The American Journal of Sports Medicine http://ajs.sagepub.com/
Length Change Behavior of Virtual Medial Patellofemoral Ligament Fibers During In Vivo Knee Flexion Si Young Song, Chae-Hyun Pang, Chan Hyoek Kim, Jeehyoung Kim, Mi Lim Choi and Young-Jin Seo Am J Sports Med published online February 3, 2015 DOI: 10.1177/0363546514567061 The online version of this article can be found at: http://ajs.sagepub.com/content/early/2015/02/02/0363546514567061
Published by: http://www.sagepublications.com
On behalf of: American Orthopaedic Society for Sports Medicine
Additional services and information for The American Journal of Sports Medicine can be found at:
P> OnlineFirst Version of Record - Feb 3, 2015 What is This?
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
AJSM PreView, published on February 3, 2015 as doi:10.1177/0363546514567061
Length Change Behavior of Virtual Medial Patellofemoral Ligament Fibers During In Vivo Knee Flexion Si Young Song,*y MD, Chae-Hyun Pang,y MD, Chan Hyoek Kim,y MD, Jeehyoung Kim,z MD, Mi Lim Choi,§ MS, and Young-Jin Seo,y MD Investigation performed at Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Republic of Korea Background: In vivo length change behavior of native medial patellofemoral ligament (MPFL) fibers throughout the range of knee motion has not been reported in vivo. Purpose: To measure the length changes of various fibers of the MPFL and to determine their length change patterns during in vivo passive knee flexion. Study Design: Descriptive laboratory study. Methods: The right knees of 11 living subjects were scanned with a high-resolution computed tomography scanner at 0°, 30°, 60°, 90°, and 120° of knee flexion, and 3-dimensional (3D) models were constructed using customized software. Five patellar points were determined: 20% (point 20), 30% (point 30), 40% (point 40), 50% (point 50), and 60% (point 60) from the superior pole of the patella. The Scho¨ttle femoral point (point F) was marked on a translucent 3D model of a true lateral view. Five virtual fibers connecting these points on the 3D knee model were created, and the lengths of various fibers were digitally measured. Results: The average length changes were 9.1 6 2.5 mm in F20, 9.1 6 2.5 mm in F30, 8.1 6 2.6 mm in F40, 6.9 6 2.4 mm in F50, and 6.9 6 1.7 mm in F60. There were significant differences in length changes of these 5 fibers (P \ .001). The lengths of 2 superior fibers (F20 and F30) increased as the knee flexed from 0° to 30° and decreased as the knee flexed over 30°. The lengths of a middle fiber (F40) and an inferior fiber (F50) increased from 0° to 30°, reached a plateau from 30° to 60°, and then decreased from 60° to 120°. F60 showed an increase from 0° to 30°, and then a plateau pattern from 30° to 90°, followed by a decrease during further flexion. Conclusion: Superior fibers exhibited their maximum lengths at low flexion angles, and inferior fibers exhibited their maximum lengths at midflexion angles. The MPFL is a complex of functionally various fibers with some taut and others slack over the whole range of knee motion. Clinical Relevance: The results for lengths and length change patterns of various MPFL fibers are expected to serve as a theoretical background for anatomic double-bundle MPFL reconstruction. Keywords: patella; medial patellofemoral ligament; fiber length change; 3-dimensional knee model
The medial patellofemoral ligament (MPFL) is known to be the primary soft tissue restraint against lateralization of the patella.2,5,26 The biomechanical behavior of the MPFL during knee motion is crucial for the maintenance of patellar stability and control of normal kinematics of the patellofemoral joint, particularly at low flexion angles.17,21,23,26 The MPFL is a triangular and broad condensation of capsular fibers that is wider at its patellar attachment than its femoral attachment.3,13,26 Despite the fact that the MPFL is composed of many ligamentous fibers, little is known about these various fibers, including their lengths, length change patterns, and function. Several studies have examined the length and length changes of the MPFL.26,27,29,32 Although these studies have provided good information about the MPFL fiber complex,
*Address correspondence to Si Young Song, MD, Department of Orthopaedic Surgery, Dongtan Sacred Heart Hospital, Hallym University College of Medicine, 7 Keunjaebong-gil, Hwaseong, Gyeonggi-do, 445170, Republic of Korea (e-mail: [email protected]). y Department of Orthopaedic Surgery, Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Republic of Korea. z Department of Orthopaedic Surgery, Seoul Sacred Heart General Hospital, Seoul, Republic of Korea. § Department of Data Statistics, Korea Culture & Tourism Institute, Seoul, Republic of Korea. The authors declared that they have no conflicts of interest in the authorship and publication of this contribution. The American Journal of Sports Medicine, Vol. XX, No. X DOI: 10.1177/0363546514567061 Ó 2015 The Author(s)
1 Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
2
Song et al
The American Journal of Sports Medicine
this knowledge is based mainly on cadaveric studies. However, in vitro studies using invasive procedures may not reflect actual kinematics in the living knee because of the removal of adjacent soft tissue around the knee. Recently, an in vivo 3-dimensional (3D) technique has been used to overcome obvious limitations of cadaveric studies and enable the analysis of in vivo motion of any joint and the more accurate measurement of ligament length noninvasively.15,21,34 In the past decade, MPFL reconstruction has become increasingly popular for the treatment of recurrent lateral patellar instability. Because malposition of a graft at the native footprint of the MPFL during reconstruction results in nonphysiological patellofemoral pressure and abnormal patellar tracking,7,22,23,27,28 anatomic graft placement may provide a better option. During the past few years, there has been an increased interest in anatomic double-bundle (DB) MPFL reconstruction, which replicates 2 functional bundles, to more closely restore normal patellofemoral stability and kinematics. To date, various surgical techniques for DB MPFL reconstruction have been described with generally favorable short- to midterm clinical outcomes.1,9,12,19,22,33 Although the concept of anatomic DB reconstruction to restore a triangular form of the MPFL represents a trend in MPFL reconstruction, the biomechanical rationale of the DB reconstruction is not well known. Most studies have focused on surgical techniques and clinical outcomes, but relatively little attention has been directed at the function of each bundle of the reconstructed MPFL. A better understanding of the length change of each bundle during knee flexion could form the basis for advanced anatomic DB MPFL reconstruction techniques. Determining the length of various MPFL fibers and their length change in vivo is a useful way to understand how the MPFL stabilizes the patella. Furthermore, this knowledge on length change behavior will contribute to assumptions about the biomechanical function of each reconstructed bundle. We hypothesized that the native MPFL does not behave as a single bundle of fibers with a constant length but as a continuum of ligament fibers with varying length change behavior during knee flexion. The purpose of this study was to measure the functional lengths of the MPFL fibers and to characterize their length change patterns during knee flexion in vivo.
