- Email: [email protected]
Collateral ligament length change patterns after joint line elevation may not explain midflexion instability following TKA
Medical Engineering & Physics 33 (2011) 1303–1308
Contents lists available at ScienceDirect
Medical Engineering & Physics journal homepage: www.elsevier.com/locate/medengphy
Collateral ligament length change patterns after joint line elevation may not explain midflexion instability following TKA Christian König, Georg Matziolis, Alexey Sharenkov, William R. Taylor, Carsten Perka, Georg N. Duda, Markus O. Heller ∗ Julius Wolff Institute and Center for Musculoskeletal Surgery, Charité – Universitätsmedizin Berlin, Center for Sports Science and Sports Medicine Berlin, Philippstr. 13, Haus 11, 10115 Berlin, Germany
a r t i c l e
i n f o
Article history: Received 27 February 2010 Received in revised form 15 June 2011 Accepted 18 June 2011 Keywords: Knee TKA Joint line Stability Collateral ligaments Midflexion instability
a b s t r a c t Midflexion instability (MFI) after TKA is a phenomenon often described as varus–valgus instability between 30◦ and 45◦ knee flexion. The exact mechanisms causing MFI remain unclear, but elevation of the joint line (JLE) may be one possible cause. In an in silico approach using 4 subject specific musculoskeletal models, the length change patterns of the collateral ligaments during knee flexion (relative to the extended knee) were calculated for the anatomically reconstructed joints as well as for JLEs of 5 and 10 mm. Analysis of the distance between the ligaments’ attachment sites (DA) in midflexion revealed a relative decrease in DA magnitude after JLE for both collateral ligaments in comparison to the anatomically reconstructed knee. This finding suggests that JLE could contribute to MFI. However, the anterior ligament regions also experienced a DA increase (MCL) or only a slight DA decrease (LCL) for each JLE simulated. From this perspective, the anterior ligament portions are unlikely to slacken in midflexion and JLE is unlikely to contribute greatly to MFI. In conclusion, our findings did not support the idea that JLE is a major contributor to midflexion instability for this particular ultra-congruent implant design. © 2011 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Careful balancing of the soft-tissue stabilizers is essential for the success of total knee arthroplasty (TKA) [1–3]. In particular in non- or semi-constrained implants, the lateral collateral ligament (LCL) and the medial collateral ligament (MCL) ensure the knee’s varus/valgus stability while the MCL also resists external rotations and anterior–posterior translations [4,5]. Even though gap and ligament balancing techniques have been established to create a stable joint in extension and flexion [1,3,6–8], joint stability related problems account for a significant proportion of TKA revisions [9–15]. Of particular concern here is the varus–valgus instability of the knee in flexion angles between 30 and 45◦ , also referred to as midflexion instability (MFI) [16]. Even knees that have been carefully balanced in extension and at 90◦ flexion can exhibit MFI. While the exact mechanisms leading to MFI are not fully understood, elevation of the joint line (JLE) after TKA has been suggested as a potential contributor to MFI [16,17]. An approach towards a better understanding of MFI can be derived from the work of Amis and Zavras [18], where it was found that the location of the ligaments’ femoral
∗ Corresponding author. Tel.: +49 30 2093 46128; fax: +49 30 2093 46001. E-mail address: [email protected] (M.O. Heller).
attachment sites relative to the knee flexion axis significantly influences the length change patterns of the cruciate ligaments during flexion. Since JLE alters the knee’s flexion axis relative to the location of the collateral ligaments’ femoral attachment sites, it is likely that JLE can induce altered lengthening or shortening characteristics of the collateral ligaments at different flexion angles [19,20]. The function of the collateral ligaments, which is critical for providing varus–valgus stability of the knee during flexion, might thus be compromised at certain points of the flexion/extension cycle. Computer modelling opens up the possibility to systematically study the influence of selected parameters. By studying multiple subjects using patient-specific models, a robust response as to whether JLE results in a substantially larger distance decrease between the femoral and tibial collateral ligament attachment sites in midflexion could be obtained and compared to provide an understanding of the support provided by the passive soft tissue structures at different knee flexion angles. We hypothesised that JLE and the subsequent change in the position of the femoral collateral ligament attachment sites relative to the femoral articulating surface alters the length change patterns of the collateral ligaments during knee flexion, specifically in midflexion. The aim of this study was therefore to analyze the length change patterns of the collateral ligaments during knee flexion for situations of JLE after TKA in order to assess whether JLE can indeed contribute to midflexion instability.
1350-4533/$ – see front matter © 2011 IPEM. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.medengphy.2011.06.008
1304
C. König et al. / Medical Engineering & Physics 33 (2011) 1303–1308
Fig. 2. The medial and lateral collateral ligaments were each modelled by 5 portions to better characterize their functional behaviour throughout flexion. To this end, the femoral origin and tibial attachment sites of the medial (left) and the lateral (right) collateral ligaments were identified using landmarks. For the MCL, static wrapping points were introduced at the proximal tibia. The current study focussed on the analyses of the most anterior and posterior ligament regions (thick lines).
the subject specific scaling factors, a single implant size was found to be adequate for all four subjects when the virtual implantation was performed and verified by an experienced surgeon. Fig. 1. The medial and lateral collateral ligaments (here displayed together with the cruciate ligaments) were modelled using high resolution cross sectional images of the Visible Human (VH) dataset.
