Patellar thickness can critically impact knee joint biomechanics after total knee arthroplasty

Abstract

Aims

Despite the potential benefits of an increased moment arm of the knee extensor mechanism with a thicker patellar button in patients undergoing total knee arthroplasty (TKA), many surgeons are reluctant to alter the natural patellar thickness (PTh) during resurfacing. Clinically, this hesitation stems from concerns about straining the extensor apparatus under passive motion exercises and the potential consequences on postoperative joint biomechanics, but also because many patients with a thinner patella experience improved functionality, range of motion, and possibly also reduced soft-tissue pain in the short term postoperatively. However, the biomechanical relationships underlying such short- and long-term consequences remain unexplored. Our aim was to investigate the influence of PTh on knee biomechanics after TKA.

Methods

This study used advanced computational modelling techniques to undertake a systematic exploration of the influence of PTh variation on knee joint kinematics, contact mechanics, and muscle and soft-tissue loading patterns during level walking and squatting.

Results

Our results indicate that increased PTh efficiency enhances the extensor mechanism, which is consistent with previous studies, but our investigation additionally demonstrates the reduction in tibiofemoral loading conditions, especially at higher flexion angles. Conversely, a thinner patella induces reduced patellofemoral contact forces and pressure, and may offer increased joint range of motion.

Conclusion

PTh was found to influence knee biomechanics in clinically relevant ways. A modest increase in thickness improved quadriceps efficiency and reduced tibiofemoral forces by up to ~10%, but also raised patellofemoral pressures by up to ~5 MPa. Conversely, thinner buttons lowered patellofemoral loading and may facilitate early comfort and flexion. These trade-offs were consistent across implant congruency designs, highlighting the need for patient-specific resurfacing strategies.

Cite this article: Bone Joint Res 2026;15(5):472–481.

Article focus

• This study investigates the biomechanical impact of patellar thickness (PTh) variation following total knee arthroplasty (TKA).

• Using advanced in silico modelling, it systematically examines how changes in PTh influence joint loading patterns and kinematics during functional activities.

• The aim is to elucidate the underlying biomechanical mechanisms that can inform clinical decision-making in patellar resurfacing, particularly weighing the trade-offs between preserving native patellar anatomy and optimizing postoperative joint performance.

Key messages

• The findings suggest that a thinner patella may lead to improved short-term outcomes, such as reduced patellofemoral contact forces and increased joint range of motion, which could contribute to early patient satisfaction.

• In contrast, increasing PTh appears to enhance extensor mechanism efficiency and reduce tibiofemoral loading, effects that may support implant longevity.

• These results indicate that PTh should not be viewed solely as a surgical preference, but rather as a biomechanically meaningful factor with implications for both immediate function and long-term clinical outcomes.

Strengths and limitations

• A key strength of this study lies in the integration of state-of-the-art in vivo biomechanical datasets with validated computational models to simulate knee joint behaviour under physiologically realistic loading conditions.

• This approach allows for the isolated investigation of PTh effects, independent of confounding surgical or anatomical variables, thereby offering a clearer understanding of cause-effect relationships in knee biomechanics.

• However, the generalizability of the results may be limited for patients with atypical anatomical or physiological characteristics, or during high-demand functional activities not represented in the current simulation framework.

Introduction

Total knee arthroplasty (TKA) is a safe and effective surgical intervention for alleviating degenerative joint pain, towards resuming patients’ normal activities with promising long-term results over several decades.¹ However, high patient dissatisfaction rates are still reported, varying from 10% to 30%.^2,3 Among the diverse factors contributing to dissatisfaction, a predominant concern is anterior knee pain (AKP), with a reported incidence of 5% to 10%.^4-6 Interestingly, even many satisfied patients also report AKP, suggesting that up to 47% of all TKA patients experience some level of pain postoperatively.⁷

