This paper analyzes the coevolution of opinion dynamics in evolving networks, through simulation methods. The Bounded-Confidence opinion dynamics model is used to represent the change of opinions in a society. The evolution of the network topology is developed using preferential attachment rules. The analysis is focused on the convergence of opinions affected by the evolution of the network topology. The simulation results present two behaviors of the coevolution: i) the presence of peaks around the convergence value; and ii) agents with opinion values out of the confidence level interval reaching the convergence.