Feasibility and reliability of shear wave elastography-guided collateral ligament balancing during total knee arthroplasty Authors: Qi Wang, Xiangdong Wu, Lei Li, Chengshuai Zhang, Hao Tang, Yixin Zhou Published: 2025 ABSTRACT AimsTo assess the feasibility and reliability of shear wave elastography (SWE) as an intraoperative tool for real-time monitoring and adjustment of collateral ligament tension during total knee arthroplasty (TKA).MethodsA total of 33 porcine knee specimens were divided into stretching (n = 13) and minimally invasive needle puncture (MNP) groups (n = 20). The stretching experiment evaluated correlations between distraction forces (20 to 60 N) and shear wave speed (SWS) in medial (MCL) and lateral collateral ligaments (LCL). The MNP experiment assessed SWS and medial gap changes during 45 punctures at 90° flexion or full extension under 45 N force. Pearson correlation, generalized linear mixed models (GLMMs), and two-way analysis of variance analyzed relationships and effects. Intra- and interobserver reliability were calculated using intraclass correlation coefficients (ICCs).ResultsIn the stretching experiment, SWS increased linearly with distraction force (r = 0.67 to 0.80, p < 0.001), unaffected by ligament thickness or structure. In the MNP experiment, SWS decreased with puncture count (r = 0.73, p < 0.001), detecting tension changes earlier than medial gap measurements (5 vs 20 punctures in flexion; 10 vs 20 in extension). GLMMs confirmed significant effects of force and puncture count on SWS (p < 0.001). Reliability was excellent (ICC > 0.88).ConclusionSWE is a reliable, noninvasive tool for monitoring ligament tension changes during TKA, offering greater sensitivity than gap measurements. It shows great potential for optimizing knee ligament tension, improving soft-tissue balancing, and enhancing TKA outcomes.Cite this article: Bone Joint Res 2025;14(12):1167–1175. Article focus This study evaluates shear wave elastography (SWE) as an intraoperative tool for monitoring collateral ligament tension during total knee arthroplasty (TKA), assessing its sensitivity in detecting tension changes in porcine models under distraction forces and during minimally invasive needle puncture (MNP), compared to medial gap measurements, for optimal soft-tissue balance. This study evaluates shear wave elastography (SWE) as an intraoperative tool for monitoring collateral ligament tension during total knee arthroplasty (TKA), assessing its sensitivity in detecting tension changes in porcine models under distraction forces and during minimally invasive needle puncture (MNP), compared to medial gap measurements, for optimal soft-tissue balance. Key messages SWE shows a strong linear correlation between distraction force and shear wave speed (SWS) for precise tension monitoring in TKA. SWE shows a strong linear correlation between distraction force and shear wave speed (SWS) for precise tension monitoring in TKA. It also detects tension changes earlier than medial gap measurements during MNP, identifying stabilization trends sooner and reducing the risk of over-release. It also detects tension changes earlier than medial gap measurements during MNP, identifying stabilization trends sooner and reducing the risk of over-release. High intra- and interobserver reliability (intraclass correlation coefficient > 0.88) supports SWE’s consistency for clinical ligament assessment. High intra- and interobserver reliability (intraclass correlation coefficient > 0.88) supports SWE’s consistency for clinical ligament assessment. Strengths and limitations A rigorous experimental design, innovative application of experimental tools, and comprehensive statistical analysis ensure reliable and robust results. A rigorous experimental design, innovative application of experimental tools, and comprehensive statistical analysis ensure reliable and robust results. Porcine models may not fully replicate human TKA, ligament viscoelasticity and shear wave dispersion may affect SWS accuracy, and probe instability necessitates improved bedside fixation for clinical use. Porcine models may not fully replicate human TKA, ligament viscoelasticity and shear wave dispersion may affect SWS accuracy, and probe instability necessitates improved bedside fixation for clinical use. Introduction Noninvasive assessment of ligament tension is crucial for improving