The increase in energy consumption associated with ever-intensifying network traffic is becoming a major problem. A number of researchers have focused on technologies that dynamically adjust the processing performance and the link speed of routers according to the network traffic load in order to achieve energy-efficient networking. However, when such power-saving routers are present in an end-to-end network path, the accuracy of existing methods for measuring the end-to-end available bandwidth may degrade because of the variable bandwidth and delays at bottleneck links. Furthermore, the energy efficiency of power-saving routers also decreases under the additional traffic load caused by bandwidth probing. In this paper, we employ a network environment with a power-saving router to evaluate the performance of Pathload, which is a popular tool for measuring the end-to-end available bandwidth. By showing detailed simulation results, we demonstrate that both the measurement accuracy of Pathload and the energy efficiency of routers degrade, particularly when the power saving function of routers is triggered in shorter cycles. We also propose Pathload parameter settings that maintain measurement accuracy without affecting the behavior of power-saving routers.