OBJECTIVE: Recent findings have cast doubt on vasospasm as the sole cause of delayed cerebral ischemia after subarachnoid hemorrhage.
METHODS: We reviewed the medical records of 29 patients who died after subarachnoid hemorrhage. Brain sections were taken from the insula, cingulate gyrus, and hippocampus. Adjacent sections were stained with hematoxylin-eosin and immunostained for thromboemboli. The density (burden) of the latter was calculated blindly and correlated with evidence for ischemia and with the amount of subarachnoid blood.
RESULTS: There is a strong correlation between microclot burden and delayed cerebral ischemia. Patients with clinical or radiological evidence of delayed ischemia had mean microclot burdens of 10.0/cm2 (standard deviation [SD], ±6.6); those without had mean burdens of 2.8 (SD, ±2.6), a highly significant difference (P = 0.002). There is also significant association (P = 0.001) between microclot burden and histological evidence of ischemia, with the mean burdens being 10.9 in sections exhibiting severe ischemia and 4.1 in those in which ischemia was absent. Microclot burden is high in patients who died within 2 days of hemorrhage, decreasing on Days 3 and 4. In delayed ischemia, the numbers rise again late in the first week and remain high until after the second week. In contrast, the average clot burden is low in patients dying without developing delayed ischemia. The amount of blood on an individual slide influenced the microclot burden on that slide to a highly significant extent (P < 0.001).
CONCLUSION: Thromboembolism after subarachnoid hemorrhage may contribute to delayed cerebral ischemia, which parallels that caused by vasospasm. The pathogenesis of thromboembolism is discussed.