While traumatic aortic rupture (TAR) continues to be a major cause of fatality in motor vehicle accidents, its underlying mechanisms are still not well identified. TAR is believed to be caused by a complex state of loading occurring in only a few milliseconds. A major step to gain insight into the mechanisms of TAR is to understand the high-rate failure properties of aortic tissue. Fresh samples from porcine isthmus region were tested using a new high rate uniaxial extension test setup. Three levels of low, medium, and high strain rates (approximately 3 s-1, 60 s-1, and 300 s-1) were applied in the longitudinal direction and the time histories of stretch ratios and Cauchy stress were determined. The maximum strain rate was increased by 10 folds compared to previous studies. Two states were identified to characterize the tissue failure namely the failure initiation (FI) and the maximum load (ML) states. The FI state was detected from high speed video images and occurred before a noticeable change in force was observed. The ML state was determined directly from force measurements. The trends of stretch ratio and stress in both states were nonlinear. Deterministic models were derived for the FI state parameters. The trends of the ML state parameters were more complex which was attributed to the statistical distribution of collagen fibers undulation. The results suggest that the discrepancies reported in the failure properties of aorta may be due to these more general nonlinear trends.