- Email: [email protected]
Mimicking anatomical condylar configuration into knee prosthesis could improve knee kinematics after TKA — A computational simulation
Clinical Biomechanics 27 (2012) 176–181
Contents lists available at SciVerse ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Mimicking anatomical condylar configuration into knee prosthesis could improve knee kinematics after TKA — A computational simulation Yu-Liang Liu a, e, Wen-Chuan Chen a, b, Wen-Ling Yeh c, Colin Joseph McClean a, b, Chun-Hsiung Huang d, Kun-Jhih Lin a, b, Cheng-Kung Cheng a, b,⁎ a
Orthopedic Biomechanics Laboratory, Institute of Biomedical Engineering, National Yang-Ming University, Taipei, Taiwan Orthopedic Device Research Center, National Yang-Ming University, Taipei, Taiwan Department of Orthopedic Surgery, Chang Gung Memorial Hospital, Taoyuan, Taiwan d Department of Orthopedic Surgery, Mackay Memorial Hospital, Taipei, Taiwan e Department of Product Development, United Orthopedic Corporation, Taipei, Taiwan b c
a r t i c l e
i n f o
Article history: Received 18 May 2011 Accepted 18 August 2011 Keywords: Knee kinematics Different condylar heights Computer simulation TKA
a b s t r a c t Background: Restoration of femoral rollback and tibial internal rotation are two of the major objectives following total knee arthroplasty. Previously, we improved prosthetic knee kinematics by replicating the convexly lateral tibial plateau of intact knee. This study attempted to regain more normal knee kinematics through a posterior cruciate ligament retaining knee, which simultaneously incorporated convexly lateral tibial plateau and anatomical condylar configuration into the prosthesis design. Methods: Computational simulation was utilized to analyze motion of three-dimensional knee models. Three total knee systems with consistent convex insert design but with different condylar heights of 0, 2.7 and 4.7 mm were investigated in present study. Magnetic resonance images of the subject were utilized to construct the bone models and to distinguish the attachment sites of ligaments and tendons. The distal femurs were modeled to rotate about designated flexion axes of femoral components, and the motion of the proximal tibia was unconstrained except further activity of flexion/extension. Movements of the medial/lateral condyles and tibial rotation were recorded and analyzed. Findings: Significant improvements in posterior movement of the lateral condyle and in tibial internal rotation were observed for knee models with different condylar heights, as compared to the knee model with consistent condylar height, when flexion exceeded 100°. Results also revealed that excessive difference in condylar height over anatomical condylar configuration provided no contribution to the restoration of normal knee kinematics. Interpretation: Replicating the morphology of anatomical condylar configuration of the intact knee into knee prostheses could improve knee kinematics during higher knee flexion. © 2011 Elsevier Ltd. All rights reserved.
1. Introduction Regaining a normal joint line is considered a critical factor in achieving successful total knee arthroplasty (TKA) (Anouchi et al., 1993; Arima et al., 1995; Griffin et al., 2000; Partington et al., 1999). One important criterion for recovering an adequate joint line is to restore an anatomical condylar configuration. Anatomical condylar configuration aids in the spiral motion of the tibia about the femur during flexion and extension (Nordin and Frankel, 2002). The spiral motion of the tibiofemoral joint could result from a combination of femoral rollback and tibial internal rotation during knee flexion. Greater femoral rollback would improve the efficiency of quadriceps muscles ⁎ Corresponding author at: Orthopedic Biomechanics Laboratory, Institute of Biomedical Engineering, National Yang-Ming University, No.155, Sec.2, Linong St., Shih-Pai, Taipei, 11221 Taiwan. E-mail address: [email protected] (C.-K. Cheng). 0268-0033/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2011.08.010
(Andriacchi and Mikosz, 1991) and facilitate higher knee flexion (Mikashima et al., 2010). Tibial rotation was associated with deepflexion knee postures, such as squatting, kneeling or lunging (Banks et al., 2003). Nevertheless, most contemporary posterior cruciate ligament retaining (PCR) knees were designed with symmetrical articular surface design and consistent condylar height that could lead to inadequate femoral rollback and insufficient tibial internal rotation (Casino et al., 2009; Dennis et al., 2004; Li et al., 2001; Yue et al., 2011). Therefore, asymmetrical prosthetic designs have been developed to restore more natural knee kinematics, such as motion guiding implants (Walker et al., 2009), medial pivot knee (Moonot et al., 2009) and fully flat insert design (Gomaa and Williams, 2009). Although the abovementioned modifications offer improvements in specific motions, they are still not able to properly recover normal knee kinematics. A possible cause is that the prosthetic geometries of TKAs are inconsistent with the morphology of intact knee. The lateral tibial plateau of intact knee is of convex shape (Bare´ et al., 2006);
Y.-L. Liu et al. / Clinical Biomechanics 27 (2012) 176–181
but most contemporary tibial inserts are symmetrically concave in design. Again, the medial condyle of intact knee is more distal than the lateral condyle. However, most TKAs employ a consistent femoral condylar height to match the symmetrical insert articulation. Previously, we demonstrated that convex lateral insert articulation could improve femoral rollback and tibial internal rotation (Liu et al., 2011). Subsequently, based on a validated knee model, current study attempted to restore greater femoral rollback and tibial internal rotation using a newly developed PCR knee which simultaneously replicated the convexly lateral tibial plateau and anatomical condylar configuration. 2. Methods Bone and cartilaginous models were from a healthy right knee of a female (age: 20 years old, weight: 50 Kg, height: 158 cm), including distal femur, proximal tibia, patella, and fibula. These models were sourced from sagittal magnetic resonance images (MR: SIEMENS MAGNETOM Trio A Tim SYSTEM 3T, Siemens, Germany). The slice interval of MR images was 1 mm with a resolution of 480 × 512 pixels. The bones of the proximal tibia and the distal femur were transected approximately 75 mm from the natural joint line. Thereafter, the bone models were smoothed using Geomagic Studio v9.0 (Parametric Technology Crop., Needham, MA, U.S.A.). Finally, the newly developed knee prostheses were implanted into a bone model and imported into MSC.ADAMS_R3 (MSC Software, Santa Ana, CA, U.S.A.) for dynamic simulation. 2.1. Assignments of ligament, tendon and meniscus of intact knee The posterior cruciate ligament (PCL), medial collateral ligament (MCL), lateral collateral ligament (LCL), patellar and quadriceps tendons were distinguished by self-footprint from MR images of the subject. Origin and insertion points of ligaments and tendons were confirmed by the collaborator who is an experienced orthopedic surgeon (C-H Huang) (Fig. 1). In accordance with the characterization of ligaments in representative literature (Abdel-Rahman and Hefzy, 1998), the PCL was considered as anterior and posterior fiber bundles; the MCL was considered as anterior, deep and oblique fiber bundles, and the LCL was considered as a single fiber bundle. All ligaments were applied to TKA models and simulated as adequate
177
force components with parabolic and linear regions according to the following equations (Abdel-Rahman and Hefzy, 1998): 8 > > <
0 2 K1 Lj −L0j ; j F¼ h i > > : K2j Lj −ð1 þ ε1 ÞL0j
εj ≤ 0 0 b εj ≤ 2ε1 εj b 2ε1
where εj is the strain in the jth element, K1j and K2j are the stiffness coefficients of the jth spring element for the parabolic and linear regions, respectively, and Lj and L0j are its current and slack lengths, respectively. The linear range threshold is specified as ε1 = 0.03. The stiffness of each ligament is shown in Table 1. Additionally, both quadriceps and patellar tendons were considered as medial and lateral fiber bundles and simulated as purely elastic tensile springs. The stiffness coefficients of quadriceps and patellar tendons were 2000 N/mm (Yu et al., 2001) and 1142 N/mm (Hashemi et al., 2005) respectively. 2.2. Construction of TKA models The TKA models included a validated anatomic-like (ALK) model (Liu et al., 2011), an ALK model with a 2.7 mm difference in medial/ lateral condylar heights (ALK_DH_2.7) and an ALK model with a 4.7 mm difference in medial/lateral condylar heights (ALK_DH_4.7). The 2.7 mm difference of condylar heights was measured on the subject in this study and the condylar twist angle was 3.8°. According to the findings of Koudela et al. (2010), the average condylar twist angle of the intact knee is 5.25° ± 1.68°. Therefore, the difference of condylar heights of the ALK_DH_4.7 model was set at 4.7 mm to produce the condylar twist angle of 6.7°, which was the maximum possible condylar twist angle of the intact knee. The thickness of the medial/ lateral femoral condyles of the ALK model was 9 mm, referring to U2 total knee system-cruciate ligament retaining (CR) type (United, Co., Hsinchu, Taiwan). For ALK_DH_2.7 and ALK_DH_4.7 models, the lateral condylar thickness was referenced from the Journey Knee design (Smith & Nephew, Inc., London, UK) and considered as 7 mm. Based on above mentioned differences of condylar heights, the medial condylar thicknesses were simulated as 9.7 mm (7 + 2.7 = 9.7) and 11.7 mm (7 + 4.7 = 11.7) in ALK_2.7 and ALK_4.7 models, respectively. The thicknesses of each medial/lateral femoral condyle of the three models are shown in Fig. 2. Except for the consideration of different
Fig. 1. ALK model, showing the definitions of the knee coordinate system in ± x (flexion/extension axis), ± y (varus/valgus rotation axis) and ± z (external/internal rotation axis). Multiple beads connected by springs were utilized to simulate the wrapping of the quadriceps tendon around the trochlear groove. Attachments of ligaments and tendons were sourced from MR images of the subject and considered as tensile spring-like force elements. Note: P_T = patellar tendon, Q_T = quadriceps tendon, MCL = medial collateral ligament, LCL = lateral collateral ligament, ACL = anterior cruciate ligament, PCL = posterior cruciate ligament.
