Postoperative alignment of TKA in patients with severe preoperative varus or valgus deformity: is there a difference between surgical techniques?

Background: There have been conflicting studies published regarding the ability of various total knee arthroplasty (TKA) techniques to correct preoperative deformity. The purpose of this study was to compare the postoperative radiographic alignment in patients with severe preoperative coronal deformity (>= 10 degrees varus/valgus) who underwent three different TKA techniques; manual instrumentation (MAN), computer navigated instrumentation (NAV) and patient specific instrumentation (PSI). Methods: Patients, who received a TKA with a preoperative coronal deformity of >= 10 degrees with available radiographs were included in this retrospective study. The groups were: MAN; n = 54, NAV; n = 52 and PSI; n = 53. The mechanical axis (varus / valgus) and the posterior tibial slope were measured and analysed using standing long legand lateral radiographs. Results: The overall mean postoperative varus / valgus deformity was 2.8 degrees (range, 0 to 9.9; SD 2.3) and 2.5 degrees (range, 0 to 14.7; SD 2.3), respectively. The overall outliers (> 3 degrees) represented 30.2% (48 / 159) of cases and were distributed as followed: MAN group: 31.5%, NAV group: 34.6%, PSI group: 24.4%. No significant statistical differences were found between these groups. The distribution of the severe outliers (> 5 degrees) was 14.8% in the MAN group, 23% in the NAV group and 5.6% in the PSI group. The PSI group had significantly (p = 0.0108) fewer severe outliers compared to the NAV group while all other pairs were not statistically significant. Conclusions: In severe varus / valgus deformity the three surgical techniques demonstrated similar postoperative radiographic alignment. However, in reducing severe outliers (> 5 degrees) and in achieving the planned posterior tibial slope the PSI technique for TKA may be superior to computer navigation and the conventional technique. Further prospective studies are needed to determine which technique is the best regarding reducing outliers in patients with severe preoperative coronal deformity.