Carl E Stafstrom1, Kevin Rostasy, Anna Minster. 1. Departments of Pediatrics and Neurology, Division of Pediatric Neurology, Floating Hospital for Children, New England Medical Center, Tufts University School of Medicine, Boston, Massachusetts, USA. stafstrom@neurology.wisc.edu
Abstract
OBJECTIVE: To determine whether drawings can aid in the differential diagnosis of headaches in children. METHODS: Before taking any history, 226 children who were seen consecutively for the evaluation of headache were asked to draw a picture to show how their headache felt. The pictures were then scored as migraine or nonmigraine by pediatric neurologists who were blinded to the clinical history. A clinical diagnosis of headache type was determined independently by another pediatric neurologist using the usual history and examination. The diagnoses of headache type based on the pictures drawn and the clinical findings obtained were then compared to calculate the sensitivity, specificity, and predictive values of the drawings for the diagnosis of migraine. RESULTS: Children produced dramatic and insightful headache drawings. Compared with the clinical diagnosis (gold standard), headache drawings had a sensitivity of 93.1%, a specificity of 82.7%, and a positive predictive value (PPV) of 87.1% for migraine. That is, drawings that contained an artistic feature consistent with migraine (eg, pounding pain, nausea/vomiting, desire to lie down, periorbital pain, photophobia, visual scotoma) predicted the clinical diagnosis of migraine in 87.1% of cases. Predictive values were also calculated for specific migraine-associated features on drawings: artistic depiction of focal neurologic signs, periorbital pain, recumbency, visual symptoms (photophobia, scotomata), or nausea/vomiting had a PPV of >90% for migraine; severe or pounding pain had a PPV of >80% for migraine. Band-like pain was not predictive of migraine (PPV of 11.1%). Features on drawings such as sadness or crying did not differentiate migraine from nonmigraine headaches. CONCLUSIONS: Children's headache drawings are a simple, inexpensive aid in the diagnosis of headache type, with a very high sensitivity, specificity, and predictive value for migraine versus nonmigraine headaches. We encourage the use of drawings in the evaluation of any child with a headache, as an adjunct to the clinical history and physical examination.
OBJECTIVE: To determine whether drawings can aid in the differential diagnosis of headaches in children. METHODS: Before taking any history, 226 children who were seen consecutively for the evaluation of headache were asked to draw a picture to show how their headache felt. The pictures were then scored as migraine or nonmigraine by pediatric neurologists who were blinded to the clinical history. A clinical diagnosis of headache type was determined independently by another pediatric neurologist using the usual history and examination. The diagnoses of headache type based on the pictures drawn and the clinical findings obtained were then compared to calculate the sensitivity, specificity, and predictive values of the drawings for the diagnosis of migraine. RESULTS:Children produced dramatic and insightful headache drawings. Compared with the clinical diagnosis (gold standard), headache drawings had a sensitivity of 93.1%, a specificity of 82.7%, and a positive predictive value (PPV) of 87.1% for migraine. That is, drawings that contained an artistic feature consistent with migraine (eg, pounding pain, nausea/vomiting, desire to lie down, periorbital pain, photophobia, visual scotoma) predicted the clinical diagnosis of migraine in 87.1% of cases. Predictive values were also calculated for specific migraine-associated features on drawings: artistic depiction of focal neurologic signs, periorbital pain, recumbency, visual symptoms (photophobia, scotomata), or nausea/vomiting had a PPV of >90% for migraine; severe or pounding pain had a PPV of >80% for migraine. Band-like pain was not predictive of migraine (PPV of 11.1%). Features on drawings such as sadness or crying did not differentiate migraine from nonmigraine headaches. CONCLUSIONS:Children's headache drawings are a simple, inexpensive aid in the diagnosis of headache type, with a very high sensitivity, specificity, and predictive value for migraine versus nonmigraine headaches. We encourage the use of drawings in the evaluation of any child with a headache, as an adjunct to the clinical history and physical examination.