- Email: [email protected]
Retropatellar contact characteristics before and after total knee arthroplasty
The Knee 12 (2005) 9 – 12 www.elsevier.com/locate/knee
Retropatellar contact characteristics before and after total knee arthroplasty Susanne Fuchs *, Adrian Skwara, Carsten O. Tibesku, Dieter Rosenbaum Orthopaedic Department, University Clinics Mu¨nster, Klinik und Poliklinik fu¨r Allgemeine Orthopa¨die, Albert-Schweitzer-Str. 33, D-48149 Mu¨nster, Germany Accepted 25 April 2002
Abstract Purpose: Qualitative analysis of the retropatellar contact characteristics after total knee arthroplasty. Material and Methods: Six cadaveric knees were investigated before and after implantation of a Genesis I knee prosthesis without and with patellar resurfacing in different positions. Joint contact characteristics were evaluated with Fuji Prescale film type ‘Super Low’ and analyzed qualitatively in nine quadrants. The pressure was determined from a 5-s loading duration in four different knee positions between 45j and 120j of flexion. The femoral component of the prosthesis was implanted in neutral as well as in internal and external rotation. A quadriceps force of 280 N with either a predominant medial or lateral pulling direction was applied. Results: Without prosthesis the largest contact area is between 60j and 90j of flexion. A lateral muscle force direction as well as an external rotation increased the frequency of loading in the medial quadrants. After implantation of the prosthesis the central and superior quadrants were predominantly contacted irrespective of the flexion angle. No marked differences between the flexion angles were found. Implantation of the patellar resurfacing led to contact in the three central quadrants. Conclusion: Implantation of the endoprosthesis leads to increased contact in slight and extreme flexion angles. Especially the central areas are increasingly loaded. No predominant influence of the rotation of the femoral component or the direction of the muscle pull was found. An improved distribution of the contact area could not be demonstrated after patellar resurfacing. D 2002 Elsevier B.V. All rights reserved. Keywords: Total knee arthroplasty; Patella; Contact area; Muscle force; Rotation
1. Introduction A general problem in total knee arthroplasty, should the patella surface be replaced or not, has been discussed for quite some time. Even though the causes for patellar problems are not conclusively determined, high retropatellar pressures in an unfavorable position appear to be a main contributory factor [2,4,5]. Potential solutions are an external rotation of the femoral component, an implantation of a patellar replacement in different positions and a lateral release of the patella in combination with strengthening exercises of the vastus medialis muscle. Lee et al. [11] investigated the influence of femoral rotation with Fuji film and demonstrated that an internal rotation of the femur caused an increased lateral contact stress of the patella, an external rotation increased the medial contact stress. Rhoads et al. [13] found a medialization of the patella between 20 and 80j of knee flexion with a 10j * Corresponding author. Tel.: +49-25183-48002; fax: +49-25183-47989. E-mail address: [email protected] (S. Fuchs). 0968-0160/$ - see front matter D 2002 Elsevier B.V. All rights reserved. doi:10.1016/S0968-0160(02)00045-5
internal rotation of the femoral component. In flexion angles beyond 80j this effect vanished. Anouchi et al. [1] found comparable values for 5j of rotation. McLain et al. [12] demonstrated that patellar resurfacing reduced the contact area. Hsu et al. [6] revealed that patellar designs with improved conformity increased the contact area. Hsu et al. [7] and Jiang et al. [9] showed that the contact area decreased between 15 and 90j in comparison to a healthy knee joint. Wackerhagen et al. [16] evaluated the influence of a splitting of the retinaculum and saw a significant decrease of the local force in the lateral and proximal quadrant that depended on the flexion angle, the direction of muscle force and the prosthetic design. A good congruence between patella and femur resulted in a marked reduction of joint contact forces. Takahashi et al. [15] demonstrated with Fuji film that a lateral retinaculum splitting does not cause major changes. Hsu et al. [8] revealed that a reduction of the contact force with retinaculum splitting could only be achieved in more extreme flexion angles. Between 30 and 75j they even found an increased contact force. The contact area moved to the lateral and proximal part of the patella.
