Object: Gamma Knife surgery (GKS) has been shown to be effective for treating many patients with brain metastasis. Some brain metastases demonstrate significant peritumoral edema; radiation may induce cerebral edema or worsening preexisting edema. This study was conducted to evaluate the imaging and neurobehavioral outcomes in patients with preexisting peritumoral edema who then undergo GKS.
Methods: Between August 2003 and January 2008, 63 cases of brain metastasis with significant peritumoral edema (> 20 cm(3)) were prospectively studied. The study inclusion criteria were as follows: 1) a single metastatic lesion with significant edema (perilesional edema signal volume on FLAIR > 20 cm(3)); and 2) inclusion of only 1 lesion > 20 cm(3) in the study (in cases of multiple lesions noted on FLAIR images). All patients received MR imaging with pulse sequences including T1-weighted imaging and FLAIR with or without contrast and T2-weighted imaging at an interval of 3 months. A neurological assessment and Brain Cancer Module (BCM-20) questionnaire were obtained every 2-3 months. Kaplan-Meier, Cox regression, and logistic regression were used for analysis of survival and associated factors.
Results: At the time of GKS, the median Karnofsky Performance Scale (KPS) score was 70 (range 50-90), and the mean BCM-20 score was 45.5 +/- 6.1. The mean tumor volume (+/- standard deviation) was 5.2 +/- 4.6 cm(3) with corresponding T2-weighted imaging and FLAIR volumes of 59.25 +/- 37.3 and 62.1 +/- 38.8 cm(3), respectively (R(2) = 0.977, p < 0.001). The mean edema index (volume of peritumoral edema/tumor volume) was 17.5 +/- 14.5. The mean peripheral and maximum GKS doses were 17.4 +/- 2.3 and 35 +/- 4.7 Gy, respectively. The median survival was 11 months. The longer survival was related to KPS scores >or= 70 (p = 0.008), age < 65 years (p = 0.022), and a reduction of > 6 in BCM-20 score (p = 0.007), but survival was not related to preexisting edema or tumor volume. A reduction in BCM-20 score of > 6 was related to decreased volume in T1-weighted and FLAIR imaging (p < 0.001). Thirty-eight (79.2%) of 48 patients demonstrated decreased tumor volume and accompanied by decreased T2-weighted imaging and FLAIR volume. Eight (16.7%) of the 48 patients exhibited increased or stable tumor volume. A margin dose > 18 Gy was more likely to afford tumor reduction and resolution of peritumoral edema (p = 0.005 and p = 0.006, respectively). However, prior external-beam radiation therapy correlated with worsened preexisting peritumoral edema (p = 0.013) and longer maintenance of corticosteroids (p < 0.001).
Conclusions: Patients demonstrating a reduction in the BCM-20 score > 6, age < 65 years, and KPS score >or= 70 exhibited longer survival. Significant preexisting edema did not influence the tumor response or clinical outcome. The resolution of edema was related to better quality of life but not to longer survival.