Hemodynamic factors may play a critical role in neurological deterioration occurring within 72 hrs after lacunar stroke

Results: Ten out of the 43 patients had END. Patients having END was significantly associated with lower chances of favorable and good outcomes at 3 months (p = 0.01 and p = 0.002, respectively). END was predicted when the non-core hypoperfused area overlapped on the corticospinal tract, which is defined as the expected END profile. Cohen's Kappa coefficient between the 2 image readers to define images of expected END profiles was 0.74. In 15 patients with expected END profile, 9 had END clinically, whereas 28 patients had no expected END profile, and only 1 patient had END (p<0.0001). After adjusting for sex, the expected END profile was still associated with END (odds ratio, 42.2; p = 0.002).

Conclusion: Our study demonstrated that the END in acute lacunar stroke is likely related to the transformation of non-core hypoperfused area into infarction in the anatomy of corticospinal tracts.

Background: Whether a perfusion defect exists in lacunar infarct and whether it is related to early neurological deterioration (END) is still under debate. The aim of this study was to evaluate whether END in lacunar infarct is related to a perfusion defect using diffusion-weighted imaging (DWI), diffusion tensor imaging (DTI) and perfusion MR imaging.

Methods: One hundred and forty-one consecutive patients had an MRI scan within 30 hours after onset of symptoms and 43 patients with acute lacunar infarct and classic lacunar syndrome were recruited. The MRI sequences included DWI, DTI and cerebral blood flow (CBF) maps to respectively represent the topographic locations of acute infarcts, the corticospinal tract and perfusion defects. The END was defined in reference to the National Institute of Health Stroke Scale (NIHSS) as an increase ≥2 within 72 hours. Cohen's Kappa coefficient was used to examine the reliability between the 2 image readers. A multivariate logistic regression model was constructed adjusting for baseline variables.