Abstract 214: Corticospinal Tract Integrity is Acutely Maintained Within Perihematoma Edema

Abstract

Background: Perihematoma edema in intracerebral hemorrhage (ICH) is thought to be associated with tissue injury. Fractional Anisotropy (FA), as measured with Diffusion Tensor Imaging (DTI), can be used to assess white matter integrity. We tested the hypotheses that sections of the Corticospinal Tract (CST) passing directly through edema would 1) have low FA relative to the entire tract and 2) predict motor score in ICH patients. Methods: Patients were prospectively imaged with DTI within 14 days of symptom onset. Hematoma volume was measured on CT using planimetric techniques. Edema volume was assessed on CT using a 5-23 Hounsfield Unit threshold. Mean FA was measured in the edematous region (‘perihematoma edema'), the entire ipsilateral CST, and in the portion of CST passing through the perihematoma edema ('edematous CST'). Motor function was evaluated with a composite of the upper and lower extremity NIHSS motor score (0=normal, 8=hemiplegia). Results: Patients (n=27, mean age 67±13) were scanned with DTI at a median of 2 (3) days. Hematoma distribution was: lobar 5 (18%), basal ganglia 21 (78%), and brainstem 1 (4%). Median acute ICH volume was 8.2 (22) ml at 2 (2)h. Acute edema volume was 0.9 (1.8) ml and grew to 1.9 (3.9) ml at 26h (26). NIHSS motor score was 3 (6) at 72h and 3 (7) at day 7. FA in the edema was significantly lower (0.23±0.06) than in contralateral mirror regions (0.37 ±0.07, p<0.0001). Mean FA in the edematous CST was lower (0.34±0.08) than FA in the entire ipsilateral CST (0.44±0.04, p<0.0001), but higher than FA in the perihematoma edema (0.25±0.06, p<0.0001). There was a weak correlation between FA in the edematous CST and 72h motor score (r= -0.40, p=0.050,) which disappeared at day 7 (r= -0.34, p=0.131). FA in the edematous CST was not related to time to scan (r=-0.027, p=0.892). Hematoma volume predicted FA in the edematous CST (ß=-0.46, 95% CI:[-0.05- -0.01]; p=0.015). Perihematoma edema volume did not predict FA in the edematous CST independently of the ICH (ß=-0.29, [-0.1-0.04]; p=0.475). Conclusion: FA is decreased in the CST where it passes through the edema, though not to the extent of the surrounding edema. The transient relationship between decreased FA and motor function suggests that edema temporarily impairs tract function but not integrity.