Metastases are the most prevalent tumors in the brain and are commonly associated with high morbidity and mortality. Previous studies have suggested that brain tumors can induce a loss of functional connectivity and alter the brain network architecture. Little is known about the effect of brain metastases on whole-brain functional and structural connectivity networks. In this study, 14 patients with brain metastases and 16 healthy controls underwent resting state functional magnetic resonance imaging (rs-fMRI) and diffusion tensor imaging (DTI). We constructed functional connectivity network using rs-fMRI signal correlations and structural connectivity network using DTI tractography. Graph theoretical analysis was employed to calculate network properties. We further evaluated the performance of brain networks after metastases resection by a simulated method. Compared to healthy controls, patients with brain metastases showed an altered "small-world" architecture in both functional and structural connectivity networks, shifting to a more randomness organization. Besides, the coupling strength of functional-structural connectivity was decreased in patients. After removing nodes infiltrated by metastases, aggravated disruptions were found in both functional and structural connectivity networks, and the alterations of network properties correlated with the removed hubs number. Our findings suggest that brain metastases interfere with the optimal network organization and relationship of functional and structural connectivity networks, and tumor resection involving hubs could cause a worse performance of brain networks. This study provides neuroimaging guidance for neurosurgical planning and postoperative assessment of brain metastases from the aspect of brain networks.