2. Methods 2.1. Musculoskeletal model The computer model that formed the basis for this study was derived from the Visible Human (VH) CT dataset [21] and literature data [22], and contained bony structures, muscles and simplified representations of the ligament insertion sites of the lower limb [23]. To determine the length change patterns of the collateral ligaments during knee flexion the musculoskeletal model was complemented with a more detailed description of the collateral ligaments. The high resolution colour images (resolution 0.32 mm × 0.32 mm × 1 mm) of the same VH dataset that was used to create the reference musculoskeletal model were employed to localize the collateral ligaments [24–29]. This allowed the 3D modelling of the medial and lateral collateral ligaments (Fig. 1) using a 3D visualisation and volume modelling Software (Amira, Visage Imaging, Berlin, Germany). In the combined 3D models of the tibia, femur and the collateral ligaments, the bony attachment sites of the medial and lateral collateral ligaments were each modelled in five portions to better characterize their functional behaviour throughout knee flexion. The origin and insertion of each portion were identified with landmarks to allow calculation of the relative change in length of the ligaments (Fig. 2) [24–30]. For the MCL, additional static wrapping points were defined at the proximal tibia. This more detailed reference model was then adapted to match the anatomy of four participating subjects using linear scaling [23,31]. As a result of adapting the reference models’ femur and tibia using subject specific scaling parameters, variation in the insertions of the knee ligaments across the four subjects was captured in the models. Subsequently, a virtual total knee arthroplasty using an ultra-congruent, fixed bearing and cruciate sacrificing implant design (Columbus UC, Aesculap AG, Tuttlingen, Germany) was performed on the same four musculoskeletal models [32]. Although the bone sizes of the four subjects varied somewhat as a result of
2.2. Variation of the joint line JLE can result from a number of different clinical conditions, including, in the case of distal femoral bone loss, a too small tibial resection, the use of a too high inlay, or combinations thereof. For the current study, it was assumed that the tibial cut was adequate and JLE was a result of modifying the proximal–distal position of the femoral component and the height of the inlay [32]. According to the manufacturer’s recommendation, the femoral component was downsized with increased JLE. Thus the component’s anterior–posterior dimension was reduced by the same amount that the JL was elevated, in order to avoid tightening of the flexion gap that would otherwise result from JLE. In this study, JLE of 5 mm and 10 mm was simulated. The leg length remained unchanged in all simulations. 2.3. Kinematic adaptation The tibio-femoral kinematics of the four subjects were adapted to the postoperative situation by using a kinematic model that reflected the geometry of the ultra-congruent prosthesis’ articulating surfaces [32]. In agreement with the geometric design of the femoral component that possesses three different radii, this model used three knee flexion angle dependent axes of rotation to describe the relative positions of the components throughout the range of knee flexion. Since the axis of rotation of these radii did not coincide, anterior–posterior, as well as superior–inferior translation was captured in the kinematic model (albeit as secondary effects) according to the prescribed relative motion of the knee around the three radii of the femoral component. 2.4. Collateral ligament length change To assess the length change patterns in the anterior and posterior regions of both the MCL and LCL, the distances between pairs of landmarks that represent the extremes of the individual ligament insertion sites (anterior, posterior) were analyzed in both collateral ligaments [30,33]. At first, in the anatomically reconstructed knee, the distance between the attachment sites (DA) was computed for all four sub-
C. König et al. / Medical Engineering & Physics 33 (2011) 1303–1308
jects during knee flexion from 0◦ to 90◦ , using an increment of 15◦ . Here, by keeping the leg length constant also the reference DA at extension remained unaltered for all JL conditions. Then the relative change in this distance in reference to the extended knee (0◦ position) was calculated. These calculations were repeated for all modifications of the JL as described previously, resulting in a simulation of 12 cases in total (4 subjects, 3 conditions each). Furthermore, at every analyzed flexion angle, the DA changes that occurred during flexion with an elevated JL were compared to the DA changes in the anatomically reconstructed knee. 3. Results
1305
erence implantation. This relative decrease was somewhat larger than the relative DA decrease observed as a result of JLE at 90◦ knee flexion (2.0 ± 0.2% at 5 mm and 2.8 ± 0.5% at 10 mm JLE). In the posterior MCL, the relative DA decrease caused by JLE was also more pronounced in midflexion than at higher flexion angles, compared to the anatomical reconstruction. For a JLE of 10 mm, a DA decrease of 4.2 ± 0.1% was observed at 45◦ flexion, while at 90◦ flexion this reduced to 1.0 ± 0.9%. In the anterior LCL, a JLE of 10 mm decreased the DA in midflexion by 2.6 ± 0.1% at 45◦ flexion compared to the anatomical reconstruction. In midflexion, there was a slight DA decrease in the posterior regions of the LCL compared to the anatomical reconstruction (1.9 ± 0.03% at 5 mm and 3.0 ± 0.1% at 10 mm JLE (both at 45◦ flexion)).
3.1. Anatomical reconstruction If the JL was anatomically reconstructed, the distance between the femoral and tibial attachments in the anterior MCL increased by 13.5 ± 0.4% compared to its reference length at extension when flexing the knee from 0◦ to 90◦ (Fig. 3). During the same knee flexion cycle the MCL’s posterior regions experienced a distance decrease relative to their reference length at full extension of 8.1 ± 1.1%. In the anterior region of the LCL the DA remained approximately constant, while the posterior LCL experienced a 14.6 ± 2.2% decrease in distance compared to its reference length at full extension when the knee was flexed to 90◦ . 3.2. DA changes after JLE in midflexion In midflexion, JLE generally resulted in a reduced DA in the anterior and posterior regions of both collateral ligaments relative to the anatomically reconstructed knee joint: while the DA increase during flexion (expressed relative to its reference length at full extension) in the anterior MCL of the anatomically reconstructed joint was 7.9 ± 0.4% at 45◦ knee flexion, the DA at 10 mm JLE was 3.9 ± 0.3%, indicating a 4.0 ± 0.1% DA decrease relative to the ref-
3.3. DA changes after JLE in relation to the conditions of the extended knee Despite the fact that with JLE the DA in the anterior MCL was reduced in midflexion when compared to its DA in the anatomically reconstructed condition, the DA during knee flexion did not fall below the reference DA obtained for the fully extended knee. For a JLE of 10 mm, the DA in the anterior LCL in midflexion fell slightly below the reference DA in extension (−2.1 ± 0.6% at 30◦ and −1.0 ± 1.1% at 45◦ flexion), while the DA for 5 mm JLE was −0.9 ± 0.6% at 30◦ and remained virtually unchanged compared to the conditions in extension for a flexion angle of 45◦ . In the posterior regions of the MCL and the LCL, the DA in midflexion was always found to be below the reference DA of the extended knee. This was true for the anatomically reconstructed knee as well as for the conditions after JLE. 4. Discussion Even though gap and ligament balancing techniques are established steps in TKA for creating a stable joint in both static and
Fig. 3. Length change patterns of the anterior (left) and posterior (right) regions of the medial (top) and lateral (bottom) collateral ligaments from full extension to 90◦ of knee flexion. The dashed curve represents the anatomical reconstruction of the joint line, while the light gray and the gray curves represent joint line elevations of 5 and 10 mm respectively.
1306
C. König et al. / Medical Engineering & Physics 33 (2011) 1303–1308
dynamic conditions [1,3,6–8], stability related problems are still a major reason for a TKA revision [9–11]. It has been speculated that an elevated joint line (JL) might contribute to midflexion instability [16,17]. However, little is known regarding effects of JLE on the collateral ligaments, which are key passive stabilizers of the knee, specifically in midflexion. Using four musculoskeletal models of the lower limb post TKA [32], we have demonstrated for an ultra-congruent implant design that in midflexion an elevation of the JL resulted in a reduced distance between the attachment sites of the anterior and posterior regions in both collateral ligaments when compared to the anatomically reconstructed condition. This finding alone would indeed suggest that JLE can contribute to MFI. Although smaller, the DA of the anterior MCL still exhibited a distance increase between the attachment sites during flexion, hence indicating support from the passive soft tissue structures at midflexion, even in the JLE condition. Thus, as in the intact knee where the MCL is known to be taut in extension and becomes more tight during flexion [5,33,34], the MCL displayed the characteristics of a tensed ligament in midflexion. The anterior LCL, which is taut throughout flexion in the intact knee [5], was also affected by JLE and experienced a distance decrease in midflexion compared to the anatomically reconstructed JL. In contrast to the anterior MCL, however, the DA in the anterior LCL decreased below the DA level in the extended knee. In the worst case, the anterior LCL is only just taut in extension and not pre-tensioned and would therefore become slack with the maximum observed 2.1% DA decrease (10 mm JLE at 30◦ ). Such a condition would result in a varus/valgus laxity of less than 1.5◦ when considering an average ligament length of 60 mm and a knee width of 55 mm. In the intact knee, the posterior fibres of the MCL and LCL slacken with increasing flexion angles [4,5,26,33], a behaviour that was also predicted by our model for the anatomically reconstructed knee. In JLE conditions this effect was intensified in midflexion. How-
ever, it is rather unlikely that a further slackening of already slack posterior fibres would contribute to any form of instability. JLE therefore did not seem to be a major contributor to midflexion instability when using the Columbus ultra-congruent prosthesis design. Nevertheless, while the observed alterations in collateral ligament length change patterns in cases of JLE are unlikely to contribute greatly to midflexion instability, they may still have an impact on the overall ligament function and should therefore be subject to further investigation. For example the generally decreased DA in lower flexion angles and in the case of the MCL also in deeper flexion, may imply that fewer regions of the ligament can contribute to load sharing, possibly increasing the risk of overloading within the remaining active structures. In addition, the observed distance increase in the anterior LCL at higher flexion angles even above the DA of the extended knee in which these areas of the ligament are already taut, may be a possible cause for complications. Further complications may also arise when the JL is moderately elevated but the femoral component is not properly downsized. Compared to the anatomically reconstructed joint at 90◦ with 5 mm JLE the DA increased in the anterior MCL by 4.7 ± 0.5% to a total of 18.2 ± 0.5% and to a total of 9.9 ± 0.8% in the anterior LCL (Fig. 4). Considering that the collateral ligaments are taut in extension [4,5,26,28], it seems safe to assume that any further increase in DA would be linked to an increase in the ligament strain. It is therefore reasonable to assume that the DA increase observed in this study, specifically in the anterior MCL, is indicative of a strain increase in this area of the ligament. While the absolute ligament strain cannot be determined with the methods presented here, it is very well possible that as a result of the DA increases of as much as 18%, JLE may cause strains in the ligaments which are approaching the magnitude of the maximum tensile strain of ligaments, reported to be between 13 and 17% [35,36].
