Anterolateral approach with tibial tubercle osteotomy versus standard medial approach for primary total knee arthroplasty: does it matter?

The purpose of this prospective consecutive multicenter study was to investigate whether the type of surgical approach (medial parapatellar (MPA) or lateral parapatellar with tibial tubercle osteotomy (TubOT)) influences the early clinical and radiological outcomes of primary total knee arthroplasty (TKA).

Ligament balancing primary TKA with a rotating platform was performed in 143 knees (m:w = 1:1.6; mean age 69 ± 8 years). The TKA was done by a lateral parapatellar subvastus approach with stepcut osteotomy of the tibial tubercle (53%; n = 76, group A) or medial parapatellar approach (47%; n = 67, group B). The outcome was assessed at 1 and 2 years postoperatively by the American Knee Society score (KSS) and the knee society total knee arthroplasty roentgenographic evaluation and scoring system (TKA-RESS). The patient's pain level and satisfaction was noted by a visual analogue scale (VAS). Data were analyzed by an independent statistician with a level of significance of p < 0.05. The Wilcoxon two sample test (two-sided) was used to investigate differences of patients between group A and B pre- and postoperatively. The paired t-test was used to evaluate differences over course of time within each group. For comparison of radiological alignment a Chi²-test was performed.

Although having a lower degree of preoperative flexion (112° ± 15° versus 115° ± 15°) patients in group A showed a significantly (p = 0.027) higher degree of flexion (118° ± 10°) at their last follow-up than patients in group B (114° ± 10°). Patients in group A showed a significantly better mean VAS pain (p = 0.0001) and satisfaction (p = 0.0058) at 2 years follow-up. The pain free walking distance was significantly (p = 0.036) longer for group A than group B. Patients treated with a lateral approach were significantly more stable in terms of valgus stress (p = 0.049). The Knee society score was significantly (p = 0.0009) higher at two years follow up in group A compared to group B. The postoperative mechanical alignment and positioning of the prosthesis were not significantly different. Patients in group B presented with significantly (p = 0.0017) more tibial radiolucencies (> 2 mm) at their last follow-up than patients in group A. There was no prosthesis related revision in either group. The revision rate in group A (4%) was higher than in group B (1.5%), which was mainly due to two cases of traumatic secondary displacement of the tibial tubercle and need for refixation.

The TubOT led to slightly better functional results and less pain two years after primary TKA. It is however not clear if the improved outcome can outweigh the longer operation time and higher risk of early complications and revisions. Long-term studies are necessary to show whether there is any difference in prosthesis longevity between both types of approach.