Post-Traumatic Headache in Children and Genetic Risk of Migraine: An Observational Cohort Study.

BACKGROUND AND

OBJECTIVES

To examine the association between genetic risk of migraine and post-traumatic headache (PTH) in children.

METHODS

This is a secondary analysis of a cohort study that recruited children aged 8-17 years within 48 hours of concussion or orthopedic injury (OI) from 5 pediatric emergency departments and followed them for 6 months. Genetic risk of migraine was estimated through (1) family history of migraine, (2) polygenic risk scores (PRSs), and (3) forty individual single-nucleotide polymorphisms (SNPs). PTH was categorized as: no PTH (no PTH), PTH nonmigraine phenotype (PTH-NM), or PTH migraine phenotype. Ordinal regression models with cluster-adjusted standard errors and restricted cubic splines were used to estimate associations between genetic risk of migraine and increasing severity of PTH (no PTH vs PTH-NM vs PTH migraine phenotype), adjusting for injury type (concussion vs OI), age, sex, time since injury, first principal component, deprivation indices, concussion history, and preinjury headache history.

RESULTS

The sample included 646 children (median [interquartile range] age = 12.38 [10.52-14.41] years, including 436 (67.5%) with concussion. Family history of migraine was associated with increasing severity of PTH in 2/5 models (rs7684253 model: χ2 = 3.99, df = 1, p = 0.046; rs75002882 model: χ2 = 4.02, df = 1, p = 0.045), while migraine PRS was not (χ2 = 12.95, df = 12, p = 0.373). Four SNPs were associated with increasing severity of PTH (rs13078967: χ2 = 103.87, df = 6, p < 0.001; rs7684253: χ2 = 12.90, df = 6, p = 0.045; rs2160875: χ2 = 17.35, df = 6, p = 0.008; and rs75002882: χ2 = 291.79, df = 6, p < 0.001) and 2 remained significant after Bonferroni correction (rs13078967 and rs75002882).

DISCUSSION: Four migraine susceptibility genetic variants were associated with PTH, as was family history of migraine in some models, but a migraine PRS was not.