1 Introduction

Over recent years, the alignment in the coronal plane of a total knee arthroplasty (TKA) has been a topic of discussion. Various principles have emerged in pursuit of optimal clinical outcomes; however, the most clinically effective approach remains uncertain.

Mechanical alignment in TKA was initially outlined by Ranawat and Insall in the 1970s. This method aims to position both the femoral and tibial components perpendicular to the mechanical axis¹ while adjusting soft tissue as necessary to achieve a neutral joint line (parallel to the floor). The objective is to establish a neutral lower limb alignment to facilitate balanced load distribution across the medial and lateral compartments.² Despite being the traditional standard, it has been suggested that mechanical alignment may be responsible for some poorer outcomes following TKA.³

The kinematic alignment principle, introduced by Howell et al., in 2008,⁴ seeks to restore the pre-arthritic hip–knee–ankle (HKA) angle and joint line obliquity, thus preserving ligament tension.¹ This approach is individualised to each patient's unique pre arthritic coronal alignment. There have been some concerns expressed when this leads to positioning implants in extreme varus-valgus alignments, potentially compromising implant survival by creating vastly different medial and lateral load distributions.

Consequently, a contemporary trend favours a restricted kinematic alignment (rKA) strategy, incorporating safe boundaries to restore a variable percentage of a patient's natural alignment, contingent on accepted limits (e.g. rKA boundaries of HKA −6° to 3°, lateral distal femoral angle (LDFA) or medial proximal tibial angle (MPTA) 84°–93°, had knee constitutional alignment restored in 85.4 % of patients).⁵

MacDessi S.J et al., 2021 (6) identified nine phenotypes of coronal plane alignment of the knee (CPAK), based on arithmetic Hip-Knee-Ankle (aHKA) measurements of constitutional limb alignment derived by subtracting the LDFA from the MPTA, and joint line obliquity (JLO) (MPTA plus LDFA). The most prevalent categories observed in the study population were Type II, Type I, and Type V, respectively, with Types VII, VIII, and IX being rare. Altering a patient's CPAK phenotype during TKA may impact gait, biomechanics, and soft tissue balance, although the extent of its influence on clinical outcomes and satisfaction remains uncertain.

This study aims to compare preoperative and postoperative coronal plane knee alignment (CPAK) in patients undergoing bilateral TKA with SAIPH implants (MatOrtho, UK). The concept was to control the implant (same knee both sides), patient (bilateral model) and assess the effect that any surgical alteration in CPAK alignment had on patient outcomes measured by patient-reported outcome measures (PROMS) and patient satisfaction.

2 Method

143 patients who underwent bilateral sequential TKA with a SAIPH implant (MatOrtho, UK), conducted by two senior orthopaedic surgeons (A.S and J.B) between January 1, 2012 to January 1, 2022 were identified. Surgery was performed using a midline incision and a medial parapatellar arthrotomy. Both knees in all patients had the SAIPH TKA (MatOrtho, UK). This design is a medially stabilized knee^7,8 with a cumulative percent revision of 2.8 (2.0, 4.1) at 10 years. This is the lowest revision rate on the Australian Joint Replacement Registry.⁹ The initial patients in the series started in 2010, prior to general adoption of the concept of kinematic or restricted kinematic alignment. Our starting goal was to achieve mechanical alignment but then intra operatively using bone resections and soft tissue releases to achieve a stable balanced knee. What is important in this study is not what we planned to achieve or how we achieved it but what we actually achieved in the coronal plane and how it affected clinical outcome.

Patients were excluded if they were uncontactable, didn't want to participate, had incomplete/unavailable radiological data, had documented previous fractures in the lower limb (femur or tibia), undergone revision surgery on their knees, or experienced complications requiring another surgery e.g., knee infection, vascular injury, or patella tendon rupture. After exclusions there were eighty-seven patients with a full data set of clinical and radiological data available for follow up and assessment (Fig. 1).

