QRS-waves integral can reflect the activities of ventricular depolarization. In this paper are analyzed the nonlinear characters in QRS-waves integral: reconstructing state space with these integral signals after nonlinear noise reduction, calculating its attractor dimension and the largest Lyapunov exponents. With the results based on MIT-BIH database we proposed that QRS-waves integral is of chaos nature. When compased with HRV nonlinear analysis, QRS-waves integral analysis was found to be of more advantages in some cases such as premature ventricular contraction. This QRS-waves integral nonlinear analysis may be of use in evaluating clinical cardiac function.