Surgical management of petroclival meningiomas: defining resection goals based on risk of neurological morbidity and tumor recurrence rates in 137 patients.

OBJECTIVE: Meningiomas arising from the petroclival region remain a challenging surgical problem. Because of the substantial risk of neurological morbidity, uniformly pursuing a gross total resection (GTR) to minimize tumor recurrence rates may not be justified. We sought to define optimal resection goals based on risk factors for postoperative neurological morbidity and tumor recurrence rates. METHODS: This series represents our experience with 137 meningiomas arising from the petroclival region resected between June 1993 and October 2002. There were 38 male and 99 female patients with a mean age of 53 years. RESULTS: GTR was achieved in 40% of patients, and near total resection (NTR) was achieved in 40% of patients. One operative death occurred. Twenty-six percent of patients experienced new postoperative cranial nerve deficits, paresis, or ataxia when assessed at a mean follow-up of 8.3 months. The risk of cranial nerve deficits increased with prior resection (P < 0.001), preoperative cranial nerve deficit (P = 0.005), tumor adherence to neurovascular structures (P = 0.046), and fibrous tumor consistency (P = 0.005). The risk of paresis or ataxia increased with prior resection (P = 0.001) and tumor adherence (P = 0.045). Selective NTR rather than GTR in patients with adherent or fibrous tumors significantly reduced the rate of neurological deficits. Radiographic recurrence or progression occurred in 17.6% of patients at a mean follow-up of 29.8 months. Tumor recurrence rates after GTR and NTR did not differ significantly (P = 0.111). CONCLUSION: Intraoperatively defined tumor characteristics played a critical role in identifying the subset of patients with an increased risk of postoperative deficits. By selectively pursuing an NTR rather than a GTR, neurological morbidity was reduced significantly without significantly increasing the rate of tumor recurrence.