In this paper, a novel control performance monitoring method using Kullback-Leibler divergence is proposed for a multi-input-multi-output (MIMO) control system. Kullback-Leibler divergence is employed to quantify the dissimilarity in the closed loop output data distributions between the monitored period and the reference period. Furthermore, a Kullback-Leibler divergence based performance index is developed to detect the control performance change. Compared with conventional covariance based control performance indices, the proposed performance index can not only cope with the closed loop output data distribution under non-Gaussian noise, but also shows greater sensitivity to the control performance change. Simulation results demonstrate the effectiveness of the proposed Kullback-Leibler divergence based control performance index.