Visual evoked potentials in the diagnosis of headache before 5 years of age

E. Lahat*, J. Barr, A. Barzilai, H. Cohen, M. Berkovitch

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


Headache is a common complaint in children. Diagnosis of the type of headache and therapeutic approach is predominantly clinical based on a detailed history and physical examination. These data are often not available or informative in young children with headache, leading clinicians to look for diagnostic studies. We reviewed our experience with 53 children under the age of 5 years with headache. Of these, 42 (75%) children underwent neuro- imaging studies including CT scan (32 cases), MRI (10 cases), or both studies (6 cases). All were within normal limits except for two cases with a small arachnoid cyst and cerebellar hypoplasia respectively which were not directly related to the headache. Visual evoked potentials were performed in all children. There were no significant differences between the mean latencies of N1P100 and N2 of the children with clinically diagnosed migraine, however, the P100-N2 amplitudes of children with migraine were significantly larger compared to those of children without migraine. Even in young children with headache, neuro-imaging studies have a very limited diagnostic value. Visual evoked potentials may also be used in this age group as a diagnostic tool, particularly when clinical symptoms are either unavailable or not characteristic. Conclusion: The diagnosis of migraine in young children remains clinical, based on history obtained from children and parents; however, visual evoked potentials may support the diagnosis in this age group.

Original languageEnglish
Pages (from-to)892-895
Number of pages4
JournalEuropean Journal of Pediatrics
Issue number11
StatePublished - 1999


  • Headache
  • Migraine
  • Visual evoked potentials


Dive into the research topics of 'Visual evoked potentials in the diagnosis of headache before 5 years of age'. Together they form a unique fingerprint.

Cite this