Objectives: To assess the utility of neuroimaging in the evaluation of children presenting with two of the most common forms of headache, migraine and chronic daily headache, and to determine the utility and pathological yield of neuroimaging in specific headache syndromes in children whose neurological examinations are normal.
Methods: We retrospectively reviewed the medical records of patients coded for headache (ICD 784) in the Pediatric Neurology Clinic at Children's Hospital of the King's Daughters between 1997 and 1999. The age range considered was between 6 and 18 years. The study focused on the two most common types of headache, uncomplicated migraine and chronic daily headache. Only patients with normal physical and neurological examinations were considered in this analysis.
Results: Three hundred two patients were coded for headache within the defined age group. One hundred seven (35.4%) patients fulfilled IHS-R criteria as having uncomplicated migraine with a normal examination, and 30 (9.9%) patients fulfilled criteria for chronic daily headache. Twenty-nine (9.6%) patients presented with migrainelike symptoms, and 6 (2.0%) presented with chronic daily symptoms, but had neurological abnormalities present on examination. The remainder of the patients with headache had the following etiologies: 50 (16.6%) with secondary headache, 22 (7.3%) with complicated migraine, 20 (6.6%) with posttraumatic headache, 13 (4.3%) with seizure-related headache, 11 (3.6%) with brain tumors, 10 (3.3%) with tension-type headache, and 4 (1.3%) with pseudotumor cerebri. Of the 107 patients with migraine, 42 (39.3%) received CT scans; 2 (4.8%) of which were considered "abnormal." One of the abnormalities was an arachnoid cyst and the other was a dilated Virchow-Robin space. Twelve (11.2%) patients with migraine received an MRI, 2 (16.7%) of which were considered abnormal. Both of the abnormal findings were Chiari type I malformations. Of the 30 patients with chronic daily headache, 17 (56.7%) received CT scans, 3 (17.6%) of which were considered abnormal. The abnormalities consisted of a maxillary opacification, a mucous retention cyst, and an occult vascular malformation. Eight (26.7%) of the patients with chronic daily headache had an MRI, 2 (25.0%) of which were abnormal. One of the abnormalities was a Chiari I malformation, and the other was an occult vascular malformation.
Conclusion: The yield of neuroimaging in children with uncomplicated migraine and normal neurological examination was 3.7%. The yield in children with chronic daily headache and normal neurological examination was higher at 16.6%. The abnormalities discovered included arachnoid cysts, Chiari I malformations, sinus disease, occult vascular malformations and "dilated Virchow-Robin spaces." While none of the neuroimaging findings were apparent clinically, their discovery did not influence the diagnosis, management, or outcome of the patients. None of the abnormalities necessitated surgical intervention or were associated with the headache presentation. Therefore, neuroimaging is not warranted in children and adolescents with defined clinical headache syndrome diagnoses whose neurological examinations are normal.