Patellar resurfacing is a cost-effective procedure that substantially reduces AKP by 4% to 17%,⁷ and lowers the revision rates.^8,9 However, an increased risk of patellar fracture, dislocation, patellar tendon injury, and patellar implant failure has been reported when the procedure is not properly executed.^8,10-13 Patellar thickness (PTh) is one of the most important surgical parameters to be considered during patellar resurfacing. However, the existing literature regarding outcomes of TKA with different PThs remains controversial.¹⁴ On one hand, a thicker patella has been suggested to have a positive effect on efficiency of the knee extensor mechanism through increasing the quadriceps moment arm.^15-17 On the other hand, a thicker patella or patella overstuffing is reported to increase patellofemoral (PF) contact pressure,^18,19 restrict the range of motion,²⁰ add strain to the extensor apparatus under passive motion exercises, and increase PF shear forces,^21,22 possibly contributing to AKP, patellar maltracking, and implant loosening.^23-25 As a result, some recent investigations advocated restoring the native PTh and subsequently the anterior offset to avoid consequences on patellar joint biomechanics.^16,26,27 Given that TKA subjects often exhibit substantially weaker quadriceps compared to their healthy, age-matched peers,^28-30 patellar thickening can become unavoidable in certain instances. Consequently, it is crucial to thoroughly comprehend the consequences of intraoperative adjustments to PTh on postoperative knee loading and kinematics, and specifically provide insights into any compromise between short-term functional outcome and long-term implant survival.

Biomechanical investigations to assess the influence of PTh on knee mechanics have been mainly focused on specific joint outcomes during a limited number of activities and do not provide a comprehensive understanding of changes to knee biomechanics due to PTh variation. Here, intraoperative measurement of the knee range of motion during passive knee flexion²⁰ and in vitro measurement of the TF (tibiofemoral) and PF kinematics in cadaveric knees during unloaded flexion indicate a positive correlation between PTh and internal tibial rotation,³¹ but a negative correlation between PTh and range of knee flexion.¹⁶ Moreover, cadaveric investigations have revealed increased shear and compressive PF contact forces,^19,21,22 along with a reduced contact area, resulting in elevated contact pressures for patellar resurfacing with thicker buttons.¹⁶ However, only a few studies have investigated functional activities with different PThs. Here, Parke et al³² performed gait analysis in 15 subjects pre- and post-TKA and found a positive correlation between knee extensor strength and PTh during walking. However, due to limitations in skin-marker motion capture techniques, such measurements could not capture PTh impacts on other degrees of freedom (DoFs) of the knee kinematics. Moreover, a clear understanding of the influence of PTh on TF and PF joint loading, as well as changes to the surrounding soft-tissue strains, remains limited by technical challenges associated with in vivo measurement of the internal loading conditions within the human body.³³

Recent computational modelling techniques enable a comprehensive exploration of the knee joint with 12 DoFs as well as muscle forces and soft-tissue loading.³⁴ Using the COMAK algorithm that allows concurrent optimization of the muscle and ligament forces as well as articular contact mechanics, knee biomechanical parameters can be systematically investigated for different TKA surgical scenarios.^34,35 This facilitates the exploration of various surgical options and intraoperative parameters on postoperative knee kinematics and internal loading conditions during functional activities. The capacity of this in silico modelling framework for predication of in vivo knee joint mechanics has been extensively investigated in previous studies indicating reliable estimation of joint parameters for level walking and squatting.³⁶

This study aims to utilize this previously developed and validated modelling framework to investigate knee joint biomechanics under different patellar resurfacing scenarios. Specifically, we explored the impact of PTh on the 12 DoF kinematics, articular contact forces of the TF and PF joints, muscle forces, and soft-tissue loading during walking and squatting.

