Surg Radiol Anat DOI 10.1007/s00276-014-1270-1
Original Article
Anatomical study of the medial patello‑femoral ligament: landmarks for its surgical reconstruction Anthony Viste · Florian Chatelet · Romain Desmarchelier · Michel‑Henri Fessy
Received: 31 July 2013 / Accepted: 6 February 2014 © Springer-Verlag France 2014
Abstract Purpose The aim of this dissection study was to describe the anatomical insertions of the medial patello-femoral ligament (MPFL), and to assess its relationship with surrounding structures to improve its surgical reconstruction. Methods Twelve knees (7 cadavers) were included for the study. Measurements and general features of the MPFL were assessed: lengths, widths and insertions. Results The MPFL was found in all knees, presenting a triangular shape, and extending from the medial part of the patella to its femoral insertion (its length was of 59 ± 6.6 mm), distal to the adductor tubercle. The mean femoral insertion of the MPFL was 7.2 ± 2.7 mm proximal and 7.4 ± 4.0 mm posterior to the medial femoral epicondyle (MFE). It was also at a mean 11 ± 2.8 mm distal and 1.3 ± 2.1 mm posterior to the adductor tubercle, and 22 ± 6.4 mm anterior to the posterior condyle. We did not find any double-bundle organization on the patellar insertion. The width of the MPFL was 8.8 ± 2.9 mm at the A. Viste · F. Chatelet · M.-H. Fessy Faculté de Médecine Lyon Sud-Charles Mérieux, Laboratoire d’Anatomie, 165 Chemin du Petit Revoyet, BP 12, 69921 Oullins Cedex, France A. Viste (*) · R. Desmarchelier · M.-H. Fessy Hospices Civils de Lyon, Centre Hospitalier Lyon-Sud, Service de Chirurgie Orthopédique, Traumatologique et Médecine du Sport, 165 Chemin Grand Revoyet, 69495 Pierre Bénite, Cedex, France e-mail: anthony.viste@chu‑lyon.fr A. Viste · M.-H. Fessy Université de Lyon, Lyon, France A. Viste · M.-H. Fessy IFSTTAR, UMRT_9406, LBMC, Université Lyon1, Villeurbanne, France
femoral insertion, 27 ± 5.9 mm at the patellar insertion, and 12 ± 3.1 mm in the middle of the MPFL. The vastus medialis obliquus was found to be inserted on the superior part of the MPFL. Conclusion The adductor tubercle appeared to be a better landmark than the MFE for the femoral tunnel positioning during surgical reconstructions of the MPFL because it was easier to identify and its relationship with the femoral insertion of the MPFL was constant (10 mm below). Keywords Medial patello-femoral ligament (MPFL) · Patellar dislocation · Reconstruction · Ligamentous knee
Introduction The medial patello-femoral ligament (MPFL) has been considered to be the main passive soft-tissue restraint to lateral displacement of the patella [3, 29]. The MPFL has been first described, but not namely mentioned, in 1957 by Kaplan [11]. The MPFL seems to be injured in almost 100 % of patellar dislocations [16] and does not reinsert afterward, leading to chronic patellar instability. Recently, the importance of the MPFL has been demonstrated biomechanically and clinically [1, 4, 9]. In 1979, Warren and Marshall [27] described the anatomy of the medial side of the knee and introduced the concept of superimposed layers system. Layer 1 was the crural fascia above the Sartorius tendon. Layer 2 was composed of the superficial medial collateral ligament and the tendons of the semimembranosus, semitendinosus and gracilis muscles. In the second layer, the MPFL was described as transverse fibers, extending from the insertion of the superficial medial collateral ligament and the medial side of the patella, deeper than the vastus medialis obliquus (VMO).
13
Surg Radiol Anat
Fig. 1 The constant triangular area between the VMO, the AMT and the MPFL. MPFL medial patello-femoral ligament, AMT adductor magnus tendon, VMO vastus medialis obliquus
Layer 3 corresponded to the joint capsule of the knee and the deep medial collateral ligament. Many knee surgeons have developed MPFL reconstruction techniques [6, 13] for treating recurrent lateral patellar dislocations. Surgeons tried to restore an anatomical and isometric medial stabilizer of the patella. However, the femoral insertion remained controversial, whereas the importance of the femoral tunnel’s location for graft has been demonstrated [7, 23]. The main issues for surgical reconstruction remain the risk of anisometry, the precise locations of its attachments on the femur and patella and, above it all, its relationships to pertinent osseous landmarks. The aim of this dissection study was to describe the anatomical insertions of the MPFL, and to assess its relationship with surrounding structures to improve its surgical reconstruction.
Materials and methods Twelve knees (5 left and 7 right) from seven embalmed cadavers (3 male and 4 female) were included for study and provided by the Department of Anatomy at the Medicine Faculty Lyon-Est. They were all intact knees and did not have scars or bone abnormality around the knee. The mean age of the donors was 75 years (range 70–83). Each knee was stored frozen at 20 °C and then thawed the day prior dissection.