our understanding of joint biomechanics in various musculoskeletal conditions.1,2 Accurate evaluation of the passive tension in different ligaments, such as the medial (MCL) and lateral collateral ligament (LCL), is essential for optimizing joint stability and function, particularly during surgical interventions such as total knee arthroplasty (TKA), a widely adopted treatment for advanced knee osteoarthritis (KOA).3,4 Patients with advanced KOA often exhibit complex anatomical changes, including osteophyte formation, soft-tissue contracture, and joint capsule adhesion, which complicate intraoperative ligament tension management considerably during TKA.5,6 Traditional methods for managing ligament tension—digital palpation, gap measurement with spacer blocks, and trial component testing—rely heavily on the surgeon’s subjective experience to guide soft-tissue release techniques like the Knife Pie-Crusting technique or minimally invasive needle puncture (MNP), with the goal of achieving a symmetrical, rectangular knee joint gap to ensure proper alignment and balance.7,8 However, these approaches lack objective assessment and are particularly challenging in patients with severe KOA because of soft-tissue abnormalities. Recent technological advances, including augmented reality-based ligament tracking, robotic-assisted TKA systems, and sensor-based technologies, have improved upon traditional methods by providing more precise measurements.9-11 However, these innovations indirectly assess ligament tension through gap size or pressure and thus fail to directly quantify internal tension changes or provide real-time visualization to guide surgeons during balancing, potentially contributing to postoperative dissatisfaction despite allowing the formation of symmetrical joint gaps.12,13 Changes in ligament tension closely reflect alterations in collagen fibre arrangement, which can be measured by shear wave elastography (SWE), a noninvasive ultrasound technique that assesses tissue stiffness through a quantity called the shear wave speed (SWS).14 SWE is increasingly used in musculoskeletal biomechanics, revealing a strong correlation between SWS and axial stress in tendons and ligaments, including the collateral ligaments of the knee, under controlled laboratory conditions.15,16 However, these findings may not directly translate to clinical settings because of the influence of the surrounding tissues and the properties of the ligaments themselves.17 Therefore, this study aimed to evaluate SWE as an intraoperative tool for assessing and adjusting passive tension in the collateral ligaments of the knee. Establishing the reliability of SWE in this clinical context is essential before optimal SWS ranges for achieving balanced soft-tissue tension can be determined. We hypothesized that SWE can reliably predict dynamic changes in collateral ligament tension, even in the presence of surrounding soft tissues and anatomical differences between the MCL and LCL, and that combining SWS with medial gap measurements would effectively reflect changes in the tension of the MCL during MNP, with SWE showing a sensitivity equal to or greater than that of medial gap assessments. Methods We have adhered to the ARRIVE guidelines and have supplied the ARRIVE checklist in the Supplementary Material. Specimen preparation and grouping A total of 33 male porcine knee joint specimens from five- to six-month-old Landrace pigs were acquired for this study. The sample size was determined through power analysis to ensure statistical robustness. TKAs were performed by an experienced orthopaedic surgeon (XW) using a posterior-stabilized knee system (A3 Total Knee System; AK Medical, China). Cruciate ligaments were resected, the distal femur cut at a 5° valgus angle to the anatomical axis, and the proximal tibia resected 9 mm below the medial plateau’s highest point; the joint capsule and surrounding soft- tissues were preserved to maintain anatomical integrity. Specimens were inspected for abnormalities and randomly assigned to two experimental groups by drawing lots: Stretching experiment (n = 13): to investigate the relationship between the incremental knee joint distraction force and SWS changes in both the MCL and LCL. MNP experiment (n = 20): to evaluate SWS changes in the MCL during minimally invasive needle puncture (MNP), split into flexion (90° flexion, n = 10) and