Methods

A previously validated musculoskeletal model³⁶ of a TKA patient (65 years, height 1.74 m, mass 95.6 kg, with an ultra-congruent Innex FIXUC implant; Zimmer Biomet, Switzerland) measured for the CAMS-Knee project³⁷ was used in this study (Figure 1). It should be noted that, while the INNEX FIXUC femoral component is one of the few designs supported by comprehensive in vivo biomechanical datasets, enabling realistic boundary conditions and robust model validation, it may differ slightly from newer implants. Specifically, although it is sided with a lateralized trochlear groove, it lacks certain refinements found in modern designs, such as thinner anterior flanges and extended trochlear geometry, that enhance patellofemoral tracking.

Fig. 1.

a) Subject-specific musculoskeletal model, where muscles are represented as red fibres. b) Passive soft-tissue surrounding the knee. c) and d) Implantation shown in the coronal and sagittal planes, reconstructing a mechanical limb alignment (1° varus) with 7° posterior tibial slope. e) and f) Geometrical difference between the INNEX (shown in light blue) and low congruency (shown in purple) inlays.

The lower-extremity model was constructed based on patient-specific medical image data with detailed representation of the knee structures (including soft tissues and articular contact) enabling six DoF kinematics and loading for each of the tibiofemoral and patellofemoral joints. Implant positioning and alignment were extracted from postoperative CT scans. The TKA alignment represented a mechanically aligned leg with 1° varus. Femoral component flexion relative to the anatomical axis was 4.5°, and internal rotation relative to the surgical transepicondylar axis was 0.6° (Figures 1c and 1d), with additional details of the femoral implant geometry provided in Supplementary Figure a. Tibial posterior slope was 7°, and the overall PTh (anterior to posterior) was 22.5 mm. The full details of this model have been previously reported.³⁶

Four trials of two activities, level walking and squatting, were simulated on this baseline model using the OpenSim-COMAK algorithm.^34,35 The algorithm used skin marker trajectories and ground reaction forces as inputs to solve the muscle redundancy problem in the presence of ligament and contact forces. In this study, the TF joint flexion/extension angle (F-E) was constrained to match the kinematics measured via fluoroscopy, but abduction/adduction (A-A), internal/external rotation (I-E), and the three tibiofemoral joint translations, namely anterior/posterior (A-P), proximal/distal (P-D), and lateral/medial (L-M), were estimated within the COMAK optimization at each timeframe. In addition to the muscle activations/forces, the total tibiofemoral contact force (TF-CF_Total) as well as the contact forces acting on each condyle (i.e. TF-CF_Lateral, TF-CF_Medial) were estimated. For the PF joint, the six DoF kinematics, including patellar flexion/extension (F-E), tilt, spin, and joint translations (anterior/posterior (A-P), proximal/distal (P-D), and lateral/medial (L-M)), were estimated using the COMAK optimization algorithm. Here, the PF joint contact force was also calculated as part of the optimization process. As standard outputs from the COMAK modelling algorithm, the trajectory of the centres of pressure (CoP) on the PF and TF joints were also determined together with their respective contact areas and mean contact pressure (MCP).

The baseline INNEX inlay features a high conformity design, known to produce highly constrained conditions within the TF joint, hence limiting our understanding of biomechanical changes that occur due to altered joint mechanics. We therefore additionally modelled a reduced conformity inlay design (contact radius arc altered from the original 36 mm to 67 mm). In both the baseline and the low-conformity inlay models, the patellar button thickness was then systematically varied by -6/+6 mm around the baseline PTh of 22.5 mm (16.5 mm to 28.5 mm, thickness measured from the most anterior to the most posterior point on the patella) in 2 mm increments to provide an understanding of the biomechanical changes that propagate through the joint after possible surgical patellar resurfacing scenarios, as well as to identify kinematic and kinetic parameters most sensitive to PTh alteration. For all simulations, the slack length of the knee ligaments was adjusted to ensure an equal residual strain within the soft tissues at full extension of the knee, thereby mimicking a consistent intraoperative ligament balancing procedure. Since the COMAK algorithm produces force-dependent kinematics and kinetics equilibrium, it was able to provide secondary modifications to joint translations and rotations to re-balance the joint boundary conditions after introducing surgical or geometrical alterations. As such, the analyses were able to iteratively balance the joint mechanics and thereby produce predictive biomechanical outcomes based on the modified inlay and PTh scenarios.