10
S. Fuchs et al. / The Knee 12 (2005) 9–12
Fig. 1. Subdivision of the patella into nine quadrants.
The review of literature did not show an investigation which compared the situation with and without prosthesis or an evaluation of the effects of femoral rotation, muscle force direction and placement of the patellar resurfacing in a single prosthetic design. The present investigation will try to fill this gap.
2. Material and methods Six cadaveric specimens with a mean age of 71.6 years (range 60 – 86 years) were investigated before and after
Fig. 3. Total number of contact areas after implantation of a Genesis prosthesis*.
Fig. 2. Total number of contact areas without prosthesis*.
implantation of a knee prosthesis. The cadavers were embalmed in a modified Thiel fixation that largely preserves bone and soft tissue characteristics and does not cause functional deficits. A prosthesis model Genesis I (Smith & Nephew, Schenefeld, Germany) and the ‘onlay’ version of the patellar resurfacing in size ‘small’ and ‘medium’ were used. We used the size ‘large’ cemented prosthesis design that retained the posterior cruciate ligament. The implantation was performed with specific instrumentation according to the manufacturer’s instructions. The measurements were performed in knee flexion angles of 45j, 60j, 90j and 120j with different placements of the femoral component and various directions of the muscle force. The femoral component was implanted in neutral as well as 10j of external and internal rotation referencing from the trans-epicondylar line. The normal muscle pull was simulated with a total force of 280 N equally distributed to the vastus medialis and lateralis, a medial (lateral) muscle pull was simulated with a combination of 200 N (80 N) to the vastus medialis and 80 N (200 N) to the vastus lateralis. These different muscle pulls distinguish the current study
S. Fuchs et al. / The Knee 12 (2005) 9–12
11
3. Results The results before prosthesis implantation (Fig. 2) demonstrated signs of contact less frequently in 45j and 120j of flexion as compared to 60j and 90j. In the inferior quadrants imprints were rarely seen. A comparison of the muscle pull directions indicates that with a laterally dominant muscle force the superior medial and central medial quadrant often had contact. With a medial force the lateral central quadrant had the most frequent contact. An external rotation of the femur caused large contact areas in the central medial and central middle quadrant. Internal rotation caused a lateral shift in the central quadrants. Independent of the flexion angle the superior middle and central middle quadrant had the most frequent contact after implantation of the prosthesis (Fig. 3). No marked differences were seen between different flexion angles even though a medial shift of the contact area was seen in 120j of flexion. A lateral muscle force caused a more frequent
Fig. 4. Total number of contact areas after implantation of the patellar replacement (size ‘medium’)*.
from the published studies that used more static designs, although it is still a rough simulation of the in vivo dynamic situation. The investigations were carried out without patellar resurfacing and repeated with patellar resurfacing in a central position as well as with a medial, lateral, proximal and distal offset of 4 mm (only in 60j flexion). Joint contact pressures and areas were determined with Fuji film type superlow (Tiedemann & Betz, GarmischPartenkirchen, Germany) with a nominal pressure range of 0.5 to 2.5 MPa that was determined as appropriate in pilot investigations. The Fuji film was cut according to the size of the patellar joint surface and was sealed with surgical film (Opsite Flexifix, Smith & Nephew, Schenefeld, Germany) to prevent the influence of tissue fluids. The Fuji film samples were inserted with tweezers and were loaded in each experimental situation for a duration of 5 s. For semiquantitative analysis the joint contact area was subdivided in nine areas to represent the total numbers of contacts in six bodies in that specific quadrant (Fig. 1).
Fig. 5. Total number of contact areas after implantation of the patellar replacement (size ‘small’)*. *The numbers in the quadrants indicate the frequency of contacts in the respective quadrant.
