The Knee 22 (2015) 117–121
Contents lists available at ScienceDirect
Minor varus alignment provides better results than neutral alignment in medial UKA Michele Vasso ⁎, Chiara Del Regno, Antonio D'Amelio, Davide Viggiano, Katia Corona, Alfredo Schiavone Panni University of Molise, Department of Medicine and Health Sciences, Via Francesco De Sanctis, 86100 Campobasso, Italy
a r t i c l e
i n f o
Article history: Received 14 July 2014 Received in revised form 4 November 2014 Accepted 4 December 2014 Keywords: Unicompartmental knee arthroplasty Alignment Varus Lower limb
a b s t r a c t Introduction: Few data exist regarding the outcome and survivorship of medial UKA in patients with minor varus alignment. The purpose of this study was therefore to analyse the clinical results of medial UKA implanted with no more than 7° of varus, and to verify whether there was a relationship between limb alignment and overall outcomes. Material and methods: One hundred and twenty ﬁve medial ﬁxed-bearing UKAs with no more than 7° of varus were retrospectively analysed. The varus/valgus inclination and thickness of the bone cuts were performed relating to the proximal tibial epiphyseal axis. Patients were assessed with the IKS scores and range of knee motion. The subjects were classiﬁed into three groups according to the postoperative femoro-tibial mechanical alignment angle (group A: −2° to 1°; group B: 2° to 4°; group C: 5° to 7°). Results: The mean follow-up was 7.6 years (range, 3.5–9.3). IKS knee scores increased proportionally with increasing varus according to a linear relationship (p ≪ 0.01). Additionally, IKS knee scores were signiﬁcantly higher in group B and still higher in group C if compared to those in group A (p = 0.003). Finally, a signiﬁcantly higher frequency of IKS function scores N 90 points in subjects with femoro-tibial mechanical alignment angle ≥ 4° was found (p = 0.009). Conclusions: Minor varus alignment does not compromise the mid- to long-term outcome of a medial UKA, and gives better results compared to neutral or close-to-neutral alignment. Level of evidence: IV — Retrospective case series study. © 2014 Elsevier B.V. All rights reserved.
1. Introduction The correct restoration of limb alignment and the appropriate positioning of the prosthetic components are the major contributors to successful unicompartmental knee arthroplasty (UKA) [1,2]. Whilst in total knee arthroplasty (TKA) ﬁnal limb alignment depends on the positioning of the components, in UKA ﬁnal limb alignment is determined by the thickness of the implant relative to the bone excised . The aim of UKA is to restore knee kinematics by restoring ligament tension to normal. This enables alignment of the knee and leg to return to its pre-disease state . In UKA correct ligament balance is restored by positioning the components accurately and inserting an appropriate thickness of bearing. As the tension is restored to normal, the intra-articular deformity secondary to arthritis is corrected . Many patients have varus alignment before they develop medial osteoarthritis or necrosis because of constitutional tibia vara (extra-articular deformity), resulting in varus alignment post-operatively . Therefore, patients with preoperative frontal deviations exceeding 10–15° are not optimal candidates for
⁎ Corresponding author. Tel.: +39 338 3121949 (Mobile). E-mail address: [email protected]
http://dx.doi.org/10.1016/j.knee.2014.12.004 0968-0160/© 2014 Elsevier B.V. All rights reserved.
UKA, since the residual postoperative axis will exceed 8–10° with possible failure due to polyethylene wear and implant loosening [3,6]. Although several studies have already reported how marked (≥ 8–10°) varus alignment is likely to increase the rate of UKA failure from wear or loosening [2–4,6–8], few data exist regarding the outcome and survivorship of medial UKA in patients with different degrees of minor varus alignment. The aim of this study was therefore to analyse the clinical results of medial UKA implanted with no more than 7° of varus, and to verify whether there was a relationship between the limb alignment and the clinical and radiological outcomes. The hypothesis was that minor varus leg alignment was essential to obtain optimal results and medium to long-term survivorship of medial UKA. 2. Materials and methods From 2005 to 2011, 164 UKAs were implanted for unicompartmental osteoarthritis or osteonecrosis at our institution, by the same senior surgeon (ASP). All patients received a cemented metal-back ZUK (Zimmer Unicompartmental High Flex Knee System, Zimmer, Warsaw, IN — USA) prosthesis with a ﬁxed bearing. Selection criteria for implantation of a UKA were; frontal deformity of less than 10°, ﬂexion contracture less
M. Vasso et al. / The Knee 22 (2015) 117–121
Table 1 The demographic data and follow-up time. Total no.