Finally, to analyze any possible relationships between PTh and knee joint mechanics, Spearman correlation coefficients were calculated between PTh and the peaks and ranges of different kinematic and kinetic parameters.

Results

For brevity, figures detailing the biomechanical effects of PTh are provided only for level walking directly within the manuscript, but for details on the squatting activity, please refer to the additional supplementary material.

Patellofemoral joint kinematics

During level walking, slight variations in patellar tilt and spin were observed with changes in PTh, with angles showing an alteration of approximately 2° in tilt and 1° in spin due to a 12 mm perturbation in PTh (Figure 2).

Fig. 2.

Patellofemoral (PF) kinematics and kinetics during level walking with different patellar button thickness (PTh) and congruency levels (violet shading shows outcomes resulting from ± 6 mm changes in PTh).

During squatting, a thicker patella button resulted in up to 11.4 mm inferior translation of the patella relative to the femur depending on the knee flexion angle (Supplementary Figure e). However, the changes in patellar tilt and spin resulting from PTh variation were minimal (Supplementary Figure b).

For the low congruency inlay, the influence of PTh on PF joint kinematics was generally consistent with that observed for the INNEX inlay, although the low congruency inlay exhibited slightly smaller ranges of variation in PF kinematics (1.4° vs. 2° variation in PF tilt for the low congruency vs INNEX inlay during level walking, Figure 2; 1.2° vs 1.1° variation in PF tilt for the low congruency vs INNEX inlay during squatting, Supplementary Figure b).

Patellofemoral joint contact mechanics

The impact of PTh on PF contact forces was moderate, with a variation of up to +8% body weight (BW) and -10% BW for the thickest compared to the thinnest PTh for level walking and squatting, respectively (Figure 2, Supplementary Figure b). However, the PF contact forces themselves were generally low for walking (< 0.6 BW), while much higher forces of > 3.5 BW were observed during squatting.

During level walking, an increase in PTh resulted in a larger PF contact area and a higher MCP (2.96 MPa increase in MCP for the thickest compared to the thinnest PTh, Figure 2). In contrast, during squatting, an increased PTh led to a smaller PF contact area and a considerably higher MCP (5.6 MPa increase in MCP for the thickest compared to the thinnest PTh, Supplementary Figure b), with this effect becoming more pronounced at deeper knee flexion angles. During both activities, patellar resurfacing scenarios with thicker patellae showed only a very subtle lateral shift in the CoP trajectory on the femoral contact surface (Figure 3, Supplementary Figure c).

Fig. 3.

Center of pressure (CoP) traces on the femoral and tibial contact during level walking with different inlay congruency and ± 6 mm changes in patellar button thickness (PTh).

Simulations with the low-congruency implant showed a very similar influence of PTh variation on PF contact area, pressure, and CoP trajectory on the femoral component, comparable to those observed in simulations with the highly congruent INNEX inlay (Figure 2 and Figure 3, Supplementary Figure b and c).

Tibiofemoral joint kinematics

For both level walking and squatting simulations, a ±6 mm modification in PTh was associated with very minimal variations in TF translations and rotations (below 1 mm and 1°; Figure 4, Supplementary Figure d).

Fig. 4.

Tibiofemoral kinematics, kinetics, and quadriceps muscle forces during level walking with different patellar button thickness (PTh) and congruency levels (violet shading shows outcomes resulting from ± 6 mm changes in PTh).

With the baseline PTh, the TF joint experienced notably greater posterior translations (9.5 mm vs 7.3 mm) and internal tibial rotations (8.5° vs 1.4°) with the low-congruency inlay compared to the INNEX inlay during level walking. Additionally, the influence of PTh variation was more pronounced on the range of TF anteroposterior translation and axial rotation with the low-congruency inlay than with the INNEX inlay (Figure 4, Supplementary Figure d). These findings were consistent across both level walking and squatting.

