Abstract
Visual disturbances are frequent symptoms in migraine. Since there is a possibility of separate damage in the magno- or parvo-cellular visual pathway in migraine patients, we performed a study including the measurement of static and dynamic spatial contrast sensitivity on 15 patients suffering from migraine without aura under photopic and scotopic conditions. Fifteen healthy volunteers without primary headache served as controls. The results revealed a marked decrease in contrast sensitivity at low spatial frequencies in the migraine patients. Spatial contrast sensitivity demonstrated some lateralization, as the sensitivity to low spatial frequencies obtained through separate eyes showed significantly larger side-differences in migraine patients than in control subjects. These findings suggest that the mechanisms responsible for vision at low spatial frequencies are impaired in migraine patients. This might indicate impaired function of the magnocellular pathways in this condition.
Introduction
Visual symptoms are common in migraine. The most striking ones appear in the aura, frequently involving transitionally blurred vision, homonymous hemianopsy, scotomas or photophobia (for reviews, see 1, 2). These phenomena obviously signal the involvement of visual retino-cortical mechanisms in the pathophysiology of migraine. The major question remains, however, of whether the parallel parvo- and magno-cellular visual pathways are equally affected in this condition or there is a preponderance of magnocellular malfunctions in migraine-associated visual phenomena (3, 4). The answer to this question is greatly hindered by the difficulty in distinguishing between magnocellular and parvocellular dysfunctions in human clinical investigations (5, 6). Nevertheless, it is generally held that the parvocellular pathway dominates the information transfer at high spatial frequency (SF) and low temporal frequency (TF), while the magnocellular pathway conveys information at low SF and high TF (7, 8). This warrants the application of contrast sensitivity measurements in migraineurs.
The use of scotopic stimulating conditions was indicated by animal experiments in which visual stimulation at low SFs and under scotopic conditions excited predominantly magnocellular ganglion cells in the retina (9, 10). Scotopic tests have been employed in several human studies in the search for pathological alterations in human magnocellular functions (11, 12). With regard to these facts, we set out to compare photopic and scotopic spatial contrast sensitivity functions in migraine patients without aura. To increase the sensitivity of these examinations, both static and dynamic contrast sensitivity functions were tested.
Subjects and methods
The patients enrolled in the study were 15 women with migraine without aura. The visual acuity was 1.0 in all cases. The age-range was 18–53 years, with a median age of 31 years. The duration of the complaints ranged between 1 and 25 year, with a median value of 10 years. The frequency of their headache was in the interval between monthly occurrence and three attacks per year. The patients were diagnosed according to the criteria of the International Headache Society (13). All subjects underwent detailed neuro-ophthalmological examinations, including physical examination, CAT scan, blood chemistry, ophthalmoscopy and visual perimetry. Only patients with no other neurological or ophthalmological diseases were included in the study. Intraocular pressure level was routinely measured in all patients. It was below 16 mmHg in all cases. In two cases the sinusoidally reversing grating caused acute headache, as described by Shepherd et al. (14). These patients were excluded from the study. Similarly, patients who had migraine attacks without aura two weeks before or after the study were also excluded.
The control group comprised 15 age-matched female volunteers with good vision and without neurological symptoms or primary headache.
Monocular static and dynamic contrast sensitivity was measured at nine spatial frequencies (SFs) (0.5, 1.2, 1.9, 2.9, 3.6, 4.8, 5.7, 7.2 and 14.3 cycle/degree) with a computerized test (Venus, NeuroScientific Corporation, USA). The stimuli were luminance contrast horizontal gratings with a sinusoidal luminance profile. For the dynamic test, the pattern was reversed at a temporal frequency of 4 Hz. The display subtended a visual angle of 13° × 13° and was viewed from a distance of 1 m. The luminance of the screen was 17 and 0.17 cd/m2 under photopic and scotopic conditions, respectively. The maximum contrast was 70.7%. Both eyes were tested under photopic conditions, while only the right eye was tested under the scotopic one. For the measurement of contrast thresholds, the contrast was initially set at 15 dB above the mean normal value. The participants were all able to detect this submaximal contrast level. The contrast level was then decreased by 3 dB every 5 s until they failed to detect the stimulus any longer (descending method). Patients gave a verbal indication of changes in contrast detection. In both the static and the dynamic test the contrast was then set at 15 dB below the threshold measured with the descending method and it was increased by 3 dB every 5 s without interstimulus interval until subjects detected the stimulus (ascending method). The whole procedure was repeated five times in order to obtain a mean contrast threshold at a given SF. The contrast sensitivity was defined as the reciprocal of the contrast threshold (14). The sequences of the descending and ascending methods, the SFs tested and the static vs. dynamic tests were counterbalanced across the subjects. Not all the patients completed all the tests. In some rare cases the testing was prematurely concluded. This is indicated by the differing degree of freedom values.
