Surgical treatment of brain metastases: Review of 124 patients

Objective This study was performed to assess the outcomes and prognostic factors in the patientswith brain metastases who underwent surgery as the primary treatment. Materials and Methods Between January 1998 and June 2008, 124 patients were surgically treated for newly diagnosed brain metastases. Surgery was recommended as an initial treatment for the patients with Karnofsky performance scale(KPS) score not less than 60 and single accessible lesion or multiple lesions including at least one largesymptomatic lesion of superficial location. We assessed preoperative status, postoperative neurological outcomes, complications and survival through retrospective review of medical record and radiological data. Results Eighty five patients presented with single lesion and 39 patients with multiple lesions. Lung was the most common site of primary tumor(55, 44%) followed by breast (8, 6%), genitouninary(GU) system(renal, ovary and prostate) (21, 17%), gastrointestinal(GI) system(stomach, pancreas, rectal) (16, 13%), others(17, 14%), unknown primaries(7, 6%), and so on. Postoperatively, whole brain radiotherapy(WBRT) was performed in 112 patients(90%) and radiosurgery in 28 patients(23%). Compared with preoperative state, 103 patients(83%) showed improved neurological state during immediate postoperative period. Postoperative morbidity and mortality related to surgery developed in 15 patients (13.2%) and 2(0.9%) respectively. Overall median survival time was 34 weeks(range, 1- 380). Patients with single cerebral metastasis had a median survival time(MST) of 35weeks(range 1 380), whereas patients with multiple lesions had a MST of 33 weeks (range 2-131) showing no statistical difference between two groups(p = 0.13). Recursive partitioning analysis(RPA) class and status of extra-cerebral tumor were significant prognostic factors identified through univariate and multivariate analysis. Median survival of RPA class I patients was 53 weeks(range 2 380) significantly longer than 30 weeks(range 1 - 321) of RPA II or III patients(p = 0.001). Local recurrence of surgically treated lesions developed in 15 patients whereas recurrence at remote site from the operated lesion developed in 17 patients(14%) during the follow up period. So, local recurrence rate was 12% and 24% at 6 and 12 months after operation. Conclusion This analysis essentially confirms that surgical resection of symptomatic metastatic lesions relieve neurological symptoms rapidly with acceptable risk. Local control rate is seemingly better than that without surgery when compared with the data from the literature. The RPA class or extracranial disease activity are important prognostic factors of survival. Surgical excision of only symptomatic lesions followed by WBRT and radiosurgery resulted in median survival not worse than that of the patients with single lesion removed completely. Operative treatment of metastatic brain lesions should be tailored according to the clinical situation and traditional guide for surgical treatment needs to be refined.