1 Introduction

Anterior cruciate ligament (ACL) deficiency leads to suboptimal knee kinematics and progressive degenerative joint disease of the knee.^1–3 Conversely, osteoarthritis of the knee causes narrowing of the femoral intercondylar notch, and chronic attrition and rupture of the ACL, leading to further degenerative changes.⁴ The natural history of ACL insufficiency from ligament rupture till the development of osteoarthritis is well-described.^1–3 Varus thrust during the stance phase of gait in a knee with constitutional varus and chronic ACL deficiency results in condylar lift-off and separation of the lateral tibio-femoral joint, abnormally high forces on the lateral soft tissues, a medial shift of the point of maximal joint pressure, excessive medial tibio-femoral loading, and progressive medial compartment osteoarthritis.^5,6 The severity of this arthritis is a function of the cartilaginous injury sustained during the initial trauma and the degree of instability.^7,8

Varus osteoarthritic knees with ACL deficiency present with posteromedial tibial wear and greater preoperative varus deformity than knees with a functional ACL.^7,9–11 This wear pattern is determined by the functional status of the ACL, and not the severity of the preoperative varus deformity.⁷ Although lateral femoral condyle hypoplasia in valgus knees is well-described,^12,13 medial femoral condyle hypoplasia is reported not to occur in varus knees.¹⁴ However, in our experience with computer-navigation assisted (CAS) total knee arthroplasty (TKA) for osteoarthritic varus knees, we noted that a subset of patients had visible hyperplasia of the posterior medial femoral condyle (PMFC) with internal rotation of the posterior condylar axis (PCA) well beyond the routinely anticipated value of 3° with respect to the perpendicular to Whiteside's axis, while another smaller subset of patients had PMFC attrition (from cartilaginous wear) and neutral or mild external rotation of the PCA in relation to the perpendicular to Whiteside's axis. To our knowledge, the morphological changes in the PMFC among varus osteoarthritic knees have not been described in the past.

We hypothesized that chronic ACL deficiency among varus osteoarthritic knees would be associated with not only larger coronal plane deformities, but also an overall increase in the internal rotation of the PCA with respect to the perpendicular to Whiteside's axis, as an extension of the continuum of changes in the knee ensuing from ACL incompetence. We also hypothesized that the PMFC in ACL-deficient knees would demonstrate a greater prevalence of attrition from cartilaginous wear, as well as compensatory hyperplastic changes in response to the posteromedial tibial bone defect, when compared to ACL-competent knees. The aims of this study were to compare ACL-deficient and competent osteoarthritic varus knees during CAS TKA in terms of (a) the preoperative varus deformity, (b) the extent of internal rotation of the PCA with respect to the perpendicular to Whiteside's axis, and (c) the prevalence of PMFC attrition or hyperplasia.

2 Materials and methods

2.2

2.2 Intraoperative assessment of coronal deformity

Following limited surgical exposure, the following anatomic landmarks were registered: the most distal points on the medial and lateral femoral condyles, the posterior-most points on the medial and lateral femoral condyles (to identify the PCA), the anterior-most point of the femoral cortex where the anterior flange of the femoral prosthesis would end proximally (to assess the size of femoral component required), the deepest points of the medial and lateral tibial plateau (to identify the tibial resection level), the centre of the anterior edge of the ACL or the point between the anterior one-third and the posterior two-thirds of the antero-posterior diametral line of the tibial plateau (to determine the tibial centre), and the medial and lateral malleoli (to determine the ankle centre). Kinematic registration of the ipsilateral hip and knee centres, and manual registration of the ankle centre were performed. The system then permitted dynamic goniometry of the knee while specifying the coronal deformity throughout the range of motion based on the relationship between the femoral and tibial mechanical axes. The stressed and corrected varus deformities in full extension were recorded in the form of screenshots from the navigation software interface (Fig. 1).

Close

Fig. 1 Screenshots from the navigation database showing assessment of varus deformity in full knee extension in a patient with a 9° flexion deformity are shown. (A) The stressed varus deformity is shown to be 6°. (B) The corrected varus deformity is shown to be 1°.

