Lack of habituation of visual-evoked potential in the interictal period is not a consistent neurophysiological marker of migraine: A cross-sectional analytical study

INTRODUCTION

Migraine is a widespread neurovascular disorder with a high prevalence resulting in socioeconomic, personal, and other repercussions. The International Headache Society classifies migraine headaches as with or without an aura. In 82% of cases, patients experience the visual symptoms of migraineous aura, which include scotoma, a central or paracentral blind spot, stars, zigzags, and flashes of light.^[1] According to studies, migraineurs exhibit alterations in their evoked potential even when they are not experiencing headaches.^[2] The brain’s sensitivity to migraine attacks appears to be mostly dependent on cortical excitability during migraine interictal periods. Patients with migraines exhibit amplification rather than habituation of stimulus-evoked brain responses between attacks. The main structures involved in the pathogenesis of migraine are the brainstem, trigeminovascular system, and cerebral cortex. It is unclear whether the change in habituation is caused by higher or lower cortical pre-activation levels. Increased or decreased cortical excitability may contribute toward the pathogenesis. Mainly two theories prevail to explain the cortical excitability in migraineurs, which are the theory of hypoexcitability and the theory of hyperexcitability. Authors, who support the idea of decreased pre-activation in migraineurs, emphasize lower initial P₁₀₀ amplitude during repetitive stimulation^[3] while some studies^[4] showed the migraineurs being characterized by an increase in neuronal excitability. There has been substantial research into the biochemical and neurophysiological abnormalities that lead to migraine attacks, but the inconspicuous element that causes the disorder, which, if present, should also be visible during a period of pain-free time, is the underlying dysfunction.^[5] Visual-evoked potentials (VEPs) are a functional option that smoothens the path of assessment of the visual pathway. Pattern-reversal VEPs can be used to explore the two parallel routes that carry visual information: The luminance and contour processing pathways. These pathways are hypothesized to be differentially affected in migraine.^[6] The mass activity of visual cortex neurons is inferred from cortical VEPs. In a previous pattern reversal (PR)- VEP study, during the interictal stage, potentiation substituted the normal habituation pattern in migraineurs.^[7]

The visual cortex appears to exhibit habituation of the VEP as a physiological phenomenon when it malfunctions during episodes of migraine. Cortical activity is habituated and potentiated by chemically regulated connections that come from the brainstem and may use transmitters such as serotonin, noradrenaline, dopamine, histamine or acetylcholine. Through the assessment of VEP in migraine patients in between attacks and comparison with normal patients, this study aims to gain insight into the pathophysiology of migraine. The brain’s vulnerability to migraine attacks appears to be mostly determined by the cerebral cortex’s excitability during the interictal state of migraine. Besides, no previous studies have analyzed the various clinical factors associated with the P₁₀₀ mean amplitude and latency differences between the first and the last block in migraine patients on consecutive VEP stimulations. This will be important in determining the clinical factors that are associated with the lack of VEP habituation in migraine patients. This study helped us to know the cortical excitability by analyzing the habituation of VEP in between attacks, which can be used as an objective measurement of prophylactic treatment response. Furthermore, it can be used to compare the efficacy of various migraine prophylactic drugs, thereby aiding in better treatment outcomes.

MATERIALS AND METHODS

RESULTS

A total of 40 subjects were included in the study out of which 20 subjects were recruited as controls and 20 subjects with migraine were recruited as cases. The mean age of the patients was 34.4 ± 8.72 years and that of controls was 26.45 ± 5.45. The mean frequency of the headache in migraine patients was 7.7 ± 4.01 episodes per month, and the duration of the headache was 6.65 ± 1.93 h. Baseline characteristics and demographic features were compared between cases and matched controls [Table 1].

The amplitude of P₁₀₀ in the fourth block decreased in both eyes of the controls compared to the first block [Table 2a], which was statistically significant (P = 0.003 [left side]; P = 0.002 [right side]). Similar to this, there was a statistically significant (P < 0.001) drop in the amplitude of P₁₀₀ in the fourth block compared to the first block in both eyes of the migraine patients [Table 2a]. The N₁₄₅ latency in the fourth block was shorter in both eyes in migraine patients compared to controls (P = 0.01) [Table 2b]. However, there was no statistically significant difference between migraine patients and controls in the amplitude or latency of P₁₀₀ in the fourth block in either eye [Table 2c and d].

