The transmission and reception of sensor measures between nodes in distributed target tracking applications of wireless sensor networks is energy expensive. This paper shows that a selective transmission policy can be used to increase the network lifetime without reducing the accuracy of the target parameter (position, velocity) estimates in a significant manner. To do so, nodes compute an importance value of every measure, and make a decision about transmitting this information or not depending on this value, the energy available at the node battery and the energetic cost of transmitting and receiving information. In this way, weak or noisy signals are discarded, saving energy for future measurements. Simulation results show that the proposed scheme applied to a target tracking scenario is energy-saving as well as accurate.