13
The skin and subcutaneous tissues of the medial part of the knee were first removed. We excised the crural fascia. Then, using a lateral force applied on the patella, and the constant [2, 10] triangular area between the VMO and the adductor tubercle, we identified the MPFL (Fig. 1). We removed the crural fascia over the MPFL and with a forceps inserted under the MPFL, and we dissected the femoral and patellar insertions (Fig. 2). The MPFL was easily individualized from the capsule. The adductor tubercle and the medial femoral epicondyle (MFE) were then identified. All measurements were taken at 30° of knee flexion with a millimetric scale ruler (±1 mm precision). General features of the MPFL were investigated. We measured distances between the femoral insertion of the MPFL and the MFE, the adductor tubercle and posterior medial femoral condyle. We also measured the transverse length of the MPFL, from the femoral attachment to the middle of the patellar insertion. We measured the length of the insertion of the VMO on the MPFL, and then we reflected the VMO proximally to identify the insertion of the superior edge of the MPFL. We measured the width of the MPFL at the patellar insertion, at the femoral insertion and in the middle of the MPFL, and we measured the overall patellar length.
Results The MPFL was found in all the knees dissected. The MPFL was situated in layer 2, under the superficial
Surg Radiol Anat
Fig. 2 MPFL and its insertions. FI femoral insertion, AMT adductor magnus tendon
retinaculum. This ligament had approximately a triangular shape. The proximal edge was easily identified, loosely attached under the VMO, and it was easy to release it from the MPFL. The distal edge of the MPFL was more difficult to dissect because of its relation with the medial retinaculum and the medial patello-meniscal ligament (MPML). On the distal part of the patella, we found the medial patello-tibial ligament (MPTL) to be inserted distal to the MPFL. Patellar insertion The patellar insertion of the MPFL was found to locate on the proximal part of the medial side of the patella. The average overall length of the patella was 51 ± 2.5 mm (range 47–55 mm), and the width of the MPFL at the patellar insertion was 27 ± 5.9 mm (range 20–39 mm). So, the MPFL was found to be inserted on the superior half part of the patella.
posterior to the MFE. It was also 11 ± 2.8 mm (range 7–15) distal and 1.3 ± 2.1 mm (range 0–6) posterior to the adductor tubercle, and 22 ± 6.4 mm (range 15– 31) anterior to the posterior condyle. The width of the MPFL at the femoral insertion was 8.8 ± 2.9 mm (range 5–14 mm). General dimensions We have not found, during our dissections, any extension of the MPFL fibers to the medial collateral ligament or to the adductor magnus tendon (Fig. 2). The length of the MPFL was 59 ± 6.6 mm (range 50–67), and the width in the middle of the MPFL was 12 ± 3.1 (range 6–16). Upper insertion of the VMO on the MPFL was found in 12 knees (100 %) and its length was 31 ± 6.0 mm (range 22–40).
Discussion (Table 1) Femoral insertion The femoral insertion of the MPFL was 7.2 ± 2.7 mm (range 3–13) proximal and 7.4 ± 4.0 mm (range 0–15)
This study was performed to describe the anatomical insertions of the MPFL and its relationship with surrounding structures. Anatomists and orthopedic surgeons
13
Surg Radiol Anat
Table 1 Anatomy of the MPFL: review of literature Authors
Number of knees (n)
Presence of MPFL (%)
Mean width (mm)
Average location of the FI (mm)
Mean length (mm)
At the FI
In the middle
At the PI
To the MFE
To the AT
– – Anterior 9 Prox 13 Post 11 Prox 9 Post 0
0 0 – –
56 – – 72
2 Ant 4 Distal –
65
9.5 Prox 5 Post 10 Prox Posterior Anterior –
–
59
11 Distal 10 Distal – 0
58 58 – 53
7 Prox 7 Post
11 Distal 1 Post
59
Aragao et al. [2] Conlan et al. [3] Feller et al. [8] Kang et al. [10]
17 33 20 12
88 88 100 100
17 13 – –
20 – – –
28 – – 22
LaPrade et al. [12]
20
100
–
–
–
Mochizuki et al. [14]
16
100
–
–
Nomura et al. [17]
20
100
–
12
24 (medial to VI) –
Philippot et al. [18] Smirk and Morris [23] Steensen et al. [25] Tuxoe et al. [26]
23 25 11 39
100 100 100 100
12 – 15 19 (10 mm lateral to AT)
– – – –
24 – 17 –
Our study
12 (7 donors)
100
9
12
27
56
PI patellar insertion, FI femoral insertion, prox proximal, post posterior, ant anterior, VI vastus intermedius, AT adductor tubercle, MFE medial femoral epicondyle
are divided about its existence as a distinct entity: Reider et al. [19] thought MPFL was a condensation of the medial capsule of the knee joint, contrarily to our opinion and the fact that MPFL was always present in our dissections. The ultimate goal was to improve its surgical reconstruction. The current landmark used to drill the femoral tunnel is the MFE (posteriorly) but we think that this wide area is more difficult to identify accurately. So, we assert that the best landmark for femoral tunnel positioning is the adductor tubercle: the MPFL femoral insertion was 10 mm distal to the adductor tubercle. Biomechanical studies have reported that the MPFL provided 50–60 % of the medial stabilizing force of the patella (primary stabilizer) [3, 4, 9]. These studies demonstrated the need for surgical reconstruction of the MPFL in case of patellar dislocations. Clinical studies have found a MPFL loose or torn in lateral patellar dislocations without bone abnormalities [15, 20, 25]. Our study found the MPFL within layer II (as described by Warren and Marshall [27]) in 100 % of knees dissected. Conlan et al. [3] found the MPFL in 90 % of knees, Aragao et al. [2] in 88 % of knees and Reider et al. [19] in only 35 % of knees. More recent studies found the MPFL in all knees [4, 10, 23] as in our study. The limitations of this study could be the paired knees included, the embalmed cadavers (vs. fresh frozen), the measurement precision (±1 mm) and the number of knees dissected but only small variations were reported.