The amplitude decreases progressively from block one to block four [Table 2e] in both the control and migraine patients, while there is no significant difference between the two groups.

Multiple regressions were run to predict the mean difference of P₁₀₀ latencies between the first and the fourth block from age, gender, frequency of migraine attacks, duration of disease, and clinical features of headache that included distribution, character, the severity of headache, photophobia, phonophobia, and nausea/vomiting. The statistical prediction of P₁₀₀ latencies based on these variables is non-significant, F (11, 8) = 1.389482, P = 0.32, R² = 0.656 [Table 3]

The mean difference of P₁₀₀ amplitude between the first and the fourth block was also used as a dependent variable for multiple regression analysis with independent variables the same as mentioned for the mean difference of P₁₀₀ latencies. The statistical prediction of P₁₀₀ amplitudes based on these variables is also non-significant, F (11, 8) = 0.616526, P = 0.77, R² = 0.4587 [Table 4].

There was a negative correlation between age and mean P100 amplitude difference between the first and fourth block (Pearson’s r = −0.08497, P < 0.001) [Figure 1a] and a positive correlation with headache frequency (r = 0.07438, P < 0.001) [Figure 1b]. However, there was a moderate negative correlation present between headache duration and mean P₁₀₀ amplitude difference (r = −0.45682, P < 0.001) [Figure 1c]. The mean P₁₀₀ latency difference between the first and fourth block correlated positively with age (r = 0.07086, P < 0.001) [Figure 1a] and negatively with headache frequency (r = −0.19043, P < 0.001) [Figure 1b] while there was a moderate positive correlation with headache duration (r = 0.33583, P < 0.001) [Figure 1c and Table 5].

Close

Figure 1:

Correlation graphs showing (a) negative correlation graph between age and mean difference (block one–block four) of P100 amplitude (Left) and positive correlation graph between age and mean latency difference (block one–block four) of P100 latency (Right). (b) Positive correlation graph between frequency of headache and mean difference (block one–block four) of P100 amplitude (Left) and negative correlation graph between frequency of headache and mean difference (block one–block four) of P100 latency (Right). (c) Negative correlation graph between duration of migraine and mean difference (block one–block four) of P100 amplitude (Left) and positive correlation graph between duration of headache and mean difference (block one–block four) of P100 latency (Right).

Export to PPT

DISCUSSION

We recruited 20 controls and 20 patients (Female [12]: Male [8]) with migraine; 85% of those patients reported migraine without aura. Photophobia (45%) and nausea (55%) were the most common symptoms accompanying migraine. A majority (80%) of the patients reported throbbing pain and 65% had moderate pain. The 55% of the total patients recruited reported unilateral pain while 40% had holocranial pain. The 40% of the patients were on prophylactic medications.

In our study, subjects with migraine showed P₁₀₀ amplitude habituation as compared to controls, which contradicts findings from other studies published.^[1,2,7] Both controls and migraineurs exhibited habituation. This can be due to methodological differences. For instance, in the current study, which we designed to evaluate habituation, the stimulation rate and the number of averaged responses on which measures were performed were modest. In addition to this, it was found that VEP parameters vary as a function of spatial frequency.^[8,9] An increase in spatial frequency results in a reduction of P₁₀₀ amplitude. It has been shown that as the spatial frequency increases, visual sensitivity gradually declines. This can be attributed to the eye’s optical quality or the higher levels of visual processing (beyond the lateral geniculate nucleus) responsible for VEP production. In addition, it is suggested that sensitivity loss at high spatial frequency may result from quantal variations in light. This might be one of the reasons behind the discrepancy in the results. Furthermore, most of our patients were already on migraine prophylactic drugs and did not have migraine attacks in the recent past before the study. This result contradicts claims made in several other review articles that the lack of habituation in interictal migraineurs is a neurophysiological characteristic and the most repeatable finding.^[2,7] A few trials with either negligible or no variation in the amplitude of the P₁₀₀ wave were found to have similar outcomes to our study.^[10-15]

