Predictors of First-Line Treatment Success in Children and Adolescents Visiting an Infusion Center for Acute Migraine.

OBJECTIVES

To characterize a population of pediatric patients visiting an infusion center for acute migraine and determine predictors of first-line treatment success in this population.

BACKGROUND

Though migraine is common in the pediatric emergency department and specialized infusion centers, little is known about this patient population and whether or not clinical data can be used to predict treatment response.

METHODS

This was an observational study involving a retrospective analysis of data from visits to the Cincinnati Children's Hospital infusion center for treatment of an acute migraine. Data were extracted from a database and chart reviews were completed for missing or outlying data. Descriptive statistics were used to outline patient: sex, age, race, primary ICHD-III diagnosis, chronic migraine, medication overuse headache (MOH), headache frequency, month of treatment, headache severity, headache duration, use of acute medication at home in the past 24 hours and treatment received (metoclopramide vs prochlorperazine and dexamethasone vs no dexamethasone). The odds of success of first-line intervention were modeled using simple logistic regression with the above characteristics used as predictors. Predictors with a P value <.05 in the multiple logistic regression model with progressive backwards selection were entered into a final multiple logistic regression model controlling for age, sex and diagnosis, where a P value <.05 was considered statistically significant.

RESULTS

Data from 837 visits (n = 837) were included in this analysis. Visits were significantly more frequent in the academic year as compared to the summer (Z = 10.716, P < .0001). In the multiple logistic regression model, the odds of first-line treatment success decreased as headache frequency increased (OR = 0.951, 95% CI = 0.934-0.969, P < .0001), were higher for patients without MOH (OR = 1.528, 95% CI = 1.097-3.714, P = .0469), and were higher for patients who did not receive dexamethasone (OR = 1.528, 95% CI = 1.097-3.714, P = .0469).

CONCLUSIONS

Higher baseline headache frequency and presence of MOH may predict lower odds of treatment success after first-line intervention for pediatric migraine in the acute setting. These predictors may be useful in stratifying patients for treatment protocols and for planning future intervention studies.