13
Femoral insertion Femoral insertion was described either at the adductor tubercle [3, 4, 8, 26], or at the anterior part of the MFE [8], or at the MFE [2, 9, 27] or between the MFE and the adductor tubercle [10, 17]. In our study, we found the femoral insertion of the MPFL to be located just distal to the adductor tubercle and proximal-posterior from the MFE. We studied the distance between the insertion of the MPFL and the posterior medial femoral condyle, but high variations in values (range of 15–31 mm with standard deviation of 6.4) demonstrated that this anatomical distance could not be used to locate precisely the femoral insertion of the MPFL. Distances between the MFE and the femoral insertion of the MPFL were similar to the literature (Table 1). The MFE has been widely used by authors as the landmark to locate the femoral insertion of the MPFL [10, 17]. In our study, the average distance (1.3 mm) and standard deviation (2.1) of the transversal length between the adductor tubercle and the femoral insertion of the MPFL were low. We also found in eight knees (67 %) a femoral insertion of the MPFL just distal from the adductor tubercle. We think that this low variation can be used in the surgical reconstruction of the MPFL, using the adductor tubercle to locate the femoral insertion of the MPFL, instead of using the MFE, whose values of the distance from the femoral
Surg Radiol Anat
insertion of the MPFL are more variable. Smirk and Morris [23] also found a femoral insertion of the MPFL most commonly located approximately 1 cm distal to the adductor tubercle and they demonstrated that better results were obtained with a graft posterior to the adductor tubercle rather than anterior. Other authors [17, 26] found a femoral insertion of the MPFL located just distally to the adductor tubercle without more precisions. In our study, we found in four knees (33 %) an insertion slightly posterior to the adductor tubercle (maximum 6 mm). Some authors found, at the femoral insertion of the MPFL, decussations to the adductor magnus tendon [2, 4] and to the medial collateral ligament [4, 25, 26]. According to LaPrade et al. [12], the adductor magnus tendon did not attach directly to the adductor tubercle (3 mm proximal and posterior to the adductor tubercle). We didn’t find any fiber of this type in our study, potentially explained by our dissection technique, focused on functional fibers of the MPFL. Decussations appeared to belong to the medial retinaculum and did not seem to have mechanical functions. Wijdicks et al. [28] and Schottle et al. [22] dealt with the radiographic landmarks of the MPFL femoral insertion. It is routinely used postoperatively in MPFL reconstruction to make sure that the femoral tunnel was accurately located. Two lines were used to define the accurate femoral location: a line tangent to the posterior cortex of the femur and a line perpendicular tangent to the posterior portion of the Blumensaat line. Wijdicks et al. [28] found that MPFL was 15.9 mm from the medial epicondyle and 8.9 mm from the adductor tubercle. In the study by Schottle et al. [22], the femoral attachment was found 1.3 mm anterior to the posterior femoral cortex and 5.5 mm proximal to the perpendicular line whereas Wijdicks found it at 8.8 mm anterior and 2.6 mm proximal. But in both cases, the femoral attachment was located in the antero-proximal quadrant.
