Clinical and radiological characteristics and recurrent event risk of cerebrovascular fibromuscular dysplasia in patients with Ischemic stroke or transient ischemic attack - a population-level study.

BACKGROUND

Data are scarce regarding the frequency, characteristics, and associated recurrent event risks of cerebrovascular fibromuscular dysplasia (cFMD) in patients presenting with ischemic stroke and transient ischemic attack (TIA).

METHODS

We included patients diagnosed with ischemic stroke/TIA in an entire Canadian province (Alberta) from 1-April-2016 to 31-March-2017, with follow-up of at least 5-years until 31-March-2022. Direct neuroimaging review and keyword searches of radiologist reports were used to identify suspected cFMD cases. Clinical and imaging characteristics and event etiology were ascertained. 5-year hazards of recurrent stroke/TIA were estimated, adjusted for age, sex, and vascular risk factors.

RESULTS

Among 4,033 patients with ischemic stroke/TIA who underwent CT/MR angiography, we found 35 (0.86%) cases of cFMD. Patients with cFMD had similar age (median:74[IQR:53-88] vs 69[IQR:20-99] years, p = 0.09) and higher female preponderance (80% vs 44.9%, p < 0.0001) versus those without. The extracranial proximal internal carotid artery (ICA) was most frequently involved. Most observed infarcts (23/26, 88.5%) occurred downstream from carotid/vertebral arteries with cFMD changes. Strokes/TIAs for most cFMD patients (27/35, 77.1%) were adjudicated to be of otherwise undetermined etiology following conventional investigations. All-cause mortality occurred in 10 (28.5%) and recurrent stroke/TIA in 8 (22.9%) patients at 5-years; this was not significantly higher than patients without FMD (aHR: 1.54, 95%CI:0.77-3.11).

CONCLUSIONS

cFMD was a rare finding in this subset of patients with stroke/TIA who underwent adequate neuroimaging, primarily affecting females and the extracranial proximal ICA. Infarcts in these patients often occurred in the territory of FMD-affected vessels and events were often classified as cryptogenic. Whereas these patients had substantial rates of recurrent stroke/TIA and mortality, the adjusted hazards versus patients without cFMD were imprecise.