Long-Term Outcomes of Gamma Knife Radiosurgery for Cerebral Cavernous Malformations: 10 Years and Beyond

Ho Sung Myeong,Sang Soon Jeong,Jung Hoon Kim,Jae Meen Lee,Kwang Hyon Park,Kawngwoo Park,Hyun Joo Park,Hye Ran Park,Byung Woo Yoon,Eun Jung Lee,Jin Wook Kim,Hyun Tai Chung,Dong Gyu Kim,Sun Ha Paek
DOI: https://doi.org/10.3346/jkms.2024.39.e229
2024-08-19
Abstract:Background: We aimed to evaluate long-term outcomes of gamma knife radiosurgery (GKS) for cerebral cavernous malformations (CCMs). Methods: Among the 233 CCM patients who underwent GKS, 79 adult patients (96 lesions) followed for over 10 years were included and analyzed retrospectively. Annual hemorrhage rate (AHR) was analyzed the entire cohort of 233 patients and the subset of 79 enrolled patients by dividing lesions into overall CCM lesions and brainstem lesions. AHR, neurologic outcome, adverse radiation effect (ARE), and changes of lesions in magnetic resonance imaging (MRI) were compared before and after GKS. Cox-regression analysis was performed to identify risk factors for hemorrhage following GKS. Results: Mean follow-up duration of 79 enrolled patients was 14 years (range, 10-23 years). The AHR of all CCMs for entire cohort at each time point was 17.8% (pre-GKS), 5.9% (≤ 2 years post-GKS), 1.8% (≤ 10 years post-GKS). The AHR of all CCM for 79 enrolled patients was 21.4% (pre-GKS), 3.8% (2 years post-GKS), 1.4% (10 years post-GKS), and 2.3% (> 10 years post-GKS). The AHR of brainstem cavernous malformation (CM) for entire cohort at each time point was 22.4% (pre-GKS), 10.1% (≤ 2 years post-GKS), 3.2% (≤ 10 years post-GKS). The AHR of brainstem CM for 79 enrolled patients was 27.2% (pre-GKS), 5.8% (2 years post-GKS), 3.4% (10 years post-GKS), and 3.5% (> 10 years post-GKS). Out of the 79 enrolled patients, 35 presented with focal neurologic deficits at the initial clinical visit. Among these patients, 74.3% showed recovery at the last follow-up. Symptomatic ARE occurred in five (6.4%) patients. No mortality occurred. Most lesions were decreased in size at the last follow-up MRI. Previous hemorrhage history (hazard ratio [HR], 8.38; 95% confidence interval [CI], 1.07-65.88; P = 0.043), and brainstem location (HR, 3.10; 95% CI, 1.26-7.64; P = 0.014) were significant risk factors for hemorrhage event. Conclusion: GKS for CCM showed favorable long-term outcomes. GKS should be considered for CCM, especially when it has a previous hemorrhage history and brainstem location.
What problem does this paper attempt to address?