Tibiofemoral joint contact mechanics

During level walking, PTh modification resulted in minimal changes in the TF axial contact force (up to 0.05 BW). Peak forces of 1.36 BW and 0.82 BW were observed on the medial and lateral condyles, respectively. In contrast, during squatting, TF-CF_Total of up to 2.75 BW were observed, with higher forces observed on the lateral (TF-CF_Lateral up to 1.91 BW) compared to the medial (TF-CF_Medial up to 0.87 BW) condyle. Both the medial and lateral condyles were significantly affected by PTh variation, where the thickest PTh led to a 0.26 BW reduction (approximately 30% of the peak, r = −1.0) on the medial side and a 0.36 BW reduction (approximately 19% of the peak, r = −1.0) on the lateral condyle (Supplementary Figure e).

During level walking, PTh variation had a minor influence on the TF-CoP excursion on the tibial inlay (Figure 3) and likewise showed only a small impact on the TF contact area and contact pressure (Figure 4, Supplementary Figure e). During squatting, a modified PTh led to only a slight A-P shift in TF-CoP (Supplementary Figure c). Additionally, the thickest compared to thinnest PTh resulted in a 0.22 cm² (15%) reduction in total TF contact area, with a 0.05 cm² (6%) reduction on the lateral side and a 0.17 cm² (22%) reduction on the medial side. This PTh increase also caused a decrease of approximately 2.72 MPa in TF-MCP (Supplementary Figure d, e).

Despite the lower contact area and higher MCP associated with the reduced congruency inlay, the influence of PTh on TF joint contact mechanics remained generally consistent between the two implant congruencies. During squatting, a thicker patella induced greater posterior femoral translation on the tibia than a thinner patella (Supplementary Figure d). However, CoP trajectories differed slightly between the two inlay congruencies, with the CoP trace for both activities remaining closer to the inlay centre for the low congruency inlay (Figure 3, Supplementary Figure c).

Muscle and ligament forces

A thicker PTh resulted in up to 0.1 BW decrease in PT and quadriceps peak forces during level walking (Figure 4, Supplementary Figure e). During squatting, however, the reduction in peak PT and quadriceps forces due to a greater PTh were more highlighted (0.55 BW and 0.3 BW; Supplementary Figure d and e). Despite the considerable increase in the reference length of MPFL and LPFL (17% and 20% for the thickest compared to the thinnest PTh), no major changes were observed in the knee ligament forces due to PTh modification for either level walking or squatting. In general, the impact of PTh on muscle and ligament forces was generally consistent between the highly congruent INNEX and the low congruency tibial inlay (Figure 4, Supplementary Figures d and e).

Discussion

In the current study, we used a comprehensive musculoskeletal modelling framework to investigate the influence of PTh modifications on knee joint biomechanics, focusing on its potential impact on both PF and TF joint kinematics as well as forces exerted on the implant, soft tissues, and muscles. While confirming that thickening the patella improves quadriceps efficiency, and hence reduces TF-CFs, our results show that an increased PTh influences both PF and TF biomechanics, with effects that were often more pronounced at higher knee flexion angles. Moreover, a thicker patella was shown to elevate the patellofemoral contact pressure, which was more prominent during squatting. Although greatly increased TF-MCP and TF axial rotation were observed, the changes resulting from a modified PTh were generally consistent between the highly congruent INNEX and the low congruency implant tested.