Fig. 4. Length change patterns of the collateral ligaments for the conditions of joint line elevation and no downsizing of the femoral component. The graphs show the length change patterns for the anterior (left) and posterior (right) regions of the medial (top) and lateral (bottom) collateral ligaments from full extension to 90◦ flexion. The dashed curve represents the anatomical reconstruction of the joint line, while the solid curve represents a joint line elevation of 5 mm. The horizontal dotted line represents an average value for maximum tensile strain of the collateral ligaments as reported in the literature.
C. König et al. / Medical Engineering & Physics 33 (2011) 1303–1308
Even though global pre-operative lower limb kinematics for walking, stair climbing and knee bend activities were originally captured for the same four subjects using a gait analysis system [23], no detailed general postoperative in vivo kinematic patterns have been presented so far in the literature that allow for subject specific modelling of postoperative kinematics [37–40]. Although ultra-congruent implant designs were introduced to provide enhanced joint stability, little is known about the actual interactions of such designs with the collateral ligaments. Investigations into such an implant design that offers a high level of constraint could therefore provide new insights into the performance of such designs, but also enable a more robust approximation of the postoperative kinematics than would have been possible by using a more conventional, less constrained implant design. The current study used a kinematic model that was limited to sagittal plane motion, since preliminary fluoroscopic analyses performed in our lab indicate that only little internal–external rotation occurs in vivo with such an ultra-congruent fixed bearing design. As a result, internal–external movement was considered secondary for these analyses and therefore not included. Here, an isolated axial rotation of 5◦ would result in a DA increase of approximately only 0.3% (assuming an average ligament length of 60 mm and a knee width of 55 mm and an eccentric rotation axis). While the knee kinematics were thus generic to the implant rather than subject specific, the individual scaling of the musculoskeletal models reflected the anatomical variation in the relative position of the ligament attachment sites with respect to the femoral component across the four subjects. Therefore, the models did include essential subject-specific variations in those musculoskeletal structures that were key for the current study. Observing the distance between ligament attachment sites to characterize the function of the ligaments [30,33] has the limitation that the zero strain conditions of the ligaments are difficult to obtain [4] and any initial strain within the ligament as well as the condition of a slack ligament is not considered. However, since the extended knee, in which the collateral ligaments are taut or strained [4,5,26,28] is used as a reference in this study, any assumptions on ligament strains would be rather under- than overestimated. Furthermore, we did not directly compare the DA length change patterns in the anatomically reconstructed compared to the healthy unimplanted knee. While we can therefore not rule out that the anatomical reconstruction resulted in MFI, the length change patterns of the collateral ligaments during flexion (i.e. after an ideal TKA that maintained the joint line) were found not to differ substantially from those patterns for the intact knee: in particular, the distances in the anterior proportions of the ligament attachments in midflexion never fell below the values in extension. Following the literature that suggests that the ligaments’ anterior portions are taut in extension [34,41–44] and also during flexion [5,33], these results suggest that the ligaments also stay taut after the implantation of the prosthesis. Although this does not provide direct proof, this reasoning strongly suggests that it is unlikely that instability in midflexion would be caused by the ideally implanted prosthesis. Amis and co-worker [18] already emphasized the importance of the locations of the ligament attachment sites relative to the femoral flexion axes, since these are closely associated with the length change patterns of the ligaments. Implant designs with different condylar radii, different sizing of the implant and patient specific variation in the location of ligament attachment zones may further alter the ligament length change patterns in midflexion and warrant further investigation. The influence of the dynamics of the tibial wrapping points was not further analyzed in this study, since the main influence of JLE on the ligaments’ DA is the changing position of the femoral ligament attachments relative to the rotation axis of the femoral component.
1307
In this study we assumed that a well balanced knee was achieved for the reference implantation in all subjects, without requiring excessive soft tissue management. Furthermore, in the specific cases of 5 and 10 mm JLE, the tightening of the flexion gap can be exactly compensated by an equivalent reduction of the femoral components’ anterior–posterior dimension through the implantation of a smaller implant size. In real life, the amount of JLE would not always exactly match the available implant sizes, and would thus very likely necessitate some additional soft tissue balancing. In practice, a surgeon would try to balance the soft tissues in both extension and 90◦ flexion during TKA by appropriate release techniques. However, the possibilities to accurately simulate these finer details of the surgical release techniques in silico are certainly limited. The conditions analyzed here therefore represent a rather specific situation for which no additional soft tissue balancing is required. However, by focusing the study on elevation levels that could be matched by appropriate changes in the anterior–posterior femoral implant dimension, the number of unknown parameters was minimized. This facilitated the study of the influence of JLE on the behaviour of the collateral ligaments under well controlled conditions, and was therefore considered an acceptable approximation of the clinical situation. As hypothesised, JLE did alter the length change patterns of the collateral ligaments during knee flexion, but our findings did not support the idea that JLE is a major contributor to midflexion instability for the Columbus ultra-congruent implant design. Acknowledgements This study was supported by a grant of the German Research Foundation (DFG SFB 760) and the European Union 7th Framework Programme (FP7/2007-2013 ICT-2009.5.2 MXL 248693). Conflict of interest No conflict of interest declared. References [1] Peters CL. Soft-tissue balancing in primary total knee arthroplasty. Instr Course Lect 2006;55:413–7. [2] Pape D, Kohn D. Soft tissue balancing in valgus gonarthrosis. Orthopäde 2007;36, 657–8, 60–66. [3] Whiteside LA. Soft tissue balancing: the knee. J Arthroplasty 2002;17:23–7. [4] Harfe DT, Chuinard CR, Espinoza LM, Thomas KA, Solomonow M. Elongation patterns of the collateral ligaments of the human knee. Clin Biomech (Bristol, Avon) 1998;13:163–75. [5] Gardiner JC, Weiss JA, Rosenberg TD. Strain in the human medial collateral ligament during valgus loading of the knee. Clin Orthop Relat Res 2001:266–74. [6] Insall JN, Binazzi R, Soudry M, Mestriner LA. Total knee arthroplasty. Clin Orthop Relat Res 1985:13–22. [7] Dorr LD, Boiardo RA. Technical considerations in total knee arthroplasty. Clin Orthop Relat Res 1986:5–11. [8] Krackow KA, Mihalko WM. The effect of medial release on flexion and extension gaps in cadaveric knees: implications for soft-tissue balancing in total knee arthroplasty. Am J Knee Surg 1999;12:222–8. [9] Yercan HS, Ait Si Selmi T, Sugun TS, Neyret P. Tibiofemoral instability in primary total knee replacement: a review. Part 1. Basic principles and classification. Knee 2005;12:257–66. [10] Gioe TJ, Killeen KK, Grimm K, Mehle S, Scheltema K. Why are total knee replacements revised? Analysis of early revision in a community knee implant registry. Clin Orthop Relat Res 2004:100–6. [11] Fisher DA, Dierckman B, Watts MR, Davis K. Looks good but feels bad: factors that contribute to poor results after total knee arthroplasty. J Arthroplasty 2007;22:39–42. [12] Vince KG, Abdeen A, Sugimori T. The unstable total knee arthroplasty: causes and cures. J Arthroplasty 2006;21:44–9. [13] Firestone TP, Eberle RW. Surgical management of symptomatic instability following failed primary total knee replacement. J Bone Joint Surg Am 2006;88(Suppl. 4):80–4. [14] Naudie DD, Rorabeck CH. Managing instability in total knee arthroplasty with constrained and linked implants. Instr Course Lect 2004;53:207–15. [15] Krackow KA. Instability in total knee arthroplasty: loose as a goose. J Arthroplasty 2003;18:45–7.
