Since the initial description of CACNA1A-related hemiplegic migraine (HM), the phenotypic spectrum has expanded from mild episodes in neurotypical individuals to potentially life-threatening events frequently seen in individuals with developmental and epileptic encephalopathies. However, the overall longitudinal course throughout childhood remains unknown. We analyzed HM and seizure history from electronic medical records in individuals with CACNA1A-related HM, delineating frequency and severity of events in monthly increments through a standardized approach. Combining these data with medication prescription information, we assessed the response of HM to different agents. Our cohort involved 15 individuals between 3 and 29 years (163 patient years) and included 11 unique and 2 recurrent variants (p.R1349Q and p.V1393M; both n = 2). The age of first confirmed HM ranged from 14 months to 13 years (average 3 years). 25% of all HM events were severe (lasting >3 days), and 73% of individuals had at least 1 severe occurrence. Spacing of HM events ranged from 1 month to 14 years, and changes in HM severity over time showed increases or decreases of >2 severity levels in 12 of 122 events. Eight individuals had epilepsy, but severity of epilepsy was only weakly correlated with frequency and severity of HM events. While levetiracetam (n = 6) and acetazolamide (n = 5) were the most frequently used medications, they did not show efficacy in HM prevention or severity reduction. However, verapamil (risk differences [RD] 0.10, CI 0.05-0.15) and valproate (RD 0.08, CI 0.04-0.12) were associated with a modest prevention of HM, but not reduction in severity. The longitudinal course of CACNA1A-related HM lacks recognizable patterns for timing and severity of HM events or strong correlation with seizure patterns. Our data underscore the unpredictability of CACNA1A-related HM, highlighting the need for close surveillance for reoccurring HM events even in individuals with symptom-free periods.
Read full abstract