In this paper we study centrality resilience, that is, how well closeness and betweenness centralities are maintained under attacks. We propose efficient attack models to disrupt the rank of the top k centrality vertices. To develop our attack models, we extend the concept of rich clubs of influential vertices to the more general framework of scattered rich clubs–dense subgraphs of high-centrality vertices that are spread across the network. To improve computational efficiency, we use snowball sampling to identify these important substructures. Our results over real-world networks demonstrate that our algorithm can identify the single or scattered rich clubs efficiently and is more effective in disrupting the centrality rankings of the network, compared to other baseline methods.