extension (full extension, n = 10) groups. Experimental device setup Surgically prepared specimens were positioned laterally on a custom rig, with the target ligament upward, secured by clamps at each end. An ultrasound probe, aligned parallel to ligament fibres, was suspended over the ligament centre (Figure 1a). Ultrasound coupling gel (Hynaut, China) ensured no direct probe-ligament contact, minimizing measurement bias. The region of interest for SWS measurements was placed at the ligament midpoint, reflecting passive tension.18 Experimental setup for shear wave elastography measurements. a) Experimental setup with a porcine knee specimen showing the application of shear wave elastography (SWE). The dashed box shows an enlargement of the tensioner system. b) Paddles of the electronic knee tensioners, with a ruler for scale. c) Electronic knee tensioners used for gap distraction in the experiment. We employed custom-designed electronic knee tensioners (Balance Solver; Tinavi, China) (Figures 1b and 1c), which connect wirelessly to a control terminal via Bluetooth 5.0 and have demonstrated desirable accuracy and reliability.19 The bases of the tensioners are securely positioned on the tibial plateau, and two independent, movable pedals ensure precise contact with the femoral condyles, allowing controlled and accurate distraction of the medial and lateral knee compartments while their respective gap values are recorded in real time (Figure 2a). Knee tensioner data, shear wave elastography (SWE) imaging, and minimally invasive needle puncture (MNP) experimental setup. a) Display interface of the knee tensioner showing gap (left) and distraction force (right). SWE images b) with and c) without ligament tension, with colour maps and Multi-Q-BOX quantification boxes. d) Measurement of ligament thickness using B-mode SWE. e) Experimental setup and puncture operation. Experimental procedures and data collection SWE imaging used a Supersonic Mach 30 ultrasound system with a SuperLinear LH 20–6 MHz probe (Aixplorer; Supersonic Imagine, France), operating within a 0 to 1,200 kPa range (0 to 20 m/s). The Multi Q-BOX function positioned a 2.0 mm diameter measurement circle at the ligament midpoint to assess SWS, selected to encompass the ligament without including surrounding tissues (Figures 2b and 2c). Ligament thickness was measured in B-mode at multiple sites perpendicular to the fibres (Figure 2d). Three measurements per site were averaged for SWS and thickness to ensure data reliability. In the stretching experiment, a 20 N distraction force was initially applied to open the knee’s bilateral compartments, then increased in 3 N increments to 60 N. After each increment, the ultrasound probe was held steady for five seconds until the shear wave pattern stabilized within the region of interest. A SWE image was captured, and SWS was calculated for the MCL and LCL. The MNP experiment maintained a 45 N distraction force for consistent joint gap measurements. Using 18-gauge needles, punctures were made at 3 to 5 mm intervals around the MCL’s marked centre (Figure 2e). Needle insertions, placed outside the SWE region of interest, minimized elastography artifacts. Puncture depth was controlled to release superficial and deep MCL fibres. Overall, 45 punctures were performed in nine sets (5 punctures/set);20 after each set, SWS and medial gap values were recorded. Reliability assessment A total of 12 specimens (six from each experiment) were randomly selected from 33 for reliability analysis. Two trained operators (QW, LL) independently measured SWS, taking three consecutive values at the region of interest. Intrarater reliability was assessed with intraclass correlation coefficients (ICCs) for each operator’s repeated measurements. Interobserver reliability was assessed with ICCs between the operators’ mean SWS values. Statistical analysis Continuous variables are reported as means (SDs), and between-group comparisons were evaluated with the two-tailed independent-samples t-test. Pearson correlation assessed relationships between force and SWS in the stretching experiment, and between puncture count and SWS/medial gap changes in the MNP experiment. Generalized linear mixed models (GLMMs) analyzed SWS effects. For the stretching experiment, GLMMs evaluated fixed effects of distraction force, ligament structure (LCL vs MCL), thickness, and