End-to-end congestion control, aiming to find out how much traffic load offered by the source can be handled by a network, is an essential function of a transport layer protocol. Although various efforts have been made to address the end-to-end congestion control issue in traditional wireless ad hoc networks, these approaches usually lead to excessively long delay for the source node to react to the congestion in cognitive radio (CR) ad hoc network because of the non-uniform channel availability. Thus, the end-to-end throughput suffers significant degradation. In this paper, a novel end-to-end congestion control scheme named ECCO is proposed that considers the unique features in multi-hop CR ad hoc networks such as spectrum sensing, channel rendezvous, and licensed user activities. The average packet round trip time (RTT) for multi-hop CR ad hoc networks is derived. In addition, through extensive simulation, it is shown that our proposed novel end-to-end congestion control scheme outperforms the existing congestion control mechanisms in terms of higher end-to-end throughput. To the best of our knowledge, this is the first paper that investigates the end-to-end congestion control issue in multi-hop CR ad hoc networks.