Abstract
The critical flicker frequencies (CFF) of individuals with migraine with and without aura were determined and compared to those of normal controls. Twenty-six migraine patients, 12 with aura and 14 without aura and 30 healthy controls were included. Migraineurs were tested during a migraine-free period, through both the continuous flicker method (CFM) and the forced choice method (FCM). Migraineurs presented a mean flickering fusion threshold lower than healthy controls with the CFM (40.45 vs. 44.33, respectively; P = 0.019) and with the FCM (34.16 Hz vs. 38.5 Hz, respectively, P = 0.019). Both groups of migraineurs had significantly lower thresholds as compared to controls, migraineurs with aura presenting the lowest thresholds for the fusion of flickering (P = 0.008 and P = 0.0001 with the CFM and the FCM, respectively). Results confirmed and extended previous observations of abnormal flicker fusion thresholds in migraineurs. We hypothesize that this finding might be related to a shorter cortical stimulation silent period.
Keywords
Introduction
The visual system harbours the greater sensorial network of the working brain (1) and, migraine being primarily a neural disorder (2), it is understandable that interictal abnormalities of the central nervous system can be most easily evaluated by accessing the steps of visual processing. Critical flicker frequency (CFF) or flicker fusion threshold (FFT) is ‘the maximum temporal frequency of a high luminance flickering target discernible by the observer that is recorded’ (3) or the ‘lowest frequency at which a subject perceives a no longer flickering but steady light’ (4). It is known to reflect the temporal responsiveness of the visual system. A previous study has described abnormally lower CFF in migraineurs without aura, but not in migraineurs with aura (3). Others have shown the CFF to be modified by the abuse of antimigraine drugs in a reversible fashion (4, 5). We hypothesized that CFF may have implications in the way light affects the migrainous brain, and that migraineurs with aura should present a more abnormal CFF than migraineurs without aura. Aiming to study the CFF in a population of migraineurs not submitted to prophylactic therapy, we have conducted this study.
Methods
The study was approved by the Ethics Committee for Research on Human Beings of the Hospital de Clínicas – Universidade Federal do Paraná. Migraineurs and healthy controls were randomly selected from the hospital's outpatient clinic, from their families and from hospital personnel. All the individuals were included only after reading and signing an informed consent form. Migraineurs were diagnosed through a semistructured interview and in accordance with the International Headache Society's criteria for Migraine with Aura and Migraine without Aura. Exclusion criteria consisted of the presence of other neurological diseases; eye disease other than refraction errors; alcoholism; recent alcohol intake or symptoms and/or signs of abstinence; lack of capacity to understand or to cooperate with the study procedures; basilar migraine, migraine with prolonged aura, acute onset aura, familial hemiplegic migraine or other types of migraine; use in the month preceding the study, of the following substances: antidepressants, anxyolitics, neuroleptics, lithium, antiepileptic drugs, drugs for Parkinson's disease, muscle relaxing drugs, anticholinergic drugs, prophylactic drugs for migraine, calcium channel blockers, dopamine antagonists, betahistine, cinarizine and flunarizine, piracetan and hormone replacement therapy. The use of antimigraine drugs other than opioids was allowed if taken before the 24 h preceding the study procedures. Twenty-six migraine patients (10 males, 16 females) aged 31.1 ± 11.3 years, 12 with aura (34.9 ± 14.2 years) and 14 without aura (27.8 ± 7.1 years), diagnosed according to IHS criteria were compared with 30 healthy controls (males: 17, females: 13) aged 30.9 ± 8.9 years were included. Of the patients with Migraine with aura, 11 reported visual aura, mostly scintillating scotomata (n: 10), one hemianopic aura and one somesthesic aura. There was no significant difference between the three groups regarding age (Mann–Whitney and unpaired Student's t-test for independent samples). The migraine