Export to PPT

2.3

2.3 Assessment of internal rotation of the PCA, evaluation of the ACL

Whiteside's axis, connecting the lowest point of the trochlea and the apex of the inter-condylar notch, was identified and marked using a sterile marker. It was then registered using the femoral orientation guide. The system now presented the angle between the PCA and the perpendicular to Whiteside's axis as the native femoral rotation (Fig. 2). ACL competence was assessed using a probe, and graded as intact, frayed, linear-striated, or torn.¹⁵ All knees with frayed, linear-striated, or torn ACLs were considered ACL-deficient.¹⁶

Close

Fig. 2 A screenshot from the navigation database showing 8° of native femoral external rotation is depicted. This indicates that the posterior condylar axis lay at 8° of internal rotation with respect to the perpendicular to Whiteside's axis in the patient's native knee.

Export to PPT

2.4

2.4 Evaluation of the posterior medial femoral condyle

Bone resection was performed next, starting with the femur. Distal femoral and proximal tibial resections were performed perpendicular to the mechanical axes of the femur and tibia, respectively, at a level based on the manufacturer's recommendation for the specific implant used. Anterior and posterior femoral resections were based on the indicated rotational alignment, perpendicular to Whiteside's axis. Following posterior femoral condylar resection, the thickness of resected bone from the medial and lateral condyles was measured using calipers.

Given that each millimetre of medio-lateral difference in posterior femoral resection thickness corresponds to a change in femoral rotation of approximately 1°,¹⁷ and that 3° of external rotation from the PCA is generally considered the acceptable position for femoral component placement, 3° ± 2° (1° to 5°) of internal rotation of the PCA from the perpendicular to Whiteside's axis was considered physiological; smaller rotational values in the absence of visible cartilaginous wear were also considered to be physiological as they were unrelated to degenerative attrition. Consequently, for the purposes of this study, PMFC hyperplasia (Fig. 3) was considered to be present when the resected posterior medial femoral fragment was thicker than the lateral fragment by ≥ 6 mm, and PMFC attrition (Fig. 4) was defined as the presence of PMFC cartilaginous wear with the thickness of the resected posterior medial femoral fragment being equal to or less than that of the corresponding lateral fragment.

Close

Fig. 3 Examples of posterior medial femoral condyle (PMFC) hyperplasia. (A) and (B) Anteroposterior and lateral radiographs of a varus osteoarthritic knee showing PMFC hyperplasia. (C) Intra-operative image following arthrotomy, showing PMFC hyperplasia. (D) Intra operative image of another knee, with a 4-way cutting jig in place, showing PMFC hyperplasia. (E) Fragments from the distal femur retrieved following posterior femoral resection. The fragment on the left is thinner and from the lateral femoral condyle, while the fragment on the right is thicker and from the medial femoral condyle.

Export to PPT

Close

Fig. 4 Examples of posterior medial femoral condyle attrition. (A) Intraoperative image, following surgical exposure of the knee. (B) Intraoperative image, following distal femoral and proximal tibial resection.

Export to PPT

The rest of the surgery was performed routinely. Soft-tissue releases were performed where necessary after assessing the extension and flexion gap tension using a gap tensioner (Fig. 5). Once the alignment and gap status were found to be satisfactory, definitive implants were cemented. Femoral notching or patellar maltracking, where present, were noted.

Close

Fig. 5 Intraoperative image depicting the use of a gap tensioner.

Export to PPT

3 Results

Among the 500 patients studied, 322 (64.4 %) were women and 178 (35.6 %) were men. Their mean age was 62.96 ± 6.88 years (range: 50–86 years). The demographic parameters of the ACL-deficient and competent patients are summarized in Table. 1. ACL deficiency was associated with significantly greater preoperative stressed and corrected varus deformities (P < 0.001) (Table. 1). ACL-deficient knees demonstrated significantly greater internal rotation of the PCA in relation to the perpendicular to Whiteside's axis (P < 0.001) compared to ACL-competent knees (Table. 1).

