Abstract
OBJECTIVE: To investigate an association between percentage lipid-rich necrotic core (LRNC) and an index ischemic stroke in an embolic stroke of undetermined source (ESUS) cohort. METHODS: A total of 167 ESUS patients with 259 non-stenotic intracranial plaques including 155 ipsilateral and 104 contralateral to stroke were finally enrolled in the current analysis. The multi-dimensional parameters involving remodeling index (RI), plaque burden (PB), LRNC, discontinuity of plaque surface (DPS), intraplaque hemorrhage (IPH), and vulnerable plaque defined as presence of complicated plaque were evaluated by high-resolution magnetic resonance imaging. RESULTS: We found that %LRNC was an independent predictor for ESUS in model 1 (OR: 2.574, 95% CI: 1.854-3.573, P < 0.001), and model 2 (OR: 2.550, 95% CI: 1.835-3.545, P < 0.001), but the association was not seen in PB. In receiver operating characteristic curve analysis, the discrimination of LRNC for ESUS was significantly superior to that of PB (absolute difference: 0.121, 95% CI: 0.056-0.205, P < 0.001). Importantly, a significantly positive synergy between the remodeling pattern and LRNC in response to plaque vulnerability was found by Sankey diagram (P for interaction = 0.001). CONCLUSION: This is the first report that LRNC, beyond PB, may be correlated with an index ESUS, and a synergistic effect between positive remodeling and larger LRNC could promote plaque vulnerability. The findings suggest that a potential target subgroup may benefit from stroke prevention with intensive statin, although this must be confirmed in future.