A 9-Year Longitudinal Study of Basilar Artery Diameter

Abstract

Background: Dilatation of the basilar artery (BA) has been recognized as a predictor of cardiovascular events (CVEs). However, it is unclear if the longitudinal change in BA diameter (ΔBA) is associated with CVEs. Methods and Results: In a cohort of Japanese participants with vascular risk factors in an observational study, we evaluated the relationship of ΔBA to CVEs and the time course of the BA diameter. The short axis of the BA diameter was measured at the midpons level in T2-weighted images. Brain magnetic resonance imaging measurements included cerebral small-vessel disease, lacunars, and white matter hyperintensities. First, 493 patients were analyzed by the time-dependent Cox proportional hazards model to evaluate the association between ΔBA and CVEs, with adjustment for age, sex, vascular risk factors, and magnetic resonance imaging parameters. Second, we assessed the longitudinal ΔBA in 164 patients who underwent long-term follow-up magnetic resonance imaging, by linear regression analysis. In the mean follow-up of 8.7 years, 105 patients developed CVEs. A smaller ΔBA was independently associated with the high incidence of CVEs (hazard ratio, 0.36; 95% CI, 0.16–0.78; P=0.010; n=493). After a mean interval of 9.4 years, the average ΔBA was 0.41±0.46 mm (excluding patients with fetal-type circle of Willis). Progression of BA dilatation was associated with men but inversely associated with initial BA diameter and fetal-type circle of Willis (n=164). Conclusions: BA diameter increased over time (excluding the patients with fetal-type circle of Willis), whereas ΔBA was inversely associated with the incidence of CVEs.