Headaches in the Pediatric Emergency Department

Abstract & Commentary

Source: Burton LJ, et al. Headache etiology in a pediatric emergency department. Pediatric Emerg Care 1997;13:104.

Burton and colleagues at the miami chil- dren’s Hospital retrospectively reviewed the records of 288 patients. Overall, headache was a relatively uncommon chief complaint, constituting just 1.3% of visits. Patients ranged in age from 2 to 18 years. Diagnoses established were viral illness (39.2%), sinusitis (16.0%), migraine (15.6%), post-traumatic headache (6.6%), viral meningitis (5.2%), streptococcal pharyngitis (5.2%), tension headache (4.5%), and "other" (7.7%). The "other" category included one child with a shunt malfunction, one with newly diagnosed hydrocephalus, another with a punctate hemorrhage secondary to head trauma, and a child with known Burkitt’s lymphoma and new CNS infiltration. These four patients, as well as those with viral meningitis, were considered to have "serious neurologic conditions," and they accounted for 16.6% of the total. In almost all of these cases, there were historical or physical findings consistent with the diagnosis.

A range of diagnostic tests were performed, including rapid strep studies (31.9%), sinus x-rays (27.8%), brain imaging (19.0%), and lumbar puncture (8.0%). Of the rapid strep assays, 16.3% were positive. Radiologists interpreted 46.5% of the sinus x-rays as positive because of maxillary thickening/opacification. Sixty-one percent of the lumbar punctures yielded abnormal results, as did 19% (4 total) of the CT scans.

Burton et al compare their findings with the few previously published similar reports from general emergency department settings. In all of these, serious neurological conditions have been quite uncommon in pediatric patients evaluated for headache in the ED.


Burton et al have made an effort to define the range of diagnoses associated with a presenting complaint of headache in a pediatric ED. Not surprisingly, they found serious conditions in a minority of patients. Their study does suffer from a few limitations. It was retrospective, which might compromise the ability to extract complete information from the records. Second, a large number of physicians treated the patients, bringing into question the consistency of the physical examination data and the diagnostic criteria used. For example, one might challenge the seemingly frequent use of sinus x-rays in this study.

More troublesome is the method used to identify patients. A computer search was used to find patients with the specific chief complaint of headache at the time of ED registration. This may explain why there were no patients with brain tumors or intracranial hemorrhages or pseudotumor cerebri or intracranial abscesses—all diagnoses I have encountered over the last few years. Patients with these conditions usually complain of headache to various degrees, although it may not be recorded as the chief complaint.

What can we learn from this study? It reinforces the notion that, in the absence of other suggestive historical or physical findings, headache is unlikely to be the result of a significant intracranial process in the great majority of pediatric ED patients complaining of headache. It also supports our teaching that computed tomography is of low yield in the work-up of pediatric patients with headache but no other findings. The limited role of brain imaging in children with headaches was also emphasized in another study, where no relevant findings were found in 78 children evaluated in a neurology clinic for headaches.1 When the emergency physician is confronted with a child with a headache, the need for a careful history, a thorough examination, and close follow-up is important. (Dr. Bachman is Director of the Pediatric Emergency Service at the Maine Medical Center in Portland.)


1. Maytal J, et al. The value of brain imaging in children with headaches. Pediatrics 1995;96:412-416.