12
S. Fuchs et al. / The Knee 12 (2005) 9–12
contact in the superior medial and superior middle quadrant as well as in the medial central and lateral central quadrant. A medial muscle force led to a frequent contact in the superior middle and lateral middle and central middle and central lateral quadrant. In comparison to the situation without prosthesis the superior quadrants were contacted more frequently. No marked differences were found between the various rotation positions. The changes with respect to the measurements without prosthesis were seen in a more frequent contact in the superior patella pole. A patellar resurfacing implant was used in the size ‘medium’ (Fig. 4) and ‘small’ (Fig. 5). In general, the implantation led to contact in the three middle quadrants that was independent of the size of the implant. In comparison to the situation without prosthesis few differences were seen with the medium size implant. With the small implant the superior pole was rarely contacted. It was noted for both implant sizes that the inferior pole was rarely contacted. A medial placement caused a slight lateral shift of the contact area and a lateral placement caused a medial shift. A distal or proximal placement has virtually no effect.
4. Discussion Contrary to the situation without prosthesis, contact was measured also at 45j and 120j of flexion after implantation of the prosthesis. In accordance with Benjamin et al. [3] no marked difference between various flexion angles was found. Generally, a slight proximal shift of the contact areas was found after implantation. With respect to the goal to achieve a centralized patellar guidance the present results appear favorable indicating a most frequent contact in the central middle and superior quadrant. Furthermore, the medial shift of the contact area in 120j of flexion can be rated as positive. The lateral muscle pull caused a more frequent contact especially in the superior medial and middle quadrant as well as in the central medial and lateral quadrant. This medialization of the patella is desirable in order to prevent a tendency towards lateralization. The usually recommended strengthening of the medial vastus muscle or the operative weakening of the lateral muscles with a lateral release or a lateral approach is questioned by the findings of our study. Hsu et al. [6] also found a lateral shift of the contact area after a lateral release whereas Takahashi et al. [15] found no effect. The recommended implantation of the femoral component in neutral or external rotation was reconfirmed with the present results [1,14,11]. However, we could not confirm the results of Rhoads et al. [13] who stated that the patella shifts medially between 20j and 80j in 10j internal rotation of the femoral component. The implantation of the patellar replacement in different positions did not cause major changes and—due to the reported complications—should be omitted. This conclusion is supported by McLain et al. [12] who found reduced contact areas after patellar resurfacing which is unfavorable for the
contact stress. Similarly disadvantageous are the results of Hsu et al. [7] and Jiang et al. [9] who reported smaller contact areas especially between 15j and 90j of flexion. Lee et al. [10] investigated only the different patellar positions of the patellar resurfacing and found the best result for the central position without comparison of the non-resurfaced situation. In summary, there is no need to routinely resurface the patella or perform a lateral release under the premise that the centralized patella is the most favorable situation. Furthermore, the recommendation for selective strengthening of the medial thigh muscles was not supported by our results. However, the recommendation for a neutral or external rotation of the femoral component was supported by the present study.