1308
C. König et al. / Medical Engineering & Physics 33 (2011) 1303–1308
[16] Parratte S, Pagnano MW. Instability after total knee arthroplasty. J Bone Joint Surg Am 2008;90:184–94. [17] Martin JW, Whiteside LA. The influence of joint line position on knee stability after condylar knee arthroplasty. Clin Orthop Relat Res 1990:146–56. [18] Amis AA, Zavras TD. Isometricity and graft placement during anterior cruciate ligament reconstruction. Knee 1995;2:5–17. [19] Nietert M. The human knee-joint as a biomechanical problem (author’s transl). Biomed Tech (Berl) 1977;22:13–21. [20] Churchill DL, Incavo SJ, Johnson CC, Beynnon BD. The transepicondylar axis approximates the optimal flexion axis of the knee. Clin Orthop Relat Res 1998:111–8. [21] Spitzer V, Ackerman MJ, Scherzinger AL, Whitlock D. The visible human male: a technical report. J Am Med Inform Assoc 1996;3:118–30. [22] Duda GN, Brand D, Freitag S, Lierse W, Schneider E. Variability of femoral muscle attachments. J Biomech 1996;29:1185–90. [23] Heller MO, Bergmann G, Deuretzbacher G, Durselen L, Pohl M, Claes L, et al. Musculo-skeletal loading conditions at the hip during walking and stair climbing. J Biomech 2001;34:883–93. [24] Amis AA, Bull AM, Gupte CM, Hijazi I, Race A, Robinson JR. Biomechanics of the PCL and related structures: posterolateral, posteromedial and meniscofemoral ligaments. Knee Surg Sports Traumatol Arthrosc 2003;11:271–81. [25] Robinson JR, Bull AM, Amis AA. Structural properties of the medial collateral ligament complex of the human knee. J Biomech 2004. [26] Robinson JR, Sanchez-Ballester J, Bull AM, Thomas Rde W, Amis AA. The posteromedial corner revisited. An anatomical description of the passive restraining structures of the medial aspect of the human knee. J Bone Joint Surg Br 2004;86:674–81. [27] Otake N, Chen H, Yao X, Shoumura S. Morphologic study of the lateral and medial collateral ligaments of the human knee. Okajimas Folia Anat Jpn 2007;83:115–22. [28] Putz R, Muhlhofer H, Ercan Y. Ligaments of the knee. Orthopäde 2007;36(612):4–9. [29] Seebacher JR, Inglis AE, Marshall JL, Warren RF. The structure of the posterolateral aspect of the knee. J Bone Joint Surg Am 1982;64:536–41. [30] Park SE, DeFrate LE, Suggs JF, Gill TJ, Rubash HE, Li G. The change in length of the medial and lateral collateral ligaments during in vivo knee flexion. Knee 2005;12:377–82. [31] Taylor WR, Heller MO, Bergmann G, Duda GN. Tibio-femoral loading during human gait and stair climbing. J Orthop Res 2004;22:625–32.
[32] König C, Sharenkov A, Matziolis G, Taylor WR, Perka C, Duda GN, et al. Joint line elevation in revision TKA leads to increased patellofemoral contact forces. J Orthop Res 2010;28:1–5. [33] Van de Velde SK, DeFrate LE, Gill TJ, Moses JM, Papannagari R, Li G. The effect of anterior cruciate ligament deficiency on the in vivo elongation of the medial and lateral collateral ligaments. Am J Sports Med 2007;35:294–300. [34] Phatak NS, Sun Q, Kim SE, Parker DL, Sanders RK, Veress AI, et al. Noninvasive determination of ligament strain with deformable image registration. Ann Biomed Eng 2007;35:1175–87. [35] Butler DL, Kay MD, Stouffer DC. Comparison of material properties in fasciclebone units from human patellar tendon and knee ligaments. J Biomech 1986;19:425–32. [36] Quapp KM, Weiss JA. Material characterization of human medial collateral ligament. J Biomech Eng 1998;120:757–63. [37] Zihlmann MS, Gerber H, Stacoff A, Burckhardt K, Szekely G, Stussi E. Threedimensional kinematics and kinetics of total knee arthroplasty during level walking using single plane video-fluoroscopy and force plates: a pilot study. Gait Posture 2006;24:475–81. [38] Nilsson KG, Karrholm J, Ekelund L. Knee motion in total knee arthroplasty. A roentgen stereophotogrammetric analysis of the kinematics of the Tricon-M knee prosthesis. Clin Orthop Relat Res 1990:147–61. [39] Dennis DA, Komistek RD, Mahfouz MR, Haas BD, Stiehl JB. Multicenter determination of in vivo kinematics after total knee arthroplasty. Clin Orthop Relat Res 2003:37–57. [40] Dennis DA, Komistek RD, Mahfouz MR, Walker SA, Tucker A. A multicenter analysis of axial femorotibial rotation after total knee arthroplasty. Clin Orthop Relat Res 2004:180–9. [41] Arms S, Boyle J, Johnson R, Pope M. Strain measurement in the medial collateral ligament of the human knee: an autopsy study. J Biomech 1983;16:491–6. [42] Grood ES, Noyes FR, Butler DL, Suntay WJ. Ligamentous and capsular restraints preventing straight medial and lateral laxity in intact human cadaver knees. J Bone Joint Surg Am 1981;63:1257–69. [43] Hull ML, Berns GS, Varma H, Patterson HA. Strain in the medial collateral ligament of the human knee under single and combined loads. J Biomech 1996;29:199–206. [44] LaPrade RF, Tso A, Wentorf FA. Force measurements on the fibular collateral ligament, popliteofibular ligament, and popliteus tendon to applied loads. Am J Sports Med 2004;32:1695–701.