MATERIALS AND METHODS This study was approved by the institutional review board at Dongtan Sacred Heart Hospital, and an informed consent was obtained from all subjects. The subjects comprised 12 male volunteers with no history of any knee disorders or previous injury. Each subject received physical examinations and radiographic evaluations (anteroposterior, lateral, and Merchant views of the knee and a full-length anteroposterior weightbearing radiograph). One volunteer with high patella (Insall-Savati ratio, 1.3) and high tibial tuberosity–trochlear groove (TTTG) distance (17.5 mm) was excluded. As a result, 11 volunteers were included in this study. The apprehension test and the J-sign were negative in all volunteers, and none
had pain on patellar compression, an effusion, or limitation of motion. The mean age was 32.0 6 3.9 years (range, 27-39 years), the mean height was 175.9 6 4.4 cm (range, 168-181 cm), and the mean body mass index was 26.1 6 2.8 kg/m2 (range, 23.1-32.1 kg/m2). The mean femorotibial angle was 4.5° 6 1.5° mm (range, 2.7°-7.4° mm), the mean sulcus angle was 137.5° 6 4.5° (range, 128.6°-144.1°), the mean congruence angle was 28.3° 6 3.9° (range, 215.6° to 22.5°), and the mean Insall-Savati ratio was 1.03 6 0.11 (range, 0.88-1.15). The mean TTTG distance obtained from computed tomography (CT) scanning was 8.8 6 1.3 mm (range, 7.3-10.9 mm). There was no case of trochlear dysplasia based on the criteria of Dejour et al.4
Construction of 3D Bone Models The right knee of each subject was scanned with a highresolution CT scanner (SOMATOM Sensation; Siemens) at 5 knee flexion angles (0°, 30°, 60°, 90°, and 120°). The flexion angle was determined by a knee surgeon using a goniometer with extendable arms (Gollehon) placed over the thigh and leg. During CT scanning, each subject was asked to take the lateral decubitus position in a relaxed state. Axial plane images with 1-mm slices were obtained, and these CT images were saved in DICOM (Digital Imaging and Communications in Medicine) files. These files were then imported into a visualization software package (Amira 5.3; Mercury Computer Systems), and 3D models of a total of 11 knees were constructed. These 3D images were then imported to a customized software package (Rapidform XOR; INUS Technology) for analysis.
Attachment Site Identification and Creation of Virtual Fibers Previous studies have defined the anatomic attachment site of the MPFL. The patellar insertion of the MPFL is at the medial upper two-thirds of the patella,14,18 and the radiographic landmark of the femoral MPFL attachment site has been defined by Scho¨ttle et al.25 Based on these studies, 5 patellar points were marked on the 3D knee model at 0° of flexion: 20% (point 20), 30% (point 30), 40% (point 40), 50% (point 50), and 60% (point 60) from the superior pole of the patella (Figure 1). To obtain a reproducible radiographic femoral point described by Scho¨ttle et al,25 the 3D surface model at 0° of flexion was converted into a translucent 3D model by using the registration function of a customized software package (Rapidform XOR). This translucent model resembled a fluoroscopic image, and a true lateral radiograph was then made by accurately superimposing the posterior portion of the medial and lateral femoral condyles. The femoral insertion point (point F) was marked on the translucent 3D model and was then reconverted into a 3D surface model (Figure 2). All points were marked by 2 expert knee surgeons jointly. To minimize technical error in the measurement process, the patellar model at 0° of flexion was superimposed on each discrete patellar model at 30°, 60°, 90°, and 120° of flexion, and then superimposed models were again divided into 5 models at different flexion
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
Vol. XX, No. X, XXXX
Length Change Behavior of MPFL Fibers
Figure 1. The patellar medial patellofemoral ligament attachment sites are marked on the 0° knee model (reference model). Five patellar points are marked: 20% (point 20), 30% (point 30), 40% (point 40), 50% (point 50), and 60% (point 60) from the superior pole of the patella. angles. This procedure was performed using the surface-tosurface matching method that was the registration function of the customized software package (Rapidform XOR). With this process, identical patellar points were automatically established on the 3D models at 30°, 60°, 90°, and 120° of flexion (Figure 3). Femoral points in 3D models at 30°, 60°, 90°, and 120° of flexion were established using the same method. A virtual medial capsule connecting the medial side of the patellofemoral joint surface was created on the 3D models (Figure 4A). The shortest line connecting these patellar and femoral points was designated as the MPFL fiber. Virtual linear fibers in the 3D space along the 3D surface of the bone models were represented by surface lines to express native MPFL paths and were detoured by bony protrusions to avoid bone penetration (Figure 4B). In this way, 5 virtual fibers were created on the 3D knee models.
Measurement of Fiber Length The 3D length of a virtual fiber was defined as the shortest linear distance curving around the bony surface between a digitalized femoral point and a patellar point. With the customized software package (Rapidform XOR), lengths of 5 fibers were digitally measured at 5 knee flexion angles. This provided 25 measurements per knee. All measurements were performed twice with an interval of 1 week by 2 knee surgeons. The intra- and interobserver reliability of the measurements was assessed using intraclass and interclass correlation coefficients. Intraclass and interclass correlation coefficients were .99 and .97, respectively. The length changes of the individual fibers at 5 flexion angles were calculated, and length change patterns were analyzed.
Statistical Analysis All data were expressed as the mean 6 standard deviation. The length changes among the 5 fibers were compared
3
Figure 2. With use of the registration function of the customized software, the 3D surface model at 0° of flexion was converted into a translucent 3D model, which can produce images resembling a fluoroscopic image. A true lateral radiograph was then made by accurately superimposing the posterior portion of the medial and lateral femoral condyles. The radiographic Scho¨ttle femoral point (point F) was marked on the translucent 3D model and then reconverted into a 3D surface model. through a 1-way repeated-measures analysis of variance (ANOVA). The level of significance was set at P \ .05. The differences in lengths for angle-by-fiber interaction, pointby-fiber interaction, and angle-by-point interaction were assessed using a 2-way repeated-measures ANOVA with both repeated factors by the Greenhouse-Geisser method. The level of significance was set at P \ .05. The length change pattern for each flexion angle within each fiber was analyzed through a 2-way repeated-measures ANOVA followed by post hoc multiple pairwise comparisons with the Holm-Bonferroni method. According to this method, P \ .00365 was considered significant. The statistical analysis was conducted using SPSS version 20.0 (SPSS Inc).
RESULTS Length Changes of Virtual Fibers All 5 fibers were not isometric throughout the flexionextension arc of the knee joint (Figure 5). During the entire range of knee motion, the average length changes were 9.1 6 2.5 mm in F20, 9.1 6 2.5 mm in F30, 8.1 6 2.6 mm in F40, 6.9 6 2.4 mm in F50, and 6.9 6 1.7 mm in F60. Length changes in these 5 fibers were significantly different (P \ .001).
Length Change Patterns of Virtual Fibers There were significant differences in lengths for angle-byfiber interaction (P = .029), point-by-fiber interaction (P \ .001), and angle-by-point interaction (P \ .001). For 2 superior fibers (F20 and F30) (Figure 5A), their lengths increased as the knee flexed from 0° to 30° and decreased as it flexed over 30°, and the lengths were longest at 30° of flexion and shortest at 120° of flexion. There were significant differences among all the adjacent knee flexion angles in the lengths of 2 superior fibers.
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
4
Song et al
The American Journal of Sports Medicine
Figure 3. Identical patellar points are automatically established on the 3D models at 30°, 60°, 90°, and 120° of flexion by the surface-to-surface matching method. The patellar model at 0° of flexion is superimposed on each discrete patellar model at 30°, 60°, 90°, and 120° of flexion, and then the superimposed models are again divided.
Figure 4. (A) Virtual medial capsule connecting the medial side of the patellofemoral joint surface is created on the 3D models. (B and C) Virtual fibers connecting the patellar and femoral points are created. These fibers are detoured by bony protrusions to avoid bone penetration. F20, the fiber connecting point F and point 20; F30, the fiber connecting point F and point 30; F40, the fiber connecting point F and point 40; F50, the fiber connecting point F and point 50; F60, the fiber connecting point F and point 60. For 1 middle fiber (F40) and 1 inferior fiber (F50) (Figure 5B), the lengths increased from 0° to 30°, then reached a plateau between 30° and 60°, and decreased from 60° to 120°. Except between 30° and 60°, there were significant differences among the adjacent knee flexion angles in its fiber length. For 1 inferior fiber (F60) (Figure 5C), the length increased as the knee flexed from 0° to 30°, then reached a plateau pattern between 30° and 90°, and decreased as it flexed over 90°.