their interactions. In the MNP experiment, GLMMs assessed puncture count, knee position (flexion vs extension), thickness, and interactions. In both models, the specimen number was included as a random effect to account for inter-specimen variability. Two-way repeated measures analysis of variance (ANOVA) examined differences in SWS and medial gap across puncture counts and groups (flexion vs extension), followed by Bonferroni-corrected pairwise comparisons for significant results. Analyses used SPSS 26.0 (IBM, USA). Significance was set at p < 0.05 (two-tailed). Results Baseline specimen characteristics In the stretching experiments (n = 13), LCL and MCL ligament thickness and initial SWS without tension showed no significant differences, indicating comparable baseline properties. In the MNP experiments (n = 20), MCL SWS at 45 N force was significantly lower in the flexion group than in the extension group (p = 0.020, two-tailed independent-samples t-test), while ligament thickness and initial medial gap were comparable between groups (p > 0.05, two-tailed independent-samples t-test) (Table I). Table I.Baseline characteristics of the specimens used in the stretching and minimally invasive needle puncture (MNP) experiments.ExperimentGroupNumber of specimensMean initial SWS, m/s (SD)p-value*Mean ligament thickness, mm (SD)p-value*Mean medial gap, mm (SD)p-value*Stretching experimentLCL135.60 (1.49)0.7902.70 (0.70)0.933N/AN/AMCL135.45 (1.30)2.73 (0.59)N/AMNP experimentFlexion1014.26 (4.14)0.0202.27 (0.41)0.25513.49 (2.77)0.468Extension1016.08 (2.55)2.49 (0.39)12.21 (3.78)*Independent-samples t-test.LCL, lateral collateral ligament; MCL, medial collateral ligament; N/A, not applicable; SWS, shear wave speed. Baseline characteristics of the specimens used in the stretching and minimally invasive needle puncture (MNP) experiments. Independent-samples t-test. LCL, lateral collateral ligament; MCL, medial collateral ligament; N/A, not applicable; SWS, shear wave speed. Relationship between the gap distraction force and ligament SWS GLMM analysis showed that distraction force significantly increased SWS in both LCL and MCL (p < 0.001), while ligament thickness and its interaction with force had no notable effect (LCL: thickness, p = 0.652; force × thickness, p = 0.693; MCL: thickness, p = 0.466; force × thickness, p = 0.170). Combined GLMM analysis indicated that force consistently influenced SWS (p < 0.001), with no significant effect from ligament structure (p = 0.957) or its interaction with force (p = 0.560), suggesting a uniform SWS response across ligaments. The model also showed strong explanatory power (marginal R²: 0.86 LCL, 0.79 MCL) (Table II). In support of this relationship, Pearson correlation analysis revealed strong positive correlations between force and SWS (LCL: r = 0.67, R² = 0.45; MCL: r = 0.80, R² = 0.65; all p < 0.001) (Figure 3). Table II.Generalized linear mixed model results for the effects of distraction force, ligament thickness, and ligament structure on lateral collateral ligament (LCL) and medial collateral ligament (MCL) shear wave speeds.ModelEffectsβ (95% CI)p-valueLCLIntercept7.22 (1.43 to 13.0)0.014Force0.19 (0.09 to 0.29)< 0.001Thickness-0.47 (-2.52 to 1.58)0.652Force × thickness-0.01 (-0.03 to 0.04)0.693MCLIntercept4.18 (-0.90 to 9.25)0.105Force0.28 (0.19 to 0.37)< 0.001Thickness0.67 (-1.13 to 2.48)0.466Force × thickness-0.02 (-0.06 to 0.01)0.170Ligament structureIntercept6.03 (4.82 to 7.24)< 0.001Force0.21 (0.19 to 0.23)< 0.001Structure-0.03 (-1.36 to 1.69)0.957Force × structure0.01 (-0.02 to 0.04)0.560Model fit metrics LCL MCL Marginal R²0.860.79Conditional R²0.950.90Random effects variance2.051.03 Generalized linear mixed model results for the effects of distraction force, ligament thickness, and ligament structure on lateral collateral ligament (LCL) and medial collateral ligament (MCL) shear wave speeds. Relationship between the joint distraction force and shear wave speed (SWS) in the collateral ligaments. a) The 13 individual force‒SWS curves are shown for the lateral collateral ligament (LCL); each dashed line represents data from one specimen. b) Corresponding force‒SWS curves for the medial collateral ligament (MCL). c) and d) Plots representing a simple linear regression model for the ligaments, R² = 0.45 (LCL) and 0.65 (MCL); points with error bars are the means and standard error intervals