This paper investigates strategies for coordinating the decisions calculated by a network-constrained multi-period auction model of hydrothermal systems with the ones calculated by medium-term problems. Initially, we discuss the necessity for implementing such coordination strategies in hydrothermal systems from the standpoint of a market operator. We investigate three alternative strategies for performing the coordination. The first strategy incorporates primal information from the medium-term problem into the auction model in order to perform coordination, while the second one makes use of dual information associated with the medium-term. A third strategy combining primal and dual information from the medium-term problem is also investigated. These strategies are compared by means of simulation results involving the IEEE 24-bus reliability test-system. The results focuses on evaluating the impact of the coordination strategies on market clearing prices and generation scheduling.