Efficient trajectory planning plays a crucial role in the development of autonomous vehicles, ensuring safe and optimized navigation in dynamic environments. This paper proposes a novel energy-efficient hybrid trajectory planning by integrating a sampling-based method with an optimizationbased path refining method. It uses the strength of the samplingbased methods to reduce the solution space and generate a reactive trajectory in a dynamic environment. Following path selection, a septic path is generated and utilized as a reference for an energy-efficient path-refining optimization problem, producing a jerk-controlled trajectory with enhanced computational efficiency. The simulations were conducted in a joint-simulation environment using Simulink/Matlab and the Scaner Studio vehicle dynamics and driving environment simulator. The findings demonstrate the effectiveness of our approach in achieving significant energy savings while adeptly addressing dynamically changing environments.