A femoral load-cell was installed in twelve fresh-frozen knee specimens from cadavera, to measure the resultant force at the femoral origin of the posterior cruciate ligament during a series of tibial-loading tests. The posterior cruciate ligament was removed, and a ten-millimeter-wide bone-patellar ligament-bone graft was inserted. The knee was flexed to 90 degrees, the graft was pre-tensioned to restore the anterior-posterior laxity to that recorded after installation of the load-cell, and the loading tests were repeated. With the tibia locked in neutral rotation and a 200-newton posterior force applied to the tibia, the mean force generated in the intact posterior cruciate ligament ranged from 220 newtons at 90 degrees of flexion to thirty-six newtons at full extension. When the tibia was locked in external rotation during the posterior drawer test, the force was reduced when the knee was flexed 10 to 70 degrees; when the tibia was locked in internal rotation, the mean force was reduced at only 30 and 45 degrees of flexion. The mean forces in the graft were not significantly different, with the numbers available, from the corresponding values for the intact ligament during application of a straight posterior tibial force (neutral tibial rotation), during application of a fifteen-newton-meter flexion or extension moment (hyperflexion or hyperextension), during application of a ten-newton-meter varus or valgus moment, or during application of a ten-newton-meter internal or external tibial torque. With the numbers available, there were no significant differences between the mean tibial rotations associated with the intact posterior cruciate ligament and those associated with the graft at any angle of flexion, without or with applied tibial torque.CLINICAL RELEVANCE: The amount of force generated in the posterior cruciate ligament during the posterior drawer test depends on the angle of flexion at which the test is performed. When the angle of flexion is near 90 degrees, all of the posterior force applied to the tibia is transmitted to the ligament and the force in the ligament is not affected by the position of tibial rotation. When the test is performed at an angle of flexion near 30 degrees and in neutral tibial rotation, other structures (such as the collateral ligaments and the posterior part of the capsule) help to resist the posterior force applied to the tibia. The position of tibial rotation is important when the test is performed with the knee at an angle of flexion near 30 degrees, as secondary structures pre-tensioned by tibial torque act to reduce the amount of force carried by the posterior cruciate ligament even more. With a few minor exceptions, we found that the forces in a graft used to replace the posterior cruciate ligament were approximately the same as those in the intact ligament. Therefore, there appears to be little justification for restricting low-level rehabilitation activities once the fixation of the graft has healed. However, forces in the graft could be quite high during hyperextension and hypertension, as they are in the intact ligament. Thus, bracing in the early postoperative period may be advisable to prevent these motions.