group had 3 blue-eyed, 5 green-eyed, 16 brown-eyed and 2 black-eyed individuals. The control group included 2 blue-eyed, 3 green-eyed, 23 brown-eyed and 2 black-eyed individuals. During testing, mean room temperature was 25.2 ± 3.08°C (77.38 ± 37.5°F). Mean pupillary diameter of the individuals was 4.8 mm (4.8 ± 0.7 mm). All the migraineurs were tested during a migraine-free period. The flickering equipment consisted of a black box measuring 100 × 12.5 × 12.5 cm (39.37 × 4.9 × 4.9 inches) with an opening in its front and with four red Light Emitting Diodes (LEDs) arranged to form a 1 cm2 square in the back wall. Those four LEDs were placed so as to result in a visual angle of 1.4° in relation to the foveal vision. The procedure was carried out under a dim light and in a temperature controlled environment. The LEDs were activated by a specially designed motherboard controlled through specific software with a notebook. For each individual tested, two methods were used, the continuous flicker method (CFM) (3) and the forced-choice method (FCM) (6). The subjects were asked to focus on the four LEDs during all the recording procedures. CFM was performed starting at flickering frequencies of 20 Hz, increasing steeply at a 0.5 Hz rate up to 70Hz, with a 5 second stimulus-free interval between each flickering stimulus. Each flickering frequency was maintained during 10 s before the stimulus-free interval. The patient was asked to press a stop switch at the frequency at which he/she could no longer perceive the flickering. After an ascending sequence, a descending sequence (from 70 Hz to 20 Hz) was done. The two sequences were performed three times for each patient and for each only the mean of the results was considered for statistical analysis. In the FCM, flickering stimuli were delivered during four seconds, two at a perceivable frequency starting from 20 Hz up, and two seconds at a nonperceptible frequency(70 Hz). The ‘perceptible’ phase of the stimuli was changed randomly in a double-blind fashion, and the investigator asked the patient if there was flickering in the first or in the second phase of the stimuli. The fusion frequency was considered that in which the patient was unable to tell which was the perceivable phase. The investigator pursued a consistent report of fusion. The results obtained from the migraineurs with aura and without aura were compared with each other and the results of the migraineurs with aura and without aura and of all the migraineurs were compared with those obtained from the controls. The Student's t-test for unpaired data was used for statistical analysis, with a minimal significance level of 5%. Comparison of the age between groups was carried out through Student's t-test for unpaired data and the Mann–Whitney U-test. The evenness in the proportion of clear and dark eyes between the groups was studied through χ2 analysis or the Fisher's exact test. Pearson's correlation analysis, one-way
Results
Migraineurs presented lower mean flicker fusion thresholds than that of healthy controls at the CFM (40.45 ± 6.38 Hz vs. 44.33 ± 6.49 Hz, respectively; P = 0.019) and at the FCM (34.16 ± 4.7 Hz vs. 38.5 ± 7.84 Hz, respectively; P = 0.019). Migraineurs with aura presented the lower thresholds for the fusion of flickering, which were significantly lower than the ones presented by the healthy controls (P = 0.008 and P = 0.0001 with the CFM and the FCM, respectively). Results are summed up in Table 1 and in the Fig. 1 (CFM) and Fig. 2 (FCM). Two of the subjects of the control group showed markedly high CFF results. Although these findings were consistent on multiple testing, we cannot exclude a possible artifact.
Mean values for the Critical Flicker Frequencies by the Continuous Flicker Method. • Controls; □ migraine without aura; ○ migraine with aura.
Mean values for the Critical Flicker Frequencies by the Forced Choice Method • Controls; □ migraine without aura; ○ migraine with aura.
Mean thresholds for Critical Flicker Frequency (CFF) obtained through the CFM and FCM in migraineurs and healthy controls
P: significance at the Student's t-test for unpaired data reflects migraineurs’ CFF results as compared to those of the healthy controls; CFM: continuous flicker method; FCM: forced choice method; sd: standard deviation.
