gms | German Medical Science

Objective: The Simpson grading has been widely established to quantify the extent of resection in meningiomas and to estimate the risk of postoperative tumor recurrence. However, grading is performed by the attending neurosurgeon and might suffer from bias and/or limited intraoperative view at the operative field. On the other hand, correlations between the postoperative tumor volume and the risk of recurrence are sparsely investigated and the value of postoperative imaging in meningiomas remains uncertain. We therefore compared the value of intraoperative assessment of the extent of resection with the tumor volume for the prediction of recurrence in a large series of meningiomas.

Methods: Tumor volume on initial postoperative MRI and the extent of resection according to the Simpson classification were determined for 939 patients with primary diagnosed intracranial WHO grade I-III meningioma (671 females, 268 males, median age of 58 years). Predictive values for recurrence were compared in uni- and multivariate analyses and the critical cut-off value of the tumor volume were determined in a tree-structured regression model.

Results: With a median follow up of 40 months, recurrence was observed in 12% of all 939 patients. The postoperative tumor volume was available in 423 patients and ranged between volumes of 0 ccm and 78.5 ccm. After designated gross total resections (Simpson grades I-III), residual tumor tissue could be detected in eight percent of the patients (range: 0.12 ccm – 33.5 ccm). In univariate analysis, both the Simpson grading (p=0.004) as well as the postoperative tumor volume (HR: 1.04, 95%CI 1.02-1.06, p<.001) strongly correlated with recurrence. Multivariate analysis adjusted for age, sex, tumor location, WHO grade, volume of tumor remnants and the extent of resection revealed high grade histology (HR: 3.95, 95%CI 2.00-7.77; p<.001) and postoperative tumor volume (HR: 1.05, 95%CI 1.02-1.08; p=0.0001), but not Simpson grade (p=0.5859) as predictive factors for recurrence. Regression Tree model revealed a postoperative tumor volume of >0.00 ccm as the significant threshold for predicting recurrence.

Conclusion: Postoperative MRI reveals residual tumor tissue in a considerable portion of meningioma surgeries. Postoperative tumor volume is superior in predicting tumor recurrence compared to the Simpson classification. Risk of recurrence increases with any residual tumor tissue.