DISCUSSION The MPFL is known to tighten in low to mid flexion and slacken in high flexion, and therefore, the contribution of the MPFL to resist patellar dislocation is greatest in low
to mid flexion.2,21,26,32 However, there is some controversy over the length change patterns of the MPFL. An in vivo biomechanical study reported that the MPFL is longest at 30° of flexion,34 on the other hand, a cadaveric study reported it to be longest at 60° of flexion.26 Victor et al32 showed that the length of the MPFL was longer at 0° to 40° of flexion than at over 40° of flexion. In an in vivo magnetic resonance imaging (MRI) study, Higuchi et al11 reported that the length of the MPFL showed a plateau pattern from 0° to 60° and then a significant decrease beyond 60° of flexion. To date, little is known about length change patterns of native MPFL fibers, and basic science about these fibers is lacking. In the present study, 2 superior fibers (F20 and F30) showed similar length change patterns (Figure 5A); lengths increased as the knee flexed from a full extension to 30° and decreased beyond 30° of flexion. This indicates that the contribution of these fibers
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
Vol. XX, No. X, XXXX
Length Change Behavior of MPFL Fibers
5
Figure 5. Length change patterns of the medial patellofemoral ligament fibers at knee flexion angles of 0°, 30°, 60°, 90°, and 120°. (A) Two superior fibers (F20 and F30). (B) One middle fiber (F40) and 1 inferior fiber (F50). (C) One inferior fiber (F60). *P \ .00365. to resist patellar dislocation would increase from 0° to 30°, peak at 30°, and become less on further flexion. On the other hand, 1 inferior fiber (F50) showed an increase in its lengths as the knee flexed from 0° to 30°, then a plateau pattern from 30° to 60°, and a decrease as it flexed over 60° (Figure 5B). The other inferior fiber (F60) showed an increase in its length from 0° to 30°, then a plateau pattern from 30° to 90°, and a decrease beyond 90° of flexion (Figure 5C); therefore, the contribution of inferior fibers would increase from 0° to 30°, peak between 30° and 90°, and become less on further flexion. According to our data, the MPFL can be functionally divided into superior and inferior portions, each of which show different length change patterns. It seems that the superior and inferior fibers act synergistically to restrain lateral force throughout the range of knee flexion, with the superior fibers being taut in low flexion and the inferior fibers in mid flexion, respectively. There may be a cascade of tightening and loosening of various MPFL fibers during the range of knee motion. The MPFL provides 50% to 60% restraint to lateral patellar displacement at 0° to 30° of flexion, whereas beyond 30° of flexion, patella stability is mainly provided by articular geometry, and surrounding soft tissue restraints become less important.5,10,21,32 As a result of our data, the superior fibers would be the major fibers compared with the inferior fibers because the superior fibers provide the more important restraint to lateral force in low flexion than in mid to high flexion. These findings can be related to length changes of 2 bundles after DB MPFL reconstruction. Recently, MPFL reconstruction techniques have focused on the anatomic and DB methods. Several in vitro studies have suggested biomechanical advantages of the DB procedure in comparison with the single-bundle procedure.20,32 However, no consensus has been reached regarding relative contributions of superior and inferior bundles in anatomic DB MPFL reconstruction, and the exact function of each bundle has not yet been described.
The results of this in vivo study imply a distinct functional role of each bundle composing the reconstructed MPFL; the role of the superior bundle to resist patellar dislocation is maximal at low flexion angles, whereas that of the inferior bundle is maximal at mid-flexion angles. Thus, each bundle may act reciprocally to restrain the lateral force throughout the range of motion. Accordingly, single-strand ligament reconstruction may not effectively restore lateral patellar instability. More specifically, superior patellar fixation may cause patellar instability at mid to high knee angles, and conversely, inferior patellar fixation may produce excessive laxity at low flexion angles. To reproduce the normal function of the MPFL, a graft should be positioned as broad as possible within the patellar footprint of the MPFL. The importance of the patellar insertion in current MPFL reconstruction remains unclear. In vitro and in vivo studies have suggested that length changes and length change patterns of the reconstructed MPFL are determined by femoral, not patellar, points.2,18,27,29,31,34 It is widely accepted that the femoral fixation site is the single most important factor influencing length change behavior of the graft. As a result of this in vivo study, length changes in 5 fibers were significantly different. That is, there was significant correlation between patellar points and length changes of the MPFL fiber. Furthermore, length change patterns of the MPFL fibers depended on patellar points. The results suggest that the patellar fixation site plays an important role in restoring native length change behavior of the graft in MPFL reconstruction and that an anatomic DB reconstruction that closely replicates the native patellar footprint may reproduce the normal function of the MPFL and stabilize the patella throughout range of knee motion. These data may provide important insights into a proper knee angle at graft fixation. An appropriate knee flexion angle at which the graft should be fixed remains controversial. According to the in vivo data, F20 was longest at 30° of flexion, whereas F60 was longest at 60° (Figure 5, A and
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
6
Song et al
The American Journal of Sports Medicine
C). These results imply appropriate knee angle for graft fixation to be between 30° and 60°. Recently, new aperture fixation techniques with femoral fixation first have been introduced.12,24,33 In these techniques with reverse fixation sequence, the authors recommend that the superior bundle be fixed to the patella at 30° of flexion and inferior bundle be fixed close to 60°. This study has several limitations. First, various points within the native femoral attachment site of the MPFL were not examined. Because the bony surface outlines of the femoral MPFL attachment site have not been well defined, the Scho¨ttle point, considered the central point of the native femoral insertion, was employed. Second, length change was assumed to reflect tensional status within ligament fibers, although actual tension was not measured. However, this assumption was based on several studies suggesting length change to be well correlated with tensional status in the ligament.8,16,30 Third, the 3D knee model from CT images did not express surrounding soft tissue, and the MPFL path was extrapolated from the bony architecture, which might have induced some measurement error. In this regard, a direct analysis of the MPFL path on the 3D model from MRI may be a more appropriate method in that MRI images clearly show the MPFL. Fourth, each subject’s flexion angle was manually measured using a goniometer, not a custom-made frame, which might have affected the results. Finally, static, not dynamic, motion was assessed, which would not reproduce a real situation during knee motion. CT scanning was performed in all volunteers with their quadriceps relaxed. Because the patella translates laterally with the quadriceps contracted,6 there may be some differences in length change pattern between the quadriceps contracted and relaxed state. In addition, the analysis was conducted only at 5 knee flexion angles because of the limitation of radiation exposure for CT scanning. The length change behavior of the MPFL in discrete knee flexion angle could not be the same as that in continuous knee motion. The MPFL stabilizes the patella during the initial 30° knee flexion,21,23 but the length of the MPFL between 0° and 30° of knee flexion was not analyzed. Further research should provide a more in-depth understanding of the functional role of native MPFL fibers during dynamic as well as continuous activity. The strength of this study was that an in vivo knee model was used to measure the length of MPFL fibers during knee motion. In this 3D model based on living tissue, the role of surrounding soft tissue, which makes a substantial contribution to patellofemoral biomechanics, was preserved across the knee. In addition, this study provides the first in vivo analysis to clarify relative contributions of various fibers of the MPFL and address the question of how they change during knee flexion. The results might provide the theoretical background for anatomic DB MPFL reconstruction.
CONCLUSION Superior fibers of the MPFL showed their maximum lengths at low flexion angles, while the inferior fibers
showed their maximum lengths at mid-flexion angles. The results suggest that the MPFL is not a single-bundle structure but a complex of functionally varying fibers with some taut and others slack throughout the range of knee motion.
ACKNOWLEDGMENT The authors thank Korea Bone Bank Co Ltd for software processing.
