Low cycle fatigue properties of porcine aorta - Pilot study

Biomechanical rupture risk assessment of aortic tissues is commonly based on computed stress to measured uniaxial static strength comparison. Loading of the arterial wall, however, is cyclic; thus, the static strength may not be a proper limit value. This study investigates the low cycle fatigue of porcine aortic samples tested in a circumferential direction.

7 porcine descending aorta (both thoracic and abdominal) were harvested and 56 dogbone-shaped samples were prepared. Static strength was measured, the limit of engineering stress was chosen and then force controlled cyclic loading was performed up to 100,000 cycles. Efforts were made to obtain a sufficient number of points across the entire range of 0-100,000 cycles. Data were fitted by both linear and logarithmic law and extrapolated towards 1 cycle for validation against static strength/ultimate tension. Data dispersion was evaluated via normalised root mean square error.

Out of 56 samples from 7pigs, 28 samples from 4 pigs were successfully tested. There was a strong negative correlation between applied stress/tension and number of cycles to failure. The fitting of both linear and logarithmic values resulted in a similar accuracy (R²=0.72 and 0.71 for stress and R²=0.62 and 0.7 for tension, respectively), while predicting static failure properties was more accurate by logarithmic law. NRMSE was lower for absolute values (20-21%) than for relative values (27-30%).

Absolute values of cyclic strength and tension are less dispersed than relative ones. Logarithmic fits are more robust in predicting static strength from cyclic data, while linear fits serve as a lower limit estimation.