Carotid Perivascular Adipose Tissue Density as a Marker of Large Artery Atherosclerotic Stroke in Patients Undergoing Mechanical Thrombectomy for Acute Middle Cerebral Artery Occlusion

Background/Objectives: Carotid perivascular adipose tissue (PVAT) density on computed tomography angiography (CTA) is a noninvasive surrogate marker of local vascular inflammation, but its relevance to stroke etiology in a homogeneous cohort of patients undergoing mechanical thrombectomy (MT) remains unclear. Methods: We retrospectively analyzed 146 consecutive patients with acute ischemic stroke treated with MT for acute middle cerebral artery (MCA) occlusion between May 2018 and August 2024. Baseline CTA was used to quantify carotid PVAT density with two 2–3 mm2 circular regions of interest per internal carotid artery (ICA), placed ≥1 mm from the vessel wall. Measurements were performed bilaterally, and the ICA ipsilateral to the occluded MCA was defined as the stroke-side ICA. Etiology was classified according to the Trial of ORG 10172 in Acute Stroke Treatment (TOAST) system and grouped as large-artery atherosclerosis (LAA), cardioembolism (CE), and other/undetermined (OD/UD). Interobserver agreement was assessed using the intraclass correlation coefficient. Results: The mean age was 72.21 ± 12.39 years; 83.6% of patients achieved successful recanalization (mTICI ≥ 2b), and 47.9% had a favorable 90-day outcome (mRS ≤ 2). In the LAA subgroup (n = 38), ipsilateral PVAT density was significantly higher (less negative) than contralateral PVAT density (−64.24 ± 11.74 vs. −78.22 ± 9.13 HU; p < 0.001). Ipsilateral PVAT density differed significantly across TOAST groups (ANOVA p = 0.004), being higher in LAA than in CE (Δ = 11.19 HU; p = 0.003) and OD/UD (Δ = 9.54 HU; p = 0.004). ROC analysis showed modest discrimination for LAA versus non-LAA stroke (AUC 0.67, 95% CI 0.58–0.75), with an optimal cutoff of −79 HU (sensitivity 92.1%, specificity 40.7%). In multivariable logistic regression, higher ipsilateral PVAT density was independently associated with LAA etiology (per 1-HU increase: OR 1.048, 95% CI 1.018–1.079; p = 0.0016). PVAT density was not associated with recanalization success or 90-day functional outcome. Conclusions: In patients with acute MCA occlusion undergoing MT, higher carotid PVAT density on the stroke side was independently associated with LAA stroke etiology but had limited value for predicting MT success or short-term clinical outcome.