Computer-Assisted Volumetric Analysis Compared With ABC/2 Method for Assessing Warfarin-Related Intracranial Hemorrhage Volumes

Abstract

Intraparenchymal hemorrhage (IPH) volume is a powerful predictor of 30-day mortality. Warfarin-related intracranial hemorrhage (ICH) has a higher mortality than ICH without anticoagulation, possibly due to continued growth after 24 h, larger average size, and extension to extraparenchymal compartments. We compared 2 methods of measuring ICH volume in patients with warfarin-related ICH. ICH volume was estimated using the ABC/2 method and a computer-assisted method (Analyze 6.0 software) applied to the initial head computed tomographic scans in a consecutive series of 8 patients with warfarin-related ICH. The 2 methods were compared for relative and absolute differences in estimated hematoma volumes. The ABC/2 method underestimated hematoma volume in 4 of 5 patients with IPH, inaccurately calculated intraventricular hemorrhage volume in 2 patients (overestimated by 9%, underestimated by 23%), and underestimated a complex subdural hematoma in 1 patient by 24% despite use of the Gebel modification. The mean percentage difference between the Analyze and ABC/2 methods was 24% in ellipsoid hemorrhages and 28% in nonellipsoid hemorrhages (P = 0.77). The mean of the absolute difference between the 2 methods was 6.7 cm3 in ellipsoid hemorrhages and 38.0 cm3 in nonellipsoid hemorrhages (P = 0.18). The ABC/2 method accurately and quickly estimates smaller, ellipsoid intraparenchymal hematomas but is inaccurate for larger, complex-shaped warfarin-related intraparenchymal, intraventricular, and subdural hematomas. Warfarin-related ICH mortality may be underestimated by the ABC/2 method because of larger, complex-shaped, and multicompartmental hematomas.