178
Y.-L. Liu et al. / Clinical Biomechanics 27 (2012) 176–181
Table 1 The stiffness of each ligament in current study (Abdel-Rahman and Hefzy, 1998).
PCL-anterior PCL-posterior MCL-anterior MCL-oblique MCL-deep LCL
K1(N/mm)
K2 (N/mm2)
31.26 19.29 10.00 5.00 5.00 10.00
125.00 60.00 91.25 27.86 21.07 72.22
Note: PCL = posterior cruciate ligament; MCL = medial collateral ligament; LCL = lateral collateral ligament.
condylar heights, the radii of the femoral components and tibial inserts of ALK_DH_2.7 and ALK_DH_4.7 models were consistent with the ALK model. The anterior/posterior dimension of the polyethylene insert and metal baseplate of the ALK model was 47 mm, and the medial/lateral dimension was 69 mm. The sagittal radius of the ALK model's femoral component was 25.7 mm (full extension to 93° flexion), which referred to the radius of the medial condyle of the subject located on the tibio-femoral contact point at full knee extension. Also, the frontal radius of the ALK model's medial femoral condyle was 25.7 mm to produce a spherical geometry in order to regain maximum contact area. A more dished medial insert articular surface also contributes to joint stability during higher knee flexion, particularly during subluxation of the lateral condyle. Conversely, the frontal radius of the lateral femoral condyle of the ALK model was 70 mm. A shallower and more flat insert articular surface allowed rolling and sliding of the lateral condyle to achieve natural knee motion. In accordance with the findings of Sathasivam and Walker (1999), a difference in femoral and tibial frontal radii of 2 mm was helpful in restoring natural knee kinematics after TKA. Therefore, the frontal radii of medial and lateral insert condyles of the ALK model were set as 27.7 mm and 72 mm, respectively. Again, the sagittal radius of the posteromedial insert condyle was 27.7 mm, and the convexly sagittal radius of posterolateral insert condyle was 78 mm, which measured the curvature of the subject's posterolateral tibial plateau located on the tibiofemoral contact point at full knee extension. The reflection point of the ALK model was set on the base of the tibial insert dish and located 19.8 mm from the posterior cliff of the proximal tibia. The radii of the femoral component and tibial insert of the ALK model are shown in Fig. 3.
(Godest et al., 2002). In order to simulate the wrapping of quadriceps tendon around the trochlear groove at higher knee flexion, multiple beads connected by springs were used (Guess et al., 2010; Kessler et al., 2008) (Fig. 1). 2.4. Data acquisition A Cartesian coordinate system was defined by the mediolateral axis (x, flexion/extension axis), the anteroposterior axis (y, varus/ valgus rotation axis), and the longitudinal axis (z, internal/external rotation axis) (Fig. 1). Movements of medial/lateral condyles were defined as the distance between original and present positions of flexion facet centers in the y direction of local coordinates during knee flexion. Internal/external tibial rotation was measured in the z direction on the local coordinate of the tibia relative to the local coordinate of the femur. All data of femoral movements and tibial rotations were recorded every 10° from full extension to 140° of flexion. 3. Results 3.1. Femoral movements Lateral and medial condylar movements of three TKA models are shown in Fig. 4A and B. Compared with the ALK model, the root mean square values of lateral condylar movements were respectively 1.4 and 0.7 mm for ALK_2.7 and ALK_4.7 models during the angular interval 0° to 100° of knee flexion. Again, the maximum differences of lateral condylar movements were 3.5 mm posteriorly (at 100° of knee flexion) and 1.3 mm anteriorly (at 70° of knee flexion) for ALK_2.7 and ALK_4.7 models at less than 100° of knee flexion as compared with the ALK model. Beyond 100° of knee flexion, the maximum increments of posterior movements of the lateral condyle by the ALK_DH_2.7 model were 9.6 and 4.2 mm at 140° of flexion when compared with ALK and ALK_DH_4.7 models, respectively. In medial condylar movements, compared with ALK model, the root mean square values were 1.1 and 1.3 mm respectively for ALK_2.7 and ALK_4.7 models during full range of flexion. Besides, the maximum differences of medial condylar movements were 2.2 mm posteriorly (at 110° of knee flexion) and 3.7 mm anteriorly (at 140° of knee flexion) respectively for ALK_2.7 and ALK_4.7 models, as compared with the ALK model. 3.2. Tibial internal rotations
2.3. Constraints of knee motion The femoral flexion axes of TKA models were in compliance with the condylar radii of femoral components. Femoral flexion was driven by a revolute joint of ADAMS software, and its control type was displacement with a speed of 15 deg/s. Motion of the tibial component was unconstrained, except in flexion–extension. An average ground reaction force of 1.5 × body weight (750 N) (D'Lima et al., 2007) was applied to the center of mass of the tibial component. The friction coefficient of metal-to-polyethylene surfaces was designated as 0.04
The results of tibial internal rotation of the three TKA models are shown in Fig. 4C. Compared with the ALK model, the root mean square values of tibial internal rotations were 0.8° and 0.6° respectively for ALK_2.7 and ALK_4.7 models through the angular interval 0° to 100° of knee flexion. Again, the maximum differences of tibial rotations were 1.7° internally (at 100° of knee flexion) and 1° externally (at 90° of knee flexion) respectively for ALK_2.7 and ALK_4.7 models before 100° of knee flexion as compared with the ALK model. Beyond 100° of knee flexion, tibial internal rotation for the ALK_DH_2.7 and ALK_DH_4.7
Fig. 2. The medial/lateral condylar heights of ALK, ALK_2.7 and ALK_DH_4.7 models.
