Video transmission plays a critical role in robotic telesurgery because of the required high bandwidth and quality requirement. We propose an adaptive video preprocessing technique to accelerate the transmission of telesurgical video. Using this technique, the bandwidth can be reallocated adaptively from non-essential surrounding regions to the region of interest when preprocessed image sequences are passed to the video encoder, ensuring excellent image quality of critical regions (e.g. surgical landscape). The algorithm first generates a physiologically appropriate attention map which can be updated dynamically. Then, an adaptive edge-preserving image smoothing filter is utilized to remove insignificant features. The parameters of the filter are adjusted according to the attention map. Our experimental result shows that, with the preprocessing technique, over half of the bandwidth can be reduced while there is no significant visual effect at the site of remote observer.