Close

Fig. 1 Exclusion and inclusion criteria.

Export to PPT

80 patients attended clinical review consultations, while 7 were reviewed remotely. Range of movement was assessed using a mobile app (Dr. Goniometer).¹⁰ The 7 patients unable to attend were provided with a protocol on how to take photos in a seated position with leg extension and maximum bend. These photos were then sent via email or post and assessed using the Dr. Goniometer app.¹⁰

Patients were sent PROMS questionnaires via email or via postage. Instructions were given stressing the importance of answering questions for each joint separately. The PROMS used were: University of California at Los Angeles (UCLA) activity level score, Visual Analogue Scale (VAS) for pain and satisfaction, Oxford Knee Score (OKS), Forgotten Joint score (FJS) and Knee Injury and Osteoarthritis Outcome Score (KOOS). All scores had assessments for each joint except the UCLA score. In addition, patients were asked which knee did they prefer and why. Afterwards, patients would bring the questions answered to the consultation or send them via email. Results were registered by the orthopaedic fellow.

Patients examined in the consultation underwent assessment of range of movement with a goniometer, anteroposterior stability, mediolateral stability in extension and at 30° flexion and hip pain. Most clinical assessments were conducted by the orthopaedic fellow. There were exceptions for three patients: one performed by an orthopaedic training registrar and two by a senior orthopaedic surgeon.

From 2012, a protocol was followed for preoperative planning and postoperative analysis using CT scans to assess lower limb alignment. This preoperative CT planning included the analysis of patient's lower limb alignment allowing retrospective calculation of both pre-operative native CPAK and post-operative CPAK classification, ie, calculation of LDFA, MPTA, JLO, aHKA preoperatively and postoperatively. All landmarking and registration were conducted by two independent medical engineers who were involved in pre-operative radiological planning for TKA. Preoperatively, all patients had plain radiology (AP, lateral and patella sunrise views) which were repeated at the time of review.

17 patients were further excluded after clinical consultation and current plain radiology for reasons of knee instability, patella wear or patella cysts associated with anterior pain, fracture shaft of femur or tibia/fibula, back pain/lower limb joint pain affecting knee assessment, knee tumor, periprosthetic joint infection, dementia and poor preoperative scan quality imaging hindering measurement calculation (Fig. 1). This leads to 70 patients being eligible for inclusion in the study group.

The mean and 95 % confidence intervals (CI) were used to describe all anatomical, clinical functional and PROMS measurements. Independent two-tailed T-Tests were conducted to compare PROMS across different patient groups stratified by the recreation of the CPAK phenotype, preference for the joint and isolated changes in either aHKA or JLO categories. A Chi-Square test was performed to assess whether patients exhibited a preference for joints where the CPAK phenotype had been recreated. All statistical analyses were performed using Posit R studio (Boston, United States) and statistical significance was defined by a P-Value below 0.05.

3 Results

A total of 70 patients and 140 knees were included in this study. The mean age for the first surgery was 66.6 ± 1.7, and for the second TKA was 69 ± 1.7. Among the 70 patients, 27 were females and 43 were males.

In the 140 TKAs, 52 % of joints underwent patellofemoral resurfacing. The mean range of movement for all TKAs was an extension of 0.2 ± 0.5 (−8 - 15) and a flexion of 116.6 ± 1.5 (95–140).

PROMS were assessed for all 140 joints individually, the mean OKS was 42.8 ± 0.9 (16–48), KOOS 85.0 ± 2.7 (18,75–100), FJS 73.3 ± 4.4 (0–100), UCLA 6.3 ± 0.3,^3–10 VAS for satisfaction 92.1 ± 2 (10–100), and VAS for pain 1.0 ± 0.3 (0–7).

