Migraine and risk of stroke and acute coronary syndrome in two case-control studies in the Danish population

Publikation: Bidrag til tidsskriftTidsskriftartikelfagfællebedømt

INTRODUCTION: Migraine has consistently been associated with increased risk of ischemic stroke, while the evidence for a relation with other types of stroke or coronary outcomes is limited. We examined the association between migraine and stroke and acute coronary syndrome (ACS) subtypes and the influence of potential confounding factors.

METHODS: All first-time hospital contacts for stroke (n=155,216) or ACS (n=97,799) were identified in Danish National Patient Registers and matched with 2 control groups of the background population. A hospital diagnosis of migraine and use of migraine medication were the main exposures and associations (odds ratios [OR]) were estimated using multiple logistic regression. Confounding was also addressed by including use of general headache medication as a negative control exposure.

RESULTS: The diagnosis of migraine was associated with increased odds of both stroke (ORcrude, age <50 years: 4.80 [95% CI: 3.75-6.21]; ORcrude, age ≥50 years:1.91 [95% CI: 1.67-2.19]) and ACS (ORcrude:1.88 [95% CI: 1.53-2.32]), while the ORs for the associations between migraine medication and stroke and ACS were lower. Patients with a diagnosis of migraine or redeemed migraine medication had increased ORs of all stroke subtypes (ischemic, hemorrhagic stroke and transient ischemic attacks). The diagnosis of migraine was also associated with both angina and myocardial infarction (ST-elevation Myocardial Infarction [STEMI], non-STEMI and unspecified) with the highest OR for angina. These associations were not fully explained by adjustment for confounding co-variables or when compared with the negative control exposure that were assumed to be influenced by similar confounding factors, but no shared pathogenesis.

CONCLUSION: Hospital-diagnosed migraine was associated with all stroke and ACS subtypes, with ischemic stroke and angina having the highest odds. Confounding did not explain the associations.

OriginalsprogEngelsk
TidsskriftClinical Epidemiology
Vol/bind9
Sider (fra-til)439-449
Antal sider11
ISSN1179-1349
DOI
StatusUdgivet - 31 aug. 2017

Antal downloads er baseret på statistik fra Google Scholar og www.ku.dk


Ingen data tilgængelig

ID: 184669107