Contents lists available at ScienceDirect
Medical Engineering & Physics journal homepage: www.elsevier.com/locate/medengphy
Collateral ligament length change patterns after joint line elevation may not explain midflexion instability following TKA Christian König, Georg Matziolis, Alexey Sharenkov, William R. Taylor, Carsten Perka, Georg N. Duda, Markus O. Heller ∗ Julius Wolff Institute and Center for Musculoskeletal Surgery, Charité – Universitätsmedizin Berlin, Center for Sports Science and Sports Medicine Berlin, Philippstr. 13, Haus 11, 10115 Berlin, Germany
a r t i c l e
i n f o
Article history: Received 27 February 2010 Received in revised form 15 June 2011 Accepted 18 June 2011 Keywords: Knee TKA Joint line Stability Collateral ligaments Midflexion instability
a b s t r a c t Midflexion instability (MFI) after TKA is a phenomenon often described as varus–valgus instability between 30◦ and 45◦ knee flexion. The exact mechanisms causing MFI remain unclear, but elevation of the joint line (JLE) may be one possible cause. In an in silico approach using 4 subject specific musculoskeletal models, the length change patterns of the collateral ligaments during knee flexion (relative to the extended knee) were calculated for the anatomically reconstructed joints as well as for JLEs of 5 and 10 mm. Analysis of the distance between the ligaments’ attachment sites (DA) in midflexion revealed a relative decrease in DA magnitude after JLE for both collateral ligaments in comparison to the anatomically reconstructed knee. This finding suggests that JLE could contribute to MFI. However, the anterior ligament regions also experienced a DA increase (MCL) or only a slight DA decrease (LCL) for each JLE simulated. From this perspective, the anterior ligament portions are unlikely to slacken in midflexion and JLE is unlikely to contribute greatly to MFI. In conclusion, our findings did not support the idea that JLE is a major contributor to midflexion instability for this particular ultra-congruent implant design. © 2011 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Careful balancing of the soft-tissue stabilizers is essential for the success of total knee arthroplasty (TKA) [1–3]. In particular in non- or semi-constrained implants, the lateral collateral ligament (LCL) and the medial collateral ligament (MCL) ensure the knee’s varus/valgus stability while the MCL also resists external rotations and anterior–posterior translations [4,5]. Even though gap and ligament balancing techniques have been established to create a stable joint in extension and flexion [1,3,6–8], joint stability related problems account for a significant proportion of TKA revisions [9–15]. Of particular concern here is the varus–valgus instability of the knee in flexion angles between 30 and 45◦ , also referred to as midflexion instability (MFI) [16]. Even knees that have been carefully balanced in extension and at 90◦ flexion can exhibit MFI. While the exact mechanisms leading to MFI are not fully understood, elevation of the joint line (JLE) after TKA has been suggested as a potential contributor to MFI [16,17]. An approach towards a better understanding of MFI can be derived from the work of Amis and Zavras [18], where it was found that the location of the ligaments’ femoral
∗ Corresponding author. Tel.: +49 30 2093 46128; fax: +49 30 2093 46001. E-mail address: [email protected] (M.O. Heller).
attachment sites relative to the knee flexion axis significantly influences the length change patterns of the cruciate ligaments during flexion. Since JLE alters the knee’s flexion axis relative to the location of the collateral ligaments’ femoral attachment sites, it is likely that JLE can induce altered lengthening or shortening characteristics of the collateral ligaments at different flexion angles [19,20]. The function of the collateral ligaments, which is critical for providing varus–valgus stability of the knee during flexion, might thus be compromised at certain points of the flexion/extension cycle. Computer modelling opens up the possibility to systematically study the influence of selected parameters. By studying multiple subjects using patient-specific models, a robust response as to whether JLE results in a substantially larger distance decrease between the femoral and tibial collateral ligament attachment sites in midflexion could be obtained and compared to provide an understanding of the support provided by the passive soft tissue structures at different knee flexion angles. We hypothesised that JLE and the subsequent change in the position of the femoral collateral ligament attachment sites relative to the femoral articulating surface alters the length change patterns of the collateral ligaments during knee flexion, specifically in midflexion. The aim of this study was therefore to analyze the length change patterns of the collateral ligaments during knee flexion for situations of JLE after TKA in order to assess whether JLE can indeed contribute to midflexion instability.
1350-4533/$ – see front matter © 2011 IPEM. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.medengphy.2011.06.008
1304
C. König et al. / Medical Engineering & Physics 33 (2011) 1303–1308
Fig. 2. The medial and lateral collateral ligaments were each modelled by 5 portions to better characterize their functional behaviour throughout flexion. To this end, the femoral origin and tibial attachment sites of the medial (left) and the lateral (right) collateral ligaments were identified using landmarks. For the MCL, static wrapping points were introduced at the proximal tibia. The current study focussed on the analyses of the most anterior and posterior ligament regions (thick lines).
the subject specific scaling factors, a single implant size was found to be adequate for all four subjects when the virtual implantation was performed and verified by an experienced surgeon. Fig. 1. The medial and lateral collateral ligaments (here displayed together with the cruciate ligaments) were modelled using high resolution cross sectional images of the Visible Human (VH) dataset.