the MPFL with this configuration in mind, but we could not find any element in favor of a double-bundle organization. Relationship with the quadriceps muscles Dirim et al. [5] found, based on MR images and histological examination, a relationship between superior fibers of the MPFL and the fibers of the VMO, impossible to separate, except at the patellar attachment, where bilaminar images had been obtained on MRI. In their anatomic study, Mochizuki et al. [14] found a main attachment between the MPFL and the vastus intermedius, without tight adhesion to the VMO. In our dissections, we found some proximal fibers of the MPFL, which ran over the superior-medial corner of the patella, under the VMO, to insert on the deep part of the quadriceps tendon, without extension toward the belly of the rectus intermedius. These results are similar to Conlan et al. [3] who found proximal fibers extending to the aponeurotic fibers of the vastus intermedius. But those fibers did not represent a functional bundle inserted on the quadriceps tendon on all knees dissected. This insertion of the VMO was not always found [2, 23]. The length of the upper insertion of the VMO on the MPFL was 31 ± 6.0 mm (range 20–40 mm) in our study, and present in 100 % of the knees. These results were larger than in the literature, Nomura et al. [17] found a length of 20.3 ± 6.1 mm (range 10–32 mm), and Philippot et al. [18] found a length of 25.7 ± 6.0 mm (range 19–38 mm). Even if our results are not significantly different from the literature, our results insist on the importance of this insertion. The existence of this insertion, even if there is no tight adhesion, could demonstrate the role in the dynamic restraint in lateral displacement of the patella played by the MPFL through the action of the VMO. General dimensions
Patellar insertion Patellar insertion has been reported at the proximal half of the patella [1, 3, 25], proximal two-thirds of the patella [23, 26, 27], at the vastus medialis tendon [8], and to the quadriceps tendon [3, 14]. In our study, we found the length of the patellar insertion was 27 ± 5.9 mm (range 20–39), and a patellar height of 51 ± 2.5 mm (range 47–55). First mentioned by Amis et al. [1], functional bundles have been described by Kang et al. [10].They found two functional bundles of the MPFL on the patellar insertion. The inferior straight bundle runs from the femoral insertion to the medial side of the patella, and is the main static soft structure restraint. The superior oblique bundle runs from the patellar insertion to the upper end of the patella and quadriceps fibers, and is a dynamic soft-tissue restraint to patellar lateral displacement. In our study, we dissected
In our study, the length of the MPFL was 59 ± 6.6 mm (range 50–67). This is similar to the literature (Table 1). Presentation could be described as a hourglass shape [4]. In our study, the width at the middle of the MPFL was greater to the width at the femoral insertion in 11 knees (92 %). In one knee (8 %) the widths at the middle of the MPFL and at its femoral insertion were equal. The width at the patellar insertion was always bigger to the width in the middle of the MPFL. This triangular shape, found in all the knees (100 %), is present in other studies [2, 23]. Implications for knee surgeons Surgical treatment for patellar instability has been related to lower rate of re-dislocation than non-operative management [24]. Some authors [21] proposed to restore the
13
triangular shape of the MPFL, using a double-bundle graft, or to combine reconstruction of MPFL and MPTL [6]. Knowledge of the anatomy of the medial part of the knee and of soft-tissue restraint to lateral displacement of the patella seems to be the key for any surgery of the MPFL. Elias and Cosgarea [7] demonstrated that the femoral insertion of the graft, in MPFL reconstructions, had a significant impact on forces applied on the medial patellofemoral cartilage, which can lead to cartilage degradation, arthritis, and pain. Smirk and Morris [23] also demonstrated that the location of the femoral tunnel had an influence on graft isometry. As a result, we advocate the use of the adductor tubercle as a landmark for surgeries instead of the most commonly used MFE. Lower variations in the distances between the adductor tubercle and the femoral insertion of the MPFL than the MFE could allow avoiding some errors. Moreover, the MFE is usually not well defined: it looks like a C-shaped ridge with a central sulcus for the insertion of the superficial medial collateral ligament [12].
Conclusion The MPFL was always present, running from the proximal and medial part of the patella to a femoral insertion distal to the adductor tubercle and posterior to the MFE. The VMO inserted on the MPFL on its superior edge, whereas the MPFL extended under the VMO to insert on the quadriceps tendon. Our results are mostly similar to the literature, but imply to use the adductor tubercle as a landmark during reconstructions of the MPFL, which would provide a more constant anatomical landmark. Conflict of interest None.