Y.-L. Liu et al. / Clinical Biomechanics 27 (2012) 176–181
179
Fig. 3. The radii of femoral component and tibial insert of ALK model. (A) sagittal radius; (B) frontal radii of femoral component; (C) frontal radii, (D) sagittal radius of lateral condyle and (E) sagittal radius of medial condyle of tibial insert.
models continuously increased, whereas tibial internal rotation for the ALK model obviously decreased with knee flexion. On average, the ALK_DH_2.7 model has a larger tibial internal rotation than the ALK_DH_4.7 and ALK models. The maximum increments of tibial internal rotation by ALK_DH_2.7 were 15.4° and 2.2° at 140° of flexion when compared with the ALK and ALK_DH_4.7 models, respectively. 4. Discussion Restoration of the femoral rollback and tibial internal rotation are two major goals for contemporary knee prosthetic design with the aim of regaining normal knee kinematics. However, many studies have demonstrated that femoral rollback and tibial internal rotation were inadequate or abnormal after replacement by contemporary PCR TKA (Casino et al., 2009; Dennis et al., 2004; Li et al., 2001; Yue et al., 2011). Our previous study demonstrated that the inclusion of convexly lateral insert articular surface would improve femoral rollback and tibial internal rotation (Liu et al., 2011). However, it was also observed that tibial internal rotation on convex insert design reduced significantly beyond 100° of flexion. Therefore, the newly developed PCR knee incorporated convexly lateral insert articular surface and anatomical condylar configuration simultaneously into the prosthetic design to enhance femoral rollback and tibial internal rotation. Previously, in order to resolve the problem of inadequate knee kinematics after replacement by contemporary TKA, asymmetrical prosthetic designs were developed, such as the motion guiding prosthetic design (Walker et al., 2009), medial pivot knee (Moonot et al., 2009) and flat insert articular surface design (Gomaa and Williams, 2009). Although the abovementioned modifications offer improvements in specific motions, they are still unable to properly recover normal knee kinematics. A possible cause is that the prosthetic geometries of TKAs are inconsistent with the morphology of intact knee. The morphology of the posterolateral tibial plateau is convex, but the abovementioned TKAs were designed with concave or flat insert articulation. Li et al. (2001) suggested that the concave posterior portion of the polyethylene component may serve as a major resistance to posterior femoral translation when compared with the morphology of lateral tibial plateau of intact knee. Again, the
flat insert articulation could improve femoral rollback but may suffer from a sense of instability due to paradoxical anterior sliding (Gomaa and Williams, 2009). Besides, the anatomical medial condyle is more distal than the lateral condyle; nevertheless, the abovementioned TKAs were designed with consistent condylar heights. Essentially, the convexly lateral insert articular surface allowed the lateral femoral condyle to sublux off the back of the tibial plateau during high flexion. This is a characteristic of normal knee kinematics (Bare´ et al., 2006). Additionally, the convex insert articular surface was able to increase the tibial posterior clearance, which allowed deeper knee flexion without interference between the distal femur and tibial insert. As for the anatomical condylar configuration, it was associated with the reconstruction of the joint line. Previous literature reported that inappropriate location of the joint line would result in higher patellofemoral contact forces (Singerman et al., 1994), an unstable knee joint after TKA (Laskin, 2002) and patellar dislocation (Insall et al., 1972). Although traditional TKA designs also restored the location of the joint line at knee full extension, they still could not completely reproduce natural motion patterns during knee flexion. That could be caused by the geometries of symmetrical femoral condyles of most contemporary TKAs being inconsistent with intact knee. Conversely, the anatomical condylar configuration was demonstrated to be helpful in reproducing the spiral motion of the tibia about the femur (Nordin and Frankel, 2002). The spiral motion of the tibiofemoral joint should be a combination of femoral movement and tibial rotation during knee flexion/extension, and normal spiral motion should be beneficial in regaining normal knee kinematics. Although many surgeons believe that soft tissue balance is one of the most important factors in achieving a successful surgery, soft tissues cannot completely recover their original performance after long-term abnormal weight bearing. Walker (2001) also took into account that characteristics of normal knee motion could be achieved by modifying prosthetic features alone. Recently, a convex insert articular surface and different condylar heights were applied to commercial knee prostheses, such as the Journey Knee. However, the Journey Knee was a bi-cruciate stabilized knee system with an insert post and femoral cam that differed from the newly developed PCR knees in the current study. Additionally, the FINE Total Knee System-PCR type (Nakashima
180
Y.-L. Liu et al. / Clinical Biomechanics 27 (2012) 176–181
(-) Posterior <---> (+) Anterior (mm)
(A) Knee Flexion (degrees) ALK
ALK_DH_2.7
ALK_DH_4.7
0.0 -5.0 0
20
40
60
80
100
120
140
-10.0 -15.0 -20.0 -25.0 -30.0 -35.0
(-) Posterior <---> (+) Anterior (mm)
(B) Knee Flexion (degrees) ALK
4.0
ALK_DH_2.7
ALK_DH_4.7
2.0 0.0 -2.0 0
20
40
60
80
100
120
140
-4.0 -6.0 -8.0 -10.0 -12.0
(C) Knee Flexion (degrees) ALK
Tibial Rotation (degrees)
25.0
ALK_DH_2.7
ALK_DH_4.7
20.0 15.0 10.0 5.0 0.0
-5.0
0
20
40
60
80
100
120
140
Fig. 4. Comparisons of (A) lateral condylar movements, (B) medial condylar movements and (C) tibial internal rotations of ALK, ALK_DH_2.7 and ALK_DH_4.7 models.
Medical, Okayama, Japan) considered an anatomical condylar configuration of the femoral component. However, the flat lateral insert articular surface was still different from the convex lateral tibial plateau of intact knee. To our knowledge, there are no commercial PCR knee designs that simultaneously mimic the convexly lateral tibial plateau and anatomical condylar configuration of intact knee. In current study, the ligament condition referred to the research of Abdel-Rahman and Hefzy (1998) which was inconsistent with other relative publications by Blankevoort and Huiskes (1996) and Li et al. (1999) regarding ligament quantity and ligament force equation. For ligament quantity, the LCL was simulated as a single bundle in our study but was simulated as three bundles by Blankevoort and Huiskes (1996) and Li et al. (1999). Furthermore, our study excluded the MCL capsule from the simulation, which was inconsistent with the research of Li et al. (1999). Although the quantity of ligaments in current study was not completely consistent with the abovementioned two literatures, it was compatible with the research of Abdel-Rahman and Hefzy (1998) and Yu et al. (2001). For ligament force equations, the measurements in our study were relative to the variation of ligament length, but
the research of Blankevoort and Huiskes (1996) and Li et al. (1999) were relative to the variation of ligament strain. Although the ligament equation in our study was inconsistent with the two abovementioned studies, the ligament performance should be similar, because they referred to the consistent reference of Wismans et al. (1980). As can be seen from the results, at knee flexion beyond 100°, the ALK_DH_2.7 and ALK_DH_4.7 models showed significantly improved posterior movement of the lateral condyle and tibial internal rotation when compared with the ALK model, particularly the ALK_DH_2.7 model. This revealed that excessive difference in condylar heights contributes nothing to the restoration of normal knee kinematics, such as in the ALK_DH_4.7 model. Even so, the results showed that incorporating an anatomical condylar configuration into the TKA prosthetic design could improve femoral rollback and tibial internal rotation at higher knee flexion. Consulting the literature review, Miyazaki et al. (2011) investigated the knee kinematics of FINE PCR Total Knee System, ADVANCE Total Knee System (Wright Medical Technology, Arlington, Tenn.), and the ADVANTIM Total Knee System (Wright Medical Technology, Arlington, Tenn.). The results demonstrated that the FINE knee
Y.-L. Liu et al. / Clinical Biomechanics 27 (2012) 176–181
displayed more suitable knee kinematics, including femoral rollback and tibial internal rotation, than the ADVANCE and ADVANTIM knees, by using flat lateral insert articulation and anatomical condylar configuration, consistent with the findings of the current study. Following the design rationale of the Journey Knee, there was a significant improvement in anterior/posterior translation over a traditional knee. Moreover, in femoral axial rotation, the Journey Knee kept the femoral external rotation beyond 110° of knee flexion. Conversely, femoral external rotation of traditional PS TKA designs decreased noticeably beyond 110° of knee flexion. Our study also revealed that an anatomical condylar configuration could improve tibial internal rotation particularly at high knee flexion, consistent with the Journey Knee. One of the limitations in current study was that the differences of condylar heights between patients were inconsistent; therefore collection of a large quantity of data on anatomical condylar configuration was suggested. Also, the balance of LCL and MCL soft tissue is another concern following TKA with a different condylar height design. Generally, the tensile strengths of LCL and MCL were considered balanced for contemporary TKAs. However, the newly developed PCR knee in current study was simulated with an anatomical condylar configuration, and the medial/lateral condylar thicknesses were inconsistent. The influence of anatomical condylar configuration prosthetic design on tensile strength of LCL and MCL is still unclear and should be resolved through further clinical study. Furthermore, the anatomical condylar configuration was only utilized in the ALK model, so the influence on different prosthetic designs is still unknown. Although the numerical simulation cannot take the mechanical properties of soft tissue, such as visco-elasticity into consideration, identical simulated criteria of models should be able to distinguish the influence of different prosthetic designs on knee kinematics. Finally, the femoral component and the tibial insert were simulated with an optimal alignment in current study. The influence of mal-alignment between femoral component and tibial insert on knee kinematics needs further study. 5. Conclusions Modifying the geometry of knee prosthetic design in order to restore normal femoral rollback and tibial internal rotation seems to be the current trend in designing commercial knee prostheses. Our previous study demonstrated that convex insert articular surface could improve knee kinematics after TKA. This study revealed that incorporating convex lateral insert articulation and anatomical condylar configuration could enhance more femoral rollback and tibial internal rotation during higher knee flexion. In order to ensure the efficacy of this newly developed CR total knee system, further investigation in clinical trials and follow-ups will be necessary. Conflict of interest statement There are no conflicts of interest in this manuscript. Acknowledgments We are pleased to acknowledge the financial support of the National Science Council, Taiwan, ROC (NSC99-2911-I-009-101) and the Administrative Bureau of Southern Taiwan Science Park (BY-03-04-17-98). References Abdel-Rahman, E.M., Hefzy, M.S., 1998. Three-dimensional dynamic behaviour of the human knee joint under impact loading. Med. Eng. Phys. 20, 276–290. Andriacchi, T.P., Mikosz, R.P., 1991. Musculoskeletal dynamics, locomotion and clinical applications. In: Mow, V.C., Hayes, W.C. (Eds.), Basic Orthopedic Biomechanics. Raven Press, New York, pp. 51–92.