REFERENCES 1. Ahmad CS, Brown GD, Stein BS. The docking technique for medial patellofemoral ligament reconstruction: surgical technique and clinical outcome. Am J Sports Med. 2009;37(10):2021-2027. 2. Amis AA, Firer P, Mountney J, Senavongse W, Thomas NP. Anatomy and biomechanics of the medial patellofemoral ligament. Knee. 2003;10(3):215-220. 3. Baldwin JL. The anatomy of the medial patellofemoral ligament. Am J Sports Med. 2009;37(12):2355-2361. 4. Dejour H, Walch G, Nove-Josserand L, Guier C. Factors of patellar instability: an anatomic radiographic study. Knee Surg Sports Traumatol Arthrosc. 1994;2(1):19-26. 5. Desio SM, Burks RT, Bachus KN. Soft tissue restraints to lateral patellar translation in the human knee. Am J Sports Med. 1998;26(1):59-65. 6. Duchman K, Mellecker C, El-Hattab AY, Albright JP. Case report: quantitative MRI of tibial tubercle transfer during active quadriceps contraction. Clin Orthop Relat Res. 2011;469(1):294-299. 7. Elias JJ, Cosgarea AJ. Technical errors during medial patellofemoral ligament reconstruction could overload medial patellofemoral cartilage: a computational analysis. Am J Sports Med. 2006;34(9):14781485. 8. Good L. In-vitro correlation between tension and length change in an anterior cruciate ligament substitute. Clin Biomech (Bristol, Avon). 1995;10(4):200-207. 9. Han H, Xia Y, Yun X, Wu M. Anatomical transverse patella double tunnel reconstruction of medial patellofemoral ligament with a hamstring tendon autograft for recurrent patellar dislocation. Arch Orthop Trauma Surg. 2011;131(3):343-351. 10. Hautamaa PV, Fithian DC, Kaufman KR, Daniel DM, Pohlmeyer AM. Medial soft tissue restraints in lateral patellar instability and repair. Clin Orthop Relat Res. 1998;349:174-182. 11. Higuchi T, Arai Y, Takamiya H, Miyamoto T, Tokunaga D, Kubo T. An analysis of the medial patellofemoral ligament length change pattern using open-MRI. Knee Surg Sports Traumatol Arthrosc. 2010;18(11):1470-1475. 12. Kang H, Cao J, Yu D, Zheng Z, Wang F. Comparison of 2 different techniques for anatomic reconstruction of the medial patellofemoral ligament: a prospective randomized study. Am J Sports Med. 2013;41(5):1013-1021. 13. Kang HJ, Wang F, Chen BC, Su YL, Zhang ZC, Yan CB. Functional bundles of the medial patellofemoral ligament. Knee Surg Sports Traumatol Arthrosc. 2010;18(11):1511-1516. 14. LaPrade RF, Engebretsen AH, Ly TV, Johansen S, Wentorf FA, Engebretsen L. The anatomy of the medial part of the knee. J Bone Joint Surg Am. 2007;89(9):2000-2010. 15. Marai GE, Laidlaw DH, Demiralp C, Andrews S, Grimm CM, Crisco JJ. Estimating joint contact areas and ligament lengths from bone kinematics and surfaces. IEEE Trans Biomed Eng. 2004;51(5):790799. 16. Moritomo H, Noda K, Goto A, Murase T, Yoshikawa H, Sugamoto K. Interosseous membrane of the forearm: length change of ligaments during forearm rotation. J Hand Surg Am. 2009;34(4):685-691.
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
Vol. XX, No. X, XXXX
Length Change Behavior of MPFL Fibers
17. Nomura E, Horiuchi Y, Kihara M. Medial patellofemoral ligament restraint in lateral patellar translation and reconstruction. Knee. 2000;7(2):121-127. 18. Nomura E, Inoue M, Osada N. Anatomical analysis of the medial patellofemoral ligament of the knee, especially the femoral attachment. Knee Surg Sports Traumatol Arthrosc. 2005;13(7):510-515. 19. Panni AS, Alam M, Cerciello S, Vasso M, Maffulli N. Medial patellofemoral ligament reconstruction with a divergent patellar transverse 2tunnel technique. Am J Sports Med. 2011;39(12):2647-2655. 20. Parker DA, Alexander JW, Conditt MA, Uzodinma ON, Bryan WJ. Comparison of isometric and anatomic reconstruction of the medial patellofemoral ligament: a cadaveric study. Orthopedics. 2008;31(4):339-343. 21. Philippot R, Boyer B, Testa R, Farizon F, Moyen B. The role of the medial ligamentous structures on patellar tracking during knee flexion. Knee Surg Sports Traumatol Arthrosc. 2012;20(2):331-336. 22. Ronga M, Oliva F, Longo UG, Testa V, Capasso G, Maffulli N. Isolated medial patellofemoral ligament reconstruction for recurrent patellar dislocation. Am J Sports Med. 2009;37(9):1735-1742. 23. Sandmeier RH, Burks RT, Bachus KN, Billings A. The effect of reconstruction of the medial patellofemoral ligament on patellar tracking. Am J Sports Med. 2000;28(3):345-349. 24. Scho¨ttle PB, Hensler D, Imhoff AB. Anatomical double-bundle MPFL reconstruction with an aperture fixation. Knee Surg Sports Traumatol Arthrosc. 2010;18(2):147-151. 25. Scho¨ttle PB, Schmeling A, Rosenstiel N, Weiler A. Radiographic landmarks for femoral tunnel placement in medial patellofemoral ligament reconstruction. Am J Sports Med. 2007;35(5):801-804. 26. Smirk C, Morris H. The anatomy and reconstruction of the medial patellofemoral ligament. Knee. 2003;10(3):221-227.
7
27. Steensen RN, Dopirak RM, McDonald WG III. The anatomy and isometry of the medial patellofemoral ligament: implications for reconstruction. Am J Sports Med. 2004;32(6):1509-1513. 28. Stephen JM, Kaider D, Lumpaopong P, Deehan DJ, Amis AA. The effect of femoral tunnel position and graft tension on patellar contact mechanics and kinematics after medial patellofemoral ligament reconstruction. Am J Sports Med. 2014;42(2):364-372. 29. Stephen JM, Lumpaopong P, Deehan DJ, Kader D, Amis AA. The medial patellofemoral ligament: location of femoral attachment and length change patterns resulting from anatomic and nonanatomic attachments. Am J Sports Med. 2012;40(8):1871-1879. 30. Tan J, Xu J, Xie RG, Deng AD, Tang JB. In vivo length and changes of ligaments stabilizing the thumb carpometacarpal joint. J Hand Surg Am. 2011;36(3):420-427. 31. Tateishi T, Tsuchiya M, Motosugi N, et al. Graft length change and radiographic assessment of femoral drill hole position for medial patellofemoral ligament reconstruction. Knee Surg Sports Traumatol Arthrosc. 2011;19(3):400-407. 32. Victor J, Wong P, Witvrouw E, Sloten JV, Bellemans J. How isometric are the medial patellofemoral, superficial medial collateral, and lateral collateral ligaments of the knee? Am J Sports Med. 2009;37(10): 2028-2036. 33. Wang CH, Ma LF, Zhou JW, et al. Double-bundle anatomical versus single-bundle isometric medial patellofemoral ligament reconstruction for patellar dislocation. Int Orthop. 2013;37(4):617-624. 34. Yoo YS, Chang HG, Seo YJ, et al. Changes in the length of the medial patellofemoral ligament: an in vivo analysis using 3dimensional computed tomography. Am J Sports Med. 2012; 40(9):2142-2148.