References 1. Amis AA, Firer P, Mountney J, Senavongse W, Thomas NP (2003) Anatomy and biomechanics of the medial patellofemoral ligament. Knee 10(3):215–220 2. Aragao JA, Reis FP, de Vasconcelos DP, Feitosa VL, Nunes MA (2008) Metric measurements and attachment levels of the medial patellofemoral ligament: an anatomical study in cadavers. Clinics (Sao Paulo) 63(4):541–544 3. Conlan T, Garth WP Jr, Lemons JE (1993) Evaluation of the medial soft-tissue restraints of the extensor mechanism of the knee. J Bone Jt Surg Am 75(5):682–693 4. Desio SM, Burks RT, Bachus KN (1998) Soft tissue restraints to lateral patellar translation in the human knee. Am J Sports Med 26(1):59–65 5. Dirim B, Haghighi P, Trudell D, Portes G, Resnick D (2008) Medial patellofemoral ligament: cadaveric investigation of anatomy with MRI, MR arthrography, and histologic correlation. AJR 191(2):490–498. doi:10.2214/AJR.07.3590
13
Surg Radiol Anat 6. Ebied AM, El-Kholy W (2012) Reconstruction of the medial patello-femoral and patello-tibial ligaments for treatment of patellar instability. Knee Surg Sports Traumatol Arthrosc 20(5):926–932. doi:10.1007/s00167-011-1659-x 7. Elias JJ, Cosgarea AJ (2006) Technical errors during medial patellofemoral ligament reconstruction could overload medial patellofemoral cartilage: a computational analysis. Am J Sports Med 34(9):1478–1485. doi:10.1177/0363546506287486 8. Feller JA, Feagin JA Jr, Garrett WE Jr (1993) The medial patellofemoral ligament revisited: an anatomical study. Knee Surg Sports Traumatol Arthrosc 1(3–4):184–186 9. Hautamaa PV, Fithian DC, Kaufman KR, Daniel DM, Pohlmeyer AM (1998) Medial soft tissue restraints in lateral patellar instability and repair. Clin Orthop Relat Res 349:174–182 10. Kang HJ, Wang F, Chen BC, Su YL, Zhang ZC, Yan CB (2010) Functional bundles of the medial patellofemoral ligament. Knee Surg Sports Traumatol Arthrosc 18(11):1511–1516. doi:10.1007/ s00167-010-1090-8 11. Kaplan EB (1957) Factors responsible for the stability of the knee joint. Bull Hosp Jt Dis 18(1):51–59 12. LaPrade RF, Engebretsen AH, Ly TV, Johansen S, Wentorf FA, Engebretsen L (2007) The anatomy of the medial part of the knee. J Bone Jt Surg Am 89(9):2000–2010. doi:10.2106/JBJ S.F.01176 13. Lind M, Jakobsen BW, Lund B, Christiansen SE (2008) Reconstruction of the medial patellofemoral ligament for treatment of patellar instability. Acta Orthop 79(3):354–360. doi:10.1080/17453670710015256 14. Mochizuki T, Nimura A, Tateishi T, Yamaguchi K, Muneta T, Akita K (2013) Anatomic study of the attachment of the medial patellofemoral ligament and its characteristic relationships to the vastus intermedius. Knee Surg Sports Traumatol Arthrosc 21(2):305–310. doi:10.1007/s00167-012-1993-7 15. Nomura E (1999) Classification of lesions of the medial patello-femoral ligament in patellar dislocation. Int Orthop 23(5):260–263 16. Nomura E, Horiuchi Y, Inoue M (2002) Correlation of MR imaging findings and open exploration of medial patellofemoral ligament injuries in acute patellar dislocations. Knee 9(2):139–143 17. Nomura E, Inoue M, Osada N (2005) Anatomical analysis of the medial patellofemoral ligament of the knee, especially the femoral attachment. Knee Surg Sports Traumatol Arthrosc 13(7):510– 515. doi:10.1007/s00167-004-0607-4 18. Philippot R, Chouteau J, Wegrzyn J, Testa R, Fessy MH, Moyen B (2009) Medial patellofemoral ligament anatomy: implications for its surgical reconstruction. Knee Surg Sports Traumatol Arthrosc 17(5):475–479. doi:10.1007/s00167-009-0722-3 19. Reider B, Marshall JL, Koslin B, Ring B, Girgis FG (1981) The anterior aspect of the knee joint. J Bone Jt Surg Am 63(3):351–356 20. Sallay PI, Poggi J, Speer KP, Garrett WE (1996) Acute dislocation of the patella. A correlative pathoanatomic study. Am J Sports Med 24(1):52–60 21. Schottle PB, Hensler D, Imhoff AB (2010) Anatomical doublebundle MPFL reconstruction with an aperture fixation. Knee Surg Sports Traumatol Arthrosc 18(2):147–151. doi:10.1007/ s00167-009-0868-z 22. Schottle PB, Schmeling A, Rosenstiel N, Weiler A (2007) Radiographic landmarks for femoral tunnel placement in medial patellofemoral ligament reconstruction. Am J Sports Med 35(5):801– 804. doi:10.1177/0363546506296415 23. Smirk C, Morris H (2003) The anatomy and reconstruction of the medial patellofemoral ligament. Knee 10(3):221–227 24. Smith TO, Song F, Donell ST, Hing CB (2011) Operative versus non-operative management of patellar dislocation. A metaanalysis. Knee Surg Sports Traumatol Arthrosc 19(6):988–998. doi:10.1007/s00167-010-1355-2
Surg Radiol Anat 25. Steensen RN, Dopirak RM, McDonald WG 3rd (2004) The anatomy and isometry of the medial patellofemoral ligament: implications for reconstruction. Am J Sports Med 32(6):1509–1513. doi:10.1177/0363546503261505 26. Tuxoe JI, Teir M, Winge S, Nielsen PL (2002) The medial patellofemoral ligament: a dissection study. Knee Surg Sports Traumatol Arthrosc 10(3):138–140. doi:10.1007/s00167-001-0261-z 27. Warren LF, Marshall JL (1979) The supporting structures and layers on the medial side of the knee: an anatomical analysis. J Bone Jt Surg Am 61(1):56–62
