Lower Limb Alignment Shifts Toward Varus a Decade After Knee Replacement

Longitudinal Stability of Coronal Alignment and Bone Morphology

Total knee arthroplasty remains the definitive intervention for end-stage knee osteoarthritis, with data indicating higher prosthetic survival rates compared to unicompartmental procedures over follow-up periods of five years or more [1]. While surgeons traditionally target a neutral mechanical axis, the relationship between the knee and the hindfoot is dynamic; for example, correcting a varus deformity (an inward angulation of the distal segment of a bone) can trigger compensatory changes in the hindfoot alignment [2]. Clinical recovery is often limited by persistent functional deficits, as patients may retain significant weakness in the quadriceps and hip abductors for up to three years postoperatively [3]. Although robotic-assisted techniques can achieve a lower hip-knee-ankle angle deviation (mean difference -0.99; 95% confidence interval [-1.24, -0.74]; p < 0.00001) compared to conventional methods, the longitudinal stability of these alignments is not well established [4]. A new study now examines how femoral geometry and limb alignment shift over a ten-year period following primary total knee arthroplasty.

Radiographic Assessment of Long-Term Skeletal Changes

To evaluate the longitudinal stability of bone morphology after surgical intervention, researchers conducted a retrospective study of patients who underwent total knee arthroplasty between 2008 and 2015. The study protocol required each participant to have a complete radiographic record consisting of standing antero-posterior whole-leg radiographs (WLRs) taken at two specific intervals: immediately after the operation and at a 10-year follow-up. From an initial screening of 419 total knee arthroplasties, the researchers identified 177 cases that met the requirement for both sets of imaging. To ensure the purity of the morphological data, the authors applied strict exclusion criteria, removing patients who had other prostheses in the lower limbs, those who underwent additional surgeries during the follow-up period, and those with inappropriate or poor-quality radiographs. After 73 such exclusions, 104 lower limbs remained for final analysis.

The primary outcomes focused on several key radiographic parameters to track structural shifts over the decade. These included the femoral lateral bowing angle (FLBA), which measures the outward curvature of the femoral shaft, and the mechanical lateral distal femoral angle (mLDFA), a measure of the orientation of the distal femur relative to its mechanical axis. The researchers also assessed the tibial lateral bowing angle (TLBA), the mechanical medial proximal tibial angle (mMPTA), the hip-knee-ankle angle (HKAA) to determine overall limb alignment, and the joint line convergence angle (JLCA), which measures the parallelism between the distal femoral and proximal tibial joint surfaces. For these measurements, larger values indicated a progression toward varus deformity (an inward angulation of the distal segment of a bone), with the exception of the mMPTA, where the inverse relationship applies. The researchers utilized paired t-tests to evaluate the statistical significance of changes in these parameters between the immediate postoperative baseline and the 10-year mark, providing a rigorous mathematical framework for observing how the skeletal architecture adapts long after the initial prosthetic implantation.

Significant Progression of Varus Deformity Over Ten Years

The longitudinal analysis revealed that the initial surgical correction of limb alignment does not remain static over the long term. At the 10-year follow-up, the researchers observed a significant shift in the overall mechanical axis of the limb. Specifically, the mean hip-knee-ankle angle (HKAA) shifted by 2.1 ± 2.5° toward varus (P < 0.001). This finding indicates that even when a neutral or desired alignment is achieved during the index total knee arthroplasty, the limb tends to drift toward a varus configuration (an inward angulation) over a decade of weight-bearing and physiological remodeling.

A primary driver of this alignment shift appears to be changes in the femoral shaft itself. The researchers documented that the mean femoral lateral bowing angle (FLBA) increased by 2.0 ± 2.3° toward lateral bowing (P < 0.001). This increase in the outward curvature of the femur suggests that bone morphology is dynamic rather than fixed post-surgery. For the clinician, this means that the femoral geometry used for preoperative planning may not represent the patient's anatomy ten years later, potentially impacting the load distribution across the prosthetic components.

The progression toward varus was further characterized by localized changes at the distal femur and proximal tibia. The mechanical lateral distal femoral angle (mLDFA) showed a varus change of 1.1 ± 2.7° (P < 0.001), reflecting an alteration in the orientation of the distal femur relative to its mechanical axis. Simultaneously, the mechanical medial proximal tibial angle (mMPTA) showed a varus change of -0.6 ± 1.9° (P < 0.001). While the magnitude of the tibial change was smaller than that of the femur, both measurements confirm a systemic trend toward varus deformity across the entire knee joint complex. These data underscore that the skeletal environment of a total knee arthroplasty is subject to significant morphological drift, which may have implications for long-term implant wear and joint stability.

