This paper studies the problem of optimal network topology reconfiguration in sensor networks for state estimation. Multiple sensors make observations of a process, which are then transmitted, possibly via intermediate sensors, to a central gateway. Transmission over each link can experience packet drops. The time-varying wireless network environment is modelled by the notion of a network state as in Quevedo et al. (2013a). For each network state, different network configurations can be used, which govern the network topology and routing of packets. Switching between different configurations incurs a cost, in that unwanted links will need to be removed before the establishment of new links, leading to a transient time in which some links may not be available. The problem is to determine the optimal network configurations to use in each network state, in order to minimize an expected error covariance measure that takes into account the cost of reconfiguration. A simpler suboptimal method which minimizes the upper bound of the expected error is also proposed, which in numerical simulations gives essentially identical results to the optimal method.