The thickness of the resected posterior medial femoral fragment was equal to or less than that of the corresponding lateral fragment in 13 (5.2 %) and 28 (11.2 %) ACL-competent and deficient knees, respectively. None of these ACL-competent knees, however, showed signs of cartilaginous wear, while PMFC attrition from wear was observed in 21 (8.4 %) of these ACL-deficient knees (Table. 1). PMFC hyperplasia was significantly more prevalent among ACL-deficient knees [OR: 3.52 (95 % CI: 2.12 to 5.83) (P < 0.001); RR: 2.83 (95 % CI: 1.84 to 4.36) (P < 0.001)]. A posteromedial tibial wear pattern was identified in all knees with ACL deficiency, but not those with an intact ACL. No instances of anterior femoral notching or patellar maltracking were noted.

4 Discussion

The primary finding of this study was that ACL-deficient knees were associated with greater stressed as well as corrected varus deformities preoperatively, compared to ACL-competent knees (P < 0.001) (Table. 1). Image-free CAS results in accurate coronal and rotational alignment that correlates with CT-based measurements.^19–21 Given that long-leg radiographs are associated with measurement errors in knees with flexion and/or severe coronal deformities,²² the authors compared the magnitude of the preoperative varus deformity between ACL-deficient and competent knees, and quantified the difference using CAS-based measurements, probably for the first time.

ACL-deficient knees in our study were associated with greater internal rotation of the PCA in relation to the perpendicular to Whiteside's axis (P < 0.001), and a higher prevalence of PMFC hyperplasia (P < 0.001) (Table. 1) (Fig. 3). The authors believe that ACL deficiency and the resulting posteromedial tibial wear pattern result in compensatory changes in the PMFC to resist knee instability and fill the proximal tibial bone defect, respectively. These could be the result of osteophyte formation, manifesting initially as increased internal rotation of the PCA with respect to Whiteside's axis, and eventually progressing to PMFC hyperplasia. The “cupola sign”, described in the literature as the presence of a posteromedial proximal tibial osteophyte as an adaptive response to anterior tibial translation in chronic ACL-deficient knees,^23,24 might therefore be a response to PMFC hyperplasia to contain the bone-osteophyte complex in these knees, which would validate our hypothesis.

The occurrence of PMFC attrition in our study, though seemingly contradictory to the findings of Matsuda et al.¹⁴ that medial femoral hypoplasia does not occur among varus knees, could be the result of several factors. Firstly, PMFC attrition was pre-defined in our study based on the rotation of the PCA and the presence of cartilaginous wear, which helped identify individual knees with this presentation. Secondly, our study included a considerably larger sample size of 500 varus osteoarthritic knees, while Matsuda et al. studied only 30 varus osteoarthritic knees. Thirdly, our study was based on identification of knees with PMFC attrition or hyperplasia, and a comparison of the proportions of knees with either morphologic pattern among ACL-deficient and competent knees, rather than a comparison of mean values of internal rotation of the PCA with respect to Whiteside's axis among the study groups, since use of the latter method alone can result in statistical comparability despite the presence of outliers on both sides of the mean.

The authors believe that progressive morphologic changes in the ACL-deficient varus osteoarthritic knee may manifest either as PMFC attrition from cartilaginous wear, or as increased internal rotation of the PCA in relation to Whiteside's axis, with the subsequent development of PMFC hyperplasia, and the eventual appearance of the cupola sign. The presence of PMFC hyperplasia in a few ACL-competent knees is likely to be physiological in the Asian population studied. The tibial joint line is constitutionally more varus-aligned among Asians when compared to Caucasians.²⁵ This constitutional varus alignment, coupled with the ability to hyper-flex the knee during activities of daily living among the Asian population, could have resulted in a subset of knees developing a posteromedial tibial deformity and compensatory PMFC hyperplasia as a physiologic variation in response to the functional demands, irrespective of the presence of osteoarthritis or ACL deficiency. The significantly higher prevalence of these changes among ACL-deficient knees in the study population, however, confirms the authors' initial hypothesis.

