Abstract
BACKGROUND: Early neurologic deterioration (END) occurs in a quarter of acute lacunar infarcts, but the underlying pathophysiological features are poorly understood. We sought to determine the association between luminal stenosis (<50%) of the parent artery and END. METHODS: This observational study included consecutive patients with lacunar stroke from the ischemic stroke registries of New York University Langone Health and Brown University. All included patients were admitted for acute lacunar stroke, which was defined as a subcortical infarct <1.5 cm on computed tomography or <2 cm on diffusion-weighted imaging without significant stenosis (>50%) in the parent vessel and no cardioembolic source. We defined END as any neurologic deterioration referable to the acute lacunar stroke and not related to a medical or noncerebrovascular neurological complication. We used univariate and logistic regression analyses to determine associations between luminal stenosis (<50%) and the odds of END. Furthermore, we attempted to validate findings using the Columbia University Medical Center stroke registry and perform a meta-analysis combining the derivation and validation groups because of the expected small samples and event rates. RESULTS: The New York University Langone Health and Brown University sample included 205 patients, of whom 41 (20%) had END. In adjusted models, we found no definite association between luminal stenosis (<50%) and END (adjusted odds ratio [OR], 1.74; 95% CI, 0.73-4.14). From Columbia University Medical Center, 361 total patients were included, of whom 59 (16%) had END. In adjusted models, we found an association between luminal stenosis (<50%) and END (adjusted OR, 2.28; 95% CI, 1.15-4.50). Meta-analysis of both cohorts found luminal stenosis (<50%) associated with END (relative risk, 1.69; 95% CI, 1.17-2.43). CONCLUSIONS: In this multicenter study, luminal stenosis (<50%) may be associated with END following an acute lacunar infarct. Larger studies using vessel wall imaging are needed to validate our findings.