Abstract WP129: Half of Acute Cortical Lesions in Minor Acute Ischemic Stroke Patients Persist on Flair at 30-day
Catrina S Sullivan, Marie Luby, Allison D Griffin, Sana Somani, Amie W Hsia, Lawrence L Latour- Advanced and Specialized Nursing
- Cardiology and Cardiovascular Medicine
- Neurology (clinical)
Introduction: Ischemic lesions seen on acute DWI often result in chronic infarction on FLAIR. However, in minor stroke a proportion of cortical lesions seem to disappear. The objective of this study was to characterize the persistence rate of acute cortical lesions on FLAIR and associated imaging markers on 3DT1 at 30-day.
Methods: Consented minor stroke patients (NIHSS<6) presented acutely (<24 hrs) received MRI at baseline, 5-day, and 30-day. One qualifying cortical lesion per patient was evaluated on baseline DWI for volume, white matter involvement, and on follow-up FLAIR for signal intensity ratio (SIR): comparison of lesion vs contralateral homologous tissue. Three independent raters blinded to hemisphere and exact location of the DWI lesion evaluated 3DT1 and FLAIR on follow-up (Figure).
Results: Of 159 consented patients, 30 met criteria, and lesions of 28 patients (18%) evolved as hyperintense on FLAIR at 5-day. The 28 patients had a median age of 70 years, 54% female, median admit NIHSS of 2, and 32% were treated with IV alteplase. Of the 28 lesions, 17 (61%) had white matter involvement at baseline, and 14 (50%) were not evident on FLAIR at 30-day. Comparing those hyperintense vs not evident on FLAIR, raters identified 79% vs 21% (p<0.003) cavitation in white matter, 36% vs 0% (p<0.01) cavitation in gray matter, 57% vs 0% (p<0.001) laminar necrosis, and 100% vs 14% (p<0.001) cortical thinning on 3DT1. Lesions that persisted on FLAIR vs lesions not evident at 30-day had larger baseline DWI lesion volume: 2.2mL vs 0.3mL (p=0.024), and differences in SIR: 1.63 vs 1.3 (p=0.009) at 5-day, and 1.56 vs 1.1 (p=0.003) at 30-day.
Conclusions: Half of the acute cortical lesions persisted on FLAIR at 30-day. Imaging markers on 3DT1 were more frequently associated with FLAIR evident lesions, but in ~1/3 of lesions not hyperintense, readers identified 3DT1 changes. Quantitative analysis is underway to confirm this observation.