For reprints and permission queries, please visit SAGE’s Web site at http://www.sagepub.com/journalsPermissions.nav
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
Length Change Behavior of Virtual Medial Patellofemoral Ligament Fibers During In Vivo Knee Flexion Si Young Song, Chae-Hyun Pang, Chan Hyoek Kim, Jeehyoung Kim, Mi Lim Choi and Young-Jin Seo Am J Sports Med published online February 3, 2015 DOI: 10.1177/0363546514567061 The online version of this article can be found at: http://ajs.sagepub.com/content/early/2015/02/02/0363546514567061
Published by: http://www.sagepublications.com
On behalf of: American Orthopaedic Society for Sports Medicine
Additional services and information for The American Journal of Sports Medicine can be found at:
P> OnlineFirst Version of Record - Feb 3, 2015 What is This?
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
AJSM PreView, published on February 3, 2015 as doi:10.1177/0363546514567061
Length Change Behavior of Virtual Medial Patellofemoral Ligament Fibers During In Vivo Knee Flexion Si Young Song,*y MD, Chae-Hyun Pang,y MD, Chan Hyoek Kim,y MD, Jeehyoung Kim,z MD, Mi Lim Choi,§ MS, and Young-Jin Seo,y MD Investigation performed at Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Republic of Korea Background: In vivo length change behavior of native medial patellofemoral ligament (MPFL) fibers throughout the range of knee motion has not been reported in vivo. Purpose: To measure the length changes of various fibers of the MPFL and to determine their length change patterns during in vivo passive knee flexion. Study Design: Descriptive laboratory study. Methods: The right knees of 11 living subjects were scanned with a high-resolution computed tomography scanner at 0°, 30°, 60°, 90°, and 120° of knee flexion, and 3-dimensional (3D) models were constructed using customized software. Five patellar points were determined: 20% (point 20), 30% (point 30), 40% (point 40), 50% (point 50), and 60% (point 60) from the superior pole of the patella. The Scho¨ttle femoral point (point F) was marked on a translucent 3D model of a true lateral view. Five virtual fibers connecting these points on the 3D knee model were created, and the lengths of various fibers were digitally measured. Results: The average length changes were 9.1 6 2.5 mm in F20, 9.1 6 2.5 mm in F30, 8.1 6 2.6 mm in F40, 6.9 6 2.4 mm in F50, and 6.9 6 1.7 mm in F60. There were significant differences in length changes of these 5 fibers (P \ .001). The lengths of 2 superior fibers (F20 and F30) increased as the knee flexed from 0° to 30° and decreased as the knee flexed over 30°. The lengths of a middle fiber (F40) and an inferior fiber (F50) increased from 0° to 30°, reached a plateau from 30° to 60°, and then decreased from 60° to 120°. F60 showed an increase from 0° to 30°, and then a plateau pattern from 30° to 90°, followed by a decrease during further flexion. Conclusion: Superior fibers exhibited their maximum lengths at low flexion angles, and inferior fibers exhibited their maximum lengths at midflexion angles. The MPFL is a complex of functionally various fibers with some taut and others slack over the whole range of knee motion. Clinical Relevance: The results for lengths and length change patterns of various MPFL fibers are expected to serve as a theoretical background for anatomic double-bundle MPFL reconstruction. Keywords: patella; medial patellofemoral ligament; fiber length change; 3-dimensional knee model
The medial patellofemoral ligament (MPFL) is known to be the primary soft tissue restraint against lateralization of the patella.2,5,26 The biomechanical behavior of the MPFL during knee motion is crucial for the maintenance of patellar stability and control of normal kinematics of the patellofemoral joint, particularly at low flexion angles.17,21,23,26 The MPFL is a triangular and broad condensation of capsular fibers that is wider at its patellar attachment than its femoral attachment.3,13,26 Despite the fact that the MPFL is composed of many ligamentous fibers, little is known about these various fibers, including their lengths, length change patterns, and function. Several studies have examined the length and length changes of the MPFL.26,27,29,32 Although these studies have provided good information about the MPFL fiber complex,
*Address correspondence to Si Young Song, MD, Department of Orthopaedic Surgery, Dongtan Sacred Heart Hospital, Hallym University College of Medicine, 7 Keunjaebong-gil, Hwaseong, Gyeonggi-do, 445170, Republic of Korea (e-mail: [email protected]). y Department of Orthopaedic Surgery, Dongtan Sacred Heart Hospital, Hallym University College of Medicine, Hwaseong, Republic of Korea. z Department of Orthopaedic Surgery, Seoul Sacred Heart General Hospital, Seoul, Republic of Korea. § Department of Data Statistics, Korea Culture & Tourism Institute, Seoul, Republic of Korea. The authors declared that they have no conflicts of interest in the authorship and publication of this contribution. The American Journal of Sports Medicine, Vol. XX, No. X DOI: 10.1177/0363546514567061 Ó 2015 The Author(s)
1 Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
2
Song et al
The American Journal of Sports Medicine
this knowledge is based mainly on cadaveric studies. However, in vitro studies using invasive procedures may not reflect actual kinematics in the living knee because of the removal of adjacent soft tissue around the knee. Recently, an in vivo 3-dimensional (3D) technique has been used to overcome obvious limitations of cadaveric studies and enable the analysis of in vivo motion of any joint and the more accurate measurement of ligament length noninvasively.15,21,34 In the past decade, MPFL reconstruction has become increasingly popular for the treatment of recurrent lateral patellar instability. Because malposition of a graft at the native footprint of the MPFL during reconstruction results in nonphysiological patellofemoral pressure and abnormal patellar tracking,7,22,23,27,28 anatomic graft placement may provide a better option. During the past few years, there has been an increased interest in anatomic double-bundle (DB) MPFL reconstruction, which replicates 2 functional bundles, to more closely restore normal patellofemoral stability and kinematics. To date, various surgical techniques for DB MPFL reconstruction have been described with generally favorable short- to midterm clinical outcomes.1,9,12,19,22,33 Although the concept of anatomic DB reconstruction to restore a triangular form of the MPFL represents a trend in MPFL reconstruction, the biomechanical rationale of the DB reconstruction is not well known. Most studies have focused on surgical techniques and clinical outcomes, but relatively little attention has been directed at the function of each bundle of the reconstructed MPFL. A better understanding of the length change of each bundle during knee flexion could form the basis for advanced anatomic DB MPFL reconstruction techniques. Determining the length of various MPFL fibers and their length change in vivo is a useful way to understand how the MPFL stabilizes the patella. Furthermore, this knowledge on length change behavior will contribute to assumptions about the biomechanical function of each reconstructed bundle. We hypothesized that the native MPFL does not behave as a single bundle of fibers with a constant length but as a continuum of ligament fibers with varying length change behavior during knee flexion. The purpose of this study was to measure the functional lengths of the MPFL fibers and to characterize their length change patterns during knee flexion in vivo.