of the data in a) and b), respectively. The shaded areas indicate the 95% CIs of the fitted lines. Relationship between puncture count, MCL SWS, and the medial gap GLMM analysis revealed a negative association between needle puncture count and MCL SWS (p < 0.001), indicating progressive ligament softening and a positive association with thickness (p < 0.001), while knee position and its interaction were non-significant (knee position: p = 0.051; interaction: p = 0.656). The model provided robust predictive accuracy (marginal R² = 0.93) (Table III). Pearson correlation showed puncture count correlated with MCL SWS and medial gap, with SWS stronger in flexion (r = 0.77, p < 0.001) than extension (r = 0.69, p < 0.001), and medial gap stronger in extension (r = 0.78, p < 0.001) than flexion (r = 0.61, p < 0.001). Overall correlations were 0.73 (SWS) and 0.68 (medial gap) (both p < 0.001). Table III.Generalized linear mixed model results for the effects of needle puncture count on medial collateral ligament shear wave speed.Effectsβ (95% CI)p-valueIntercept11.72 (1.73 to 21.72)< 0.001Times-0.21 (-0.33 to -0.10)< 0.001Knee position0.90 (-0.20 to 2.00)0.109Times × knee position0.01 (-0.03 to 0.05)0.656Ligament thickness2.48 (1.12 to 3.83)< 0.001Times × thickness0.03 (-0.02 to 0.08)0.195 Model fit metrics Value Marginal R²0.93Conditional R²0.96Random effects variance0.85 Generalized linear mixed model results for the effects of needle puncture count on medial collateral ligament shear wave speed. Two-way repeated-measures ANOVA confirmed differential effects of needle punctures on SWS and medial gap (Figure 4). In the flexion group, SWS decreased significantly after five punctures (p = 0.018), while in the extension group this occurred after ten punctures (p = 0.022). The flexion group showed a steeper SWS decline (0.69 m/s per five-puncture interval) than the extension group (0.60 m/s). Medial gap increased significantly after 20 punctures in both groups (flexion group: p = 0.027; extension group: p = 0.018), with a mean per-interval increase of 0.25 mm (flexion) and 0.16 mm (extension). SWS stabilized after 25 punctures (flexion group: p = 0.077) and 35 punctures (extension group: p = 0.055). Medial gap stabilized after 35 punctures (flexion group: p = 0.546) and 40 punctures (extension group: p = 0.202). Effects of minimally invasive needle puncture (MNP) on the medial collateral ligament (MCL) shear wave speed (SWS) and medial gap. a) Mean cumulative reduction in MCL SWS from baseline after every five punctures. b) Mean increase in the medial gap (mm) from baseline with the number of MNP punctures. Asterisks (*) denote statistically significant differences from baseline by two-way repeated-measures analysis of variance with Bonferroni-adjusted pairwise comparisons (p < 0.05). Brackets indicate the plateau range, within which all values are not significantly different from the final (puncture 45) measurement (p > 0.05), indicating stabilization. Measurement reliability In the stretching experiment, the intrarater ICCs were 0.999 (95% CI 0.996 to 1.000) for Operator 1 (QW) and 0.989 (95% CI 0.966 to 1.000) for Operator 2 (LL), while the interobserver ICC was 0.944 (95% CI 0.700 to 0.990). For the MNP experiment, the intrarater ICCs were 0.958 (95% CI 0.666 to 0.990) for Operator 1 and 0.982 (95% CI 0.930 to 1.000) for Operator 2, while the interobserver ICC was 0.887 (95% CI 0.150 to 0.980). Discussion Our first key finding was a significant linear relationship between joint distraction force and the SWS of the knee collateral ligaments, which remained robust in predicting dynamic ligament tension changes despite potential force sharing by surrounding soft-tissues and inherent differences between the MCL and LCL, supporting our first hypothesis. Additionally, during the needle puncture procedure, the changes in the MCL SWS were more sensitive than the medial gap measurements in detecting tension changes, with the former showing a stronger correlation with puncture count than the latter, aligning with our second hypothesis. These findings support the potential of SWE as a real-time intraoperative tool for monitoring and adjusting ligament tension during TKA. The rationale for using mechanical gap tensioner to achieve knee joint tension balance remains a subject of debate, as the tension applied by these tools to the collateral ligaments