Preoperative alignment assessment of the CPAK phenotypes of the 140 joints revealed 40.0 % (56 joints) type I, 32.1 % (45 joints) type II, 15.0 % (21 joints) type III, 4.3 % (6 joints) type IV, 2.9 % (4 joints) type V, and 5.7 % (8 joints) type VI (Fig. 2A).

Close

Fig. 2 Preoperative and postoperative CPAK distribution of study joints (n = 140 joints).

Export to PPT

Postoperative CPAK phenotype recreation was achieved in 27.9 % (39 joints), while in 72.1 % (101 joints) it was not (Fig. 2). In the 101 joints where CPAK was not recreated, 70 joints moved one CPAK category either vertically or horizontally, 28 joints moved two categories, and 3 joints moved three categories (Fig. 2B). PROMS did not significantly differ when CPAK phenotype was not recreated; however, there was a tendency for decreasing outcome scores with increasing displacement from the preoperative box (p > 0.05) (Tables 1 and 2).

If patients reported having a preferred side, it correlated significantly with better PROMS (p < 0.05) (Table 3). However, there was no clear distinction between the patients PROMS when one knee had CPAK recreated and the other did not (Table 3). In addition, in the 39 joints with recreated CPAK, 35.9 % were preferred, 43.6 % were non-preferred, and in 20.5 % the patient had no preference (p = 0.44) (Table 4).

In patients who had CPAK recreation in both their joints (16 joints, 8 patients) compared to those with neither joint having CPAK recreation (78 joints, 39 patients), no significant difference in PROMs was observed (Table 5). Similarly, no significant difference in PROMs was found between patients with one joint recreated and the other not (Table 6).

Of the 140 joints, 56 joints had isolated changes in the aHKA but not in the JLO (horizontal movement in the CPAK classification). When PROMS from these patients were compared to PROMS from patients with CPAK restoration (39 joints), no statistically significant difference were found. (Table 7).

Furthermore, 20 joints from the 140, had isolated changes in the JLO but not in the aHKA (vertical movement in the CPAK classification). We compared PROMS from these patients with another 39 patients who had recreation of native CPAK alignment and no significant differences were found (Table 8).

4 Discussion

In modern TKA, various target alignments are employed which vary from strict mechanical alignment to more recent concepts of kinematic and restricted kinematic alignment. The CPAK classification has given surgeons a strategy to analyse individual patient's native coronal alignment phenotype which allows consideration of recreating that alignment when performing the TKA. This may allow more optimal soft tissue balancing. Kinematically aligned TKAs have shown in some studies to lead to better functional outcomes compared to mechanically aligned TKAs. However, there is still ongoing debate about “personalized alignment strategies'’.

Our results suggest that failure to recreate the native CPAK phenotype does not adversely affect patient outcomes when comparing patients in a bilateral TKA model. Similar findings were reported by Sappey-Marinier et al., 2022.¹¹ Additionally, isolated changes to aHKA or JLO did not significantly impact outcomes, nor was a recreated joint more likely to be the patient's preferred side. CPAK recreation was only achieved in 27.9 % of joints; yet, mean PROMS observed high satisfaction scores. This result highlights the fact that patient's satisfaction is multifactorial^12–15 and the role of coronal alignment recreation remains unclear. Minor changes in CPAK classification do not affect outcomes and may be considered as safe boundaries to achieve improved satisfaction following TKA surgery. MacDessi S.J et al., 2021 (6)cautioned against altering native coronal alignment which may result in knee imbalance. This, in turn, may result in an increase in the intercompartmental pressure difference (ICPD) which may alter clinical outcomes including the longevity of the prosthesis.⁶