MATERIALS AND METHODS This study was approved by the institutional review board at Dongtan Sacred Heart Hospital, and an informed consent was obtained from all subjects. The subjects comprised 12 male volunteers with no history of any knee disorders or previous injury. Each subject received physical examinations and radiographic evaluations (anteroposterior, lateral, and Merchant views of the knee and a full-length anteroposterior weightbearing radiograph). One volunteer with high patella (Insall-Savati ratio, 1.3) and high tibial tuberosity–trochlear groove (TTTG) distance (17.5 mm) was excluded. As a result, 11 volunteers were included in this study. The apprehension test and the J-sign were negative in all volunteers, and none
had pain on patellar compression, an effusion, or limitation of motion. The mean age was 32.0 6 3.9 years (range, 27-39 years), the mean height was 175.9 6 4.4 cm (range, 168-181 cm), and the mean body mass index was 26.1 6 2.8 kg/m2 (range, 23.1-32.1 kg/m2). The mean femorotibial angle was 4.5° 6 1.5° mm (range, 2.7°-7.4° mm), the mean sulcus angle was 137.5° 6 4.5° (range, 128.6°-144.1°), the mean congruence angle was 28.3° 6 3.9° (range, 215.6° to 22.5°), and the mean Insall-Savati ratio was 1.03 6 0.11 (range, 0.88-1.15). The mean TTTG distance obtained from computed tomography (CT) scanning was 8.8 6 1.3 mm (range, 7.3-10.9 mm). There was no case of trochlear dysplasia based on the criteria of Dejour et al.4
Construction of 3D Bone Models The right knee of each subject was scanned with a highresolution CT scanner (SOMATOM Sensation; Siemens) at 5 knee flexion angles (0°, 30°, 60°, 90°, and 120°). The flexion angle was determined by a knee surgeon using a goniometer with extendable arms (Gollehon) placed over the thigh and leg. During CT scanning, each subject was asked to take the lateral decubitus position in a relaxed state. Axial plane images with 1-mm slices were obtained, and these CT images were saved in DICOM (Digital Imaging and Communications in Medicine) files. These files were then imported into a visualization software package (Amira 5.3; Mercury Computer Systems), and 3D models of a total of 11 knees were constructed. These 3D images were then imported to a customized software package (Rapidform XOR; INUS Technology) for analysis.
Attachment Site Identification and Creation of Virtual Fibers Previous studies have defined the anatomic attachment site of the MPFL. The patellar insertion of the MPFL is at the medial upper two-thirds of the patella,14,18 and the radiographic landmark of the femoral MPFL attachment site has been defined by Scho¨ttle et al.25 Based on these studies, 5 patellar points were marked on the 3D knee model at 0° of flexion: 20% (point 20), 30% (point 30), 40% (point 40), 50% (point 50), and 60% (point 60) from the superior pole of the patella (Figure 1). To obtain a reproducible radiographic femoral point described by Scho¨ttle et al,25 the 3D surface model at 0° of flexion was converted into a translucent 3D model by using the registration function of a customized software package (Rapidform XOR). This translucent model resembled a fluoroscopic image, and a true lateral radiograph was then made by accurately superimposing the posterior portion of the medial and lateral femoral condyles. The femoral insertion point (point F) was marked on the translucent 3D model and was then reconverted into a 3D surface model (Figure 2). All points were marked by 2 expert knee surgeons jointly. To minimize technical error in the measurement process, the patellar model at 0° of flexion was superimposed on each discrete patellar model at 30°, 60°, 90°, and 120° of flexion, and then superimposed models were again divided into 5 models at different flexion
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
Vol. XX, No. X, XXXX
Length Change Behavior of MPFL Fibers
Figure 1. The patellar medial patellofemoral ligament attachment sites are marked on the 0° knee model (reference model). Five patellar points are marked: 20% (point 20), 30% (point 30), 40% (point 40), 50% (point 50), and 60% (point 60) from the superior pole of the patella. angles. This procedure was performed using the surface-tosurface matching method that was the registration function of the customized software package (Rapidform XOR). With this process, identical patellar points were automatically established on the 3D models at 30°, 60°, 90°, and 120° of flexion (Figure 3). Femoral points in 3D models at 30°, 60°, 90°, and 120° of flexion were established using the same method. A virtual medial capsule connecting the medial side of the patellofemoral joint surface was created on the 3D models (Figure 4A). The shortest line connecting these patellar and femoral points was designated as the MPFL fiber. Virtual linear fibers in the 3D space along the 3D surface of the bone models were represented by surface lines to express native MPFL paths and were detoured by bony protrusions to avoid bone penetration (Figure 4B). In this way, 5 virtual fibers were created on the 3D knee models.
Measurement of Fiber Length The 3D length of a virtual fiber was defined as the shortest linear distance curving around the bony surface between a digitalized femoral point and a patellar point. With the customized software package (Rapidform XOR), lengths of 5 fibers were digitally measured at 5 knee flexion angles. This provided 25 measurements per knee. All measurements were performed twice with an interval of 1 week by 2 knee surgeons. The intra- and interobserver reliability of the measurements was assessed using intraclass and interclass correlation coefficients. Intraclass and interclass correlation coefficients were .99 and .97, respectively. The length changes of the individual fibers at 5 flexion angles were calculated, and length change patterns were analyzed.
Statistical Analysis All data were expressed as the mean 6 standard deviation. The length changes among the 5 fibers were compared
3
Figure 2. With use of the registration function of the customized software, the 3D surface model at 0° of flexion was converted into a translucent 3D model, which can produce images resembling a fluoroscopic image. A true lateral radiograph was then made by accurately superimposing the posterior portion of the medial and lateral femoral condyles. The radiographic Scho¨ttle femoral point (point F) was marked on the translucent 3D model and then reconverted into a 3D surface model. through a 1-way repeated-measures analysis of variance (ANOVA). The level of significance was set at P \ .05. The differences in lengths for angle-by-fiber interaction, pointby-fiber interaction, and angle-by-point interaction were assessed using a 2-way repeated-measures ANOVA with both repeated factors by the Greenhouse-Geisser method. The level of significance was set at P \ .05. The length change pattern for each flexion angle within each fiber was analyzed through a 2-way repeated-measures ANOVA followed by post hoc multiple pairwise comparisons with the Holm-Bonferroni method. According to this method, P \ .00365 was considered significant. The statistical analysis was conducted using SPSS version 20.0 (SPSS Inc).
RESULTS Length Changes of Virtual Fibers All 5 fibers were not isometric throughout the flexionextension arc of the knee joint (Figure 5). During the entire range of knee motion, the average length changes were 9.1 6 2.5 mm in F20, 9.1 6 2.5 mm in F30, 8.1 6 2.6 mm in F40, 6.9 6 2.4 mm in F50, and 6.9 6 1.7 mm in F60. Length changes in these 5 fibers were significantly different (P \ .001).
Length Change Patterns of Virtual Fibers There were significant differences in lengths for angle-byfiber interaction (P = .029), point-by-fiber interaction (P \ .001), and angle-by-point interaction (P \ .001). For 2 superior fibers (F20 and F30) (Figure 5A), their lengths increased as the knee flexed from 0° to 30° and decreased as it flexed over 30°, and the lengths were longest at 30° of flexion and shortest at 120° of flexion. There were significant differences among all the adjacent knee flexion angles in the lengths of 2 superior fibers.
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
4
Song et al
The American Journal of Sports Medicine
Figure 3. Identical patellar points are automatically established on the 3D models at 30°, 60°, 90°, and 120° of flexion by the surface-to-surface matching method. The patellar model at 0° of flexion is superimposed on each discrete patellar model at 30°, 60°, 90°, and 120° of flexion, and then the superimposed models are again divided.
Figure 4. (A) Virtual medial capsule connecting the medial side of the patellofemoral joint surface is created on the 3D models. (B and C) Virtual fibers connecting the patellar and femoral points are created. These fibers are detoured by bony protrusions to avoid bone penetration. F20, the fiber connecting point F and point 20; F30, the fiber connecting point F and point 30; F40, the fiber connecting point F and point 40; F50, the fiber connecting point F and point 50; F60, the fiber connecting point F and point 60. For 1 middle fiber (F40) and 1 inferior fiber (F50) (Figure 5B), the lengths increased from 0° to 30°, then reached a plateau between 30° and 60°, and decreased from 60° to 120°. Except between 30° and 60°, there were significant differences among the adjacent knee flexion angles in its fiber length. For 1 inferior fiber (F60) (Figure 5C), the length increased as the knee flexed from 0° to 30°, then reached a plateau pattern between 30° and 90°, and decreased as it flexed over 90°.
