gms | German Medical Science

Objective: Intraoperative MRI (iMRI) is increasingly used to extend radicality during glioma resections. We have previously reported results showing that 99,6% mean final extend of resection (EOR) can be achieved by routine iMRI guided surgery. Residual tumor was diagnosed in 70% of cases on iMRI and prompted additional resection. We now sought to identify tumor related factors that are associated with residual tumor diagnosed on iMRI.

Method: Our prospective volumetric data base listed 204 consecutive patients that received iMRI-guided glioma surgery. Using ANOVA, association of preoperative tumor volume, tumor location and WHO grade with tumor residuals on iMRI was analyzed. Also, their impact on residual tumor volume on postoperative MRI was assessed.

Results: In our cohort mean preoperative tumor volume was 31,35±29,61cm³. Stratified for tumor volume, quartiles of cases were created. Preoperative tumor volume was strongly associated with residual iMRI tumor volume (p<0.0001) but after additional resection a significant association could no longer be found when residual tumor on postoperative imaging was analyzed (p=0.11). Hemispheric affiliation of tumors showed no significant association with residual iMRI volumes but affection of the left temporal lobe did, when testing for precise tumor location (p<0,05). Again, no significant association of tumor location and residual volume was seen on postoperative MRI. WHO grading was a highly significant covariate for residual tumor volume on iMRI (p<0.0001). Residual volumes were largest in WHO°II and °III (4,25±4,18cm³ and 4,66±7,35cm³, respectively) compared to 0.24±0,47cm³ and 2,44±5,22cm³ in WHO°I and °IV tumors, respectively. After additional resection this significant correlation was no longer evident on postoperative MRI. Resection was continued after iMRI in 68,1% (139/204) of cases in this cohort.

Conclusions: In this series we show that size, location and tumor entity significantly affect glioma resections with regards to residual tumor volumes detected on iMRI. Additional resection after iMRI seems to enable the surgeon to overcome those challenges in order to achieve a gross total tumor resection regardless of WHO grade and independent of initial tumor volume. This unique knowledge about factors impeding a complete tumor resection derived from iMRI data can be valuable for many neurosurgeons, in particular those who are not using iMRI. Larger scale investigations are needed to confirm possible covariates of residual tumor.