Predictors of Morphological Remodeling

To identify which clinical and radiographic factors immediately following surgery could forecast the eventual progression of femoral bowing, the researchers utilized a multivariate statistical approach. Specifically, postoperative factors related to the 10-year increase in the femoral lateral bowing angle (FLBA) were examined using multiple linear regression (a statistical method used to determine the relationship between several independent variables and a single dependent outcome). This predictive model achieved an R-squared value of 0.30, indicating that the identified variables account for 30% of the variance in long-term femoral remodeling. By identifying these factors, clinicians may better anticipate which patients are at higher risk for significant alignment drift over the decade following their procedure.

The analysis identified four significant independent predictors of the 10-year increase in femoral lateral bowing. The postoperative ipsilateral femoral lateral bowing angle (FLBA) was a significant independent predictor (b = -0.21), suggesting that the initial degree of bowing in the operated limb directly influences its subsequent remodeling trajectory. Furthermore, the contralateral femoral lateral bowing angle (FLBA) also served as a significant independent predictor (b = 0.18). This finding implies that the baseline morphology of the opposite, non-operated limb provides important clinical context for how the surgical limb may change over time, perhaps reflecting systemic or genetic predispositions to specific skeletal configurations.

Global limb alignment and joint geometry also played critical roles in predicting morphological drift. The hip-knee-ankle angle (HKAA) was identified as a significant independent predictor (b = 0.13), highlighting that the overall mechanical axis of the limb post-surgery correlates with future femoral changes. Additionally, the joint line convergence angle (JLCA) was a significant independent predictor (b = 0.23). The joint line convergence angle, which is a measure of the parallelism between the distal femoral and proximal tibial surfaces, reflects the degree of joint space narrowing or ligamentous laxity. For the practicing surgeon, these data suggest that achieving precise postoperative alignment and joint line symmetry is not merely a short-term goal but a factor in mitigating long-term skeletal remodeling.

Clinical Implications for Surgical Planning

The longitudinal data from this cohort of 104 lower limbs demonstrate that the skeletal structure of the leg is not a static framework following prosthetic implantation. The study concluded that both bone morphology and coronal alignment can change after total knee arthroplasty, challenging the traditional surgical assumption that postoperative geometry remains fixed throughout the lifespan of the implant. Specifically, the researchers observed a significant mean increase in the femoral lateral bowing angle (the degree of outward curvature in the femur) of 2.0 ± 2.3 degrees (p < 0.001) over the 10-year follow-up period. This remodeling of the femoral shaft, combined with a 1.1 ± 2.7 degree increase in the mechanical lateral distal femoral angle (p < 0.001) and a 0.6 ± 1.9 degree decrease in the mechanical medial proximal tibial angle (p < 0.001), resulted in a cumulative shift of the hip-knee-ankle angle (the global mechanical axis of the limb) by 2.1 ± 2.5 degrees toward varus (p < 0.001).

For the practicing orthopedic surgeon, these findings suggest that initial alignment strategies may need to account for a predictable decade-long drift toward a varus configuration. Because the hip-knee-ankle angle and the joint line convergence angle (a measure of the parallelism between the femoral and tibial joint surfaces) were identified as significant independent predictors of femoral remodeling, achieving precise initial balance is critical. The regression model, which yielded an R-squared value of 0.30, indicates that nearly one-third of the variance in long-term femoral bowing can be anticipated by evaluating the postoperative state of both the surgical and contralateral limbs. Consequently, clinicians should consider the necessity of long-term radiographic monitoring to detect significant morphological shifts. Understanding that the limb may naturally migrate toward varus over time allows for more personalized surgical planning and may help in mitigating long-term prosthetic wear or functional decline.

References

1. Hu H, Li P, Liu Z, Lv H, Yang X, Liu P. Comparative long-term outcomes of unicompartmental and total knee arthroplasty in knee osteoarthritis patients: a systematic review and meta-analysis.. Frontiers in surgery. 2024. doi:10.3389/fsurg.2024.1405025

2. Butler JJ, Mercer NP, Hurley ET, Azam MT, Kennedy JG. Alignment of the hindfoot following total knee arthroplasty: A systematic review.. World journal of orthopedics. 2021. doi:10.5312/wjo.v12.i10.791

3. Daher M, Estephan M, Ghoul A, Tarchichi J, Mansour J. Hip Strengthening After Total Knee Arthroplasty: A Meta-analysis and Systematic Review.. The archives of bone and joint surgery. 2024. doi:10.22038/ABJS.2024.76202.3520

4. Yue H, Ding G, Li H, et al. Does Robotic-Assisted Total Knee Arthroplasty Improve Outcomes of Adult Osteoarthritis Patients-A Systematic Review and Trial Sequential Meta-Analysis.. Orthopaedic surgery. 2025. doi:10.1111/os.70007