References [1] Anouchi YS, Whiteside LA, Kaiser AD, Milliano MT. The effects of axial rotational alignement of the femoral component on knee stability and patellar tracking in total knee arthroplasty demonstrated on autopsy specimens. Clin Orthop 1993;287:170 – 7. [2] Ayers DC, Dennis DA, Johanson NA, Pellegrini VD. Common complications of total knee arthroplasty. J Bone Jt Surg A 1997; 79:278 – 311. [3] Benjamin JB, Szivek JA, Hammond AS, Kubchandhani Z , Matthews AI Jr., Anderson P. Contact areas and pressures between native patellas and prosthetic femoral components. J Arthroplasty 1998;13:693 – 8. [4] Healy WL, Wasilewski SA, Takei R, Oberlander M. Patellofemoral complications following total knee arthroplasty. Correlation with implant design and patient risk factors. J Arthropl 1995;10:197 – 201. [5] Hernandez Vaquero D, Alvarez Gonzalez UJ, Fernandez Corona C, Garcia Sandoval MA, Rubio Gonzalez A. Patellar complications after total knee arthroplasty. Int Orthop 1996;20:103 – 6. [6] Hsu HC, Luo ZP, Rand JA, An KN. Influence of lateral release on patellar tracking and patellofemoral contact characteristics after total knee arthroplasty. J Arthroplasty 1997;12:74 – 83. [7] Hsu HC, Luo ZP, Rand JA, An KN. Influence of patellar thickness on patellar tracking and patellofemoral contact characteristics after total knee arthroplasty. J Arthopl 1996;11:69 – 78. [8] Hsu H-P, Walker PS. Wear and deformation of patellar components in total knee arthroplasty. Clin Orthop 1989;246:260 – 9. [9] Jiang CC, Yip KM, Liu DH. Patellar thickness in knee replacement. J Formos Med Assoc 1994;93:417 – 20. [10] Lee TQ, Budoff JE, Glaser FE. Patellar component positioning in total knee arthroplasty. Clin Orthop 1999;366:274 – 81. [11] Lee TQ, Anzel SH, Bennett KA, Pang D, Kim WC. The influence of fixed rotational deformities of the femur on the patellofemoral contact pressures in human cadaver knees. Clin Orthop 1994;302:69 – 74. [12] McLain RF, Barger WF. The effect of total knee design on patellar strain. J Arthropl 1986;1:91 – 8. [13] Rhoads DD, Noble PC, Reuben JD, Mahoney OM, Tullos HS. The effect of femoral component position on patellar tracking after total knee arthroplasty. Clin Orthop 1990;260:43 – 7. [14] Singerman R, Pagan HD, Peyser AB, Goldberg VM. Effect of femoral component rotation and patellar design on patellar forces. Clin Orthop 1997;334:345 – 53. [15] Takahashi T, Wada Y, Yamamoto H. Soft-tissue balancing with pressure distribution during total knee arthroplasty. J Bone Jt Surg B 1997;79:235 – 9. [16] Wackerhagen A, Bodem F, Hopf Ch, Palme E. The influence of lateral release on patello-femoral joint loading in knee arthroplasty. Int Orthop 1992;16:19 – 24.
Retropatellar contact characteristics before and after total knee arthroplasty Susanne Fuchs *, Adrian Skwara, Carsten O. Tibesku, Dieter Rosenbaum Orthopaedic Department, University Clinics Mu¨nster, Klinik und Poliklinik fu¨r Allgemeine Orthopa¨die, Albert-Schweitzer-Str. 33, D-48149 Mu¨nster, Germany Accepted 25 April 2002
Abstract Purpose: Qualitative analysis of the retropatellar contact characteristics after total knee arthroplasty. Material and Methods: Six cadaveric knees were investigated before and after implantation of a Genesis I knee prosthesis without and with patellar resurfacing in different positions. Joint contact characteristics were evaluated with Fuji Prescale film type ‘Super Low’ and analyzed qualitatively in nine quadrants. The pressure was determined from a 5-s loading duration in four different knee positions between 45j and 120j of flexion. The femoral component of the prosthesis was implanted in neutral as well as in internal and external rotation. A quadriceps force of 280 N with either a predominant medial or lateral pulling direction was applied. Results: Without prosthesis the largest contact area is between 60j and 90j of flexion. A lateral muscle force direction as well as an external rotation increased the frequency of loading in the medial quadrants. After implantation of the prosthesis the central and superior quadrants were predominantly contacted irrespective of the flexion angle. No marked differences between the flexion angles were found. Implantation of the patellar resurfacing led to contact in the three central quadrants. Conclusion: Implantation of the endoprosthesis leads to increased contact in slight and extreme flexion angles. Especially the central areas are increasingly loaded. No predominant influence of the rotation of the femoral component or the direction of the muscle pull was found. An improved distribution of the contact area could not be demonstrated after patellar resurfacing. D 2002 Elsevier B.V. All rights reserved. Keywords: Total knee arthroplasty; Patella; Contact area; Muscle force; Rotation