28. Wijdicks CA, Griffith CJ, LaPrade RF, Johansen S, Sunderland A, Arendt EA, Engebretsen L (2009) Radiographic identification of the primary medial knee structures. J Bone Jt Surg Am 91(3):521–529. doi:10.2106/JBJS.H.00909 29. Zaffagnini S, Colle F, Lopomo N, Sharma B, Bignozzi S, Dejour D, Marcacci M (2012) The influence of medial patellofemoral ligament on patellofemoral joint kinematics and patellar stability. Knee Surg Sports Traumatol Arthrosc. doi:10.1007/ s00167-012-2307-9
13
Original Article
Anatomical study of the medial patello‑femoral ligament: landmarks for its surgical reconstruction Anthony Viste · Florian Chatelet · Romain Desmarchelier · Michel‑Henri Fessy
Received: 31 July 2013 / Accepted: 6 February 2014 © Springer-Verlag France 2014
Abstract Purpose The aim of this dissection study was to describe the anatomical insertions of the medial patello-femoral ligament (MPFL), and to assess its relationship with surrounding structures to improve its surgical reconstruction. Methods Twelve knees (7 cadavers) were included for the study. Measurements and general features of the MPFL were assessed: lengths, widths and insertions. Results The MPFL was found in all knees, presenting a triangular shape, and extending from the medial part of the patella to its femoral insertion (its length was of 59 ± 6.6 mm), distal to the adductor tubercle. The mean femoral insertion of the MPFL was 7.2 ± 2.7 mm proximal and 7.4 ± 4.0 mm posterior to the medial femoral epicondyle (MFE). It was also at a mean 11 ± 2.8 mm distal and 1.3 ± 2.1 mm posterior to the adductor tubercle, and 22 ± 6.4 mm anterior to the posterior condyle. We did not find any double-bundle organization on the patellar insertion. The width of the MPFL was 8.8 ± 2.9 mm at the A. Viste · F. Chatelet · M.-H. Fessy Faculté de Médecine Lyon Sud-Charles Mérieux, Laboratoire d’Anatomie, 165 Chemin du Petit Revoyet, BP 12, 69921 Oullins Cedex, France A. Viste (*) · R. Desmarchelier · M.-H. Fessy Hospices Civils de Lyon, Centre Hospitalier Lyon-Sud, Service de Chirurgie Orthopédique, Traumatologique et Médecine du Sport, 165 Chemin Grand Revoyet, 69495 Pierre Bénite, Cedex, France e-mail: anthony.viste@chu‑lyon.fr A. Viste · M.-H. Fessy Université de Lyon, Lyon, France A. Viste · M.-H. Fessy IFSTTAR, UMRT_9406, LBMC, Université Lyon1, Villeurbanne, France
femoral insertion, 27 ± 5.9 mm at the patellar insertion, and 12 ± 3.1 mm in the middle of the MPFL. The vastus medialis obliquus was found to be inserted on the superior part of the MPFL. Conclusion The adductor tubercle appeared to be a better landmark than the MFE for the femoral tunnel positioning during surgical reconstructions of the MPFL because it was easier to identify and its relationship with the femoral insertion of the MPFL was constant (10 mm below). Keywords Medial patello-femoral ligament (MPFL) · Patellar dislocation · Reconstruction · Ligamentous knee
Introduction The medial patello-femoral ligament (MPFL) has been considered to be the main passive soft-tissue restraint to lateral displacement of the patella [3, 29]. The MPFL has been first described, but not namely mentioned, in 1957 by Kaplan [11]. The MPFL seems to be injured in almost 100 % of patellar dislocations [16] and does not reinsert afterward, leading to chronic patellar instability. Recently, the importance of the MPFL has been demonstrated biomechanically and clinically [1, 4, 9]. In 1979, Warren and Marshall [27] described the anatomy of the medial side of the knee and introduced the concept of superimposed layers system. Layer 1 was the crural fascia above the Sartorius tendon. Layer 2 was composed of the superficial medial collateral ligament and the tendons of the semimembranosus, semitendinosus and gracilis muscles. In the second layer, the MPFL was described as transverse fibers, extending from the insertion of the superficial medial collateral ligament and the medial side of the patella, deeper than the vastus medialis obliquus (VMO).
13
Surg Radiol Anat
Fig. 1 The constant triangular area between the VMO, the AMT and the MPFL. MPFL medial patello-femoral ligament, AMT adductor magnus tendon, VMO vastus medialis obliquus
Layer 3 corresponded to the joint capsule of the knee and the deep medial collateral ligament. Many knee surgeons have developed MPFL reconstruction techniques [6, 13] for treating recurrent lateral patellar dislocations. Surgeons tried to restore an anatomical and isometric medial stabilizer of the patella. However, the femoral insertion remained controversial, whereas the importance of the femoral tunnel’s location for graft has been demonstrated [7, 23]. The main issues for surgical reconstruction remain the risk of anisometry, the precise locations of its attachments on the femur and patella and, above it all, its relationships to pertinent osseous landmarks. The aim of this dissection study was to describe the anatomical insertions of the MPFL, and to assess its relationship with surrounding structures to improve its surgical reconstruction.
Materials and methods Twelve knees (5 left and 7 right) from seven embalmed cadavers (3 male and 4 female) were included for study and provided by the Department of Anatomy at the Medicine Faculty Lyon-Est. They were all intact knees and did not have scars or bone abnormality around the knee. The mean age of the donors was 75 years (range 70–83). Each knee was stored frozen at 20 °C and then thawed the day prior dissection.