181
Anouchi, Y.S., Whiteside, L.A., Kaiser, A.D., Milliano, M.T., 1993. The effects of axial rotational alignment of the femoral component on knee stability and patellar tracking in total knee arthroplasty demonstrated on autopsy specimens. Clin. Orthop. Relat. Res. 287, 170–177. Arima, J., Whiteside, L.A., McCarthy, D.S., White, S.E., 1995. Femoral rotational alignment, based on the anteroposterior axis, in total knee arthroplasty in a valgus knee. A technical note. J. Bone Joint Surg. Am. 77, 1331–1334. Banks, S.A., Bellemans, J., Nozaki, H., Whiteside, L.A., Harman, M., Hodge, W.A., 2003. Knee motions during maximum flexion in fixed and mobile-bearing arthroplasties. Clin. Orthop. Relat. Res. 410, 131–138. Bare´, J.V., Gill, H.S., Beard, D.J., Murray, D.W., 2006. A convex lateral tibial plateau for knee replacement. Knee 13, 122–126. Blankevoort, L., Huiskes, R., 1996. Validation of a three-dimensional model of the knee. J. Biomech. 29, 955–961. Casino, D., Zaffagnini, S., Martelli, S., Lopomo, N., Bignozzi, S., Iacono, S., Russo, A., Marcacci, M., 2009. Intraoperative evaluation of total knee replacement: kinematic assessment with a navigation system. Knee Surg. Sports Traumatol. Arthrosc. 17, 369–373. Dennis, D.A., Komistek, R.D., Mahfouz, M.R., Walker, S.A., Tucker, A., 2004. A multicenter analysis of axial femorotibial rotation after total knee arthroplasty. Clin. Orthop. Relat. Res. 428, 180–189. D'Lima, D.D., Patil, S., Steklov, N., Chien, S., Colwell Jr., C.W., 2007. In vivo knee moments and shear after total knee arthroplasty. J. Biomech. 40, 11–17. Godest, A.C., Beaugonin, M., Haug, E., Taylor, M., Gregson, P.J., 2002. Simulation of a knee joint replacement during a gait cycle using explicit finite element analysis. J. Biomech. 35, 267–275. Gomaa, S.T., Williams, J.L., 2009. Effect of tibial insert geometry on kinematics following fixed-bearing cruciate-retaining total knee arthroplasty. 55th Annual Meeting of the Orthopedic Research Society. Paper No.191. Griffin, F.M., Math, K., Scuderi, G.R., Insall, J.N., Poilvache, P.L., 2000. Anatomy of the epicondyles of the distal femur: MRI analysis of normal knees. J. Arthroplasty 15, 354–359. Guess, T.M., Thiagarajan, G., Kia, M., Mishra, M., 2010. A subject specific multibody model of the knee with menisci. Med. Eng. Phys. 32, 505–515. Hashemi, J., Chandrashekar, N., Slauterbeck, J., 2005. The mechanical properties of the human patellar tendon are correlated to its mass density and are independent of sex. Clin. Biomech. 20, 645–652. Insall, J., Goldberg, V., Salvati, E., 1972. Recurrent dislocation and the high-riding patella. Clin. Orthop. Relat. Res. 88, 67–69. Kessler, O., Patil, S., Colwell Jr., C.W., D'Lima, D.D., 2008. The effect of femoral component malrotation on patellar biomechanics. J. Biomech. 41, 3332–3339. Koudela Jr., K., Koudelová, J., Koudela Sr., K., Kunešová, M., Křen, J., Pokorný, J., 2010. Radiographic measurements in total knee arthroplasty and their role in clinical practice. Acta Chir. Orthop. Traumatol. Cech. 77, 304–311. Laskin, R.S., 2002. Joint line position restoration during revision total knee replacement. Clin. Orthop. Relat. Res. 2002, 169–171. Li, G., Gil, J., Kanamori, A., Woo, S.L., 1999. A validated three-dimensional computational model of a human knee joint. J. Biomech. Eng. 121, 657–662. Li, G., Zayontz, S., Most, E., Otterberg, E., Sabbag, K., Rubash, H.E., 2001. Cruciate-retaining and cruciate-substituting total knee arthroplasty: an in vitro comparison of the kinematics under muscle loads. J. Arthroplasty 16, 150–156. Liu, Y.L., Lin, K.J., Huang, C.H., Chen, W.C., Chen, C.H., Chang, T.W., Lai, Y.S., Cheng, C.K., 2011. Anatomic-like polyethylene insert could improve knee kinematics after total knee arthroplasty — a computational assessment. Clin. Biomech. 26, 612–619. Mikashima, Y., et al., 2010. In vivo deep-flexion kinematics in patients with posteriorcruciate retaining and anterior-cruciate substituting total knee arthroplasty. Clin. Biomech. 25, 83–87. Miyazaki, Y., Nakamura, T., Kogame, T., Saito, M., Yamamoto, K., Suguro, T., 2011. Analysis of the kinematics of total knee prostheses with a medial pivot design. J. Arthroplasty 26, 1038–1044. Moonot, P., Mu, S., Railton, G.T., Field, R.E., Banks, S.A., 2009. Tibiofemoral kinematic analysis of knee flexion for a medial pivot knee. Knee Surg. Sports Traumatol. Arthrosc. 17, 927–934. Nordin, M., Frankel, V.H., 2002. Basic Biomechanics of the Musculoskeletal System, third ed. Lippincott Williams & Wilkins, New York. Partington, P.F., Sawhney, J., Rorabeck, C.H., Barrack, R.L., Moore, J., 1999. Joint line restoration after revision total knee arthroplasty. Clin. Orthop. Relat. Res. 367, 165–171. Sathasivam, S., Walker, P.S., 1999. The conflicting requirements of laxity and conformity in total knee replacement. J. Biomech. 32, 239–247. Singerman, R., Davy, D.T., Goldberg, V.M., 1994. Effects of patella alta and patella infera on patellofemoral contact forces. J. Biomech. 27, 1059–1065. Walker, P.S., 2001. A new concept in guided motion total knee arthroplasty. J. Arthroplasty 16, 157–163. Walker, P.S., Sussman-Fort, J.M., Yildirim, G., Boyer, J., 2009. Design features of total knees for achieving normal knee motion characteristics. J. Arthroplasty 24, 475–483. Wismans, J., Veldpaus, F., Janssen, J., Huson, A., Strulen, P., 1980. A three dimensional mathematical model of the knee joint. J. Biomech. 13, 677–686. Yu, C.H., Walker, P.S., Dewar, M.E., 2001. The effect of design variables of condylar total knees on the joint forces in step climbing based on a computer model. J. Biomech. 34, 1011–1021. Yue, B., Varadarajan, K.M., Moynihan, A.L., Liu, F., Rubash, H.E., Li, G., 2011. Kinematics of medial osteoarthritic knees before and after posterior cruciate ligament retaining total knee arthroplasty. J. Orthop. Res. 29, 40–46.