2. Methods 2.1. Musculoskeletal model The computer model that formed the basis for this study was derived from the Visible Human (VH) CT dataset [21] and literature data [22], and contained bony structures, muscles and simplified representations of the ligament insertion sites of the lower limb [23]. To determine the length change patterns of the collateral ligaments during knee flexion the musculoskeletal model was complemented with a more detailed description of the collateral ligaments. The high resolution colour images (resolution 0.32 mm × 0.32 mm × 1 mm) of the same VH dataset that was used to create the reference musculoskeletal model were employed to localize the collateral ligaments [24–29]. This allowed the 3D modelling of the medial and lateral collateral ligaments (Fig. 1) using a 3D visualisation and volume modelling Software (Amira, Visage Imaging, Berlin, Germany). In the combined 3D models of the tibia, femur and the collateral ligaments, the bony attachment sites of the medial and lateral collateral ligaments were each modelled in five portions to better characterize their functional behaviour throughout knee flexion. The origin and insertion of each portion were identified with landmarks to allow calculation of the relative change in length of the ligaments (Fig. 2) [24–30]. For the MCL, additional static wrapping points were defined at the proximal tibia. This more detailed reference model was then adapted to match the anatomy of four participating subjects using linear scaling [23,31]. As a result of adapting the reference models’ femur and tibia using subject specific scaling parameters, variation in the insertions of the knee ligaments across the four subjects was captured in the models. Subsequently, a virtual total knee arthroplasty using an ultra-congruent, fixed bearing and cruciate sacrificing implant design (Columbus UC, Aesculap AG, Tuttlingen, Germany) was performed on the same four musculoskeletal models [32]. Although the bone sizes of the four subjects varied somewhat as a result of
2.2. Variation of the joint line JLE can result from a number of different clinical conditions, including, in the case of distal femoral bone loss, a too small tibial resection, the use of a too high inlay, or combinations thereof. For the current study, it was assumed that the tibial cut was adequate and JLE was a result of modifying the proximal–distal position of the femoral component and the height of the inlay [32]. According to the manufacturer’s recommendation, the femoral component was downsized with increased JLE. Thus the component’s anterior–posterior dimension was reduced by the same amount that the JL was elevated, in order to avoid tightening of the flexion gap that would otherwise result from JLE. In this study, JLE of 5 mm and 10 mm was simulated. The leg length remained unchanged in all simulations. 2.3. Kinematic adaptation The tibio-femoral kinematics of the four subjects were adapted to the postoperative situation by using a kinematic model that reflected the geometry of the ultra-congruent prosthesis’ articulating surfaces [32]. In agreement with the geometric design of the femoral component that possesses three different radii, this model used three knee flexion angle dependent axes of rotation to describe the relative positions of the components throughout the range of knee flexion. Since the axis of rotation of these radii did not coincide, anterior–posterior, as well as superior–inferior translation was captured in the kinematic model (albeit as secondary effects) according to the prescribed relative motion of the knee around the three radii of the femoral component. 2.4. Collateral ligament length change To assess the length change patterns in the anterior and posterior regions of both the MCL and LCL, the distances between pairs of landmarks that represent the extremes of the individual ligament insertion sites (anterior, posterior) were analyzed in both collateral ligaments [30,33]. At first, in the anatomically reconstructed knee, the distance between the attachment sites (DA) was computed for all four sub-
C. König et al. / Medical Engineering & Physics 33 (2011) 1303–1308
jects during knee flexion from 0◦ to 90◦ , using an increment of 15◦ . Here, by keeping the leg length constant also the reference DA at extension remained unaltered for all JL conditions. Then the relative change in this distance in reference to the extended knee (0◦ position) was calculated. These calculations were repeated for all modifications of the JL as described previously, resulting in a simulation of 12 cases in total (4 subjects, 3 conditions each). Furthermore, at every analyzed flexion angle, the DA changes that occurred during flexion with an elevated JL were compared to the DA changes in the anatomically reconstructed knee. 3. Results
1305
erence implantation. This relative decrease was somewhat larger than the relative DA decrease observed as a result of JLE at 90◦ knee flexion (2.0 ± 0.2% at 5 mm and 2.8 ± 0.5% at 10 mm JLE). In the posterior MCL, the relative DA decrease caused by JLE was also more pronounced in midflexion than at higher flexion angles, compared to the anatomical reconstruction. For a JLE of 10 mm, a DA decrease of 4.2 ± 0.1% was observed at 45◦ flexion, while at 90◦ flexion this reduced to 1.0 ± 0.9%. In the anterior LCL, a JLE of 10 mm decreased the DA in midflexion by 2.6 ± 0.1% at 45◦ flexion compared to the anatomical reconstruction. In midflexion, there was a slight DA decrease in the posterior regions of the LCL compared to the anatomical reconstruction (1.9 ± 0.03% at 5 mm and 3.0 ± 0.1% at 10 mm JLE (both at 45◦ flexion)).
3.1. Anatomical reconstruction If the JL was anatomically reconstructed, the distance between the femoral and tibial attachments in the anterior MCL increased by 13.5 ± 0.4% compared to its reference length at extension when flexing the knee from 0◦ to 90◦ (Fig. 3). During the same knee flexion cycle the MCL’s posterior regions experienced a distance decrease relative to their reference length at full extension of 8.1 ± 1.1%. In the anterior region of the LCL the DA remained approximately constant, while the posterior LCL experienced a 14.6 ± 2.2% decrease in distance compared to its reference length at full extension when the knee was flexed to 90◦ . 3.2. DA changes after JLE in midflexion In midflexion, JLE generally resulted in a reduced DA in the anterior and posterior regions of both collateral ligaments relative to the anatomically reconstructed knee joint: while the DA increase during flexion (expressed relative to its reference length at full extension) in the anterior MCL of the anatomically reconstructed joint was 7.9 ± 0.4% at 45◦ knee flexion, the DA at 10 mm JLE was 3.9 ± 0.3%, indicating a 4.0 ± 0.1% DA decrease relative to the ref-
3.3. DA changes after JLE in relation to the conditions of the extended knee Despite the fact that with JLE the DA in the anterior MCL was reduced in midflexion when compared to its DA in the anatomically reconstructed condition, the DA during knee flexion did not fall below the reference DA obtained for the fully extended knee. For a JLE of 10 mm, the DA in the anterior LCL in midflexion fell slightly below the reference DA in extension (−2.1 ± 0.6% at 30◦ and −1.0 ± 1.1% at 45◦ flexion), while the DA for 5 mm JLE was −0.9 ± 0.6% at 30◦ and remained virtually unchanged compared to the conditions in extension for a flexion angle of 45◦ . In the posterior regions of the MCL and the LCL, the DA in midflexion was always found to be below the reference DA of the extended knee. This was true for the anatomically reconstructed knee as well as for the conditions after JLE. 4. Discussion Even though gap and ligament balancing techniques are established steps in TKA for creating a stable joint in both static and
Fig. 3. Length change patterns of the anterior (left) and posterior (right) regions of the medial (top) and lateral (bottom) collateral ligaments from full extension to 90◦ of knee flexion. The dashed curve represents the anatomical reconstruction of the joint line, while the light gray and the gray curves represent joint line elevations of 5 and 10 mm respectively.
1306
C. König et al. / Medical Engineering & Physics 33 (2011) 1303–1308
dynamic conditions [1,3,6–8], stability related problems are still a major reason for a TKA revision [9–11]. It has been speculated that an elevated joint line (JL) might contribute to midflexion instability [16,17]. However, little is known regarding effects of JLE on the collateral ligaments, which are key passive stabilizers of the knee, specifically in midflexion. Using four musculoskeletal models of the lower limb post TKA [32], we have demonstrated for an ultra-congruent implant design that in midflexion an elevation of the JL resulted in a reduced distance between the attachment sites of the anterior and posterior regions in both collateral ligaments when compared to the anatomically reconstructed condition. This finding alone would indeed suggest that JLE can contribute to MFI. Although smaller, the DA of the anterior MCL still exhibited a distance increase between the attachment sites during flexion, hence indicating support from the passive soft tissue structures at midflexion, even in the JLE condition. Thus, as in the intact knee where the MCL is known to be taut in extension and becomes more tight during flexion [5,33,34], the MCL displayed the characteristics of a tensed ligament in midflexion. The anterior LCL, which is taut throughout flexion in the intact knee [5], was also affected by JLE and experienced a distance decrease in midflexion compared to the anatomically reconstructed JL. In contrast to the anterior MCL, however, the DA in the anterior LCL decreased below the DA level in the extended knee. In the worst case, the anterior LCL is only just taut in extension and not pre-tensioned and would therefore become slack with the maximum observed 2.1% DA decrease (10 mm JLE at 30◦ ). Such a condition would result in a varus/valgus laxity of less than 1.5◦ when considering an average ligament length of 60 mm and a knee width of 55 mm. In the intact knee, the posterior fibres of the MCL and LCL slacken with increasing flexion angles [4,5,26,33], a behaviour that was also predicted by our model for the anatomically reconstructed knee. In JLE conditions this effect was intensified in midflexion. How-
ever, it is rather unlikely that a further slackening of already slack posterior fibres would contribute to any form of instability. JLE therefore did not seem to be a major contributor to midflexion instability when using the Columbus ultra-congruent prosthesis design. Nevertheless, while the observed alterations in collateral ligament length change patterns in cases of JLE are unlikely to contribute greatly to midflexion instability, they may still have an impact on the overall ligament function and should therefore be subject to further investigation. For example the generally decreased DA in lower flexion angles and in the case of the MCL also in deeper flexion, may imply that fewer regions of the ligament can contribute to load sharing, possibly increasing the risk of overloading within the remaining active structures. In addition, the observed distance increase in the anterior LCL at higher flexion angles even above the DA of the extended knee in which these areas of the ligament are already taut, may be a possible cause for complications. Further complications may also arise when the JL is moderately elevated but the femoral component is not properly downsized. Compared to the anatomically reconstructed joint at 90◦ with 5 mm JLE the DA increased in the anterior MCL by 4.7 ± 0.5% to a total of 18.2 ± 0.5% and to a total of 9.9 ± 0.8% in the anterior LCL (Fig. 4). Considering that the collateral ligaments are taut in extension [4,5,26,28], it seems safe to assume that any further increase in DA would be linked to an increase in the ligament strain. It is therefore reasonable to assume that the DA increase observed in this study, specifically in the anterior MCL, is indicative of a strain increase in this area of the ligament. While the absolute ligament strain cannot be determined with the methods presented here, it is very well possible that as a result of the DA increases of as much as 18%, JLE may cause strains in the ligaments which are approaching the magnitude of the maximum tensile strain of ligaments, reported to be between 13 and 17% [35,36].