The patella primarily functions as a pulley-like system for the quadriceps muscle, facilitating knee extension. In theory, a thicker patella is equivalent to increasing the radius of this pulley, enhancing its mechanical efficiency. Thus, for patients undergoing TKA who suffer from weak quadriceps, opting for a thicker patella rather than restoring the anterior offset during patellar resurfacing may offer considerable biomechanical advantages. However, previous studies have mostly recommended restoring the thickness of the native patella for anterior offset restoration, even though no serious concerns have been demonstrated for moderate patella overstuffing.³² In such configurations, our results confirm up to a 10% reduction in the quadriceps force needed to perform a squat activity (Supplementary Figure d), mainly achieved through the anterior shift of the quadriceps tendon attachment and partially due to the slight posterior shift of the tibiofemoral contact points (Supplementary Figure d), both of which increase the lever arm of the knee extensor mechanism and hence its efficiency in producing joint moments. With lower muscle forces required to perform the activity, the tibiofemoral joint experiences lower contact forces (Supplementary Figure d), potentially providing a more beneficial biomechanical situation with long-term benefits for component longevity.³⁸ Unexpectedly, however, with increasing PTh, the patellofemoral compressive force during walking showed a slight increase (Figure 2). This aligns with the findings of Brivio et al²² and Hsu et al,¹⁶ who reported that a thinner patella is associated with reduced PF compressive forces. PF compressive force is generated to balance the quadriceps and patellar tendon forces. As a result, while a thicker patella reduces the magnitudes of the quadriceps and patellar tendon forces, the decrease in angle between these forces tends to raise the PF compressive force. This plausibly explains why PF compressive force during squat remains almost unchanged, even though larger PThs led to a considerable reduction in the quadriceps and patellar tendon forces (Supplementary Figures b, d and e).

Post-TKA, the stability of the patellofemoral joint primarily depends on the implant design and to a lesser extent on the surrounding soft tissues.^39,40 While the MPFL and LPFL may not be explicitly removed in all cases, their natural function is usually compromised due to the nature of TKA surgery, resulting in subsequent changes in joint mechanics. The ligament loading patterns observed in the current study generally concur with patterns reported for the healthy knee as well as those reported for the replaced knees.^16,41-45 While as expected, the reference length of the MPFL and LPFL was directly affected by PTh variations, these length-changes were consistent over the range of knee flexion. Assuming appropriate intraoperative ligament balancing, these variations would therefore not induce major changes in the ligament strain/force patterns, as is confirmed by findings of a previous cadaveric study.⁴⁶ Additionally, the minor changes in tibiofemoral kinematics caused by PTh variations suggest that the knee’s centre of rotation, as well as the strain patterns of the LCL, MCL, and ITB, remain largely unaffected. This situation was similar even for the low-congruency inlay (Supplementary Figure e). Consequently, aside from the patellar tendon, the loading patterns of knee soft tissues would be minimally impacted by moderate over- or under-stuffing of the patella.

In addition to the alterations observed in TF and PF joint kinetics, we found that the knee joint kinematics were also affected by changes in the PTh, with a larger impact on the patellofemoral joint. Given a constant patellar tendon length, an increased PTh leads to a discernible distal positioning of the patella relative to the femoral component (Supplementary Figure e). This slight distal positioning of the thicker patella results in altered interface conditions as the patella moves further down the trochlear groove, leading to smaller contact areas (Supplementary Figure b). Since the PF joint is subject to high contact pressures during deep knee bend type activities, which already approach the recognized yield strength of PE materials (ca.35 MPa), substantially increased wear in the PF joint can plausibly be expected with a thicker patella, as indicated previously in the literature.¹⁶ An increase in PTh was also associated with a slight posterior, and lateral shift of the patellofemoral CoP, observed particularly at larger flexion angles (Supplementary Figure c). Altered joint kinematics, as well as variations in the contact area and pressure, may adversely affect patellar button polyethylene wear⁴⁷ and thus negatively impact the longevity of the implant. It therefore seems plausible that a thicker PTh provides clear biomechanical benefits for the TF joint, but possibly at the expense of the PF joint (potentially increased risk of fracture, and increased loading conditions with long-term consequences).

