NIMG-64. SURVIVAL IMPLICATIONS OF POSTOPERATIVE RESTRICTED DIFFUSION IN HIGH-GRADE GLIOMA AND LIMITATIONS OF INTRAOPERATIVE MRI DETECTION

Abstract AIMS In glioma surgery, surgically induced cerebral ischemia may poorly affect overall survival. While intraoperative MRI (iMRI) is an established tool used to maximize extent of resection, its ability to detect cerebral ischemia during surgery has not been studied. Here we evaluate both the association of post-surgical cerebral ischemia on survival independent of new functional deficits and the sensitivity of iMRI in detecting cerebral ischemia. METHODS We retrospectively reviewed 361 cranial surgeries that used a 3 Tesla iMRI. 165 patients met all inclusion criteria and were included in the final analysis. Diffusion weighted imaging (DWI) obtained during iMRI was compared to postoperative DWI obtained within 7 days of the operation in cases where no further resection occurred after the iMRI. RESULTS 42 of 165 patients (25%) showed evidence of restricted diffusion on postoperative (poMRI); 37 of these 42 (88%) lacked evidence of restricted diffusion on iMRI (false-negative rate of 88%, sensitivity of 12%). In high-grade gliomas, poMRI restricted diffusion volume was predictive of overall survival independent of new functional deficit (p = 0.011). CONCLUSION Here we present the largest case series to date analyzing the sensitivity of iMRI in detecting ischemia. In high-grade gliomas, increased volume of ischemia correlated with worsening median overall survival (OS) irrespective of postoperative neurologic deficits. Future work will focus on improving intraoperative detection of ischemia during the hyperacute phase when interventions such as blood pressure modulation or direct application of vasodilator agents may be effective.