13
The skin and subcutaneous tissues of the medial part of the knee were first removed. We excised the crural fascia. Then, using a lateral force applied on the patella, and the constant [2, 10] triangular area between the VMO and the adductor tubercle, we identified the MPFL (Fig. 1). We removed the crural fascia over the MPFL and with a forceps inserted under the MPFL, and we dissected the femoral and patellar insertions (Fig. 2). The MPFL was easily individualized from the capsule. The adductor tubercle and the medial femoral epicondyle (MFE) were then identified. All measurements were taken at 30° of knee flexion with a millimetric scale ruler (±1 mm precision). General features of the MPFL were investigated. We measured distances between the femoral insertion of the MPFL and the MFE, the adductor tubercle and posterior medial femoral condyle. We also measured the transverse length of the MPFL, from the femoral attachment to the middle of the patellar insertion. We measured the length of the insertion of the VMO on the MPFL, and then we reflected the VMO proximally to identify the insertion of the superior edge of the MPFL. We measured the width of the MPFL at the patellar insertion, at the femoral insertion and in the middle of the MPFL, and we measured the overall patellar length.
Results The MPFL was found in all the knees dissected. The MPFL was situated in layer 2, under the superficial
Surg Radiol Anat
Fig. 2 MPFL and its insertions. FI femoral insertion, AMT adductor magnus tendon
retinaculum. This ligament had approximately a triangular shape. The proximal edge was easily identified, loosely attached under the VMO, and it was easy to release it from the MPFL. The distal edge of the MPFL was more difficult to dissect because of its relation with the medial retinaculum and the medial patello-meniscal ligament (MPML). On the distal part of the patella, we found the medial patello-tibial ligament (MPTL) to be inserted distal to the MPFL. Patellar insertion The patellar insertion of the MPFL was found to locate on the proximal part of the medial side of the patella. The average overall length of the patella was 51 ± 2.5 mm (range 47–55 mm), and the width of the MPFL at the patellar insertion was 27 ± 5.9 mm (range 20–39 mm). So, the MPFL was found to be inserted on the superior half part of the patella.
posterior to the MFE. It was also 11 ± 2.8 mm (range 7–15) distal and 1.3 ± 2.1 mm (range 0–6) posterior to the adductor tubercle, and 22 ± 6.4 mm (range 15– 31) anterior to the posterior condyle. The width of the MPFL at the femoral insertion was 8.8 ± 2.9 mm (range 5–14 mm). General dimensions We have not found, during our dissections, any extension of the MPFL fibers to the medial collateral ligament or to the adductor magnus tendon (Fig. 2). The length of the MPFL was 59 ± 6.6 mm (range 50–67), and the width in the middle of the MPFL was 12 ± 3.1 (range 6–16). Upper insertion of the VMO on the MPFL was found in 12 knees (100 %) and its length was 31 ± 6.0 mm (range 22–40).
Discussion (Table 1) Femoral insertion The femoral insertion of the MPFL was 7.2 ± 2.7 mm (range 3–13) proximal and 7.4 ± 4.0 mm (range 0–15)
This study was performed to describe the anatomical insertions of the MPFL and its relationship with surrounding structures. Anatomists and orthopedic surgeons
13
Surg Radiol Anat
Table 1 Anatomy of the MPFL: review of literature Authors
Number of knees (n)
Presence of MPFL (%)
Mean width (mm)
Average location of the FI (mm)
Mean length (mm)
At the FI
In the middle
At the PI
To the MFE
To the AT
– – Anterior 9 Prox 13 Post 11 Prox 9 Post 0
0 0 – –
56 – – 72
2 Ant 4 Distal –
65
9.5 Prox 5 Post 10 Prox Posterior Anterior –
–
59
11 Distal 10 Distal – 0
58 58 – 53
7 Prox 7 Post
11 Distal 1 Post
59
Aragao et al. [2] Conlan et al. [3] Feller et al. [8] Kang et al. [10]
17 33 20 12
88 88 100 100
17 13 – –
20 – – –
28 – – 22
LaPrade et al. [12]
20
100
–
–
–
Mochizuki et al. [14]
16
100
–
–
Nomura et al. [17]
20
100
–
12
24 (medial to VI) –
Philippot et al. [18] Smirk and Morris [23] Steensen et al. [25] Tuxoe et al. [26]
23 25 11 39
100 100 100 100
12 – 15 19 (10 mm lateral to AT)
– – – –
24 – 17 –
Our study
12 (7 donors)
100
9
12
27
56
PI patellar insertion, FI femoral insertion, prox proximal, post posterior, ant anterior, VI vastus intermedius, AT adductor tubercle, MFE medial femoral epicondyle
are divided about its existence as a distinct entity: Reider et al. [19] thought MPFL was a condensation of the medial capsule of the knee joint, contrarily to our opinion and the fact that MPFL was always present in our dissections. The ultimate goal was to improve its surgical reconstruction. The current landmark used to drill the femoral tunnel is the MFE (posteriorly) but we think that this wide area is more difficult to identify accurately. So, we assert that the best landmark for femoral tunnel positioning is the adductor tubercle: the MPFL femoral insertion was 10 mm distal to the adductor tubercle. Biomechanical studies have reported that the MPFL provided 50–60 % of the medial stabilizing force of the patella (primary stabilizer) [3, 4, 9]. These studies demonstrated the need for surgical reconstruction of the MPFL in case of patellar dislocations. Clinical studies have found a MPFL loose or torn in lateral patellar dislocations without bone abnormalities [15, 20, 25]. Our study found the MPFL within layer II (as described by Warren and Marshall [27]) in 100 % of knees dissected. Conlan et al. [3] found the MPFL in 90 % of knees, Aragao et al. [2] in 88 % of knees and Reider et al. [19] in only 35 % of knees. More recent studies found the MPFL in all knees [4, 10, 23] as in our study. The limitations of this study could be the paired knees included, the embalmed cadavers (vs. fresh frozen), the measurement precision (±1 mm) and the number of knees dissected but only small variations were reported.