113 patients (125 medial UKAs)
Age (year), mean (range) Gender Male Female Diagnosis Primary osteoarthritis Osteonecrosis Side Right Left Bilateral unicompartmental arthroplasty BMI (kg/m2), mean (range) Follow-up time (y), mean (range)
70.4 (60–84) 25 patients 88 patients 106 knees 19 knees 61 knees 64 knees 12 patients 24.7 (20.7–29.1) 6.8 (2.4–9.3)
than 10°, intact ligaments, no intraoperative evidence of severe degeneration of the patellofemoral joint, patient age younger than 50 associated with high activity level, and body mass index (BMI) higher than 35 kg/m2 (obesity-II and -III ) were considered contraindications for UKA implantation. Of these 164 UKAs, 33 were excluded from the present study. Twenty two were lateral UKAs and 11 had a postoperative mechanical axis ≥8° of varus (11 cases). Four patients were lost to follow up and two patients died post-operatively. Therefore, the study cohort comprised 125 medial UKAs in 113 patients. The preoperative demographic data and
follow-up time are listed in Table 1. The ethics committee of the University of Molise does not require its approval for the review of patient records or images. The surgical technique has been already described . A medial parapatellar quadriceps-sparing approach allowed the surgeon direct access to the medial compartment. The tibial resection was ﬁrst performed assisted by an extramedullary guide. The frontal inclination and thickness of the tibial cut were performed according to the proximal tibial epiphyseal axis (PTEA) in order to restore, respectively, the correct obliquity of the joint line and the pre-degeneration mechanical axis of the lower limb. In particular, a standard 4-mm resection of the tibial bone was initially planned in all patients, slightly adjusted, if necessary, according to individual PTEA. In the sagittal plane, the tibial cut was set to the preoperative value of tibial slope, however at not more than 10° as advised by previous reports [1,7]. Based on the frontal tibial resection, the femur was successively resected through a spacer-block that allowed a distal and posterior femoral cut parallel to the tibial cut. Once the correct femoral and tibial components had been selected, a trial was made with the appropriate bearing. If the knee was slightly lax by approximately 2 mm on varus and valgus stress both in extension and in ﬂexion, the soft tension was considered to be appropriate and the corresponding deﬁnitive polyethylene implanted. No ligament release was undertaken. Patients were assessed through the International Knee Society (IKS) knee and function scores , and range of knee motion (ROM). Preoperative weight-bearing long-leg AP view radiographs as well as
Fig. 1. PTEA. The PTEA is determined by the perpendicular to the line connecting the lateral and medial physis (a). The angle between PTEA and the tibial mechanical axis represents the value of the extra-articular constitutional deformity. This angle should dictate the frontal tibial cut inasmuch as the postoperative obliquity of the joint line and the residual femoro-tibial mechanical axis should be correspondent to this value. In some cases, it may be very difﬁcult to determine the line connecting the lateral and medial physis, although the PTEA can be also deﬁned using a line parallel with the unaffected compartment (b).
M. Vasso et al. / The Knee 22 (2015) 117–121
Rosenberg, Merchant and lateral projections were taken in all patients. The PTEA was pre-operatively determined by the perpendicular to the line connecting the lateral and medial physis; when it was not possible to draw the line connecting the lateral and medial physis, then the PTEA was deﬁned as the perpendicular to the line parallel with the unaffected compartment (Fig. 1) . Post-operatively, knee radiographs were repeated at 1, 6 and 12 months after surgery, and then yearly. Femorotibial mechanical alignment (FTMA) angle (the angle between the centres of the hip–knee–ankle joints) was measured through the long-leg radiographs. Radiographic measurements of FTMA angle were performed twice by a single radiologist (who was blinded to the aim of the study), both on the preoperative and on the last follow-up radiographs. For the purposes of clarity, a varus angle was indicated as positive, and any valgus angle negative. Radiographs were also used to calculate the angle between PTEA and postoperative tibial cut surface, and to highlight the presence of radiolucent lines, osteolysis, component loosening or subsidence, and progression of arthritis in unreplaced compartments.