In the migraine without aura group only, an unexpected weak correlation between higher ages and higher flicker fusion thresholds obtained through the FCM was found (r = 0.5591, P = 0.0376). The one-way
A relationship between clear eyes and lower CFF was found with the CFM (P = 0.035), when considering patients and controls altogether.
There was no significant difference of the results of the migraineurs with aura as compared to those found in the migraineurs without aura (P = 0.963 and P = 0.214; CFM and FCM, respectively). Two migraineurs reported migraine attacks in the 24 h following the procedure and were excluded in order to avoid the influence of a prodromal state in the results of the tests. A migraineur complained of irritability and nausea during the procedure, and three others referred only mild irritability, which was not enough to interrupt the procedure.
Discussion
Although migraine prevalence is not affected by blindness (7), the visual system seems to be not only involved in the migraine mechanism, but also in many of its clinical aspects (8). The results herein described add on to previous observations of increased responses of the visual system to several types of visual stimuli in migraineurs (3, 8–11). Chronicle et al. (11) demonstrated in an elegant study of the ST-1 and ST-2 responses that migraineurs with and without aura peaked at lower illumination values for targets and at lower background temporal frequencies than controls, suggesting increased sensitivity of the parvocellular and magnocellular pathways (11). In monkeys, critical flickering frequency is conveyed through the magnocellular system (12). More recent evidence of a dysfunction of the parvocellular system suggests as well a more widespread dysfunction of the visual system as a whole, particularly in migraine with aura individuals. In this study, specific visual stimuli activated the occipital cortex as opposed to migraineurs without aura and controls (13). The parvocellular pathways probably mediate this phenomenon. Light-induced discomfort and photophobia however, are also enhanced in migraineurs, and reflect more probably an increased subcortical perception (14). In pattern-induced attacks the involvement of the occipital cortex in the migraine process occurs only after activation of the rubral nucleus and of the periaqueductal grey matter (15). Migraine attacks can be triggered by checkerboard stimuli, especially in migraine with aura patients (16). Recent evidence in migraineurs with aura, although limited, points to an inverse correlation between the cortical stimulation silent period (CSSP) and a higher frequency of migraine attacks (17). A shorter CSSP could perhaps explain our findings of lower thresholds for flickering fusion in migraineurs, a finding expectedly more pronounced in those with aura. Although Coleston and Kennard (3) did not find lower CFF for the migraineurs with aura as compared to controls, at least some of their patients had not been withdrawn from therapy. Since CFF decreases in a steady fashion after the age of 16 years (18) an age-related disparity could be expected, but there was no statistical difference of the corrected CFM and the FCM results between the groups. Although the FCM has been regarded as less prone to influences of a ‘learning effect’, its use is more recent than the CFM and its reliability deserves further testing.
Perhaps the most important aspect relating to critical flicker fusion in migraineurs, rather than the evidence of an increased excitability, may relate to an overloading of the system with triggering visual impulses, as it happens in photosensitive epilepsy. TV-induced seizures are reported to be more common in Europe than in America, since in Europe the light-emitting sources work with a frequency closer to the CFF (50 Hz and 60 Hz, respectively) (19). In the contemporary era, humans live in environments in which several types of artificial light stimuli are displayed at different frequencies from different types of light sources. Although CFF is described to occur at a frequency bracket around 40 Hz, it may go up to 60 Hz with moving flickering targets, or even to 80–90 Hz with high intensity screens (20). Indeed, we have seen a single patient whose headaches were triggered by lower screen frequencies and abolished by a higher screen frequency (21). A tempting and provocative hypothesis is that migraineurs’ brains may behave like those of photosensitive epileptics, but reacting through migraine attacks. Perhaps modifications in the patterns of light emitting sources may bypass the impact of light flickering in the migraineurs’ nervous system.
Aknowledgements
Footnotes
Acknowledgements
The authors thank Prof Juarez Gabardo, from the Department of Genetics – UFPR, for the statistical reanalysis of the results.