study was supported by the Russian Linguistic Subcommittee of the International Headache Society.