Contents lists available at SciVerse ScienceDirect
Clinical Biomechanics journal homepage: www.elsevier.com/locate/clinbiomech
Mimicking anatomical condylar configuration into knee prosthesis could improve knee kinematics after TKA — A computational simulation Yu-Liang Liu a, e, Wen-Chuan Chen a, b, Wen-Ling Yeh c, Colin Joseph McClean a, b, Chun-Hsiung Huang d, Kun-Jhih Lin a, b, Cheng-Kung Cheng a, b,⁎ a
Orthopedic Biomechanics Laboratory, Institute of Biomedical Engineering, National Yang-Ming University, Taipei, Taiwan Orthopedic Device Research Center, National Yang-Ming University, Taipei, Taiwan Department of Orthopedic Surgery, Chang Gung Memorial Hospital, Taoyuan, Taiwan d Department of Orthopedic Surgery, Mackay Memorial Hospital, Taipei, Taiwan e Department of Product Development, United Orthopedic Corporation, Taipei, Taiwan b c
a r t i c l e
i n f o
Article history: Received 18 May 2011 Accepted 18 August 2011 Keywords: Knee kinematics Different condylar heights Computer simulation TKA
a b s t r a c t Background: Restoration of femoral rollback and tibial internal rotation are two of the major objectives following total knee arthroplasty. Previously, we improved prosthetic knee kinematics by replicating the convexly lateral tibial plateau of intact knee. This study attempted to regain more normal knee kinematics through a posterior cruciate ligament retaining knee, which simultaneously incorporated convexly lateral tibial plateau and anatomical condylar configuration into the prosthesis design. Methods: Computational simulation was utilized to analyze motion of three-dimensional knee models. Three total knee systems with consistent convex insert design but with different condylar heights of 0, 2.7 and 4.7 mm were investigated in present study. Magnetic resonance images of the subject were utilized to construct the bone models and to distinguish the attachment sites of ligaments and tendons. The distal femurs were modeled to rotate about designated flexion axes of femoral components, and the motion of the proximal tibia was unconstrained except further activity of flexion/extension. Movements of the medial/lateral condyles and tibial rotation were recorded and analyzed. Findings: Significant improvements in posterior movement of the lateral condyle and in tibial internal rotation were observed for knee models with different condylar heights, as compared to the knee model with consistent condylar height, when flexion exceeded 100°. Results also revealed that excessive difference in condylar height over anatomical condylar configuration provided no contribution to the restoration of normal knee kinematics. Interpretation: Replicating the morphology of anatomical condylar configuration of the intact knee into knee prostheses could improve knee kinematics during higher knee flexion. © 2011 Elsevier Ltd. All rights reserved.
1. Introduction Regaining a normal joint line is considered a critical factor in achieving successful total knee arthroplasty (TKA) (Anouchi et al., 1993; Arima et al., 1995; Griffin et al., 2000; Partington et al., 1999). One important criterion for recovering an adequate joint line is to restore an anatomical condylar configuration. Anatomical condylar configuration aids in the spiral motion of the tibia about the femur during flexion and extension (Nordin and Frankel, 2002). The spiral motion of the tibiofemoral joint could result from a combination of femoral rollback and tibial internal rotation during knee flexion. Greater femoral rollback would improve the efficiency of quadriceps muscles ⁎ Corresponding author at: Orthopedic Biomechanics Laboratory, Institute of Biomedical Engineering, National Yang-Ming University, No.155, Sec.2, Linong St., Shih-Pai, Taipei, 11221 Taiwan. E-mail address: [email protected] (C.-K. Cheng). 0268-0033/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2011.08.010
(Andriacchi and Mikosz, 1991) and facilitate higher knee flexion (Mikashima et al., 2010). Tibial rotation was associated with deepflexion knee postures, such as squatting, kneeling or lunging (Banks et al., 2003). Nevertheless, most contemporary posterior cruciate ligament retaining (PCR) knees were designed with symmetrical articular surface design and consistent condylar height that could lead to inadequate femoral rollback and insufficient tibial internal rotation (Casino et al., 2009; Dennis et al., 2004; Li et al., 2001; Yue et al., 2011). Therefore, asymmetrical prosthetic designs have been developed to restore more natural knee kinematics, such as motion guiding implants (Walker et al., 2009), medial pivot knee (Moonot et al., 2009) and fully flat insert design (Gomaa and Williams, 2009). Although the abovementioned modifications offer improvements in specific motions, they are still not able to properly recover normal knee kinematics. A possible cause is that the prosthetic geometries of TKAs are inconsistent with the morphology of intact knee. The lateral tibial plateau of intact knee is of convex shape (Bare´ et al., 2006);
Y.-L. Liu et al. / Clinical Biomechanics 27 (2012) 176–181
but most contemporary tibial inserts are symmetrically concave in design. Again, the medial condyle of intact knee is more distal than the lateral condyle. However, most TKAs employ a consistent femoral condylar height to match the symmetrical insert articulation. Previously, we demonstrated that convex lateral insert articulation could improve femoral rollback and tibial internal rotation (Liu et al., 2011). Subsequently, based on a validated knee model, current study attempted to restore greater femoral rollback and tibial internal rotation using a newly developed PCR knee which simultaneously replicated the convexly lateral tibial plateau and anatomical condylar configuration. 2. Methods Bone and cartilaginous models were from a healthy right knee of a female (age: 20 years old, weight: 50 Kg, height: 158 cm), including distal femur, proximal tibia, patella, and fibula. These models were sourced from sagittal magnetic resonance images (MR: SIEMENS MAGNETOM Trio A Tim SYSTEM 3T, Siemens, Germany). The slice interval of MR images was 1 mm with a resolution of 480 × 512 pixels. The bones of the proximal tibia and the distal femur were transected approximately 75 mm from the natural joint line. Thereafter, the bone models were smoothed using Geomagic Studio v9.0 (Parametric Technology Crop., Needham, MA, U.S.A.). Finally, the newly developed knee prostheses were implanted into a bone model and imported into MSC.ADAMS_R3 (MSC Software, Santa Ana, CA, U.S.A.) for dynamic simulation. 2.1. Assignments of ligament, tendon and meniscus of intact knee The posterior cruciate ligament (PCL), medial collateral ligament (MCL), lateral collateral ligament (LCL), patellar and quadriceps tendons were distinguished by self-footprint from MR images of the subject. Origin and insertion points of ligaments and tendons were confirmed by the collaborator who is an experienced orthopedic surgeon (C-H Huang) (Fig. 1). In accordance with the characterization of ligaments in representative literature (Abdel-Rahman and Hefzy, 1998), the PCL was considered as anterior and posterior fiber bundles; the MCL was considered as anterior, deep and oblique fiber bundles, and the LCL was considered as a single fiber bundle. All ligaments were applied to TKA models and simulated as adequate
177
force components with parabolic and linear regions according to the following equations (Abdel-Rahman and Hefzy, 1998): 8 > > <
0 2 K1 Lj −L0j ; j F¼ h i > > : K2j Lj −ð1 þ ε1 ÞL0j
εj ≤ 0 0 b εj ≤ 2ε1 εj b 2ε1
where εj is the strain in the jth element, K1j and K2j are the stiffness coefficients of the jth spring element for the parabolic and linear regions, respectively, and Lj and L0j are its current and slack lengths, respectively. The linear range threshold is specified as ε1 = 0.03. The stiffness of each ligament is shown in Table 1. Additionally, both quadriceps and patellar tendons were considered as medial and lateral fiber bundles and simulated as purely elastic tensile springs. The stiffness coefficients of quadriceps and patellar tendons were 2000 N/mm (Yu et al., 2001) and 1142 N/mm (Hashemi et al., 2005) respectively. 2.2. Construction of TKA models The TKA models included a validated anatomic-like (ALK) model (Liu et al., 2011), an ALK model with a 2.7 mm difference in medial/ lateral condylar heights (ALK_DH_2.7) and an ALK model with a 4.7 mm difference in medial/lateral condylar heights (ALK_DH_4.7). The 2.7 mm difference of condylar heights was measured on the subject in this study and the condylar twist angle was 3.8°. According to the findings of Koudela et al. (2010), the average condylar twist angle of the intact knee is 5.25° ± 1.68°. Therefore, the difference of condylar heights of the ALK_DH_4.7 model was set at 4.7 mm to produce the condylar twist angle of 6.7°, which was the maximum possible condylar twist angle of the intact knee. The thickness of the medial/ lateral femoral condyles of the ALK model was 9 mm, referring to U2 total knee system-cruciate ligament retaining (CR) type (United, Co., Hsinchu, Taiwan). For ALK_DH_2.7 and ALK_DH_4.7 models, the lateral condylar thickness was referenced from the Journey Knee design (Smith & Nephew, Inc., London, UK) and considered as 7 mm. Based on above mentioned differences of condylar heights, the medial condylar thicknesses were simulated as 9.7 mm (7 + 2.7 = 9.7) and 11.7 mm (7 + 4.7 = 11.7) in ALK_2.7 and ALK_4.7 models, respectively. The thicknesses of each medial/lateral femoral condyle of the three models are shown in Fig. 2. Except for the consideration of different
Fig. 1. ALK model, showing the definitions of the knee coordinate system in ± x (flexion/extension axis), ± y (varus/valgus rotation axis) and ± z (external/internal rotation axis). Multiple beads connected by springs were utilized to simulate the wrapping of the quadriceps tendon around the trochlear groove. Attachments of ligaments and tendons were sourced from MR images of the subject and considered as tensile spring-like force elements. Note: P_T = patellar tendon, Q_T = quadriceps tendon, MCL = medial collateral ligament, LCL = lateral collateral ligament, ACL = anterior cruciate ligament, PCL = posterior cruciate ligament.
