[Cinematic in vivo analysis of the knee: a comparative study of 4 types of total knee prostheses].

PURPOSE OF THE STUDY

The goal of the study was to assess in vivo kinematics for four designs of knee prosthesis during level walking, stair climbing and non weight-bearing flexion-extension.

PATIENT AND METHODS

19 patients with unilateral total knee arthroplasty (TKA) were included [5 bicruciate sparing prosthesis (BI), 5 posterior cruciate sparing prosthesis with flat tibial polyethylene (PP), 5 posterior cruciate sparing prosthesis with congruent tibial polyethylene (PC), 4 postero stabilised (PS)]. These 19 patients had no prosthesis nor pathological situation in any other joint of the lower limbs. Each of these 19 prosthesis had an HSS score greater than 80 and no radiographic signs of loosening. Magnitudes of the knee rotations (flexion-extension axial rotation valgus-varus) were evaluated with a 6 degrees freedom of motion electromagnetic-goniometer during level walking, stair climbing and non weight-bearing flexion-extension. The magnitudes of the three rotations were recorded for the 19 prosthetic knees and for the 19 controlateral non prosthetic knees of the patients. Reproducibility of the method was also evaluated on 12 healthy subjects by comparison of magnitudes observed during two different recordings.

RESULTS

Reproducibility was excellent for magnitudes of flexion (r = 0.95/p = 0.0001) and axial rotation (r = 0.55/p = 0.002) but less satisfactory for valgus-varus movements (r = 0.46/p = 0.005). The magnitudes of the three rotations were inferior for TKA in comparison with healthy knees for any activities. By comparison of the 19 prosthetic and non prosthetic knees we recorded smaller magnitudes of axial rotation during swing phase for level and stair climbing and during non weight-bearing flexion extension movements. Between the four kind of prosthesis we observed: greater magnitudes of flexion for BI and PC prosthesis during stair climbing (p < 0.05) and greater magnitudes of flexion for BI PC and PS prosthesis during stair descending (p < 0.05). PC prosthesis instead of a congruent polyethylene tibial plateau had greater magnitudes of axial rotation than non constrained BI prosthesis during stair climbing (p = 0.009). In spite of a high femoro-tibial congruency we recorded axial rotation in PS prosthesis during each activities.

DISCUSSION AND CONCLUSION

Our method evaluating in vivo knee kinematics was reproducible. These four knee prosthetic designs in spite of a good functional results were unable to reproduce magnitudes of movements recorded in healthy subjects. The small number of prosthesis included in the study prompt us to consider as no definitive the differences observed between the 4 designs. Anyway the influence of design on kinematics should be considered as relative since we recorded axial rotation for all four cruciate substituting prostheses although they had high femoro-tibial congruency. Influence of femoro-tibial congruency and cruciate ligament sparing could be assessed in vivo by means of this reproducible method on a larger population.