Preconditioning of soft tissues has become a common procedure in tensile testing to assess the history dependence of these viscoelastic materials. To our knowledge, this is the first study comparing tensile properties of soft tissues-before and after cyclic preconditioning with high loads. Sixteen quadriceps tendon-bone (QT-B) complexes and 16 patellar ligament-bone (PL-B) complexes from a young population (mean age 24.9 +/- 4.4 years) were loaded to failure with a deformation rate of 1 mm/s. Half of the QT-B and the PL-B complexes underwent 200 uniaxial preconditioning cycles from 75 to 800 N at 0.5 Hz before ultimate failure loading. High-load preconditioning was made possible by the development of a highly reliable and easy-to-use cryofixation device to attach the free tendon end. PL-B complexes were more influenced by preconditioning than the QT-B complexes. Ultimate failure load, stiffness at 200 N and stiffness at 800 N were significantly higher for PL-B complexes after preconditioning, while the structural properties of QT-B complexes exhibited no significant alterations. The values of the mechanical properties like Young's modulus at 200 N and 800 N were much higher for both preconditioned specimen groups. In addition, ultimate stress was augmented by preconditioning for PL-B complexes. Hysteresis and creep effects were highest during the first few loading cycles. More than 160 cycles were needed to reach a steady state. Beyond 160 cycles there was no further creep, and hysteresis was almost constant. Creep values were 2.2% of the initial testing length for the QT-B and 3.2% of the initial testing length for the PL-B complexes. The effect of cyclic preconditioning seems to be caused by progressive fiber recruitment and by alterations of the interstitial fluid milieu.