The objective of the present study was to simultaneously investigate tibiofemoral and patellofemoral kinematics in patients with PCL-deficiency in comparison to healthy knee joints of the control group. At full extension and 30° flexion no significant difference was observed in PCL-deficient knee joints neither for the tibiofemoral nor for the patellofemoral kinematics. However, at 90° flexion the femur of PCL-deficient patients was positioned significantly more anteriorly in relation to the tibia. The patellar tilt angle as well as the patellar shift to the lateral side significantly increased compared to the healthy knee joints at 90° flexion, too. While no significant effect of isometric flexing muscle activity was observed in healthy individuals, in PCL-deficient knee joints an increased paradoxical anterior translation of the femur was observed at 90° flexion compared to the status of muscle relaxation. The results might underline the concept of the PCL being an important restraint to posterior tibial translation at flexion angles of more than 30°
[3, 9, 23]. Since it`s known that increased tibiofemoral translation to the medial side may cause the development of medial OA these results might provide an explanation for the relatively high incidence of medial tibiofemoral and patellofemoral OA in patients with PCL-deficiency
One limitation of the used technique is the fact, that weight bearing simulations are performed under static conditions. Since acquisition of a complete 3D MR data set requires a time period of approximately 4–5 minutes, dynamic studies can currently not be performed using the described 3D technique. “Real time” MRI, which means continuous MR imaging of moving objects in real time delivers actually only 2D images which implicates the problem of limited reproducibility and does not permit to measure displacement patterns three-dimensionally
[24, 25]. However, our results in healthy volunteers are consistent with findings in the literature of current in-vivo and in-vitro studies
[26–31]. Therefore, we have the opinion that our results obtained under static loading conditions are not likely to interfere with the conclusions. One possibility to overcome this problem is the improvement of MRI sequences or the combination of MRI and orthogonal fluoroscopy allowing image acquisition under quasi-static weight-bearing flexion
. However, the latter technique is – admittedly to a small extent – suffering from radiation exposure to the patients.
Regarding tibiofemoral kinematics we found a posterior translation of the femur during knee flexion in healthy knee joints. The femoral roll back occurred predominantly on the lateral side, resulting in a medial pivoting motion. These findings are consistent with previously published kinematic analyses
[26–31]. Furthermore, results in PCL-deficient knee joints show a more anterior position of the femur with respect to the tibia especially at higher degrees of flexion which is in keeping with previous in-vitro and in-vivo studies
[2, 32, 33]. This paradoxical anterior translation of the femur at 90° flexion might be an expression of the clinically known fixed posterior position of the tibia in relation to the femur in patients with chronic PCL-deficiency
. In addition, our in vivo data confirm the findings of previous in vitro studies
, that contraction of flexor muscle groups in PCL-deficient knees tends to result in an increased paradoxical anterior translation of the femur relative to the tibia at higher degrees of flexion.
Concerning patellar kinematics in healthy knees, Lee et al.
 reported on a patellofemoral angle in the sagittal plane of about 22° at 30° flexion increasing to about 60° at 90° knee flexion, which is in accordance to our findings. Our results for patellar shifting and tilting are also consistent with previously published data
[21, 35, 36]. However, many authors have reported on patellar kinematics in healthy knee joints, but only few in PCL-deficiency
[10–12, 14], although the influence of tibiofemoral kinematics on patellar motion in knee instability has been demonstrated
[10, 14–17]. Patellofemoral symptoms were described in about 50% of these patients
[7, 37]. Consistent to our findings van de Velde et al.
 observed significant changes in patellofemoral kinematics at higher degrees of flexion especially in the transversal plane, too. However, in contrast to van de Velde et al.
 who described a decrease of patellar shift (to the lateral side) we observed a significant increase of patellar shift and tilt angle at 90° flexion which tended to increase during isometric flexor muscle activity. Differences might occur due to different muscle loadings, thus applying 3 kg to the lower third of the shank in our experiment and supporting patient’s body weight on the leg being scanned in van de Velde’s experiment
. However, the increased patellar shift may be caused by the changes of the tibiofemoral joint with a decreased externally rotated and increased anteriorly positioned femur at 90° flexion which leads to a lateralization of the patella.