Fig. 4. Length change patterns of the collateral ligaments for the conditions of joint line elevation and no downsizing of the femoral component. The graphs show the length change patterns for the anterior (left) and posterior (right) regions of the medial (top) and lateral (bottom) collateral ligaments from full extension to 90◦ flexion. The dashed curve represents the anatomical reconstruction of the joint line, while the solid curve represents a joint line elevation of 5 mm. The horizontal dotted line represents an average value for maximum tensile strain of the collateral ligaments as reported in the literature.
C. König et al. / Medical Engineering & Physics 33 (2011) 1303–1308
Even though global pre-operative lower limb kinematics for walking, stair climbing and knee bend activities were originally captured for the same four subjects using a gait analysis system [23], no detailed general postoperative in vivo kinematic patterns have been presented so far in the literature that allow for subject specific modelling of postoperative kinematics [37–40]. Although ultra-congruent implant designs were introduced to provide enhanced joint stability, little is known about the actual interactions of such designs with the collateral ligaments. Investigations into such an implant design that offers a high level of constraint could therefore provide new insights into the performance of such designs, but also enable a more robust approximation of the postoperative kinematics than would have been possible by using a more conventional, less constrained implant design. The current study used a kinematic model that was limited to sagittal plane motion, since preliminary fluoroscopic analyses performed in our lab indicate that only little internal–external rotation occurs in vivo with such an ultra-congruent fixed bearing design. As a result, internal–external movement was considered secondary for these analyses and therefore not included. Here, an isolated axial rotation of 5◦ would result in a DA increase of approximately only 0.3% (assuming an average ligament length of 60 mm and a knee width of 55 mm and an eccentric rotation axis). While the knee kinematics were thus generic to the implant rather than subject specific, the individual scaling of the musculoskeletal models reflected the anatomical variation in the relative position of the ligament attachment sites with respect to the femoral component across the four subjects. Therefore, the models did include essential subject-specific variations in those musculoskeletal structures that were key for the current study. Observing the distance between ligament attachment sites to characterize the function of the ligaments [30,33] has the limitation that the zero strain conditions of the ligaments are difficult to obtain [4] and any initial strain within the ligament as well as the condition of a slack ligament is not considered. However, since the extended knee, in which the collateral ligaments are taut or strained [4,5,26,28] is used as a reference in this study, any assumptions on ligament strains would be rather under- than overestimated. Furthermore, we did not directly compare the DA length change patterns in the anatomically reconstructed compared to the healthy unimplanted knee. While we can therefore not rule out that the anatomical reconstruction resulted in MFI, the length change patterns of the collateral ligaments during flexion (i.e. after an ideal TKA that maintained the joint line) were found not to differ substantially from those patterns for the intact knee: in particular, the distances in the anterior proportions of the ligament attachments in midflexion never fell below the values in extension. Following the literature that suggests that the ligaments’ anterior portions are taut in extension [34,41–44] and also during flexion [5,33], these results suggest that the ligaments also stay taut after the implantation of the prosthesis. Although this does not provide direct proof, this reasoning strongly suggests that it is unlikely that instability in midflexion would be caused by the ideally implanted prosthesis. Amis and co-worker [18] already emphasized the importance of the locations of the ligament attachment sites relative to the femoral flexion axes, since these are closely associated with the length change patterns of the ligaments. Implant designs with different condylar radii, different sizing of the implant and patient specific variation in the location of ligament attachment zones may further alter the ligament length change patterns in midflexion and warrant further investigation. The influence of the dynamics of the tibial wrapping points was not further analyzed in this study, since the main influence of JLE on the ligaments’ DA is the changing position of the femoral ligament attachments relative to the rotation axis of the femoral component.
1307
In this study we assumed that a well balanced knee was achieved for the reference implantation in all subjects, without requiring excessive soft tissue management. Furthermore, in the specific cases of 5 and 10 mm JLE, the tightening of the flexion gap can be exactly compensated by an equivalent reduction of the femoral components’ anterior–posterior dimension through the implantation of a smaller implant size. In real life, the amount of JLE would not always exactly match the available implant sizes, and would thus very likely necessitate some additional soft tissue balancing. In practice, a surgeon would try to balance the soft tissues in both extension and 90◦ flexion during TKA by appropriate release techniques. However, the possibilities to accurately simulate these finer details of the surgical release techniques in silico are certainly limited. The conditions analyzed here therefore represent a rather specific situation for which no additional soft tissue balancing is required. However, by focusing the study on elevation levels that could be matched by appropriate changes in the anterior–posterior femoral implant dimension, the number of unknown parameters was minimized. This facilitated the study of the influence of JLE on the behaviour of the collateral ligaments under well controlled conditions, and was therefore considered an acceptable approximation of the clinical situation. As hypothesised, JLE did alter the length change patterns of the collateral ligaments during knee flexion, but our findings did not support the idea that JLE is a major contributor to midflexion instability for the Columbus ultra-congruent implant design. Acknowledgements This study was supported by a grant of the German Research Foundation (DFG SFB 760) and the European Union 7th Framework Programme (FP7/2007-2013 ICT-2009.5.2 MXL 248693). Conflict of interest No conflict of interest declared. References [1] Peters CL. Soft-tissue balancing in primary total knee arthroplasty. Instr Course Lect 2006;55:413–7. [2] Pape D, Kohn D. Soft tissue balancing in valgus gonarthrosis. Orthopäde 2007;36, 657–8, 60–66. [3] Whiteside LA. Soft tissue balancing: the knee. J Arthroplasty 2002;17:23–7. [4] Harfe DT, Chuinard CR, Espinoza LM, Thomas KA, Solomonow M. Elongation patterns of the collateral ligaments of the human knee. Clin Biomech (Bristol, Avon) 1998;13:163–75. [5] Gardiner JC, Weiss JA, Rosenberg TD. Strain in the human medial collateral ligament during valgus loading of the knee. Clin Orthop Relat Res 2001:266–74. [6] Insall JN, Binazzi R, Soudry M, Mestriner LA. Total knee arthroplasty. Clin Orthop Relat Res 1985:13–22. [7] Dorr LD, Boiardo RA. Technical considerations in total knee arthroplasty. Clin Orthop Relat Res 1986:5–11. [8] Krackow KA, Mihalko WM. The effect of medial release on flexion and extension gaps in cadaveric knees: implications for soft-tissue balancing in total knee arthroplasty. Am J Knee Surg 1999;12:222–8. [9] Yercan HS, Ait Si Selmi T, Sugun TS, Neyret P. Tibiofemoral instability in primary total knee replacement: a review. Part 1. Basic principles and classification. Knee 2005;12:257–66. [10] Gioe TJ, Killeen KK, Grimm K, Mehle S, Scheltema K. Why are total knee replacements revised? Analysis of early revision in a community knee implant registry. Clin Orthop Relat Res 2004:100–6. [11] Fisher DA, Dierckman B, Watts MR, Davis K. Looks good but feels bad: factors that contribute to poor results after total knee arthroplasty. J Arthroplasty 2007;22:39–42. [12] Vince KG, Abdeen A, Sugimori T. The unstable total knee arthroplasty: causes and cures. J Arthroplasty 2006;21:44–9. [13] Firestone TP, Eberle RW. Surgical management of symptomatic instability following failed primary total knee replacement. J Bone Joint Surg Am 2006;88(Suppl. 4):80–4. [14] Naudie DD, Rorabeck CH. Managing instability in total knee arthroplasty with constrained and linked implants. Instr Course Lect 2004;53:207–15. [15] Krackow KA. Instability in total knee arthroplasty: loose as a goose. J Arthroplasty 2003;18:45–7.