Polyethylene wear, osteolysis, component subsidence or loosening did not occur in any case. In 11 (9%) cases, 12 months after surgery, radiolucent lines (incomplete and less than 1 mm) were observed around the tibial component and did not progress at further follow-up, nor were associated with subsidence of the tibial platform. One of these 11 patients experienced moderate pain during loading, but refused revision of his prosthesis; this patient was overweight (BMI: 28.2), and in the postoperative varus of 4° (group B). The incidence of radiolucent lines was similar in all three groups. At the last follow-up, no patient had progression of the osteoarthritis in the un-replaced compartments. Two patients (one in group A and one in group C) experienced a deep infection and were revised through a two-stage reimplantation to a postero-stabilized TKA. Thus the ﬁnal survivorship for this cohort was 98.4% at the latest follow-up.
5. Discussion The principal ﬁnding of this study was that minor varus alignment (≤ 7°) was associated with a better outcome and medium to longterm survivorship of medial UKAs. The overall IKS scores were signiﬁcantly higher with increasing postoperative varus. The IKS knee scores were signiﬁcantly higher in group B and still higher in group C when
3. Statistical analysis IKS knee scores were analysed with a regression analysis using FTMA angle as the explanatory variable. Conversely, IKS function scores could not be analysed through a regression analysis, because IKS function scores are not continuous, assuming mainly values of 80, 90 or 100. IKS knee and function scores were also analysed as a function of two covariates, age and BMI. To further validate the relationship between IKS scores and FTMA angle, the subjects were classiﬁed into thirds thus deﬁning three groups according to the FTMA angle (group A: − 2° to 1°; group B: 2° to 4°; group C: 5° to 7°) and an univariate ANOVA of the IKS knee and function scores was performed, using the three FTMA groups as factor. Finally, since the limited number of the different outcomes, the IKS function scores were divided into two categories (IKS function score below and above 90 points) and the resulting incidence of these scores were analysed through a contingency table, dividing the FTMA into two groups (below or above 4°). The signiﬁcance threshold to reject the null hypothesis was set at p b 0.05. 4. Results Intra-observer reliability of the radiographic measurements resulted 0.97 (p = 0.001). Mean correction angles and mean clinical and functional results are listed in Table 2. Furthermore, the mean angle between PTEA and postoperative tibial cut surface resulted 89.7° ± (standard deviation) 1.12, 90.3 ± 0.98 and 91.0 ± 1.03 respectively in groups A, B and C (p = 0.254). A linear regression analysis of the IKS knee scores as a function of the FTMA angle showed how IKS knee scores increased proportionally with increasing varus according to a linear relationship (p b 0.01). The equation of the linear regression was: IKS knee score = 83.2 + 2.2 x FTMA°. IKS knee and function scores were not signiﬁcantly affected by covariates age and BMI (Fig. 2). The univariate ANOVA of the IKS knee score using the three FTMA groups as factor showed a signiﬁcant effect for the FTMA (p = 0.003) (Fig. 3). The univariate ANOVA of the IKS function score using the three FTMA groups as factor did not show a signiﬁcant effect for the FTMA (p = 0.576) (Fig. 4). However, this result may mainly depend on the limited number of outcomes of the IKS function score (which assumed mainly values of 80, 90 or 100, as already mentioned). However, a Chi-square test showed a signiﬁcant association between FTMA and IKS function scores (p = 0.009) (Table 3).
Table 2 Mean correction angles and mean clinical and functional results.
Mean FTMA Mean correction angle Mean correction angle in group A, B, C Mean IKS knee score Mean IKS function score Mean ROM
2.8° (−2°–7°) 3.9° (2.7°–5.3°) 3.8°, 4.1°, 3.9°
89 points (72–100) 92.6 points (82–100) 129° (117°–141°)
b 0.05 b 0.05 b 0.01
47.2 points (37–63) 52.7 points (40–69) 109° (100°–123°)
Fig. 2. IKS knee score as a function of the FTMA, and effect of the covariates on IKS scores. Linear regression between IKS knee score using the FTMA angle as predictor (a). The IKS knee and function scores were plotted as a function of the two covariates considered in the study, age (b) and BMI (c). In this case, the linear relationship between IKS scores and respectively age and BMI did not result statistically signiﬁcant. This result suggests that the IKS scores depended only on the FTMA angle and was not inﬂuenced by age and BMI. In all panels the 95% conﬁdence intervals are also reported.
