Abstract 2618: Blood Brain Barrier Permeability Increases in Intracerebral Hemorrhage do not Predict Perihematoma Edema Growth

Background: The etiology of edema in intracerebral hemorrhage (ICH) is controversial. Most studies are consistent with a vasogenic rather than cytotoxic pathophysiology. It is unknown, however, if the blood brain barrier (BBB) is compromised in ICH and to what extent this is related to either hematoma expansion or edema development. BBB permeability was measured with CT perfusion (CTP) in ICH patients. We hypothesized that perihematoma edema volume and growth would be associated with elevated permeability. We also tested the hypothesis that elevated PS would be associated with hematoma growth. Methods: Patients with ICH were assessed with CTP within 24 hours of symptom onset. Maps of BBB permeability-surface area product (PS), a measure of extravascular contrast leakage rate, were calculated from CTP source images. Mean PS was measured in regions of interest (ROI) defining the hematoma, a 1cm radial peri-hematoma region, and ipsilateral and contralateral hemispheres. The total perihematoma edema volume was also measured objectively, based on Hounsfield unit thresholds of ≤24 at the time of the CTP and 24 h later. Results: PS maps were generated in 50 patients (13 female). The median (IQR) time from symptom onset to CTP was 8.2 (12.4) h. Median hematoma volume at baseline was 16.4 (28.2) mL. Mean PS was higher in the hematoma (3.24±1.34 ml/100g/min), relative to contralateral regions (1.82±0.84 ml/100g/min, P<0.0001). PS was also elevated in the perihematoma region (2.50±1.13 ml/100g/min vs. 1.78±0.81 ml/100g/min, P<0.0001). Ipsilateral hemispheric PS was higher (2.08±1.00 ml/100g/min) than in the contralateral hemisphere (1.82±0.82 ml/100g/min, P=0.002). Acute hematoma volume did not predict PS elevations in the hematoma itself (R=0.004 [-0.01, 0.02], P=0.07) or perihematoma region (R=0.002 [-0.01, 0.01], P=0.74). Linear regression indicated that acute edema volume was not predicted by perihematoma PS (R=-0.68 [-2.15, 0.79], P=0.36). Follow-up CT scans at a median of 22.5 (IQR=2.4) h after the baseline CTP were obtained in 43 patients. Median perihematoma edema volume increased from 1.9 (2.6) mL to 3.2 (5.0) mL (P=<0.0001). This edema volume growth was not predicted by the acute perihematoma PS (R=0.03 [-1.62, 1.68], P=0.98). Significant hematoma expansion (>6 ml) was seen in 5 patients. There was a non-significant trend to higher median hematoma PS values in these 5 patients (3.7 (1.9) vs. 2.7 (1.7) in those with stable hematoma volumes, P=0.09). Conclusions: Small elevations in BBB permeability in and around the hematoma are common in ICH and do not appear to be related to hematoma volume. Neither acute perihematoma edema volume or edema growth is explained by BBB compromise. A relationship between elevated intra-hematoma permeability and significant hematoma expansion cannot be excluded and should be explored in a larger ICH population.