Statistical evaluation was carried out by means of
Results
Spatial contrast sensitivity functions revealed significant impairments in migraine patients (Fig. 1). The decrease in contrast sensitivities was most marked at low SFs and specially in measurements under scotopic conditions. Significant differences in contrast sensitivity under photopic static condition were observed by diagnosis (F[3, 40]= 25.01, P< 0.001), by frequency (F[8, 320]= 138.1, P< 0.001) and for the interactions (F[24, 320]= 10.4, P< 0.001) (Fig. 1a). Significant differences in contrast sensitivity under photopic dynamic condition were also observed by diagnosis (F[3, 42]= 8.85, P< 0.001), by frequency (F[8, 336]= 102.7, P< 0.001) and for the interactions (F[24, 336]= 6.84, P< 0.001) (Fig. 1b). The post hoc comparison showed significant differences between the migraine and control values for both eyes at the five lowest spatial frequencies in the photopic static situation and in the three lowest frequency band in dynamic situation.
Spatial contrast sensitivity functions of migraine patients and control subjects. (a) photopic contrast sensitivity estimated by static stimulation and (b) photopic contrast sensitivity functions estimated by dynamic method. Control subjects, open symbols; migraine patients, filled symbols; ▪,□ through the more sensitive eye, ▴,▵ through the less sensitive eye. (c) scotopic spatial contrast sensitivity estimated by static stimulation and (d) scotopic spatial contrast sensitivity estimated by dynamic stimulation. Control subjects ○, migraine patients •. ∗ and ∗∗ mark significant differences to the corresponding eyes of control subjects on the 0.05 and 0.001 probability levels, respectively. † and ‡ denote significant interocular contrast sensitivity differences on the 0.05 and 0.001 probability levels, respectively. Values are shown as mean±
As regards the
Interocular contrast sensitivity difference values were calculated in measurements performed under photopic circumstances at the four lowest frequency band. As regards the
Discussion
Our results showed significant changes of spatial contrast sensitivity in patients with migraine without aura. A particularly noteworthy reduction was found in the low SF range of spatial contrast sensitivity. Our finding therefore seems to support the notion that there could be an asymmetric disturbance in the function of the parallel visual pathways of migraine patients without aura. Quite recently, McKendrick et al. (15) reported a similar dysfunction in migraine with aura. In this type of migraine patients a definite reduction was found in temporal contrast sensitivity (4) that is also in concert with our conclusions.
In view of the well-known dominance of the magnocellular pathway in the transmission of visual information in the low SF range (7, 8), our finding may suggest impaired magnocellular visual functions in migraine patients. Further, the magnocellular ganglion cells in the retina (16) and in the dorsal lateral geniculate nucleus (17, 18), and the cells in layer 4C (19) display a similar contrast sensitivity function to that observed in humans under scotopic circumstances (20). There is also clinical evidence that the scotopic vision is impaired after magnocellular damage (12, 21). Therefore, finding that the decrease in the contrast sensitivity in the migraine patients was especially marked under scotopic circumstances further substantiates the notion of magnocellular damage in this condition.
It remains to be settled whether the weaker contrast sensitivity at low SF represents a cortical or precortical mechanism. Cortical neuronal changes have been extensively investigated during the last years. A definite visual cortical hyperexcitability was reported in migraine patients (22). Cortical site of action is warranted by studies investigating sensory habituation processes, too (2, 23, 24). These results are in line with the interhemispheric differences that has been reported concerning a series of parameters measured interictally or ictally in migraine patients (25). The visual evoked potential studies by Tagliati et al. (26) and Shibata et al. (27) demonstrated significant hemispheric side differences in hemianopic migraine aura. Positron emission tomography, however, failed to indicate any side difference during the visual aura of classic migraine, although there was a 40% overall decrease in the regional cerebral blood flow (28). A precortical site of action in the pathophysiology of migraine has been raised earlier (3). Interictally persisting dysfunctions of the precortical visual processing have been substantiated by the evoked potential studies of Oelkers et al. (29). The eye to eye side difference in the interictal measurements of contrast sensitivity in our migraine patients could indicate an additional evidence for some precortical sites of pathophysiological actions in migraine. It is possible that the interocular and interhemispheric differences represent two separate pathophysiological processes. Although, cortical afflictions in migraine patients have already been widely investigated, studies on retinal or optic tract abnormalities are, however, rather rare. Our findings lead us to suggest that a thorough investigation of the pathophysiological phenomena in the whole visual pathway of migraine patients might promote our knowledge concerning the pathomechanism of this painful condition.
Footnotes
Acknowledgements
The authors are greatly indebted to Ms Gabriella Dósai Molnár for her skilful assistance and to Dr Andrea Antal for her suggestions and help in the use of the VENUS equipment. The study was supported by grants OTKA 29817, ETT 57404, 56411 and FKFP 0455/2000.