Department of Clinical Neurosciences, Division of Neurosurgery, Geneva University Hospitals, 1205, Genève, Switzerland.
Faculty of Medicine, University of Oslo, Oslo, Norway.
Acta Neurochir (Wien). 2019 May;161(5):1003-1011. doi: 10.1007/s00701-019-03862-z. Epub 2019 Mar 11.
Posterior fossa meningiomas (PFMs) often represent surgical challenges due to their proximity to neurovascular structures. Factors predicting the extent of resection (EOR), overall survival (OS), and progression-free survival (PFS) were identified and integrated in a prediction tool to offer evidence-based personalized therapeutic strategies.
All meningiomas managed surgically from 1990 to 2010 from a single-center were reviewed. A classification tree was created using the classification and regression tree recursive partitioning analysis that incorporated patient and tumor data available before surgery in order to predict the rates of gross total resection (GTR).
A total of 198 patients were identified (female-to-male ratio, 2.7; mean age, 59.1 years) and compared with 1271 supratentorial meningiomas (STMs) operated in the same institution during the same time period. GTR was achieved less often (59.6% versus 81.9%; p < 0.01) in PFMs than STMs. Preoperative neurological symptoms were predictive of higher Simpson grades (OR, 2.19 [1.05; 4.58]; p = 0.04). Age was associated with reduced OS (OR, 1.08 [1.04;1.12]; p < 0.001). A KPS ≥ 70 was associated with higher survival rates (OR, 2.70 [2.19;2.92]; p = 0.02). Higher WHO grades were associated with reduced OS (OR, 3.56 [1.02;12.47]; p = 0.05). The GTR rate varies from 80% in patients without a preoperative deficit to 40% patients with a preoperative deficit, younger than 60 years old, and with adjacent bone invasion.
This study provides a classification tree of the predictors of EOR in PFMs, based upon preoperative demographic, clinical, and radiological variables. An evidence-based management protocol with estimated EORs may guide the decision-making process in PFMs.
由于后颅窝脑膜瘤(PFMs)与神经血管结构毗邻,因此通常具有手术挑战性。确定了预测切除程度(EOR)、总生存率(OS)和无进展生存率(PFS)的因素,并将其整合到预测工具中,以提供基于证据的个性化治疗策略。
回顾了 1990 年至 2010 年期间在一家中心接受手术治疗的所有脑膜瘤。使用分类树创建了一个分类和回归树递归分区分析,该分析纳入了手术前可获得的患者和肿瘤数据,以便预测大体全切除(GTR)的比率。
共确定了 198 例患者(女男比为 2.7;平均年龄为 59.1 岁),并与同期在同一机构接受手术的 1271 例幕上脑膜瘤(STMs)进行了比较。GTR 的完成率(59.6% 对 81.9%;p < 0.01)明显低于 STMs。术前神经症状可预测更高的辛普森分级(OR,2.19 [1.05;4.58];p = 0.04)。年龄与 OS 降低相关(OR,1.08 [1.04;1.12];p < 0.001)。KPS ≥ 70 与生存率提高相关(OR,2.70 [2.19;2.92];p = 0.02)。较高的 WHO 分级与 OS 降低相关(OR,3.56 [1.02;12.47];p = 0.05)。GTR 率从无术前缺陷的患者的 80%到术前有缺陷的患者的 40%不等,这些患者年龄小于 60 岁,且伴有邻近骨侵犯。
本研究基于术前人口统计学、临床和影像学变量,提供了一个预测 PFMs EOR 的分类树。基于术前预测 EOR 的证据的管理方案可能有助于指导 PFMs 的决策过程。