Early edema in warfarin-related intracerebral hemorrhage.

OBJECTIVE

The pathophysiology and clinical significance of perihematomal edema (PHE), a cause of secondary neuronal injury after intracerebral hemorrhage (ICH), is poorly understood. A leading theory proposes that early PHE results from activation of the clotting cascade. We sought to test this theory by examining the relationship between early PHE and warfarin use in ICH patients.

METHODS

ICH and PHE volumes were measured in consecutive patients with warfarin-related ICH and compared to those of controls with non-coagulopathic ICH. Subjects were identified from a prospective database of ICH patients. Clinical and radiological predictors of PHE volume and relative PHE (PHE volume/ICH volume) were identified. The relationship between PHE volume and 90-day mortality was determined.

RESULTS

For the 49 consecutive warfarin-related ICH patients and 49 matched controls: median INRs (interquartile ranges) were 3.2 (2.3, 4.1) and 1.1 (1.08, 1.2); median hematoma volumes were 37.8 cm(3) (6.7, 102.9) and 18.1 cm(3) (9, 51) (P = 0.18); median PHE volumes were 12 cm(3) (3.7, 36.7), and 11 cm(3) (4.1, 24) (P = 0.87); and median relative PHE was 0.38 (0.28, 0.52) and 2 (1.37, 3.06), respectively. In multivariable analysis, ICH volume and warfarin use independently predicted PHE volume. There was an association between higher PHE volume and decreased 90-day mortality.

CONCLUSIONS

Warfarin-related ICH is associated with less early relative edema than non-coagulopathic ICH. This is consistent with the theory that coagulation contributes to early edema. Early edema may be associated with improved functional outcome.