Our results show that a thinner patella reduces the PF contact pressure (Figure 2, Supplementary Figure b). Considering the increase in TF range of motion suggested in the literature,²⁰ this variation possibly improves the short-term success of TKA surgery based on improved joint functionality and patient-reported outcome measures. However, the higher muscle forces potentially expose a thinner patella to a higher risk of creep and component material fatigue and fracture (although these risks are somewhat mitigated in modern ageing-resistant materials such as Vitamin E-stabilized polyethylenes). Weighed against the longer-term benefits of a thicker patella, including greater extensor mechanism efficiency and reduced tibial inlay wear, surgeons should therefore consider the subject-specific needs of the patient,^48,49 including their age, physical aspirations, muscular sufficiency, and long-term requirements, when deciding on button thickness, particularly in patellar resurfacing scenarios involving substantial deviations from the natural PTh. Such choices clearly present a challenge in clinical decision-making as surgeons will often naturally tend towards the perceived short-term success of an implant, including improved joint ranges of motion and patient-reported outcome measures. For younger patients, however, longevity of the TKA is possibly of more consequence than short-term gains, and may require surgeons to exercise restraint and manage the patient’s expectations by discussing possible outcome scenarios (including short- and long-term benefits, motivation, and rehabilitation times) with patients preoperatively.

There are a few limitations of the current study. Firstly, while we tried to investigate the interaction between PTh, implant congruency, and joint mechanics by simulating both high and low congruency inlay scenarios, generalization of our findings to other implant designs with specific features such as medial-stabilized implants should be exercised with caution. Secondly, newer implant designs may differ in trochlear groove topology compared to the INNEX FIXUC investigated in this study, which may influence implant-specific patellofemoral kinematics by altering sagittal-plane contact location and frontal-plane tracking. However, while the absolute magnitude of the results reported here may differ slightly across implant geometries, within each design, the biomechanical effects of PTh are primarily driven by the anterior and distal shift of the patella in the sagittal plane and the resulting changes in muscle moment arms. Therefore, the fundamental biomechanical impact of altered PTh on knee biomechanics is expected to remain consistent across designs. As another limitation, our modelling framework did not allow direct estimation of changes in tibiofemoral flexion range, as the knee flexion angle was prescribed from fluoroscopic recordings in all simulations. Nevertheless, our results demonstrated reduced patellofemoral contact pressures with a thinner patella, which aligns with previous in vivo and in vitro studies reporting improved postoperative flexion range following patellar downsizing.^16,19,25 Finally, we did not study the influence of PTh on stress distributions within the patellar bone and therefore our investigation does not provide an insight into the impact of PTh on the risk of patellar fracture. In addition, lack of in vivo biomechanical data collected from TKA subjects with different PThs precludes any direct validation of our results. However, even if such data were available, the numerous confounding factors (e.g. intersubject anatomical variability) would diminish their usefulness for our validation purposes. To mitigate these limitations, we focused on intra-study comparison of outcomes across multiple PTh scenarios that allowed a clear understanding of the isolated impact of PTh on post-TKA joint biomechanics.

In conclusion, PTh is a clinically relevant surgical parameter. In our simulations, increasing PTh reduced quadriceps and tibiofemoral contact forces (by ~10% during squat) but raised patellofemoral pressures (by ~3 MPa). Conversely, reducing PTh increased quadriceps loading (by ~15% during squat) but lowered patellofemoral pressures (by ~2.5 MPa). These trade-offs were consistent across inlay designs and highlight the need for patient-specific resurfacing decisions. Clinically, a modestly thicker button may benefit patients with weak quadriceps or those prioritizing long-term implant survival, whereas restoring or slightly reducing thickness may be preferable for patients seeking improved early comfort and relief of AKP. Surgeons should individualize patellar resurfacing decisions according to patient age, muscle strength, activity goals, and long-term expectations.

Data Availability

The data for this study are publicly available at https://cams-knee.orthoload.com/data/data-request/.