often exceeds the threshold necessary for ligament stability.21,22 Previous studies have indicated that a total tension of 40 to 50 N is sufficient to stabilize an intact knee joint, whereas notably greater tension, up to 60 N, may be required if the cruciate ligaments are absent.4,23 Within the defined optimal tension range, our study demonstrated a positive linear association between distraction force and SWS, which aligns with prior laboratory findings by Blank et al,16 who reported exceptionally high R² values (R² = 0.98 for the LCL, R² = 0.94 for the MCL) using simple linear regression. This discrepancy likely stems from methodological and statistical differences. Their study measured ligament SWS using laser vibrometry on isolated specimens with cyclic preconditioning (100 cycles, 10 to 250 N at 1 Hz), which reduced viscoelastic effects, enhanced axial compliance, and simplified the load–SWS relationship. In contrast, our study emphasized clinical feasibility by preserving the periarticular soft-tissues to simulate realistic intraoperative TKA conditions. These surrounding structures inevitably shared part of the distraction load, attenuating the direct linear relationship between applied force and ligament SWS, a limitation further compounded by the inability of simple linear regression to account for biological variability across specimens.24 To address this, we applied a GLMM incorporating ligament type, thickness, and inter-specimen variability (variance: 1.03 to 2.05), achieving markedly improved model performance (marginal R² = 0.79 to 0.86, conditional R² = 0.90 to 0.95). These results indicate that even under anatomically complex conditions, SWS remains a reliable predictor of ligament tension changes. We believe that as long as the distraction force is applied in a stable, controlled, and incremental manner, the ligament tension will increase accordingly, and SWS—as a measurable mechanical response—can still robustly reflect the underlying relationship. Further analysis showed no significant difference in force–SWS slopes between the MCL and LCL (p = 0.56), indicating similar responses to incremental distraction.25 However, a significant intercept was observed for the LCL (p < 0.05) but not for the MCL (p > 0.05), suggesting a higher baseline SWS in the LCL under unloaded conditions. This likely reflects anatomical and mechanical differences: the LCL’s thinner, cord-like structure and smaller cross-sectional area enhance its effective density and stiffness, while its superficial position increases sensitivity to probe angle or pressure variations.26 In contrast, the MCL’s broader, laminar morphology promotes uniform stress distribution and damping, reducing interference from surrounding tissues and variability due to its deeper location.27 Though preserved periarticular tissues may have reduced slope differences, intercept disparities highlight intrinsic biomechanical distinctions. These findings confirm SWE’s clinical feasibility and the MCL’s superior measurement stability. On the basis of these findings, we further applied SWE to MNP procedures on the MCL and made several notable observations. This application of SWE leverages the well-established reliability of MNP, initially described by Bellemans et al,8 which has been refined through methods such as flexion-only puncture,28 spacer block use,29 and integration with repetitive knee manipulation and pressure sensor technology.30 Unlike these static methods, SWE allows real-time tension monitoring, detecting early MCL tension changes after five to ten punctures—in contrast to medial gap measurements, through which changes can be detected only after 20 punctures—and identifying stabilization trends earlier in both the flexion (25 punctures) and extension (35 punctures) groups, thereby increasing precision in controlling the extent of release and preventing over-release from an excessive number of punctures. Notably, in two specimens, a sudden colour shift to blue in the SWE monitoring region, accompanied by a marked increase in the medial gap, indicated tearing of the ligament fibres and excessive release, resembling the pressure or gap mutation phenomena associated with over-release in prior studies.31 These dynamic colour changes, particularly evident in high-risk flexion positions,32 offer a significant advantage over traditional stretch-measure-release