Our study illustrates the prevalence and characteristics of tibia valga in a cohort of patients with knee osteoarthritis using EOS imaging technology. Firstly, we showed that tibia valga morphology is common in valgus knees, with 35% of knees demonstrating this anatomical variant. Secondly, we showed that the distal femur in valgus knees with tibia valga is within the normal anatomical range. Thirdly, we demonstrated that the degree of tibia valga is positively correlated to the overall angular deformity of patients with valgus knees. Therefore, we can conclude that a significant proportion of valgus knees have an extra-articular deformity of the tibia which might be the primary contributor of this deformity rather than the distal femoral anatomy.
The results of our study are consistent with the only two other studies13,14 which have radiographically analysed tibia valga morphology in valgus knees. Alghamdi et al’s14 (2014) retrospective review of 97 osteoarthritic valgus knees prior to TKA via full lower limb weightbearing x-rays demonstrated a similar prevalence of tibia valga to ours, with 53% of valgus knees illustrating a tibia valga. Furthermore, they also demonstrated that the extra-articular tibia valga deformity was an isolated deformity, with mLDFA reported to be within normal limits and contributed significantly to the overall mechanical femorotibial alignment.14,20 This is also consistent with Eberbach et al’s19 (2016) study challenging the previously accepted dogma that the primary deformity of valgus knees originates in the femur. Our results not only demonstrate the characteristics of tibia valga morphology, but we also demonstrate that there is a positive correlation between the magnitude of the deformity (CORA-tibia) and the overall HKA valgus deformity of the knee.
Our results demonstrate the importance of screening for tibia valga as part of pre-operative planning as it can have implications for the TKA procedure. Restoring the mechanical axis of the lower limb and ensuring accurate cuts of the femur and tibia are critical in determining the success and survivorship of a TKA.23 Restoration of the mechanical axis is achieved through bone cuts perpendicular to the mechanical axes of the femur and tibia respectively. It has been previously demonstrated that usual tolerance of error for bone cuts is approximately 3o from its ideal position.24 The correct placement of cutting jigs and alignment guides are critical in ensuring accurate placement of femoral and tibial components in TKA.25 Although intra-articular causes of valgus deformities can be managed intra-operatively, extra-articular deformities of the lower limb may be overlooked both clinically and surgically. Figure 6 illustrates such an example, where tibia valga can be overlooked if only knee x-rays for operative planning. Given the small window for error for component placement in TKA means that presence of a tibia valga deformity can significantly mislead the orthopaedic surgeon in their tibial cuts.24 Therefore, we recommend that the surgeon adequately plan their TKA with pre-operative lower limb weightbearing radiographs to screen for extra-articular deformities which may contribute to the angular deformity of the knee, given their common prevalence. Previous studies have demonstrated the benefit of standing lower limb films compared to short-leg films which many orthopaedic surgeons still utilise in the pre-operative planning phase, with their greater accuracy in determining overall tibiofemoral alignment.26–28
Furthermore, the common prevalence of tibia valga morphology in valgus knees suggests implications regarding the optimal approach for tibial component placement in TKA. On average in non-arthritic knees, the articular surface of the tibia is 3o varus to its mechanical axis.13 However studies have demonstrated that tibial component placement in TKA should be 90o to the mechanical axis as this allows the prosthetic component to evenly distribute its weight across the tibia and therefore prolong the survivorship of the implant.14,25,29 The literature has differing evidence regarding the use of intramedullary and extramedullary tibial cutting guides in TKA. Although Reed et al’s30 (2002) randomised-controlled trial comparing intramedullary and extramedullary cutting guides for tibial resection in 135 TKAs demonstrated that intramedullary cutting guides resulted in a significantly better tibial alignment compared to extramedullary cutting guides, Ko et al (2007)31 and Yau et al (2007)32 demonstrated that intramedullary cutting guides in varus knees performed poorly with up to 30% of patients receiving unacceptable tibial cuts. Additionally, Dennis et al33 (1993) demonstrated that the use of extramedullary cutting guides in TKA resulted in better tibial component alignment compared to intramedullary cutting guides (88% versus 72%). Given the increased likelihood of cortical impingement of the intramedullary cutting guide which can result in less accurate tibial cuts in a valgus knee with significant tibia valga, we suggest the utilisation of extramedullary cutting guides in these knees. Palanisami et al’s13 (2019) study reaffirms this suggestion, demonstrating that excellent prosthetic component placement can be achieved in valgus knees with tibia valga using extramedullary cutting guides. Additionally, we recommend post-TKA lower limb weightbearing radiographs to ensure adequate component placement, especially in cases of tibia valga where tibial component placement may appear in varus relative to the proximal tibia.
This study must be viewed in consideration of its limitations. We included all patients with EOS scans, irrespective of rotation, therefore scans where the patella was not visible over the femoral condyles where included. This may reduce the accuracy of the geometric values which were obtained in this study; however, we used an experienced arthroplasty fellow to ensure the most consistent and reproducible measurements were made. Secondly, given the cross-sectional nature of this study, it does not highlight the aetiology and progression of tibia valga (for example, if it is due to lateral compartment knee osteoarthritis or a congenital or developmental deformity). Furthermore, our results demonstrated that tibia valga is also an uncommon occurrence in varus knees. What we know from the literature is that when lower-limb alignment changes, the of distribution of stress forces on the tibia can change and result in bone-remodelling which may account for this observation.34 Therefore, longitudinal radiographic studies of patients experiencing knee osteoarthritis can help to address this gap in the literature.