M. Vasso et al. / The Knee 22 (2015) 117–121
Fig. 3. IKS knee and FTMA. Mean IKS knee score in the different three FTMA groups.
Fig. 4. IKS function and FTMA. Mean IKS function score in the different three FTMA groups.
compared to those in group A (Fig. 3). On the other hand, IKS function scores increased progressively in the three groups (Fig. 4), and were not statistically signiﬁcantly different, probably because the scores tended to be valued at 80, 90 or 100. A non-parametric chi-square test found a signiﬁcantly higher frequency of IKS function scores N90 points in subjects with FTMA angle ≥4° (Table 2). Certainly, these results were not inﬂuenced by preoperative FTMA values, since the mean correction angle was not signiﬁcantly different in the three groups at a mean of 3.9° (Fig. 5). Although in TKA a neutral mechanical axis of 0° should be always obtained, in UKA the surgeon should correct only that part of the deformity due to wear, thus restoring the original (pre-disease) mechanical axis. Preserving the PTEA and avoiding excessive or insufﬁcient corrections of the pre-operative limb alignment are predictors of a successful UKA. Knee pain, polyethylene wear and lateral compartment deterioration are the most common complications affecting UKA survivorship [2, 8], but were not present in this cohort. However patients with extreme post-operative malalignment were excluded from the study. Many studies have focused on the varus/valgus inclination of the tibial component, rather than on the overall limb alignment in UKA. Collier et al.  reported a reduced risk of failure when the tibial component was implanted more valgus than the pre-operative coronal alignment of the tibial plateau. Chatellard et al.  found a decreased survival of the medial UKAs with a greater than 3° change in tibial component obliquity. Recently, Zambianchi et al.  found that slight varus alignment of the tibial component on the coronal plane relative to tibial longitudinal axis may be preferable over valgus deviations. Finally, authors suggested that excessive deviation of the tibial component orientation from the pre-operative angle contributes to early UKA failure. Anyway, one should keep in mind that varus/valgus inclination of the UKA components does not affect all the lower limb alignment, but only the obliquity of the joint line. As already mentioned, in UKA ﬁnal alignment of the lower limb is independent from the alignment of the prosthetic components . In fact, in UKA femoro-tibial alignment is determined by the height of the contact point between the medial femoral condyle and the tibial component, which is dependent on resection level of the proximal tibia, ligament stability, preoperative deformity, implant thickness and surgical technique [15–17]. Although many authors have already reported how marked varus alignment (≥ 8–10°) of the lower limb is associated with increased polyethylene wear and implant loosening [2–4,6–8], the range for an adequate lower limb alignment after (medial) is UKA still debated [6, 18,19]. Some authors have suggested that the ideal limb alignment
after UKA should be still considered the close-to-neutral one [2,20,21]. Others have suggested to avoid limb alignment higher than 5° and 6° varus ; in particular, a 2.5 times higher bone subsidence area has been reported when the prosthesis was 6° varus, if compared to a neutrally aligned UKA . However, these studies have focused primarily on the tibial component positioning too, being the overall limb alignment considered secondly. Recently, Kim et al. , at a mid-term follow-up, found no signiﬁcant relationship between the anatomic femoro-tibial angle (the angle between the shafts of the bones) and the IKS scores and ROM, whilst the survival rate was higher in patients with a neutral varus/valgus limb alignment. Gulati et al.  reported that the mean American Knee Society score worsened signiﬁcantly with increasing varus, whereas no other score deteriorated with increasing varus, and the occurrence of radiolucent lines was the same in all patients. However, the follow-up of their study was short (5 years), and patients with varus alignment were found to be signiﬁcantly younger and signiﬁcantly more likely to be male and active people. To the authors' knowledge, there are no similar reports to this study with a mean follow-up greater than 5 years for ﬁxed-bearing medial UKA that compares the different degrees of minor varus limb alignment in a large cohort of patients. This study showed how clinical and functional outcomes of medial UKAs signiﬁcantly improved with increasing varus. There is a possible biomechanical explanation for this. In UKA, whatever the alignment of the leg, the centre of the force is likely to be near the centre of the component . This is more true as the value of mechanical axis is closer to 5–7° of varus. Therefore the implantbone interface will be directly compressed. This, theoretically will reduce the risk of loosening, as found here, and may result in better clinical outcomes . The ﬁrst limitation of this study was that it was a retrospective analysis. In addition, the radiographic measurements were taken by a single radiologist; anyway, this minimized the inter-observer variability. Another limitation was that the amount of the bone resected from the medial joint was not quantiﬁed. The amount of bone resected alters the bearing thickness and therefore ﬁnal limb alignment. However, all procedures were performed by a single surgeon (ASP), so that the same surgical criteria of bone preservation was uniform throughout the entire study In all patients, a standard 4-mm resection of tibial bone was initially planned, then adjusting it according to the individual PTEA. No analysis of the postoperative tibial slope was carried out: all patients of this study presented integrity of the cruciate ligaments and received restoration of the preoperative slope (7–10°) simply by setting the sagittal inclination of the tibial cutting-guide. Therefore, the sagittal
M. Vasso et al. / The Knee 22 (2015) 117–121 Table 3 Contingency table of the IKS function score and FTMA angle. Pearson Chi-square = 0.009.
