The argument for functional alignment in total knee arthroplasty, within parameters and with the aid of robotics

Although the way we have been doing TKA has changed over the past 20 years, there isn’t consensus among orthopaedic specialists on which approach guarantees the most success. In addition, there is overlap between the approaches and it is not always easy to differentiate one concept from the other.

What we do know is that three key elements must be considered during TKA, namely: morphology, soft-tissue and alignment (Hiranaka, 2022). It is vital to maintain harmony between the elements, bearing in mind that each element impacts the other.

As I cover the pros and cons of the different TKA approaches, it’s important to remind ourselves of why we do knee replacements: to relieve pain and restore function. This should always remain our primary objective.

Mechanical alignment

Traditionally, total knee arthroplasty followed the mechanical alignment approach, the goals of which are ensuring a straight line from the middle of the hip, to the middle of the knee, to the middle of the ankle.

In this approach, the femoral cut is made perpendicular to the mechanical axis of the femur and the tibial cut is made perpendicular to the mechanical axis of the tibia (Cherian, 2014, p. 91). This creates rectangular gaps in the coronal plane during extension and flexion. If the gaps are not rectangular, a soft tissue release is performed to create rectangular gaps.

However, as Cherian (2014, p.91) notes, while there will be even distribution of joint loading forces between compartments during the stance phase, there is likely to be uneven loading of components during the gait phase due to a ‘laterally’ directed ground reaction force. In addition, Bonnin et al. (in Lustig, 2021, p 2) found that 75 – 89% of patients with mechanical alignment TKA reported significant discomfort during activities of daily living – a major cause of patient dissatisfaction after TKA.

Kinematic alignment

When kinematic alignment became the more popular approach for surgeons, patient satisfaction increased. In kinematic alignment, the femur is replaced first and then the tibia is cut to balance for the rest of the knee. However, I am not a fan of this approach, because it increases varus in the tibia and there are certain knee types which one must be particularly careful with.

My first concern is that the amount of varus that the tibia is put in with kinematic alignment increases sheer forces and the wear on plastic implant components – potentially increasing failure rates. Secondly, kinematic alignment uses medially constrained prostheses and builds around the medial side. But it can be difficult to ensure a stable medial side when doing kinematic alignment on a valgus knee, because normally the medial side is slightly attenuated or stretched.

The reason the medial side of the knee and medial components are so important is because, when in stance phase, the tibia is sloped in one or two degrees of varus and most weight bearing goes through the medial side of the knee. As a person takes a step, the tibia comes out of varus and is parallel to the floor, with the weight bearing point moving from the centre of the hip to L1.

To counteract the extreme amounts of varus-valgus, some surgeons do inverse kinematic alignment, cutting the tibia symmetrically and referencing the femur off that. However, there will still be sheer forces acting on the implant with inverse kinematic alignment. In my opinion, there are many variables at play when cutting a tibia in varus as slope needs to be considered. This raises the possibility of an oblique plane cut with subsequent increase in sheer forces.

Functional alignment with robotic assistance

The more accurately we place a prosthesis in relation to what the patient’s body needs (in terms of their soft tissue, bone and alignment), the better chance their prosthesis will survive and the patient will have a good outcome. According to Lustig (2021, p.8), the aim of functional alignment is to place the implants in the position that least compromises the soft tissue envelope, thus restoring the obliquity and plane of the joint to that which the ligaments dictate. If the deformities are fixed, a soft-tissue release may be required to balance the gaps, although the extent and frequency of such releases are smaller when compared with the mechanical alignment technique.

It is for this reason that I favour the functional alignment approach, navigated with robotic assistance.

Functional alignment is an evolution of kinematic alignment based on the progression of technology and increased precision (Lustig, 2021, p. 6). During functional alignment, we operate within a ‘safe zone’. For a tibia, this means a mechanical alignment approach (a straight line running from the middle of the hip, to middle of the knee, and middle of the ankle) with specific adjustments. On a varus knee, I can relatively predictably dial in 1 to 2 degrees varus into all tibial cuts. Valgus knees will always be cut in neutral to avoid having a prosthesis in valgus.

A TKA that is performed within this safe zone reproduces a tibial component that is stable and will not fail over the years. If there is valgus, or more than 3˚ of varus, the excessive forces on the lateral side and the pressure that goes through the knee will cause joints to fail.

When I perform knee arthroplasties, robotic assistance is used and safe zone parameters are adhered to, never using more than 2˚ of internal rotation or 5˚ of external rotation. The rotation of the tibia is done by floating the tibial tray in and orienting it according to the tibial tubercle, as well as looking at rotation according to the femoral prosthesis. It is essential to avoid excessive internal and external rotation.