When the VEP is recorded continuously for a long period (18–20 min), latency may increase while amplitude decreases. Such a prolongation of latency was noted in both migraine patients as well as in controls. However, the mean change in latencies between the migraine patients and controls was non-significant for the N₇₅ and P₁₀₀ waves, while the difference was highly significant for the N₁₄₅ wave. Numerous studies that found a non-significant prolongation of P₁₀₀ latency reported similar outcomes.^[11-15] The VEPs can show changes in neurological function brought on by ischemic injury, aberrant neurotransmitter levels or both. The prolonged P₁₀₀ latency may be brought on by the long-term repercussions of assaults, which include cerebral ischemia and edema. However, we included individuals in our study who had been suffering from migraine for only two years, which may be the cause of the non-significant lengthening of the P₁₀₀ latency [Table 1]. In addition, 85% of the participants in our survey claimed to have migraine without aura. Since only 15% of migraineurs with aura were recruited in our study, a subgroup analysis was not statistically feasible. However in a study, migraine sufferers with an aura, in particular, were found to have greater P₁₀₀ latency.^[10]

A common occurrence in the neural system, habituation has intricated spatially and functionally dependent mechanisms. It is proposed that the cerebral cortex is regulated by excitatory interneurons that receive input from thalamocortical neurons, inhibitory intra-cortical interneurons, and subcortical connections in the brainstem involving neurotransmitters such as histamine, noradrenaline, serotonin, and dopamine. These mechanisms typically guard against cortical overstimulation.^[16,17] Serotonin has extensive sensory cortex innervation, tonic pacemaker activity, and modulatory effects on cortical information processing.^[18] A 1 Hz reversal rate was used in the current study, and habituation appeared to be evident in both the controls and the migraineurs. Although there is mounting evidence that migraines modify cortical excitability, it is unclear whether these alterations are the result of changed excitatory connections, damaged inhibitory networks or altered subcortical preactivation.^[19] The inconsistent results in earlier VEP amplitude studies were initially thought to be due to habituation², but this explanation now seems less likely in light of the current findings. Several investigations have indicated that migraineurs do not exhibit a lack of habituation.^[6,8,20-23] In our study, we found that the duration of migraine is strongly associated with the mean P₁₀₀ amplitude difference. This may imply that the severity and duration of migraines over time may be a risk factor for impaired visual cortex function. In addition to this, duration is also moderately associated with the mean P₁₀₀ latency difference. We found that VEP amplitude habituation increased with migraine frequency. The average P₁₀₀ latency difference was found to be decreasing with an increase in frequency. This is possibly due to the shorter time between attacks in patients with a high frequency of the condition.

There was no discernible relationship between age and the P100 wave parameter. This might be a result of the smaller sample size and dispersed age data plot.

In our study, no correlation was observed between VEP and photophobia. It may be possible that photophobia depends more on subcortical than cortical visual pathways.^[22,24] These findings confirm the association between the severity of migraine clinical presentation and the behavior of neurophysiological responses.

This study has a few strengths and limitations. The primary strength of the study is its adequate sample size. Another strength is the inclusion of healthy controls for proper comparison between the two groups. The stimulation rate and low number of averaged responses on which measurements were performed in the present study, in which we aimed to evaluate habituation, are probably the primary limitations of the investigation. Second, most of our patients were already on migraine prophylactic drugs and did not have migraine attacks in the recent past before the study.

CONCLUSION

Our study results suggest that, at least in the interictal stage, lack of habituation cannot be used as a reliable neurophysiological migraine marker in different laboratories. The disparate findings in previous research are probably due to methodological discrepancies such as the stimulation rate, the number of averaged responses, and the patient recruitment strategy. It seems that migraineurs’ habituation behavior is not always hindered, but rather is affected in a subtle way that depends on stimulating stimuli. To properly understand the varying results of VEP habituation studies in migraineurs and the effects of the migraine cycle, particularly the pre-ictal phase, a long-term study is required. Drug-naïve people and those who have recently had episodes must be the subjects for future research.