Incidence, risk factors, and outcome of venous infarction after meningioma surgery in 705 patients

Central to safe and effective surgical resection of meningiomas is consideration of the venous anatomy both near and intrinsic to the tumor. The exact incidence of venous infarction following meningioma surgery has not been established. To determine this incidence, we present a large multivariate analysis of 705 patients undergoing craniotomy for resection of a histologically proven meningioma at our institution between 1991 and 2007. Clinical information was retrospectively reconstructed using patient medical records and radiologic data. Venous infarctions were identified by postoperative CT scans or MRI that demonstrated the typical imaging findings. Stepwise multivariate logistic regression analysis was performed to test the association with approach used and the rate of venous infarction, controlling for multiple independent variables. The overall rate of venous infarction (n = 705) was 2.0% of all patients (95% confidence interval [CI], 0.9-3.0%). Interestingly, on multivariate logistic regression analysis, we found the use of a bifrontal craniotomy was the sole independent predictor of venous infarction in this regression model (odds ratio, 3.18; 95% CI, 1.03-9.77; p < 0.05). We found that the rate of venous infarction was significantly reduced in the extended bifrontal group compared to the group not receiving biorbital osteotomies (0% versus 8.9%, ??²p < 0.05). We demonstrated that the most important factor determining the risk of venous infarction is the approach used to access the tumor.