1. Introduction A general problem in total knee arthroplasty, should the patella surface be replaced or not, has been discussed for quite some time. Even though the causes for patellar problems are not conclusively determined, high retropatellar pressures in an unfavorable position appear to be a main contributory factor [2,4,5]. Potential solutions are an external rotation of the femoral component, an implantation of a patellar replacement in different positions and a lateral release of the patella in combination with strengthening exercises of the vastus medialis muscle. Lee et al. [11] investigated the influence of femoral rotation with Fuji film and demonstrated that an internal rotation of the femur caused an increased lateral contact stress of the patella, an external rotation increased the medial contact stress. Rhoads et al. [13] found a medialization of the patella between 20 and 80j of knee flexion with a 10j * Corresponding author. Tel.: +49-25183-48002; fax: +49-25183-47989. E-mail address: [email protected] (S. Fuchs). 0968-0160/$ - see front matter D 2002 Elsevier B.V. All rights reserved. doi:10.1016/S0968-0160(02)00045-5
internal rotation of the femoral component. In flexion angles beyond 80j this effect vanished. Anouchi et al. [1] found comparable values for 5j of rotation. McLain et al. [12] demonstrated that patellar resurfacing reduced the contact area. Hsu et al. [6] revealed that patellar designs with improved conformity increased the contact area. Hsu et al. [7] and Jiang et al. [9] showed that the contact area decreased between 15 and 90j in comparison to a healthy knee joint. Wackerhagen et al. [16] evaluated the influence of a splitting of the retinaculum and saw a significant decrease of the local force in the lateral and proximal quadrant that depended on the flexion angle, the direction of muscle force and the prosthetic design. A good congruence between patella and femur resulted in a marked reduction of joint contact forces. Takahashi et al. [15] demonstrated with Fuji film that a lateral retinaculum splitting does not cause major changes. Hsu et al. [8] revealed that a reduction of the contact force with retinaculum splitting could only be achieved in more extreme flexion angles. Between 30 and 75j they even found an increased contact force. The contact area moved to the lateral and proximal part of the patella.
10
S. Fuchs et al. / The Knee 12 (2005) 9–12
Fig. 1. Subdivision of the patella into nine quadrants.
The review of literature did not show an investigation which compared the situation with and without prosthesis or an evaluation of the effects of femoral rotation, muscle force direction and placement of the patellar resurfacing in a single prosthetic design. The present investigation will try to fill this gap.
2. Material and methods Six cadaveric specimens with a mean age of 71.6 years (range 60 – 86 years) were investigated before and after
Fig. 3. Total number of contact areas after implantation of a Genesis prosthesis*.
Fig. 2. Total number of contact areas without prosthesis*.
implantation of a knee prosthesis. The cadavers were embalmed in a modified Thiel fixation that largely preserves bone and soft tissue characteristics and does not cause functional deficits. A prosthesis model Genesis I (Smith & Nephew, Schenefeld, Germany) and the ‘onlay’ version of the patellar resurfacing in size ‘small’ and ‘medium’ were used. We used the size ‘large’ cemented prosthesis design that retained the posterior cruciate ligament. The implantation was performed with specific instrumentation according to the manufacturer’s instructions. The measurements were performed in knee flexion angles of 45j, 60j, 90j and 120j with different placements of the femoral component and various directions of the muscle force. The femoral component was implanted in neutral as well as 10j of external and internal rotation referencing from the trans-epicondylar line. The normal muscle pull was simulated with a total force of 280 N equally distributed to the vastus medialis and lateralis, a medial (lateral) muscle pull was simulated with a combination of 200 N (80 N) to the vastus medialis and 80 N (200 N) to the vastus lateralis. These different muscle pulls distinguish the current study
S. Fuchs et al. / The Knee 12 (2005) 9–12
11
3. Results The results before prosthesis implantation (Fig. 2) demonstrated signs of contact less frequently in 45j and 120j of flexion as compared to 60j and 90j. In the inferior quadrants imprints were rarely seen. A comparison of the muscle pull directions indicates that with a laterally dominant muscle force the superior medial and central medial quadrant often had contact. With a medial force the lateral central quadrant had the most frequent contact. An external rotation of the femur caused large contact areas in the central medial and central middle quadrant. Internal rotation caused a lateral shift in the central quadrants. Independent of the flexion angle the superior middle and central middle quadrant had the most frequent contact after implantation of the prosthesis (Fig. 3). No marked differences were seen between different flexion angles even though a medial shift of the contact area was seen in 120j of flexion. A lateral muscle force caused a more frequent
Fig. 4. Total number of contact areas after implantation of the patellar replacement (size ‘medium’)*.