1308
C. König et al. / Medical Engineering & Physics 33 (2011) 1303–1308
[16] Parratte S, Pagnano MW. Instability after total knee arthroplasty. J Bone Joint Surg Am 2008;90:184–94. [17] Martin JW, Whiteside LA. The influence of joint line position on knee stability after condylar knee arthroplasty. Clin Orthop Relat Res 1990:146–56. [18] Amis AA, Zavras TD. Isometricity and graft placement during anterior cruciate ligament reconstruction. Knee 1995;2:5–17. [19] Nietert M. The human knee-joint as a biomechanical problem (author’s transl). Biomed Tech (Berl) 1977;22:13–21. [20] Churchill DL, Incavo SJ, Johnson CC, Beynnon BD. The transepicondylar axis approximates the optimal flexion axis of the knee. Clin Orthop Relat Res 1998:111–8. [21] Spitzer V, Ackerman MJ, Scherzinger AL, Whitlock D. The visible human male: a technical report. J Am Med Inform Assoc 1996;3:118–30. [22] Duda GN, Brand D, Freitag S, Lierse W, Schneider E. Variability of femoral muscle attachments. J Biomech 1996;29:1185–90. [23] Heller MO, Bergmann G, Deuretzbacher G, Durselen L, Pohl M, Claes L, et al. Musculo-skeletal loading conditions at the hip during walking and stair climbing. J Biomech 2001;34:883–93. [24] Amis AA, Bull AM, Gupte CM, Hijazi I, Race A, Robinson JR. Biomechanics of the PCL and related structures: posterolateral, posteromedial and meniscofemoral ligaments. Knee Surg Sports Traumatol Arthrosc 2003;11:271–81. [25] Robinson JR, Bull AM, Amis AA. Structural properties of the medial collateral ligament complex of the human knee. J Biomech 2004. [26] Robinson JR, Sanchez-Ballester J, Bull AM, Thomas Rde W, Amis AA. The posteromedial corner revisited. An anatomical description of the passive restraining structures of the medial aspect of the human knee. J Bone Joint Surg Br 2004;86:674–81. [27] Otake N, Chen H, Yao X, Shoumura S. Morphologic study of the lateral and medial collateral ligaments of the human knee. Okajimas Folia Anat Jpn 2007;83:115–22. [28] Putz R, Muhlhofer H, Ercan Y. Ligaments of the knee. Orthopäde 2007;36(612):4–9. [29] Seebacher JR, Inglis AE, Marshall JL, Warren RF. The structure of the posterolateral aspect of the knee. J Bone Joint Surg Am 1982;64:536–41. [30] Park SE, DeFrate LE, Suggs JF, Gill TJ, Rubash HE, Li G. The change in length of the medial and lateral collateral ligaments during in vivo knee flexion. Knee 2005;12:377–82. [31] Taylor WR, Heller MO, Bergmann G, Duda GN. Tibio-femoral loading during human gait and stair climbing. J Orthop Res 2004;22:625–32.
[32] König C, Sharenkov A, Matziolis G, Taylor WR, Perka C, Duda GN, et al. Joint line elevation in revision TKA leads to increased patellofemoral contact forces. J Orthop Res 2010;28:1–5. [33] Van de Velde SK, DeFrate LE, Gill TJ, Moses JM, Papannagari R, Li G. The effect of anterior cruciate ligament deficiency on the in vivo elongation of the medial and lateral collateral ligaments. Am J Sports Med 2007;35:294–300. [34] Phatak NS, Sun Q, Kim SE, Parker DL, Sanders RK, Veress AI, et al. Noninvasive determination of ligament strain with deformable image registration. Ann Biomed Eng 2007;35:1175–87. [35] Butler DL, Kay MD, Stouffer DC. Comparison of material properties in fasciclebone units from human patellar tendon and knee ligaments. J Biomech 1986;19:425–32. [36] Quapp KM, Weiss JA. Material characterization of human medial collateral ligament. J Biomech Eng 1998;120:757–63. [37] Zihlmann MS, Gerber H, Stacoff A, Burckhardt K, Szekely G, Stussi E. Threedimensional kinematics and kinetics of total knee arthroplasty during level walking using single plane video-fluoroscopy and force plates: a pilot study. Gait Posture 2006;24:475–81. [38] Nilsson KG, Karrholm J, Ekelund L. Knee motion in total knee arthroplasty. A roentgen stereophotogrammetric analysis of the kinematics of the Tricon-M knee prosthesis. Clin Orthop Relat Res 1990:147–61. [39] Dennis DA, Komistek RD, Mahfouz MR, Haas BD, Stiehl JB. Multicenter determination of in vivo kinematics after total knee arthroplasty. Clin Orthop Relat Res 2003:37–57. [40] Dennis DA, Komistek RD, Mahfouz MR, Walker SA, Tucker A. A multicenter analysis of axial femorotibial rotation after total knee arthroplasty. Clin Orthop Relat Res 2004:180–9. [41] Arms S, Boyle J, Johnson R, Pope M. Strain measurement in the medial collateral ligament of the human knee: an autopsy study. J Biomech 1983;16:491–6. [42] Grood ES, Noyes FR, Butler DL, Suntay WJ. Ligamentous and capsular restraints preventing straight medial and lateral laxity in intact human cadaver knees. J Bone Joint Surg Am 1981;63:1257–69. [43] Hull ML, Berns GS, Varma H, Patterson HA. Strain in the medial collateral ligament of the human knee under single and combined loads. J Biomech 1996;29:199–206. [44] LaPrade RF, Tso A, Wentorf FA. Force measurements on the fibular collateral ligament, popliteofibular ligament, and popliteus tendon to applied loads. Am J Sports Med 2004;32:1695–701.