An awareness of the possibility of PMFC hyperplasia or attrition is clinically relevant since it can prevent inadvertent placement of the femoral component in inappropriate rotation during TKA when using the PCA as a reference. The findings of this study reiterate that rotational alignment should be individualized on a case-to-case basis and that multiple references for rotational alignment need to be considered intraoperatively.^26–28 Although Whiteside's axis is generally considered a reliable reference in valgus knees,^29,30 Nagamine et al.¹⁸ considered it unreliable in varus knees. In a CT-based study, they found that the perpendicular to Whiteside's axis was internally rotated to the trans-epicondylar axis by a mean of 2.3° in non-arthritic knees, but that trochlear changes resulted in this angle being merely 0.1° in knees with medial tibio-femoral osteoarthritis. This external rotation of Whiteside's axis with respect to the PCA in osteoarthritic varus knees, however, results in its perpendicular being nearly parallel to the trans-epicondylar axis, which makes it an equally reliable guide for setting femoral component rotation. This is the reason why Whiteside's axis was chosen as the primary rotational reference for hypothesis testing in this study.

Our study has certain limitations. The study population comprised Indian patients only. Inter- and intra-observer variability were not tested. Lateral femoral condyle involvement was not studied, as the posterior lateral femoral condyle was assumed to be normal in varus osteoarthritic knees. Preoperative and postoperative computed tomography scans were not obtained, as CAS was considered to be accurate. Finally, postoperative clinical outcomes were not analyzed since the primary objective of this study was to identify morphological differences in the PMFC between ACL-deficient and competent knees. The greatest strength of this study lies in its presentation of morphologic variations of the PMFC among varus osteoarthritic knees, which have not been described in the past. Although CAS has been reported to be associated with the incidence of errors of judgment related to the registration of the trans-epicondylar axis, kinematic registration and the use of Whiteside's axis as the rotational reference prevented these errors from occurring. The absence of femoral notching or patellar maltracking in any of the study subjects validates our technique.

5 Conclusion

ACL deficiency is associated with progressive osteoarthritis of the knee and an increase in the severity of varus deformity. ACL-deficient osteoarthritic varus knees demonstrate increased internal rotation of the PCA with respect to the perpendicular to Whiteside's axis, which can eventually result in PMFC hyperplasia. This phenomenon must be kept in mind while setting femoral rotation during TKA, to prevent inadvertently placing the component in internal rotation. Although CAS was used in this study, the conclusions drawn are equally applicable during conventional and robot-assisted TKA.

CRediT authorship contribution statement

Krishna Kiran Eachempati: Conceptualization, Methodology, Validation, Formal analysis, Investigation, Resources, Writing – original draft, Writing – review & editing, Supervision. Apurve Parameswaran: Conceptualization, Methodology, Software, Formal analysis, Investigation, Data curation, Writing – original draft, Writing – review & editing. Sunil Apsingi: Conceptualization, Methodology, Formal analysis, Investigation, Resources, Writing – original draft, Writing – review & editing. Chandra Sekhar Dannana: Conceptualization, Methodology, Formal analysis, Investigation, Resources, Writing – original draft, Writing – review & editing. Deepak Gautam: Conceptualization, Methodology, Formal analysis, Investigation, Writing – original draft, Writing – review & editing. Neil P. Sheth: Methodology, Formal analysis, Investigation, Writing – original draft, Writing – review & editing.

Patient consent statement

Informed consent was obtained from all study participants.

Data availability statement

The data that support the findings of this study are available on reasonable request from the corresponding author, AP. The data are not publicly available due to concerns regarding the privacy of research participants.

Ethics approval statement

This study received prior approval from the Institutional Ethics Committee, Medicover Hospitals.

Ethical statement

This study has been carried out in accordance with the Code of Ethics of the World Medical Association (Declaration of Helsinki).

This study received prior approval from the Institutional Ethics Committee, Medicover Hospitals.

Guardian’s/patient’s consent

•Informed consent was obtained from all individual participants included in the study.•Patients signed informed consent regarding publishing their data.

Funding statement

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