Impact of Aura and Status Migrainosus on Readmissions for Vascular Events After Migraine Admission

• Published in: Headache 2018-07, Vol.58 (7), p.964-972
• Main Authors: Velickovic Ostojic, Lili, Liang, John W., Sheikh, Huma U., Dhamoon, Mandip S.
• Format: Article
• Language: English
• Subjects: Age; Aneurysms; cerebrovascular disease; Cerebrovascular system; Hazards; Headache; Health risk assessment; Hemorrhage; Ischemia; Migraine; Regression analysis; Risk analysis; Risk factors; stroke; Subarachnoid hemorrhage; Transient ischemic attack
• ISSN: 0017-8748; 1526-4610
• DOI: 10.1111/head.13347

Objective —To estimate readmission rates for acute ischemic stroke (AIS), transient ischemic attack (TIA), subarachnoid hemorrhage, and intracerebral hemorrhage after an index admission for migraine, using nationally representative data. Methods —The Nationwide Readmissions Database was designed to analyze readmissions for all payers and uninsured, with data on >14 million US admissions in 2013. We used International Classification of Diseases, Ninth Revision, Clinical Modification codes to identify index migraine admissions with and without aura or status migrainosus, and readmissions for cerebrovascular events. Cox proportional hazards regression was performed for each outcome with aura and status migrainosus as main predictors, adjusting for age and vascular risk factors. Results —Out of 12,448 index admissions for migraine, 9972 (80.1%) were women, mean age was 45.5 ± 14.8 years, aura was present in 3038 (24.41%), and status migrainosus in 1798 (14.44%). The 30‐day readmission rate (per 100,000 index admissions) was 154 for ischemic stroke, 86 for TIA, 42 for subarachnoid hemorrhage, and 17 for intracranial hemorrhage. In unadjusted models, aura was significantly associated with TIA (hazard ratio 2.43, 95% CI 1.39‐4.24), but not AIS (1.26, 0.73‐2.18), intracranial hemorrhage (1.86, 0.45‐7.79) or subarachnoid hemorrhage (1.85, 0.44‐7.75). When adjusting for age and vascular risk factors, aura remained significantly associated with TIA (2.13, 1.22‐3.74). Status, in adjusted models, was significantly associated with subarachnoid hemorrhage readmission (4.83, 1.09‐21.42). Conclusions —In this large, nationally representative retrospective cohort study, migraine admission with aura was independently associated with TIA readmission, and status migrainosus was independently associated with subarachnoid hemorrhage. Further research would clarify the role of misdiagnosis and causal relationships underlying these strong associations.