178
Y.-L. Liu et al. / Clinical Biomechanics 27 (2012) 176–181
Table 1 The stiffness of each ligament in current study (Abdel-Rahman and Hefzy, 1998).
PCL-anterior PCL-posterior MCL-anterior MCL-oblique MCL-deep LCL
K1(N/mm)
K2 (N/mm2)
31.26 19.29 10.00 5.00 5.00 10.00
125.00 60.00 91.25 27.86 21.07 72.22
Note: PCL = posterior cruciate ligament; MCL = medial collateral ligament; LCL = lateral collateral ligament.
condylar heights, the radii of the femoral components and tibial inserts of ALK_DH_2.7 and ALK_DH_4.7 models were consistent with the ALK model. The anterior/posterior dimension of the polyethylene insert and metal baseplate of the ALK model was 47 mm, and the medial/lateral dimension was 69 mm. The sagittal radius of the ALK model's femoral component was 25.7 mm (full extension to 93° flexion), which referred to the radius of the medial condyle of the subject located on the tibio-femoral contact point at full knee extension. Also, the frontal radius of the ALK model's medial femoral condyle was 25.7 mm to produce a spherical geometry in order to regain maximum contact area. A more dished medial insert articular surface also contributes to joint stability during higher knee flexion, particularly during subluxation of the lateral condyle. Conversely, the frontal radius of the lateral femoral condyle of the ALK model was 70 mm. A shallower and more flat insert articular surface allowed rolling and sliding of the lateral condyle to achieve natural knee motion. In accordance with the findings of Sathasivam and Walker (1999), a difference in femoral and tibial frontal radii of 2 mm was helpful in restoring natural knee kinematics after TKA. Therefore, the frontal radii of medial and lateral insert condyles of the ALK model were set as 27.7 mm and 72 mm, respectively. Again, the sagittal radius of the posteromedial insert condyle was 27.7 mm, and the convexly sagittal radius of posterolateral insert condyle was 78 mm, which measured the curvature of the subject's posterolateral tibial plateau located on the tibiofemoral contact point at full knee extension. The reflection point of the ALK model was set on the base of the tibial insert dish and located 19.8 mm from the posterior cliff of the proximal tibia. The radii of the femoral component and tibial insert of the ALK model are shown in Fig. 3.
(Godest et al., 2002). In order to simulate the wrapping of quadriceps tendon around the trochlear groove at higher knee flexion, multiple beads connected by springs were used (Guess et al., 2010; Kessler et al., 2008) (Fig. 1). 2.4. Data acquisition A Cartesian coordinate system was defined by the mediolateral axis (x, flexion/extension axis), the anteroposterior axis (y, varus/ valgus rotation axis), and the longitudinal axis (z, internal/external rotation axis) (Fig. 1). Movements of medial/lateral condyles were defined as the distance between original and present positions of flexion facet centers in the y direction of local coordinates during knee flexion. Internal/external tibial rotation was measured in the z direction on the local coordinate of the tibia relative to the local coordinate of the femur. All data of femoral movements and tibial rotations were recorded every 10° from full extension to 140° of flexion. 3. Results 3.1. Femoral movements Lateral and medial condylar movements of three TKA models are shown in Fig. 4A and B. Compared with the ALK model, the root mean square values of lateral condylar movements were respectively 1.4 and 0.7 mm for ALK_2.7 and ALK_4.7 models during the angular interval 0° to 100° of knee flexion. Again, the maximum differences of lateral condylar movements were 3.5 mm posteriorly (at 100° of knee flexion) and 1.3 mm anteriorly (at 70° of knee flexion) for ALK_2.7 and ALK_4.7 models at less than 100° of knee flexion as compared with the ALK model. Beyond 100° of knee flexion, the maximum increments of posterior movements of the lateral condyle by the ALK_DH_2.7 model were 9.6 and 4.2 mm at 140° of flexion when compared with ALK and ALK_DH_4.7 models, respectively. In medial condylar movements, compared with ALK model, the root mean square values were 1.1 and 1.3 mm respectively for ALK_2.7 and ALK_4.7 models during full range of flexion. Besides, the maximum differences of medial condylar movements were 2.2 mm posteriorly (at 110° of knee flexion) and 3.7 mm anteriorly (at 140° of knee flexion) respectively for ALK_2.7 and ALK_4.7 models, as compared with the ALK model. 3.2. Tibial internal rotations
2.3. Constraints of knee motion The femoral flexion axes of TKA models were in compliance with the condylar radii of femoral components. Femoral flexion was driven by a revolute joint of ADAMS software, and its control type was displacement with a speed of 15 deg/s. Motion of the tibial component was unconstrained, except in flexion–extension. An average ground reaction force of 1.5 × body weight (750 N) (D'Lima et al., 2007) was applied to the center of mass of the tibial component. The friction coefficient of metal-to-polyethylene surfaces was designated as 0.04
The results of tibial internal rotation of the three TKA models are shown in Fig. 4C. Compared with the ALK model, the root mean square values of tibial internal rotations were 0.8° and 0.6° respectively for ALK_2.7 and ALK_4.7 models through the angular interval 0° to 100° of knee flexion. Again, the maximum differences of tibial rotations were 1.7° internally (at 100° of knee flexion) and 1° externally (at 90° of knee flexion) respectively for ALK_2.7 and ALK_4.7 models before 100° of knee flexion as compared with the ALK model. Beyond 100° of knee flexion, tibial internal rotation for the ALK_DH_2.7 and ALK_DH_4.7
Fig. 2. The medial/lateral condylar heights of ALK, ALK_2.7 and ALK_DH_4.7 models.