In the case of severe deformities, I tend to opt for a straighter knee, however not fully corrected. A patient that comes in with severe varus, will leave surgery with a straighter leg – within the abovementioned safe zone, to avoid putting the tibia and femur at extreme degrees.

Rotation in functional alignment

The knee replacement that I use is based around the medial side. Rotation is gauged by patella-femoral tracking and concerns in this respect are a tight lateral side of the knee, lateral retinaculum, or patella tilt: A lateral facetectomy will need to be performed to prevent lateral patella-femoral tightness. I try to make sure that whatever is resected medially matches what’s built into the prosthesis thus keeping the medial-collateral ligament isometric throughout its range of motion.

The normal, unreplaced knee is slightly more constrained medially than laterally and there’s more laxity in the lateral compartment. Post-operatively, the knee needs to be constrained medially through the full range of motion, but 1-2mm of gapping laterally is not a concern. This is where functional alignment differs from mechanical alignment. In the latter, we wanted a rectangle gap. With functional alignment, we can have a trapezoidal gap, as long as the trapezoid is widest in the lateral compartment and not medially. The trapezoid can also be wider in flexion than in extension.

If there is a tight lateral compartment, we’d need to reassess femoral rotation by internally rotating the component or doing a ligament release. Therefore, one needs to assess whether there’s tightness laterally in flexion, extension, or both.

If there is lateral tightness in both flexion and extension, the tibia cut needs to be revised. If it is tight in extension, but loose in flexion, more of the distal femur needs to be cut, helping to create a limb that is aligned without doing soft tissue releases. Tightness in either flexion or extension needs to be addressed through femoral positioning, with the aid of robotic navigation. Too much femoral rotation, however, can cause patella maltracking problems.

Throughout this process, the knee needs to be observed in all planes – coronally, sagittally, and axially. As per Lustig (2021, p.7), in the coronal plane, femoral component positioning is modified from a starting point of 0˚ to the mechanical axis to balance the extension gap. In the sagittal plane, the femoral component is positioned to optimise the component sizing and to avoid femoral notch by flexing up to 3˚. In the axial plane, the femoral implant is aligned to the transepicondylar axis with 3˚ of freedom to balance the flexion gap.

Robotic navigation and implant design

The easiest way to accurately place an implant, within safe parameters, is robotically. According to Lustig (2021, p.7), “the precision offered by robotic assistance may make achieving non-neutral alignment targets more reproducible, reducing the risk of missing the target and producing significant outliers of the limb alignment”.

We use robotics to get an idea of how the limb tracks through its range of motion and how the soft tissue envelope functions. Using robotics, we are able to access the tracking of the knee before and after operating, to see how it tracks, whether the patella is moving well, and if we are happy with the implant-positioning for this patient. The crux of a successful knee replacement is to do no, or minimal releases, and to respect each compartment of the knee. Lustig (2021, p.7) notes that functional TKA reduces the need for periarticular soft-tissue releases, while restoring the patient’s native knee kinematics.

The design of the implant also plays a role in the alignment approach that we take. When thinking about the philosophy of the knee, the easiest way to get a reproducible medial gap is to ensure we’re using a single radius type design with soft tissue constraints taken into account. This means cutting the knee to become a single radius by cutting the same amount of distal femur and posterior femur. This allows us to plan that the medial collateral ligament remains isometric and doesn’t elongate – remaining the same length through the range of motion and reducing the chance of developing mid-flexion instability. This will provide stability but not over-constrain.

Using a robotic tensioner and sensor allows us to place it, plan the cuts, do the cuts and then do an objective assessment of what the knee is moving like, with real-time data feedback.

However, it remains imperative, when trial components have been placed, to ensure that the ligaments clinically feel balanced and that the patella femoral joint is tracking well. This, along with real-time feedback from a robotic tensioner should hopefully allow us to decrease the dissatisfaction rate in knee replacements.

Reference list

Cherian, JJ., Kapadia, BH., Banerjee, S., Jauregui, JJ., et al. (2014) ‘Mechanical, Anatomical, and Kinematic Axis in TKA: Concepts and Practical Applications,’ Current Reviews in Muskuloskeletal Medicine, pp89-95. DOI 10.1007/s12178-014-9218-y

Hiranaka, T., Suda, Y., Saitoh, A., Koide, M., et al. (2022) ‘Infographic: Three key elements of kinematic alignment total knee arthroplasty for clarified understanding of its approaches,’ Bone & Joint Research, 11(4). https://doi.org/10.1302/2046-3758.114.BJR-2021-0543.R1

Lustig, S., Sappey-Marinier, E., Fary, C., Servien, E., et al. (2021) ‘Personalized alignment in total knee arthroplasty: current concepts,’ SICOT-J, pp1-8. https://doi.org/10.1051/sicotj/2021021