Predicting recurrence of intracranial meningiomas. A multivariate clinicopathologic model--interim report of the New York University Medical Center Meningioma Project.

A series of 105 primary intracranial meningiomas, a subset of a larger project on over 800 such tumors, was assessed for clinical and histopathologic factors that might be predictive of recurrence. Predicting recurrence of meningiomas after a surgical resection and the criteria for and clinical importance of a diagnosis of "malignant" meningiomas are controversial issues. Using a multivariate logistic regression analysis, the data from these 105 patients suggest that the single most powerful predictor of recurrence within 5 years after tumor resection is the mitotic index of the tumor. The other powerful factor for predicting recurrence was the extent of resection. Other factors traditionally associated with recurrence or a diagnosis of malignancy were not significant factors for predicting recurrence according to this study.