SUMMARY Brain hemorrhage from an intracranial tumor was encountered in 7 males and 6 females during a 4-year period. In 5 patients, hemorrhage was responsible for the first signs of a previously unsuspected neoplasm. The intracranial lesion was demonstrated by computed tomography (CT scanning) in each patient. Characteristic CT scan findings included: a neoplastic core (high or low density); small, multifocal clots usually at the margin of the tumor; and, surrounding, often extensive, edema. Enhancement of the tumor tissue with intravenous injection of 60% Hypaque was observed in the 8 patients so studied. The regions which were enhanced had a peripheral distribution corresponding to the site of hemorrhage. Microscopic examination demonstrated 7 glioblastoma multiforme, 1 oligodendroglioma, 4 metastatic carcinomas (including 1 each of bronchogenic carcinoma, melanoma, hypernephroma, and adrenal carcinoma), and 1 hemangiopericytoma. High-grade malignancy and extensive, abnormal vascularity appeared to be predisposing factors.

Object The goal of this study was to determine whether the quantity of peritumoral brain edema displayed on computerized tomography (CT) scanning could be correlated with brain invasion and subsequent recurrence of meningiomas. Methods One hundred thirty-five patients who underwent resection of intracranial meningiomas at the Ottawa Civic Hospital were followed during the period 1980 to 1998. A complete resection was defined as one in which tumor, invaded bone, and involved dura were removed. Tumors were examined microscopically for evidence of brain invasion. The mean follow-up period was 9 years (± 4 years, standard deviation [SD]) and the mean time to recurrence was 5 years (± 4 years, SD). The authors used a simple grading system based on the average thickness (in centimeters) of edema seen on an axial CT slice showing the most tumor. Edema grade was linearly related to edema volume determined by digitizing the scans (r = 0.96; 29 cases). The chance of brain invasion increased by 20% for each centimeter of edema (rs = 1, p < 0.0001; 124 cases). The presence of brain invasion was predictive of recurrence after complete resection with an accuracy of 83%, a sensitivity of 89%, and a specificity of 82%. The chance of recurrence within 10 years after complete resection was given by the equation: percentage chance of recurrence = (centimeter of edema)3 X 0.7, which can be used to predict the chance of recurrence based on findings on CT scans (rs = 1.00, p < 0.0001; 86 patients). Statistical significance was confirmed using Kaplan-Meier and univariate and multivariate analyses. Completeness of resection was the most powerful predictor of recurrence (p < 0.00001, r = 0.6), followed by edema grade and brain invasion (both p = 0.02, r = 0.1). Patient age and gender and tumor location, size, and histological subtype were nonsignificant factors. Conclusions Brain invasion causes peritumoral edema. Invaded brain tissue is also the source of residual cells in cases of tumor recurrence after gross-total resection.

Brain invasion causes peritumoral edema. Invaded brain tissue is also the source of residual cells in cases of tumor recurrence after gross-total resection.