FTMA b 4° FTMA ≥ 4° Total
IKS b 90 points
IKS N 90 points
47.1% 14.7% 61.8%
11.7% 26.5% 38.2%
58.8% 41.2% 100%
inclination of the tibial component was not considered affecting clinical outcome and implant survivorship in this cohort. No analysis of the frontal inclination of the tibial component was carried out too. However, it has been clearly reported that varus/valgus inclination of the prosthetic components does not affect limb alignment signiﬁcantly . In this study, the frontal inclination of the tibial cut was planned orthogonal to PTEA, in order to restore native joint line obliquity; the angle between PTEA and postoperative tibial cut surface did not result signiﬁcantly different amongst the three FTMA groups, and close to the desired 90°. Despite these limitations, this study provides valuable information and clinical relevance. This study shows how limb alignment after UKA should be returned to the pre-disease state through the appropriate bone resections, and that postoperative minor varus alignment (due to a preoperative tibia vara) does not compromise overall outcomes and ﬁnal survivorship of medial UKAs. Anyway, one should keep in mind that UKA remains a quite demanding procedure, so that UKA may not be conducive to an optimal FTMA restoration, whereas meticulous care should be taken on proper implant selection (based
Fig. 5. Mean correction angle. The mean correction angle resulted homogenous in the three groups of patients with different FTMA values (p = 0.345), with an average of 3.9° that generally corresponds to the intra-articular deformity secondary to the cartilaginous wear and degeneration. In the picture, the angle between the preoperative FTMA (red dashed line) and the postoperative FTMA (red continuous line) is very close to the mean correction angle found in this series. Furthermore, the ﬁgure shows also the desired parallelism between the joint line, the tibial cut surface and the line connecting the lateral and medial physis (yellow lines).
on the patient's preoperative condition and planning) and on correct execution of surgery, in order to achieve the desired postoperative limb alignment. 6. Conclusions Post-operative mechanical femoro-tibial varus alignment of 7° or less does not compromise the mid-term outcome of a medial ﬁxedbearing metal-back UKA. Indeed, minor varus alignment provides better results compared to neutral or close-to-neutral alignment. Longer follow-up is needed to verify if minor varus alignment does not cause polyethylene wear or loosening in this patient cohort. References  Dao Trong ML, Diezi C, Goerres G, Helmy N. Improved positioning of the tibial component in unicompartmental knee arthroplasty with patient-speciﬁc cutting blocks. Knee Surg Sports Traumatol Arthrosc 2014. http://dx.doi.org/10.1007/s00167-014-2839-2.  Valenzuela GA, Jacobson NA, Geist DJ, Valenzuela RG, Teitge RA. Implant and limb alignment outcomes for conventional and navigated unicompartmental knee arthroplasty. J Arthroplasty 2013;28:463–8.  Deschamps G, Chol C. Fixed-bearing unicompartmental knee arthroplasty. Patients' selection and operative technique. Orthop Traumatol Surg Res 2011;97:648–61.  Gulati A, Pandit H, Jenkins C, Chau R, Dodd CA, Murray DW. The effect of leg alignment on the outcome of unicompartmental knee replacement. J Bone Joint Surg Br 2009;91:469–74.  Levine B, Rosenberg AG. The simple unicondylar knee: extramedullary technique. Clin Sports Med 2014;33:77–85.  