Y.-L. Liu et al. / Clinical Biomechanics 27 (2012) 176–181
179
Fig. 3. The radii of femoral component and tibial insert of ALK model. (A) sagittal radius; (B) frontal radii of femoral component; (C) frontal radii, (D) sagittal radius of lateral condyle and (E) sagittal radius of medial condyle of tibial insert.
models continuously increased, whereas tibial internal rotation for the ALK model obviously decreased with knee flexion. On average, the ALK_DH_2.7 model has a larger tibial internal rotation than the ALK_DH_4.7 and ALK models. The maximum increments of tibial internal rotation by ALK_DH_2.7 were 15.4° and 2.2° at 140° of flexion when compared with the ALK and ALK_DH_4.7 models, respectively. 4. Discussion Restoration of the femoral rollback and tibial internal rotation are two major goals for contemporary knee prosthetic design with the aim of regaining normal knee kinematics. However, many studies have demonstrated that femoral rollback and tibial internal rotation were inadequate or abnormal after replacement by contemporary PCR TKA (Casino et al., 2009; Dennis et al., 2004; Li et al., 2001; Yue et al., 2011). Our previous study demonstrated that the inclusion of convexly lateral insert articular surface would improve femoral rollback and tibial internal rotation (Liu et al., 2011). However, it was also observed that tibial internal rotation on convex insert design reduced significantly beyond 100° of flexion. Therefore, the newly developed PCR knee incorporated convexly lateral insert articular surface and anatomical condylar configuration simultaneously into the prosthetic design to enhance femoral rollback and tibial internal rotation. Previously, in order to resolve the problem of inadequate knee kinematics after replacement by contemporary TKA, asymmetrical prosthetic designs were developed, such as the motion guiding prosthetic design (Walker et al., 2009), medial pivot knee (Moonot et al., 2009) and flat insert articular surface design (Gomaa and Williams, 2009). Although the abovementioned modifications offer improvements in specific motions, they are still unable to properly recover normal knee kinematics. A possible cause is that the prosthetic geometries of TKAs are inconsistent with the morphology of intact knee. The morphology of the posterolateral tibial plateau is convex, but the abovementioned TKAs were designed with concave or flat insert articulation. Li et al. (2001) suggested that the concave posterior portion of the polyethylene component may serve as a major resistance to posterior femoral translation when compared with the morphology of lateral tibial plateau of intact knee. Again, the
flat insert articulation could improve femoral rollback but may suffer from a sense of instability due to paradoxical anterior sliding (Gomaa and Williams, 2009). Besides, the anatomical medial condyle is more distal than the lateral condyle; nevertheless, the abovementioned TKAs were designed with consistent condylar heights. Essentially, the convexly lateral insert articular surface allowed the lateral femoral condyle to sublux off the back of the tibial plateau during high flexion. This is a characteristic of normal knee kinematics (Bare´ et al., 2006). Additionally, the convex insert articular surface was able to increase the tibial posterior clearance, which allowed deeper knee flexion without interference between the distal femur and tibial insert. As for the anatomical condylar configuration, it was associated with the reconstruction of the joint line. Previous literature reported that inappropriate location of the joint line would result in higher patellofemoral contact forces (Singerman et al., 1994), an unstable knee joint after TKA (Laskin, 2002) and patellar dislocation (Insall et al., 1972). Although traditional TKA designs also restored the location of the joint line at knee full extension, they still could not completely reproduce natural motion patterns during knee flexion. That could be caused by the geometries of symmetrical femoral condyles of most contemporary TKAs being inconsistent with intact knee. Conversely, the anatomical condylar configuration was demonstrated to be helpful in reproducing the spiral motion of the tibia about the femur (Nordin and Frankel, 2002). The spiral motion of the tibiofemoral joint should be a combination of femoral movement and tibial rotation during knee flexion/extension, and normal spiral motion should be beneficial in regaining normal knee kinematics. Although many surgeons believe that soft tissue balance is one of the most important factors in achieving a successful surgery, soft tissues cannot completely recover their original performance after long-term abnormal weight bearing. Walker (2001) also took into account that characteristics of normal knee motion could be achieved by modifying prosthetic features alone. Recently, a convex insert articular surface and different condylar heights were applied to commercial knee prostheses, such as the Journey Knee. However, the Journey Knee was a bi-cruciate stabilized knee system with an insert post and femoral cam that differed from the newly developed PCR knees in the current study. Additionally, the FINE Total Knee System-PCR type (Nakashima
180
Y.-L. Liu et al. / Clinical Biomechanics 27 (2012) 176–181
(-) Posterior <---> (+) Anterior (mm)
(A) Knee Flexion (degrees) ALK
ALK_DH_2.7
ALK_DH_4.7
0.0 -5.0 0
20
40
60
80
100
120
140
-10.0 -15.0 -20.0 -25.0 -30.0 -35.0
(-) Posterior <---> (+) Anterior (mm)
(B) Knee Flexion (degrees) ALK
4.0
ALK_DH_2.7
ALK_DH_4.7
2.0 0.0 -2.0 0
20
40
60
80
100
120
140
-4.0 -6.0 -8.0 -10.0 -12.0
(C) Knee Flexion (degrees) ALK
Tibial Rotation (degrees)
25.0
ALK_DH_2.7
ALK_DH_4.7
20.0 15.0 10.0 5.0 0.0
-5.0
0
20
40
60
80
100
120
140
Fig. 4. Comparisons of (A) lateral condylar movements, (B) medial condylar movements and (C) tibial internal rotations of ALK, ALK_DH_2.7 and ALK_DH_4.7 models.
Medical, Okayama, Japan) considered an anatomical condylar configuration of the femoral component. However, the flat lateral insert articular surface was still different from the convex lateral tibial plateau of intact knee. To our knowledge, there are no commercial PCR knee designs that simultaneously mimic the convexly lateral tibial plateau and anatomical condylar configuration of intact knee. In current study, the ligament condition referred to the research of Abdel-Rahman and Hefzy (1998) which was inconsistent with other relative publications by Blankevoort and Huiskes (1996) and Li et al. (1999) regarding ligament quantity and ligament force equation. For ligament quantity, the LCL was simulated as a single bundle in our study but was simulated as three bundles by Blankevoort and Huiskes (1996) and Li et al. (1999). Furthermore, our study excluded the MCL capsule from the simulation, which was inconsistent with the research of Li et al. (1999). Although the quantity of ligaments in current study was not completely consistent with the abovementioned two literatures, it was compatible with the research of Abdel-Rahman and Hefzy (1998) and Yu et al. (2001). For ligament force equations, the measurements in our study were relative to the variation of ligament length, but
the research of Blankevoort and Huiskes (1996) and Li et al. (1999) were relative to the variation of ligament strain. Although the ligament equation in our study was inconsistent with the two abovementioned studies, the ligament performance should be similar, because they referred to the consistent reference of Wismans et al. (1980). As can be seen from the results, at knee flexion beyond 100°, the ALK_DH_2.7 and ALK_DH_4.7 models showed significantly improved posterior movement of the lateral condyle and tibial internal rotation when compared with the ALK model, particularly the ALK_DH_2.7 model. This revealed that excessive difference in condylar heights contributes nothing to the restoration of normal knee kinematics, such as in the ALK_DH_4.7 model. Even so, the results showed that incorporating an anatomical condylar configuration into the TKA prosthetic design could improve femoral rollback and tibial internal rotation at higher knee flexion. Consulting the literature review, Miyazaki et al. (2011) investigated the knee kinematics of FINE PCR Total Knee System, ADVANCE Total Knee System (Wright Medical Technology, Arlington, Tenn.), and the ADVANTIM Total Knee System (Wright Medical Technology, Arlington, Tenn.). The results demonstrated that the FINE knee
Y.-L. Liu et al. / Clinical Biomechanics 27 (2012) 176–181
displayed more suitable knee kinematics, including femoral rollback and tibial internal rotation, than the ADVANCE and ADVANTIM knees, by using flat lateral insert articulation and anatomical condylar configuration, consistent with the findings of the current study. Following the design rationale of the Journey Knee, there was a significant improvement in anterior/posterior translation over a traditional knee. Moreover, in femoral axial rotation, the Journey Knee kept the femoral external rotation beyond 110° of knee flexion. Conversely, femoral external rotation of traditional PS TKA designs decreased noticeably beyond 110° of knee flexion. Our study also revealed that an anatomical condylar configuration could improve tibial internal rotation particularly at high knee flexion, consistent with the Journey Knee. One of the limitations in current study was that the differences of condylar heights between patients were inconsistent; therefore collection of a large quantity of data on anatomical condylar configuration was suggested. Also, the balance of LCL and MCL soft tissue is another concern following TKA with a different condylar height design. Generally, the tensile strengths of LCL and MCL were considered balanced for contemporary TKAs. However, the newly developed PCR knee in current study was simulated with an anatomical condylar configuration, and the medial/lateral condylar thicknesses were inconsistent. The influence of anatomical condylar configuration prosthetic design on tensile strength of LCL and MCL is still unclear and should be resolved through further clinical study. Furthermore, the anatomical condylar configuration was only utilized in the ALK model, so the influence on different prosthetic designs is still unknown. Although the numerical simulation cannot take the mechanical properties of soft tissue, such as visco-elasticity into consideration, identical simulated criteria of models should be able to distinguish the influence of different prosthetic designs on knee kinematics. Finally, the femoral component and the tibial insert were simulated with an optimal alignment in current study. The influence of mal-alignment between femoral component and tibial insert on knee kinematics needs further study. 5. Conclusions Modifying the geometry of knee prosthetic design in order to restore normal femoral rollback and tibial internal rotation seems to be the current trend in designing commercial knee prostheses. Our previous study demonstrated that convex insert articular surface could improve knee kinematics after TKA. This study revealed that incorporating convex lateral insert articulation and anatomical condylar configuration could enhance more femoral rollback and tibial internal rotation during higher knee flexion. In order to ensure the efficacy of this newly developed CR total knee system, further investigation in clinical trials and follow-ups will be necessary. Conflict of interest statement There are no conflicts of interest in this manuscript. Acknowledgments We are pleased to acknowledge the financial support of the National Science Council, Taiwan, ROC (NSC99-2911-I-009-101) and the Administrative Bureau of Southern Taiwan Science Park (BY-03-04-17-98). References Abdel-Rahman, E.M., Hefzy, M.S., 1998. Three-dimensional dynamic behaviour of the human knee joint under impact loading. Med. Eng. Phys. 20, 276–290. Andriacchi, T.P., Mikosz, R.P., 1991. Musculoskeletal dynamics, locomotion and clinical applications. In: Mow, V.C., Hayes, W.C. (Eds.), Basic Orthopedic Biomechanics. Raven Press, New York, pp. 51–92.