cycles, underscoring the superior sensitivity and early warning potential of SWE in detecting over-release and in capturing subtle soft-tissue release effects that traditional gap measurements often fail to identify accurately in early stages. The GLMM fit indices (marginal R² = 0.934, conditional R² = 0.964) confirmed the strong relationship between SWS changes and needle punctures while accounting for inter-specimen variability, and surpassed the linear regression fit (R²avg = 0.76) reported by Blomquist et al.20 Interestingly, the results of model analysis revealed that ligament thickness significantly influenced the baseline SWS (β = 2.475, p < 0.001), likely due to the denser collagen structures in the thicker ligaments, but its interaction with puncture count was not significant (β = 0.033, p = 0.195), indicating that the softening effect of needle punctures on the SWS remained consistent across different ligament thicknesses. Finally, motivated by these observations and the need to prevent potential over-release from multiple punctures, we attempted to use SWE to precisely locate high-tension fibre bundles within the ligament to improve puncture efficiency. However, high-tension regions, displayed as deep red (indicating high stiffness), yielded artifacts that interfered with surrounding fibres, ultimately limiting the applicability of this approach.33 An expanded SWS measurement threshold may be required in future studies to enable more accurate assessment of high-stiffness ligament regions. Despite these technical challenges, the overall methodological reliability of SWE was clearly established in this study. Specifically, this study is the first to integrate a precise joint distraction device with a portable SWE ultrasound system during a simulated TKA procedure involving porcine knee specimens. The notably high intra- and interobserver reliabilities underscore the consistency and accuracy of this standardized approach in directly measuring passive tension in the knee ligaments, thereby establishing a robust framework for the future application of SWE technology in clinical TKA procedures. Based on our current findings, SWE appears best suited for intraoperative use where surgeons need to dynamically monitor ligament tension changes in response to specific surgical steps. Additionally, pre- and postoperative SWE measurements may help correlate collateral ligament tension patterns with postoperative outcomes, supporting individualized rehabilitation strategies. Although determining definitive SWS threshold values of ligaments requires further biomechanical studies, our results strongly support its immediate utility in monitoring relative tension changes during TKA, thus enabling real-time feedback for surgeons. This study has several limitations. First, while the porcine knee model is validated,34 it may not fully replicate human primary TKA, especially in varus deformities. Second, the small sample size limits generalizability, despite statistical robustness. Third, probe stability remains a challenge; a refined bedside fixation method is needed for clinical use. Additionally, the impact of ligament viscoelastic properties35 and shear wave dispersion (SWD)36 on SWS was not explicitly assessed. Since these factors may significantly affect SWS measurements, future studies should incorporate SWD and tissue viscoelasticity into analysis, validate findings with larger cohorts, and use serial dissection to quantify surrounding tissue effects. These efforts could improve measurement accuracy, establish reliable SWS thresholds, and enhance intraoperative guidance and patient outcomes. In conclusion, our study confirms that SWE is a reliable, noninvasive tool for monitoring ligament tension changes during TKA, especially within the optimal force range (< 60 N). SWE effectively tracks tension changes during both stretching and release procedures. In the stretching experiment, it provided valuable insights into tension dynamics under distraction forces. The MNP experiments further demonstrated SWE’s ability to detect early tension changes and stabilization trends, outperforming traditional gap measurements. These capabilities enable precise soft-tissue release control and help prevent over-release. Overall, SWE shows great potential for optimizing knee ligament tension, improving soft-tissue balancing and enhancing TKA outcomes.