from the published studies that used more static designs, although it is still a rough simulation of the in vivo dynamic situation. The investigations were carried out without patellar resurfacing and repeated with patellar resurfacing in a central position as well as with a medial, lateral, proximal and distal offset of 4 mm (only in 60j flexion). Joint contact pressures and areas were determined with Fuji film type superlow (Tiedemann & Betz, GarmischPartenkirchen, Germany) with a nominal pressure range of 0.5 to 2.5 MPa that was determined as appropriate in pilot investigations. The Fuji film was cut according to the size of the patellar joint surface and was sealed with surgical film (Opsite Flexifix, Smith & Nephew, Schenefeld, Germany) to prevent the influence of tissue fluids. The Fuji film samples were inserted with tweezers and were loaded in each experimental situation for a duration of 5 s. For semiquantitative analysis the joint contact area was subdivided in nine areas to represent the total numbers of contacts in six bodies in that specific quadrant (Fig. 1).
Fig. 5. Total number of contact areas after implantation of the patellar replacement (size ‘small’)*. *The numbers in the quadrants indicate the frequency of contacts in the respective quadrant.
12
S. Fuchs et al. / The Knee 12 (2005) 9–12
contact in the superior medial and superior middle quadrant as well as in the medial central and lateral central quadrant. A medial muscle force led to a frequent contact in the superior middle and lateral middle and central middle and central lateral quadrant. In comparison to the situation without prosthesis the superior quadrants were contacted more frequently. No marked differences were found between the various rotation positions. The changes with respect to the measurements without prosthesis were seen in a more frequent contact in the superior patella pole. A patellar resurfacing implant was used in the size ‘medium’ (Fig. 4) and ‘small’ (Fig. 5). In general, the implantation led to contact in the three middle quadrants that was independent of the size of the implant. In comparison to the situation without prosthesis few differences were seen with the medium size implant. With the small implant the superior pole was rarely contacted. It was noted for both implant sizes that the inferior pole was rarely contacted. A medial placement caused a slight lateral shift of the contact area and a lateral placement caused a medial shift. A distal or proximal placement has virtually no effect.
4. Discussion Contrary to the situation without prosthesis, contact was measured also at 45j and 120j of flexion after implantation of the prosthesis. In accordance with Benjamin et al. [3] no marked difference between various flexion angles was found. Generally, a slight proximal shift of the contact areas was found after implantation. With respect to the goal to achieve a centralized patellar guidance the present results appear favorable indicating a most frequent contact in the central middle and superior quadrant. Furthermore, the medial shift of the contact area in 120j of flexion can be rated as positive. The lateral muscle pull caused a more frequent contact especially in the superior medial and middle quadrant as well as in the central medial and lateral quadrant. This medialization of the patella is desirable in order to prevent a tendency towards lateralization. The usually recommended strengthening of the medial vastus muscle or the operative weakening of the lateral muscles with a lateral release or a lateral approach is questioned by the findings of our study. Hsu et al. [6] also found a lateral shift of the contact area after a lateral release whereas Takahashi et al. [15] found no effect. The recommended implantation of the femoral component in neutral or external rotation was reconfirmed with the present results [1,14,11]. However, we could not confirm the results of Rhoads et al. [13] who stated that the patella shifts medially between 20j and 80j in 10j internal rotation of the femoral component. The implantation of the patellar replacement in different positions did not cause major changes and—due to the reported complications—should be omitted. This conclusion is supported by McLain et al. [12] who found reduced contact areas after patellar resurfacing which is unfavorable for the
contact stress. Similarly disadvantageous are the results of Hsu et al. [7] and Jiang et al. [9] who reported smaller contact areas especially between 15j and 90j of flexion. Lee et al. [10] investigated only the different patellar positions of the patellar resurfacing and found the best result for the central position without comparison of the non-resurfaced situation. In summary, there is no need to routinely resurface the patella or perform a lateral release under the premise that the centralized patella is the most favorable situation. Furthermore, the recommendation for selective strengthening of the medial thigh muscles was not supported by our results. However, the recommendation for a neutral or external rotation of the femoral component was supported by the present study.