Our study has a number of limitations:•We only considered the coronal plane alignment. If any patient had a large variation from normal in sagittal alignment they were excluded from the study group. Similarly, any patient with significant patellofemoral maltracking was also excluded. These exclusions acknowledge that sagittal and axial alignment is also important in the planning and outcomes of TKA.•PROMS have important benefits such as promoting active patient involvement or enabling standardized monitoring of patient outcomes but they have limitations. PROMs may lack clinically meaningful information and may not be suited for all patients. McEwen et al., 2020¹⁶ found that by comparing rKA in one knee and mechanical alignment in the contralateral knee of the same patient undergoing TKA, clinical outcomes were equivalent at 2 years. However, significantly more participants preferred their rKA joint.•Not all patients had the patellofemoral joint resurfaced. Selective patella resurfacing is a standard technique in many centres including ours. This SAIPH Total Knee implant is a medially stabilized design which encourages internal rotation of the tibia hence medialisation of the tibial tubercle with flexion and in addition, a lateralised patella. Both of these features make it a patella friendly implant (A. Shimmin et al., 2015⁷), (Katchky. A.M et al., 2019⁸). The implant has cumulative percent revision of 2.8 (2.0, 4.1) at 10 years. This is the lowest revision rate on the Australian Joint Replacement for implants implanted over 10 years suggesting that selective patella resurfacing seems to be reasonable with this implant. In addition, any patients with patellofemoral joint symptoms were excluded from the study group.•Although this series is small, by controlling for patient factors and implant geometry (bilateral TKR with the same implant) it allows us to study in more detail the effect of coronal alignment variations in isolation. In addition, we have excluded many other factors which may affect clinical outcome; for example: infection, sagital alignment variations, fractures, etc which has also added to smaller patient numbers.•Patients that were lost to follow-up (35.6 %) may alter the results of this study.•In some patients we had to assume that both knees were the same CPAK as they only had radiology to assess CPAK preoperatively on only one knee. S.J. Macdessi et al., 2020 ¹⁷ showed that there was no significant difference between aHKA of a patient's osteoarthritic knee and that patient's contralateral knee mechanical HKA. Nevertheless, we acknowledge that other papers such as Eckhoff et al., 2016¹⁸ conclude that human limbs aren't absolutely symmetric. However, coronal alignment tends to be more symmetrical than the rotational alignment.

The strength of this study was that we were able to control patients’ variability by only selecting patients with bilateral TKAs with both knees receiving the same implant (SAIPH TKA).

5 Conclusion

Tailoring alignment approaches to individual patient characteristics may be crucial for optimal functional outcomes but in this study, we were not able to show a difference by using PROMs as an outcome measure between those patients who had native CPAK alignment recreated following TKA and those who did not.

Further research is needed to determine the clinical relevance and optimal alignment techniques to be used for TKA.

CRediT authorship contribution statement

Aida Orce Rodríguez: Investigation, Writing – original draft, Writing – review & editing, Visualization. Ishaan Jagota: Formal analysis, Visualization, Writing – review & editing. Jonathan Baré: Investigation, Resources, Writing – review & editing, Supervision. Andrew Shimmin: Conceptualization, Methodology, Validation, Resources, Writing – review & editing, Supervision, Project administration.

Guardian/patient's consent

We carried out a research study entitled ‘Impact of changes in native coronal plane alignment of the knee (CPAK) on patient-reported outcome measures (PROMS). A bilateral single implant study’ at Melbourne Orthopaedic Group, Melbourne, Victoria, Australia.

This study aims to compare preoperative and postoperative coronal plane knee alignment (CPAK) in patients undergoing bilateral TKA with SAIPH implants (MatOrtho, UK). The concept was to control the implant (same prosthesis both sides), patient (bilateral model) and assess what effect any surgical alteration in alignment had on patient's outcomes measured by patient-reported outcome measures (PROMS) and patient satisfaction.

The ethics committee approval was not required for this study as post-operative CT scans, Xray imagery and follow-up between 5 and 10 year after total knee replacements are routine practice in our consultations. Patients were aware that data would be used for this study.

Ethical statement

The ethics committee approval was not required for this study as post-operative CT scans, Xray imagery and follow-up between 5 and 10 year after total knee replacements are routine practice in our consultations. Patients were aware that data would be used for this study.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.