P21.01.A EPIDEMIOLOGY OF ISCHEMIC LESIONS AFTER RESECTION OF LOW-GRADE GLIOMAS

Abstract

Abstract BACKGROUND Diffuse low-grade gliomas are slow-growing malignant brain tumors that mainly affect people in their 30s and 40s, usually resulting in death within a decade. The standard treatment is surgery, which has a moderate risk of neurological sequelae. Clinical experience indicates that not all sequels are directly related to the resected area but rather caused by collateral damage, such as tissue- and vascular damages at the surgical site, causing ischemic lesions in the surrounding tissue. In this study, the epidemiology of these lesions is investigated, in terms of how common they are and how they are related to the size, type and location of the tumor. MATERIAL AND METHODS One hundred and sixty-six patients who underwent resection for grade II glioma (WHO 2007 or 2016) between 2012 and 2017 at any of nine hospitals in Sweden and Norway and had a postoperative MRI scan including diffusion-weighted imaging within three days after surgery, were included in this study. Lesions were segmented on the b1000 image, using the ADC and T1 image to exclude hyperintense areas caused by other factors than ischemia. Statistical analyses were performed using IBM SPSS Statistics version 27.0 (IBM Corp., Armonk, NY, USA). Analyses were carried out with ischemic lesion size as a continuous variable, and since the volumes were not normally distributed, non-parametric tests were used. RESULTS Ischemic lesions were found in 146 out of 166 (88%) cases, the median volume of the segmented lesions was 1.9 ml, with an interquartile range (IQR) of 0.95-4.5 ml. Patients with new or worsened deficits did have significantly larger ischemic volumes than does who did not (1.7 vs 2.8 ml, p=0.03). Ischemic volume was significantly correlated to both pre- and post-operative tumor volume (Spearmans rho = 0.370 and 0.304, p<0.001). No difference in ischemic volume was found between astrocytoma and oligodendroglioma, based on reported histopathologic diagnosis (p=0.43) or molecular status (p=0.44), neither between main tumor local (frontal (60%) vs temporal (19%), parietal (8%) or insular (12%), p>0.37). CONCLUSION Ischemic lesions after glioma resection are very common, but mostly small and restricted to a region close to the resection cavity. Large ischemic regions were associated with new or worsened neurological deficits and were mostly found in association with large tumors. More fine-tuned measures might be needed to find potential associations between tumor local and ischemic lesions.