References [1] Anouchi YS, Whiteside LA, Kaiser AD, Milliano MT. The effects of axial rotational alignement of the femoral component on knee stability and patellar tracking in total knee arthroplasty demonstrated on autopsy specimens. Clin Orthop 1993;287:170 – 7. [2] Ayers DC, Dennis DA, Johanson NA, Pellegrini VD. Common complications of total knee arthroplasty. J Bone Jt Surg A 1997; 79:278 – 311. [3] Benjamin JB, Szivek JA, Hammond AS, Kubchandhani Z , Matthews AI Jr., Anderson P. Contact areas and pressures between native patellas and prosthetic femoral components. J Arthroplasty 1998;13:693 – 8. [4] Healy WL, Wasilewski SA, Takei R, Oberlander M. Patellofemoral complications following total knee arthroplasty. Correlation with implant design and patient risk factors. J Arthropl 1995;10:197 – 201. [5] Hernandez Vaquero D, Alvarez Gonzalez UJ, Fernandez Corona C, Garcia Sandoval MA, Rubio Gonzalez A. Patellar complications after total knee arthroplasty. Int Orthop 1996;20:103 – 6. [6] Hsu HC, Luo ZP, Rand JA, An KN. Influence of lateral release on patellar tracking and patellofemoral contact characteristics after total knee arthroplasty. J Arthroplasty 1997;12:74 – 83. [7] Hsu HC, Luo ZP, Rand JA, An KN. Influence of patellar thickness on patellar tracking and patellofemoral contact characteristics after total knee arthroplasty. J Arthopl 1996;11:69 – 78. [8] Hsu H-P, Walker PS. Wear and deformation of patellar components in total knee arthroplasty. Clin Orthop 1989;246:260 – 9. [9] Jiang CC, Yip KM, Liu DH. Patellar thickness in knee replacement. J Formos Med Assoc 1994;93:417 – 20. [10] Lee TQ, Budoff JE, Glaser FE. Patellar component positioning in total knee arthroplasty. Clin Orthop 1999;366:274 – 81. [11] Lee TQ, Anzel SH, Bennett KA, Pang D, Kim WC. The influence of fixed rotational deformities of the femur on the patellofemoral contact pressures in human cadaver knees. Clin Orthop 1994;302:69 – 74. [12] McLain RF, Barger WF. The effect of total knee design on patellar strain. J Arthropl 1986;1:91 – 8. [13] Rhoads DD, Noble PC, Reuben JD, Mahoney OM, Tullos HS. The effect of femoral component position on patellar tracking after total knee arthroplasty. Clin Orthop 1990;260:43 – 7. [14] Singerman R, Pagan HD, Peyser AB, Goldberg VM. Effect of femoral component rotation and patellar design on patellar forces. Clin Orthop 1997;334:345 – 53. [15] Takahashi T, Wada Y, Yamamoto H. Soft-tissue balancing with pressure distribution during total knee arthroplasty. J Bone Jt Surg B 1997;79:235 – 9. [16] Wackerhagen A, Bodem F, Hopf Ch, Palme E. The influence of lateral release on patello-femoral joint loading in knee arthroplasty. Int Orthop 1992;16:19 – 24.