Document resource
Introduction Systemic inflammation is associated with poor outcomes in symptomatic intracranial atherosclerotic stenosis (sICAS); however, the underlying mechanisms remain unclear. We investigated whether urinary albumin-to-creatinine ratio (UACR), as a clinically accessible marker of endothelial dysfunction-related microvascular injury, may mediate the association between inflammation and prognosis.Patients and methods We analysed 2,267 patients with sICAS from the Third China National Stroke Registry. High-sensitivity C-reactive protein (hsCRP) ≥ 2 mg/L and UACR ≥ 30 mg/g were exposures, and poor outcome (modified Rankin Scale (mRS) 3–6) at 90 days was the endpoint. Multivariable logistic regression assessed independent associations, and mediation analysis quantified the contribution of UACR to the hsCRP–outcome association. In a 217-patient sICAS-Computational Fluid Dynamics (sICAS-CFD) cohort, patient-specific models examined links between UACR and post-stenotic perfusion.Results Both high-sensitivity C reactive protein (hsCRP) ≥2 mg/L (adjusted OR (aOR) 1.43, 95% CI 1.12 to 1.84) and UACR ≥30 mg/g (aOR 1.91, 95% CI 1.49 to 2.45) were independently associated with 90-day mRS 3–6. In a prespecified mediation framework, UACR accounted for an estimated 22.2% of the association between elevated hsCRP and 90-day poor outcome (p=0.008), with larger indirect effects in patients with elevated systolic blood pressure (proportion mediated (PM) 17.5%) or diabetes (PM 26.4%; p<0.05), consistent with systemic microvascular vulnerability. In the computational fluid dynamics cohort, hypoperfused patients with poststenotic mean arterial pressure ≥70 mm Hg exhibited higher UACR than those with lower pressure (p=0.040).Discussion and conclusion In sICAS, UACR statistically mediated the association between inflammation and functional outcome, and higher UACR was exploratorily associated with poststenotic perfusion measures, providing exploratory haemodynamic context that may help explain variation in perfusion vulnerability.