Routing methods in mega-constellation networks are critical and challenging for network performance. Dynamic topology changes incur a large amount of state interaction and routing update overhead, and require more routing convergence time. In this paper, from the perspective of complex network theory, we propose a pinning control-based routing method applicable to mega-constellation networks, where a small number of key nodes are selected based on the Maximum Local Traffic(MLT) in a distributed control framework for mega-constellation networks, and scalable routing is realized by controlling the states of these nodes. Simulation results demonstrate that the proposed routing selection method achieves performance comparable to or surpassing existing routing methods while reducing routing information exchange overhead. This validates its effectiveness as a novel routing control paradigm.