DISCUSSION The MPFL is known to tighten in low to mid flexion and slacken in high flexion, and therefore, the contribution of the MPFL to resist patellar dislocation is greatest in low
to mid flexion.2,21,26,32 However, there is some controversy over the length change patterns of the MPFL. An in vivo biomechanical study reported that the MPFL is longest at 30° of flexion,34 on the other hand, a cadaveric study reported it to be longest at 60° of flexion.26 Victor et al32 showed that the length of the MPFL was longer at 0° to 40° of flexion than at over 40° of flexion. In an in vivo magnetic resonance imaging (MRI) study, Higuchi et al11 reported that the length of the MPFL showed a plateau pattern from 0° to 60° and then a significant decrease beyond 60° of flexion. To date, little is known about length change patterns of native MPFL fibers, and basic science about these fibers is lacking. In the present study, 2 superior fibers (F20 and F30) showed similar length change patterns (Figure 5A); lengths increased as the knee flexed from a full extension to 30° and decreased beyond 30° of flexion. This indicates that the contribution of these fibers
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
Vol. XX, No. X, XXXX
Length Change Behavior of MPFL Fibers
5
Figure 5. Length change patterns of the medial patellofemoral ligament fibers at knee flexion angles of 0°, 30°, 60°, 90°, and 120°. (A) Two superior fibers (F20 and F30). (B) One middle fiber (F40) and 1 inferior fiber (F50). (C) One inferior fiber (F60). *P \ .00365. to resist patellar dislocation would increase from 0° to 30°, peak at 30°, and become less on further flexion. On the other hand, 1 inferior fiber (F50) showed an increase in its lengths as the knee flexed from 0° to 30°, then a plateau pattern from 30° to 60°, and a decrease as it flexed over 60° (Figure 5B). The other inferior fiber (F60) showed an increase in its length from 0° to 30°, then a plateau pattern from 30° to 90°, and a decrease beyond 90° of flexion (Figure 5C); therefore, the contribution of inferior fibers would increase from 0° to 30°, peak between 30° and 90°, and become less on further flexion. According to our data, the MPFL can be functionally divided into superior and inferior portions, each of which show different length change patterns. It seems that the superior and inferior fibers act synergistically to restrain lateral force throughout the range of knee flexion, with the superior fibers being taut in low flexion and the inferior fibers in mid flexion, respectively. There may be a cascade of tightening and loosening of various MPFL fibers during the range of knee motion. The MPFL provides 50% to 60% restraint to lateral patellar displacement at 0° to 30° of flexion, whereas beyond 30° of flexion, patella stability is mainly provided by articular geometry, and surrounding soft tissue restraints become less important.5,10,21,32 As a result of our data, the superior fibers would be the major fibers compared with the inferior fibers because the superior fibers provide the more important restraint to lateral force in low flexion than in mid to high flexion. These findings can be related to length changes of 2 bundles after DB MPFL reconstruction. Recently, MPFL reconstruction techniques have focused on the anatomic and DB methods. Several in vitro studies have suggested biomechanical advantages of the DB procedure in comparison with the single-bundle procedure.20,32 However, no consensus has been reached regarding relative contributions of superior and inferior bundles in anatomic DB MPFL reconstruction, and the exact function of each bundle has not yet been described.
The results of this in vivo study imply a distinct functional role of each bundle composing the reconstructed MPFL; the role of the superior bundle to resist patellar dislocation is maximal at low flexion angles, whereas that of the inferior bundle is maximal at mid-flexion angles. Thus, each bundle may act reciprocally to restrain the lateral force throughout the range of motion. Accordingly, single-strand ligament reconstruction may not effectively restore lateral patellar instability. More specifically, superior patellar fixation may cause patellar instability at mid to high knee angles, and conversely, inferior patellar fixation may produce excessive laxity at low flexion angles. To reproduce the normal function of the MPFL, a graft should be positioned as broad as possible within the patellar footprint of the MPFL. The importance of the patellar insertion in current MPFL reconstruction remains unclear. In vitro and in vivo studies have suggested that length changes and length change patterns of the reconstructed MPFL are determined by femoral, not patellar, points.2,18,27,29,31,34 It is widely accepted that the femoral fixation site is the single most important factor influencing length change behavior of the graft. As a result of this in vivo study, length changes in 5 fibers were significantly different. That is, there was significant correlation between patellar points and length changes of the MPFL fiber. Furthermore, length change patterns of the MPFL fibers depended on patellar points. The results suggest that the patellar fixation site plays an important role in restoring native length change behavior of the graft in MPFL reconstruction and that an anatomic DB reconstruction that closely replicates the native patellar footprint may reproduce the normal function of the MPFL and stabilize the patella throughout range of knee motion. These data may provide important insights into a proper knee angle at graft fixation. An appropriate knee flexion angle at which the graft should be fixed remains controversial. According to the in vivo data, F20 was longest at 30° of flexion, whereas F60 was longest at 60° (Figure 5, A and
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
6
Song et al
The American Journal of Sports Medicine
C). These results imply appropriate knee angle for graft fixation to be between 30° and 60°. Recently, new aperture fixation techniques with femoral fixation first have been introduced.12,24,33 In these techniques with reverse fixation sequence, the authors recommend that the superior bundle be fixed to the patella at 30° of flexion and inferior bundle be fixed close to 60°. This study has several limitations. First, various points within the native femoral attachment site of the MPFL were not examined. Because the bony surface outlines of the femoral MPFL attachment site have not been well defined, the Scho¨ttle point, considered the central point of the native femoral insertion, was employed. Second, length change was assumed to reflect tensional status within ligament fibers, although actual tension was not measured. However, this assumption was based on several studies suggesting length change to be well correlated with tensional status in the ligament.8,16,30 Third, the 3D knee model from CT images did not express surrounding soft tissue, and the MPFL path was extrapolated from the bony architecture, which might have induced some measurement error. In this regard, a direct analysis of the MPFL path on the 3D model from MRI may be a more appropriate method in that MRI images clearly show the MPFL. Fourth, each subject’s flexion angle was manually measured using a goniometer, not a custom-made frame, which might have affected the results. Finally, static, not dynamic, motion was assessed, which would not reproduce a real situation during knee motion. CT scanning was performed in all volunteers with their quadriceps relaxed. Because the patella translates laterally with the quadriceps contracted,6 there may be some differences in length change pattern between the quadriceps contracted and relaxed state. In addition, the analysis was conducted only at 5 knee flexion angles because of the limitation of radiation exposure for CT scanning. The length change behavior of the MPFL in discrete knee flexion angle could not be the same as that in continuous knee motion. The MPFL stabilizes the patella during the initial 30° knee flexion,21,23 but the length of the MPFL between 0° and 30° of knee flexion was not analyzed. Further research should provide a more in-depth understanding of the functional role of native MPFL fibers during dynamic as well as continuous activity. The strength of this study was that an in vivo knee model was used to measure the length of MPFL fibers during knee motion. In this 3D model based on living tissue, the role of surrounding soft tissue, which makes a substantial contribution to patellofemoral biomechanics, was preserved across the knee. In addition, this study provides the first in vivo analysis to clarify relative contributions of various fibers of the MPFL and address the question of how they change during knee flexion. The results might provide the theoretical background for anatomic DB MPFL reconstruction.
CONCLUSION Superior fibers of the MPFL showed their maximum lengths at low flexion angles, while the inferior fibers
showed their maximum lengths at mid-flexion angles. The results suggest that the MPFL is not a single-bundle structure but a complex of functionally varying fibers with some taut and others slack throughout the range of knee motion.
ACKNOWLEDGMENT The authors thank Korea Bone Bank Co Ltd for software processing.