181
Anouchi, Y.S., Whiteside, L.A., Kaiser, A.D., Milliano, M.T., 1993. The effects of axial rotational alignment of the femoral component on knee stability and patellar tracking in total knee arthroplasty demonstrated on autopsy specimens. Clin. Orthop. Relat. Res. 287, 170–177. Arima, J., Whiteside, L.A., McCarthy, D.S., White, S.E., 1995. Femoral rotational alignment, based on the anteroposterior axis, in total knee arthroplasty in a valgus knee. A technical note. J. Bone Joint Surg. Am. 77, 1331–1334. Banks, S.A., Bellemans, J., Nozaki, H., Whiteside, L.A., Harman, M., Hodge, W.A., 2003. Knee motions during maximum flexion in fixed and mobile-bearing arthroplasties. Clin. Orthop. Relat. Res. 410, 131–138. Bare´, J.V., Gill, H.S., Beard, D.J., Murray, D.W., 2006. A convex lateral tibial plateau for knee replacement. Knee 13, 122–126. Blankevoort, L., Huiskes, R., 1996. Validation of a three-dimensional model of the knee. J. Biomech. 29, 955–961. Casino, D., Zaffagnini, S., Martelli, S., Lopomo, N., Bignozzi, S., Iacono, S., Russo, A., Marcacci, M., 2009. Intraoperative evaluation of total knee replacement: kinematic assessment with a navigation system. Knee Surg. Sports Traumatol. Arthrosc. 17, 369–373. Dennis, D.A., Komistek, R.D., Mahfouz, M.R., Walker, S.A., Tucker, A., 2004. A multicenter analysis of axial femorotibial rotation after total knee arthroplasty. Clin. Orthop. Relat. Res. 428, 180–189. D'Lima, D.D., Patil, S., Steklov, N., Chien, S., Colwell Jr., C.W., 2007. In vivo knee moments and shear after total knee arthroplasty. J. Biomech. 40, 11–17. Godest, A.C., Beaugonin, M., Haug, E., Taylor, M., Gregson, P.J., 2002. Simulation of a knee joint replacement during a gait cycle using explicit finite element analysis. J. Biomech. 35, 267–275. Gomaa, S.T., Williams, J.L., 2009. Effect of tibial insert geometry on kinematics following fixed-bearing cruciate-retaining total knee arthroplasty. 55th Annual Meeting of the Orthopedic Research Society. Paper No.191. Griffin, F.M., Math, K., Scuderi, G.R., Insall, J.N., Poilvache, P.L., 2000. Anatomy of the epicondyles of the distal femur: MRI analysis of normal knees. J. Arthroplasty 15, 354–359. Guess, T.M., Thiagarajan, G., Kia, M., Mishra, M., 2010. A subject specific multibody model of the knee with menisci. Med. Eng. Phys. 32, 505–515. Hashemi, J., Chandrashekar, N., Slauterbeck, J., 2005. The mechanical properties of the human patellar tendon are correlated to its mass density and are independent of sex. Clin. Biomech. 20, 645–652. Insall, J., Goldberg, V., Salvati, E., 1972. Recurrent dislocation and the high-riding patella. Clin. Orthop. Relat. Res. 88, 67–69. Kessler, O., Patil, S., Colwell Jr., C.W., D'Lima, D.D., 2008. The effect of femoral component malrotation on patellar biomechanics. J. Biomech. 41, 3332–3339. Koudela Jr., K., Koudelová, J., Koudela Sr., K., Kunešová, M., Křen, J., Pokorný, J., 2010. Radiographic measurements in total knee arthroplasty and their role in clinical practice. Acta Chir. Orthop. Traumatol. Cech. 77, 304–311. Laskin, R.S., 2002. Joint line position restoration during revision total knee replacement. Clin. Orthop. Relat. Res. 2002, 169–171. Li, G., Gil, J., Kanamori, A., Woo, S.L., 1999. A validated three-dimensional computational model of a human knee joint. J. Biomech. Eng. 121, 657–662. Li, G., Zayontz, S., Most, E., Otterberg, E., Sabbag, K., Rubash, H.E., 2001. Cruciate-retaining and cruciate-substituting total knee arthroplasty: an in vitro comparison of the kinematics under muscle loads. J. Arthroplasty 16, 150–156. Liu, Y.L., Lin, K.J., Huang, C.H., Chen, W.C., Chen, C.H., Chang, T.W., Lai, Y.S., Cheng, C.K., 2011. Anatomic-like polyethylene insert could improve knee kinematics after total knee arthroplasty — a computational assessment. Clin. Biomech. 26, 612–619. Mikashima, Y., et al., 2010. In vivo deep-flexion kinematics in patients with posteriorcruciate retaining and anterior-cruciate substituting total knee arthroplasty. Clin. Biomech. 25, 83–87. Miyazaki, Y., Nakamura, T., Kogame, T., Saito, M., Yamamoto, K., Suguro, T., 2011. Analysis of the kinematics of total knee prostheses with a medial pivot design. J. Arthroplasty 26, 1038–1044. Moonot, P., Mu, S., Railton, G.T., Field, R.E., Banks, S.A., 2009. Tibiofemoral kinematic analysis of knee flexion for a medial pivot knee. Knee Surg. Sports Traumatol. Arthrosc. 17, 927–934. Nordin, M., Frankel, V.H., 2002. Basic Biomechanics of the Musculoskeletal System, third ed. Lippincott Williams & Wilkins, New York. Partington, P.F., Sawhney, J., Rorabeck, C.H., Barrack, R.L., Moore, J., 1999. Joint line restoration after revision total knee arthroplasty. Clin. Orthop. Relat. Res. 367, 165–171. Sathasivam, S., Walker, P.S., 1999. The conflicting requirements of laxity and conformity in total knee replacement. J. Biomech. 32, 239–247. Singerman, R., Davy, D.T., Goldberg, V.M., 1994. Effects of patella alta and patella infera on patellofemoral contact forces. J. Biomech. 27, 1059–1065. Walker, P.S., 2001. A new concept in guided motion total knee arthroplasty. J. Arthroplasty 16, 157–163. Walker, P.S., Sussman-Fort, J.M., Yildirim, G., Boyer, J., 2009. Design features of total knees for achieving normal knee motion characteristics. J. Arthroplasty 24, 475–483. Wismans, J., Veldpaus, F., Janssen, J., Huson, A., Strulen, P., 1980. A three dimensional mathematical model of the knee joint. J. Biomech. 13, 677–686. Yu, C.H., Walker, P.S., Dewar, M.E., 2001. The effect of design variables of condylar total knees on the joint forces in step climbing based on a computer model. J. Biomech. 34, 1011–1021. Yue, B., Varadarajan, K.M., Moynihan, A.L., Liu, F., Rubash, H.E., Li, G., 2011. Kinematics of medial osteoarthritic knees before and after posterior cruciate ligament retaining total knee arthroplasty. J. Orthop. Res. 29, 40–46.