CTA-Based Plaque-RADS to Assess Etiologic Role of Nonstenotic Carotid Plaques in Patients With Embolic Stroke of Undetermined Source

Background and Objectives – Nonstenotic (<50%) carotid plaques (NSCPs) are increasingly recognized as potential embolic sources in embolic stroke of undetermined source (ESUS), but their etiologic role remains uncertain because of limited data and the lack of standardized definition of high-risk plaque. We evaluated the clinical applicability and reliability of a modified CT angiography (CTA)–based Plaque-Reporting and Data System (RADS) classification for characterizing NSCPs in ESUS and examined the association between high-risk NSCPs and stroke laterality, the coexistence of competing stroke mechanisms, and the risk of stroke recurrence.Methods – We conducted a single-center cohort study (January 2016–June 2025) including consecutive ESUS patients with unilateral anterior circulation stroke and neck CTA, retrospectively and prospectively enrolled. NSCPs were classified using a modified CTA-based Plaque-RADS: (1) normal vessel wall; (2) maximum wall thickness (MWT) <3 mm; (3) MWT ≥3 mm (3ab) or ulceration (3c); (4) intraplaque hemorrhage (4a) or intraluminal thrombus (4c). Patients were stratified according to the presence of ipsilateral high-risk Plaque-RADS (categories 3–4) vs ipsilateral low-risk Plaque-RADS (categories 1–2). Inter-rater agreement for CTA-based Plaque-RADS was assessed, and associations between high-risk Plaque-RADS and stroke side, competing embolic sources, and recurrent ischemic events were evaluated with multivariable regression and survival analyses.Results – Among 512 patients (median age 70 years, 47.8% women), 109 (21.3%) had ipsilateral high-risk and 403 (78.7%) had low-risk Plaque-RADS scores. Inter-rater agreement was excellent (Cohen κ = 0.83, p < 0.001). High-risk Plaque-RADS scores were more frequent ipsilaterally than contralaterally (21.3% vs 11.9%; adjusted odds ratio [aOR] 2.65, 95% CI 1.73–4.06, p < 0.001). Ipsilateral high-risk Plaque-RADS scores were inversely associated with cardioembolic sources, including atrial cardiopathy (34.9% vs 39.2%; aOR 0.59, 95% CI 0.36–0.96, p = 0.036), high-risk patent foramen ovale (2.7% vs 21.8%; aOR 0.14, 95% CI 0.03–0.60, p = 0.008), and atrial fibrillation detected after stroke (8.4% vs 17.5%; adjusted subdistribution hazard ratio [aSHR] 0.46, 95% CI 0.23–0.95, p = 0.037). During follow-up, overall stroke recurrence did not significantly differ, but ipsilateral high-risk Plaque-RADS scores were significantly associated with an increased risk of ipsilateral stroke recurrence (11.2% vs 3.0%; aSHR 3.19, 95% CI 1.15–8.83, p = 0.026).Discussion – CTA-based Plaque-RADS can be reliably applied in routine clinical practice to categorize NSCPs in ESUS. Ipsilateral high-risk NSCPs are strongly associated with the index stroke side, lower prevalence of competing cardioembolic sources, and increased ipsilateral recurrence, supporting their role as a distinct causal mechanism within ESUS and underscoring the need for randomized trials on secondary prevention.