REFERENCES 1. Ahmad CS, Brown GD, Stein BS. The docking technique for medial patellofemoral ligament reconstruction: surgical technique and clinical outcome. Am J Sports Med. 2009;37(10):2021-2027. 2. Amis AA, Firer P, Mountney J, Senavongse W, Thomas NP. Anatomy and biomechanics of the medial patellofemoral ligament. Knee. 2003;10(3):215-220. 3. Baldwin JL. The anatomy of the medial patellofemoral ligament. Am J Sports Med. 2009;37(12):2355-2361. 4. Dejour H, Walch G, Nove-Josserand L, Guier C. Factors of patellar instability: an anatomic radiographic study. Knee Surg Sports Traumatol Arthrosc. 1994;2(1):19-26. 5. Desio SM, Burks RT, Bachus KN. Soft tissue restraints to lateral patellar translation in the human knee. Am J Sports Med. 1998;26(1):59-65. 6. Duchman K, Mellecker C, El-Hattab AY, Albright JP. Case report: quantitative MRI of tibial tubercle transfer during active quadriceps contraction. Clin Orthop Relat Res. 2011;469(1):294-299. 7. Elias JJ, Cosgarea AJ. Technical errors during medial patellofemoral ligament reconstruction could overload medial patellofemoral cartilage: a computational analysis. Am J Sports Med. 2006;34(9):14781485. 8. Good L. In-vitro correlation between tension and length change in an anterior cruciate ligament substitute. Clin Biomech (Bristol, Avon). 1995;10(4):200-207. 9. Han H, Xia Y, Yun X, Wu M. Anatomical transverse patella double tunnel reconstruction of medial patellofemoral ligament with a hamstring tendon autograft for recurrent patellar dislocation. Arch Orthop Trauma Surg. 2011;131(3):343-351. 10. Hautamaa PV, Fithian DC, Kaufman KR, Daniel DM, Pohlmeyer AM. Medial soft tissue restraints in lateral patellar instability and repair. Clin Orthop Relat Res. 1998;349:174-182. 11. Higuchi T, Arai Y, Takamiya H, Miyamoto T, Tokunaga D, Kubo T. An analysis of the medial patellofemoral ligament length change pattern using open-MRI. Knee Surg Sports Traumatol Arthrosc. 2010;18(11):1470-1475. 12. Kang H, Cao J, Yu D, Zheng Z, Wang F. Comparison of 2 different techniques for anatomic reconstruction of the medial patellofemoral ligament: a prospective randomized study. Am J Sports Med. 2013;41(5):1013-1021. 13. Kang HJ, Wang F, Chen BC, Su YL, Zhang ZC, Yan CB. Functional bundles of the medial patellofemoral ligament. Knee Surg Sports Traumatol Arthrosc. 2010;18(11):1511-1516. 14. LaPrade RF, Engebretsen AH, Ly TV, Johansen S, Wentorf FA, Engebretsen L. The anatomy of the medial part of the knee. J Bone Joint Surg Am. 2007;89(9):2000-2010. 15. Marai GE, Laidlaw DH, Demiralp C, Andrews S, Grimm CM, Crisco JJ. Estimating joint contact areas and ligament lengths from bone kinematics and surfaces. IEEE Trans Biomed Eng. 2004;51(5):790799. 16. Moritomo H, Noda K, Goto A, Murase T, Yoshikawa H, Sugamoto K. Interosseous membrane of the forearm: length change of ligaments during forearm rotation. J Hand Surg Am. 2009;34(4):685-691.
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
Vol. XX, No. X, XXXX
Length Change Behavior of MPFL Fibers
17. Nomura E, Horiuchi Y, Kihara M. Medial patellofemoral ligament restraint in lateral patellar translation and reconstruction. Knee. 2000;7(2):121-127. 18. Nomura E, Inoue M, Osada N. Anatomical analysis of the medial patellofemoral ligament of the knee, especially the femoral attachment. Knee Surg Sports Traumatol Arthrosc. 2005;13(7):510-515. 19. Panni AS, Alam M, Cerciello S, Vasso M, Maffulli N. Medial patellofemoral ligament reconstruction with a divergent patellar transverse 2tunnel technique. Am J Sports Med. 2011;39(12):2647-2655. 20. Parker DA, Alexander JW, Conditt MA, Uzodinma ON, Bryan WJ. Comparison of isometric and anatomic reconstruction of the medial patellofemoral ligament: a cadaveric study. Orthopedics. 2008;31(4):339-343. 21. Philippot R, Boyer B, Testa R, Farizon F, Moyen B. The role of the medial ligamentous structures on patellar tracking during knee flexion. Knee Surg Sports Traumatol Arthrosc. 2012;20(2):331-336. 22. Ronga M, Oliva F, Longo UG, Testa V, Capasso G, Maffulli N. Isolated medial patellofemoral ligament reconstruction for recurrent patellar dislocation. Am J Sports Med. 2009;37(9):1735-1742. 23. Sandmeier RH, Burks RT, Bachus KN, Billings A. The effect of reconstruction of the medial patellofemoral ligament on patellar tracking. Am J Sports Med. 2000;28(3):345-349. 24. Scho¨ttle PB, Hensler D, Imhoff AB. Anatomical double-bundle MPFL reconstruction with an aperture fixation. Knee Surg Sports Traumatol Arthrosc. 2010;18(2):147-151. 25. Scho¨ttle PB, Schmeling A, Rosenstiel N, Weiler A. Radiographic landmarks for femoral tunnel placement in medial patellofemoral ligament reconstruction. Am J Sports Med. 2007;35(5):801-804. 26. Smirk C, Morris H. The anatomy and reconstruction of the medial patellofemoral ligament. Knee. 2003;10(3):221-227.
7
27. Steensen RN, Dopirak RM, McDonald WG III. The anatomy and isometry of the medial patellofemoral ligament: implications for reconstruction. Am J Sports Med. 2004;32(6):1509-1513. 28. Stephen JM, Kaider D, Lumpaopong P, Deehan DJ, Amis AA. The effect of femoral tunnel position and graft tension on patellar contact mechanics and kinematics after medial patellofemoral ligament reconstruction. Am J Sports Med. 2014;42(2):364-372. 29. Stephen JM, Lumpaopong P, Deehan DJ, Kader D, Amis AA. The medial patellofemoral ligament: location of femoral attachment and length change patterns resulting from anatomic and nonanatomic attachments. Am J Sports Med. 2012;40(8):1871-1879. 30. Tan J, Xu J, Xie RG, Deng AD, Tang JB. In vivo length and changes of ligaments stabilizing the thumb carpometacarpal joint. J Hand Surg Am. 2011;36(3):420-427. 31. Tateishi T, Tsuchiya M, Motosugi N, et al. Graft length change and radiographic assessment of femoral drill hole position for medial patellofemoral ligament reconstruction. Knee Surg Sports Traumatol Arthrosc. 2011;19(3):400-407. 32. Victor J, Wong P, Witvrouw E, Sloten JV, Bellemans J. How isometric are the medial patellofemoral, superficial medial collateral, and lateral collateral ligaments of the knee? Am J Sports Med. 2009;37(10): 2028-2036. 33. Wang CH, Ma LF, Zhou JW, et al. Double-bundle anatomical versus single-bundle isometric medial patellofemoral ligament reconstruction for patellar dislocation. Int Orthop. 2013;37(4):617-624. 34. Yoo YS, Chang HG, Seo YJ, et al. Changes in the length of the medial patellofemoral ligament: an in vivo analysis using 3dimensional computed tomography. Am J Sports Med. 2012; 40(9):2142-2148.
For reprints and permission queries, please visit SAGE’s Web site at http://www.sagepub.com/journalsPermissions.nav
Downloaded from ajs.sagepub.com at COLUMBIA UNIV on February 14, 2015
