Impact of postoperative tumor volume and relative volume reduction to predict meningioma recurrence - a retrospective analysis.

PURPOSE: In meningiomas, the value of the Simpson classification system to estimate the risk of tumor relapse is increasingly discussed. While raising the question about the indication of routine postoperative magnetic resonance imaging (MRI), the value of the residual tumor volume and of the absolute and relative volume reduction to predict tumor relapse is largely unexplored.

METHODS

Extent of resection (EOR) according to the Simpson classification system, postoperative tumor volume, absolute and relative volume reduction on MRI after surgery were analyzed in 475 patients operated for intracranial WHO grade 1-3 meningiomas. Predictors for recurrence were compared in uni- and multivariate analyses.

RESULTS

Despite designated gross total resection in 374 cases, postoperative MRI revealed residual tumor in 7%. Conversely, no residual tumor was detected on postoperative MRI despite intraoperatively assessed subtotal resection in 13% (κ = 0.766, p < 0.001). In univariate analysis, the EOR (p < 0.001), the postoperative tumor volume (HR: 1.002 per ccm, 95%CI 1.001-1.003; p < 0.001) and the relative (HR: 1.002 per %, 95%CI 1.001-1.003; p < 0.001) but not the absolute meningioma volume reduction correlated with recurrence. ROC analyses confirmed the postoperative tumor volume (AUC = 0.73, p < 0.001) and the relative volume reduction (AUC = 0.73, p < 0.001) to correlate with tumor relapse. Multivariate analysis revealed only high-grade histology (HR: 4.458, 95%CI 1.684-11.800; p < 0.003) and, notably, the relative volume reduction (HR: 1.002, 95%CI 1.001-1.003; p < 0.001) as predictors for recurrence.

CONCLUSION

Routine postoperative MRI improves detection of residual tumor tissue in selected cases and contributes important information to estimate the risk of recurrence.