The effects of low levels of quadriceps tendon tension on patellar kinematics and knee injury were characterized by impacting five pairs of left and right isolated cadaver knees in a 90°-flexed posture with and without 250 N of quadriceps tendon tension. Each test was performed by impacting the anterior surface of the flexed knee along the axis of the femur with a lightly padded knee-bolster-like impactor while holding the distal shaft of the femur fixed. Lateral-view high-speed x-ray was used to track patellar motion relative to the femoral condyles during impact loading and determine the time of fractures. Results from tests conducted on left and right knees from five subjects indicate that quadriceps tendon tension reduces the tendency for the patella to move down on the femoral condyles during impact loading, but the direction of patella movement is also highly dependent on the initial position of the patella relative to the femoral condyles. Importantly, the type of knee injury is dependent on the location of the patella relative to the femoral condyles. Four out of five tests for which the patella was high on the femoral condyles with or without upward movement of the patella resulted in patellar fractures and the fifth test resulted in no injury. Three out of five tests for which the patella was at a mid or low position on the femoral condyles, either because it moved down in the absence of quadriceps tendon tension or started out low and stayed in that position, resulted in split condylar fractures and the other two tests did not result in any knee fractures. Thus, if the patella is at a mid or low position during impact loading, due either to its initial position or to downward movement in the absence of quadriceps tendon tension, a split condylar fracture is likely to occur. If the patella is high relative the condyles and/or moves up, a patella fracture is most likely.