Manzotti A, Cerveri P, Pullen C, Confalonieri N. Computer-assisted unicompartmental knee arthroplasty using dedicated software versus a conventional technique. Int Orthop 2014;38:457–63.  Hernigou P, Deschamps G. Alignment inﬂuences wear in the knee after medial unicompartmental arthroplasty. Clin Orthop Relat Res 2004;423:161–5.  Zambianchi F, Digennaro V, Giorgini A, Grandi G, Fiacchi F, Mugnai R, et al. Surgeon's experience inﬂuences UKA survivorship: a comparative study between all-poly and metal back designs. Knee Surg Sports Traumatol Arthrosc 2014. http://dx.doi.org/10. 1007/s00167-014-2958-9.  Schmiegelow MD, Andersson C, Køber L, Andersen SS, Norgaard ML, Jensen TB, et al. Associations between body mass index and development of metabolic disorders in fertile women—a nationwide cohort study. J Am Heart Assoc 2014;3:e000672.  Panni AS, Vasso M, Cerciello S, Felici A. Unicompartmental knee replacement provides early clinical and functional improvement stabilizing over time. Knee Surg Sports Traumatol Arthrosc 2012;20:579–85.  Insall JN, Dorr LD, Scott RD, Scott WN. Rationale of the Knee Society clinical rating system. Clin Orthop Relat Res 1989;248:13–4.  Jenny JY, Boéri C, Ballonzoli L, Meyer N. Difﬁculties and reproducibility of radiological measurement of the proximal tibial axis according to Lévigne. Rev Chir Orthop Reparatrice Appar Mot 2005;91:658–63.  Collier MB, Eickmann TH, Sukezaki F, McAuley JP, Engh GA. Patient, implant, and alignment factors associated with revision of medial compartment unicondylar arthroplasty. J Arthroplasty 2006;21:108–15.  Chatellard R, Sauleau V, Colmar M, Robert H, Raynaud G, Brilhault J. Société d'Orthopédie et de Traumatologie de l'Ouest (SOO). Medial unicompartmental knee arthroplasty: does tibial component position inﬂuence clinical outcomes and arthroplasty survival?Orthop Traumatol Surg Res 2013;99(4 Suppl.):S219–25.  Kim SJ, Bae JH, Lim HC. Factors affecting the postoperative limb alignment and clinical outcome after Oxford unicompartmental knee arthroplasty. J Arthroplasty 2012; 27:1210–5.  Jung KA, Kim SJ, Lee SC, Hwang SH, Ahn NK. Accuracy of implantation during computer-assisted minimally invasive Oxford unicompartmental knee arthroplasty: a comparison with a conventional instrumented technique. Knee 2010;17:387–91.  Kim KT, Lee S, Kim TW, Lee JS, Boo KH. The inﬂuence of postoperative tibiofemoral alignment on the clinical results of unicompartmental knee arthroplasty. Knee Surg Relat Res 2012;4:85–90.  Pandit H, Jenkins C, Gill HS, Barker K, Dodd CA, Murray DW. Minimally invasive Oxfor phase 3 unicompartmental knee replacement: results of 1000 cases. J Bone Joint Surg Br 2011;93:198–204.  Price AJ, Svard U. A second decade lifetable survival analysis of the Oxford unicompartmental knee arthroplasty. Clin Orthop Relat Res 2011;469:174–9.  Perkins TR, Gunckle W. Unicompartmental knee arthroplasty: 3- to 10-year results in a community hospital setting. J Arthroplasty 2002;17:293–7.  Rosenberger RE, Fink C, Quirbach S, Attal R, Tecklenburg K, Hoser C. The immediate effect of navigation on implant accuracy in primary mini-invasive unicompartmental knee arthroplasty. Knee Surg Sports Traumatol Arthrosc 2008;16:1133–40.  Sawatari T, Tsumura H, Iesaka K, Furushiro Y, Torisu T. Three-dimensional ﬁnite element analysis of unicompartmental knee arthroplasty—the inﬂuence of tibial component inclination. J Orthop Res 2005;23:549–54.  Small SR, Berend ME, Rogge RD, Archer DB, Kingman AL, Ritter MA. Tibial loading after UKA: evaluation of tibial slope, resection depth, medial shift and component rotation. J Arthroplasty 2013;28(9 Suppl.):179–83.