The clarification of factors that contribute to hematoma expansion in the setting of intracerebral hemorrhage (ICH) and the relevant physical dynamics are implemental for development of management strategies. Herein, we assessed the interplay between hematoma expansion and surface regularity of intracerebral bleeds. To do so, hematoma contours were outlined on admission and follow-up computed tomography (CT) studies using semi-automated thresholding algorithms in 133 ICH patients. Hematoma volume, surface area and surface regularity [SR=6√πvolumesurfacearea3, ranging from 0 (very irregular surface) to 1 (perfectly regular surface suggestive of 3D spherical structure)] were determined by 3D Slicer software (www.slicer.org). Hematoma growth was defined as ≥33% relative growth, or ≥ 6 mL absolute growth. Our results are as follows: The median (IQR) hematoma volume was 14.2 (6.0-34.9) mL on admission CT obtained 2.4 (1.5-4.4) hours after symptom onset; the mean ± SD SR value was calculated as 0.62 ± 0.14. Patients who underwent imaging at earlier time points were more likely to have higher SR (r = 0.18; p = 0.035). The median hematoma volume at follow-up, 35 (21-47) hours after the initial scan, was 19.7 (6.9-44.4) mL. The regularity index decreased significantly at this time point to 0.58 ± 0.13 (p < 0.001) and corresponding increase of surface irregularity was independent of change in hematoma volume. Baseline hematoma volume, INR, and time to initial imaging were significant predictors of hematoma expansion. In conclusion, our findings suggest that hematomas evolve into more irregular 3D shapes during follow-up. These observations are consistent with the 'domino' hypothesis put forward for ICH expansion.
Keywords: 3D modeling; ICH expansion; Intracerebral hemorrhage; Island sign; Prediction bio-marker; Spot sign.
Copyright © 2020 Elsevier Ltd. All rights reserved.