The influence of the surgical approach on the position of the prosthetic components in revision of total knee arthroplasties (rTKA) is still not clear. This study compared the medial parapatellar approach (MPA) and lateral parapatellar subvastus approach with tibial tubercle osteotomy (LPA).
Forty-two consecutive patients underwent rTKA from 2006 to 2016 with either MPA (n = 21) or LPA (n = 21) because of aseptic loosening, infection, malposition or instability (mean follow-up: 41.5 months). Revision TKA component position and leg alignment were assessed on three-dimensional CT (3D-CT) images and compared between groups using a t-test (p < 0.05). Rotation of rTKA components graded into internal, neutral or external rotation and rates of postoperative complications and revisions were compared between groups with a chi2-test (p < 0.05).
Group MPA and LPA showed no significant differences with respect to the femoral component (coronal: 0.6° vs 0.9°; sagittal: 7.5° vs 7.7°; transversal: 0.4° vs −0.2°), tibial component (coronal: 0.1° vs 0.3°; sagittal: 3.1° vs 1.6°; transversal: 6.9° vs 9.3°) and leg alignment (varus: 0.2° vs 1.1°) (p < 0.05). Group LPA showed a non-significant (p = 0.25) higher incidence of neutrally (28.6% vs 9.5%) and less internally (23.8% vs 38.1%) rotated tibial components. Rates of complication and revisions did not differ significantly.
In contrast with primary TKA, the two approaches did not influence postoperative rTKA position. This can be explained with the more extensive approach at rTKA and means that no corrections of the orientation of the components are needed performing either MPA or LPA.