Search tips
Search criteria 


Logo of jrsocmedLink to Publisher's site
J R Soc Med. 2001 June; 94(6): 286–287.
PMCID: PMC1281524

A diagnostic sign in migraine?

R N de Silva, MD MRCP(UK)


At the bedside it was noted that, after ocular fundoscopy, patients with migraine complained more often of an after-image than did non-migraineurs. This phenomenon was then investigated in consecutive patients attending a general neurology outpatient clinic.

The relative risk for the diagnosis of migraine in patients reporting an after-image was 2.91 (95% confidence interval 1.96 to 4.34), and the sensitivity, specificity and positive predictive value of this observation for the diagnosis of migraine were 0.63, 0.75 and 0.55 respectively. After-images were equally likely to be reported by migraineurs with and without aura, and by patients with migraine equivalents.

The after-image phenomenon probably reflects the heightened sensitivity to visual stimuli of patients with migraine. Although a diagnosis of migraine is primarily established by the patient's history, the presence of an after-image following ocular fundoscopy may support this diagnosis.


Migraineurs frequently describe visual symptoms during the aura phase or while experiencing headache. These include photopsias, scotomas, scintillating scotomas, fortification spectra, blurred vision and photophobia. Positive and negative visual phenomena can arise from migrainous disturbances of the retina (producing monocular symptoms), but are more commonly caused by occipital lobe dysfunction (producing bilateral abnormalities). Migraineurs also seem sensitive to various physical stimuli, including light, noise, smell and movement. Visual stimuli that trigger migraine attacks include sunlight, flickering or flashing lights and striped patterns. During the routine clinical examination of neurological patients it was noted that patients who had migraine reported visual after-images following routine ocular fundoscopy more commonly than non-migraineurs. The current study was designed to determine the utility of this observation in establishing a diagnosis of migraine in an unselected clinic population.


Consecutive new patients attending the general neurological outpatient clinic over six months completed a questionnaire on visual symptoms after fundoscopy. Patients were asked whether fundoscopy had been performed, what if anything had been noted afterwards and how long any after-images had lasted. The questionnaires were issued and collected by clinic nurses, and the contents were not discussed during the consultation. Pupils were not dilated and the maculae were not examined specifically. Migraine with and without aura and migraine equivalents were diagnosed by application of International Headache Society criteria1. Other diagnoses were established on clinical grounds.


A total of 263 patients were seen. No patient reported insufficient visual acuity to perceive light, and the pupillary light responses were normal in all cases. Fundoscopy was not done in 41 patients, and 2 were unable to complete the questionnaire (severe dementia and severe learning disability). The results are in Table 1.

Table 1
Numbers of patients with and without migraine, reporting the presence and absence of a visual after-image following fundoscopy

The after-image was usually a bright spot, but black spots and dots were also reported. The duration was usually between 30s and 2 min. The observation was equally likely to be made by migraineurs with and without aura and patients with migraine equivalents. All migraineurs were interictal at the time the examination took place.

The relative risk for the diagnosis of migraine in individuals reporting an after-image following ocular fundoscopy was 2.91 (95% confidence interval 1.96 to 4.34). The sensitivity, specificity and positive predictive value of the observation for the diagnosis of migraine were 0.63, 0.75 and 0.55 respectively. As expected, migraine was diagnosed more commonly in women than in men, the ratio of female to male being 2:1 when an after-image was reported and 1.6:1 when it was not. There was no difference in median age between individuals with migraine reporting and not reporting an after-image. In those patients in whom migraine was not diagnosed clinically, the ratio of female to male was slightly higher in the group reporting an after-image (1.3:1). Patients without migraine and not reporting an after-image were older than those reporting this phenomenon (P=0.0021).


Whilst the diagnosis of migraine will always be based on the patient's history, examination of the ocular fundus is usually part of the physical evaluation of a patient with headache and an after-image following this procedure may support such a diagnosis. It is not known how the sensitivity and specificity of this observation will be influenced if applied to a specialist ‘headache clinic’ population, or for that matter a general practice population with a variety of complaints including non-neurological ones. Also, the study needs to be repeated with different practitioners, preferably without their knowledge, in case the current investigator has influenced the identification of after-images by altering his technique subconsciously. It is noteworthy that in the non-migraine group of the current study, those reporting an after-image were younger than those not reporting it. This may reflect a tendency for this phenomenon to disappear with ageing. In migraineurs (where no similar age-related trend was seen) the observation was made almost three times as commonly, and this probably reflects the heightened sensitivity of such patients to visual stimuli. Using the term ‘flight of colours’, Feldman and colleagues described the reduced or abolished after-images after stimulation of each eye by a bright pocket-light in patients with defective central vision2. The observation of a heightened response in patients with migraine seems to be novel. Unlike the critical flicker fusion threshold3, where migraineurs without aura are reported to have a lower threshold than those with aura, the current observation does not discriminate between the two subgroups.

How might the phenomenon arise? Presumably it is mediated at the cortical level, since it is bilateral. Mechanisms such as those responsible for ‘spreading depression’ can be evoked to account for it, but await more detailed evaluation by studies of cortical blood flow. It would also be interesting to study susceptible subjects to see whether the presence of after-images depends on different colours, strengths and durations of ocular illumination. Electrophysiological studies with transcranial magnetic stimulation confirm cortical hyperexcitability in migraineurs. 13 of 15 migraine patients (14 migraine with aura, 12 women, mean age 39.9 SD 8.2 years) developed phosphenes after standardized occipital cortex stimulation by magnet, compared with only 2 of 8 normal controls (5 women, mean age 37.3 SD 6.0 years)4.


I thank Wan-Thai Weissman for help with data collection. The questionnaires were distributed and collected after completion by Anne Humphrays, Helen Burke, Kathy Denby and Norma Morgan. Christopher Hawkes gave helpful advice on data interpretation and regarding the discussion.


1. Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalagia 1988;8(suppl 7): 1-96 [PubMed]
2. Feldman M, Todman L, Bender MB. “Flight of colours” in lesions of the visual system. J Neurol Neurosurg Psychiatry 1974;37: 1265-72 [PMC free article] [PubMed]
3. Coleston DM, Kennard C. Responses to temporal visual stimuli in migraine: the critical flicker fusion test. Cephalalgia 1995;15: 396-8 [PubMed]
4. Aurora SK, Cao Y, Bowyer SM, Welch KMA. The occipital cortex is hyperexcitable in migraine: experimental evidence. Headache 1999; 39: 469-78 [PubMed]

Articles from Journal of the Royal Society of Medicine are provided here courtesy of Royal Society of Medicine Press