13
Femoral insertion Femoral insertion was described either at the adductor tubercle [3, 4, 8, 26], or at the anterior part of the MFE [8], or at the MFE [2, 9, 27] or between the MFE and the adductor tubercle [10, 17]. In our study, we found the femoral insertion of the MPFL to be located just distal to the adductor tubercle and proximal-posterior from the MFE. We studied the distance between the insertion of the MPFL and the posterior medial femoral condyle, but high variations in values (range of 15–31 mm with standard deviation of 6.4) demonstrated that this anatomical distance could not be used to locate precisely the femoral insertion of the MPFL. Distances between the MFE and the femoral insertion of the MPFL were similar to the literature (Table 1). The MFE has been widely used by authors as the landmark to locate the femoral insertion of the MPFL [10, 17]. In our study, the average distance (1.3 mm) and standard deviation (2.1) of the transversal length between the adductor tubercle and the femoral insertion of the MPFL were low. We also found in eight knees (67 %) a femoral insertion of the MPFL just distal from the adductor tubercle. We think that this low variation can be used in the surgical reconstruction of the MPFL, using the adductor tubercle to locate the femoral insertion of the MPFL, instead of using the MFE, whose values of the distance from the femoral
Surg Radiol Anat
insertion of the MPFL are more variable. Smirk and Morris [23] also found a femoral insertion of the MPFL most commonly located approximately 1 cm distal to the adductor tubercle and they demonstrated that better results were obtained with a graft posterior to the adductor tubercle rather than anterior. Other authors [17, 26] found a femoral insertion of the MPFL located just distally to the adductor tubercle without more precisions. In our study, we found in four knees (33 %) an insertion slightly posterior to the adductor tubercle (maximum 6 mm). Some authors found, at the femoral insertion of the MPFL, decussations to the adductor magnus tendon [2, 4] and to the medial collateral ligament [4, 25, 26]. According to LaPrade et al. [12], the adductor magnus tendon did not attach directly to the adductor tubercle (3 mm proximal and posterior to the adductor tubercle). We didn’t find any fiber of this type in our study, potentially explained by our dissection technique, focused on functional fibers of the MPFL. Decussations appeared to belong to the medial retinaculum and did not seem to have mechanical functions. Wijdicks et al. [28] and Schottle et al. [22] dealt with the radiographic landmarks of the MPFL femoral insertion. It is routinely used postoperatively in MPFL reconstruction to make sure that the femoral tunnel was accurately located. Two lines were used to define the accurate femoral location: a line tangent to the posterior cortex of the femur and a line perpendicular tangent to the posterior portion of the Blumensaat line. Wijdicks et al. [28] found that MPFL was 15.9 mm from the medial epicondyle and 8.9 mm from the adductor tubercle. In the study by Schottle et al. [22], the femoral attachment was found 1.3 mm anterior to the posterior femoral cortex and 5.5 mm proximal to the perpendicular line whereas Wijdicks found it at 8.8 mm anterior and 2.6 mm proximal. But in both cases, the femoral attachment was located in the antero-proximal quadrant.
the MPFL with this configuration in mind, but we could not find any element in favor of a double-bundle organization. Relationship with the quadriceps muscles Dirim et al. [5] found, based on MR images and histological examination, a relationship between superior fibers of the MPFL and the fibers of the VMO, impossible to separate, except at the patellar attachment, where bilaminar images had been obtained on MRI. In their anatomic study, Mochizuki et al. [14] found a main attachment between the MPFL and the vastus intermedius, without tight adhesion to the VMO. In our dissections, we found some proximal fibers of the MPFL, which ran over the superior-medial corner of the patella, under the VMO, to insert on the deep part of the quadriceps tendon, without extension toward the belly of the rectus intermedius. These results are similar to Conlan et al. [3] who found proximal fibers extending to the aponeurotic fibers of the vastus intermedius. But those fibers did not represent a functional bundle inserted on the quadriceps tendon on all knees dissected. This insertion of the VMO was not always found [2, 23]. The length of the upper insertion of the VMO on the MPFL was 31 ± 6.0 mm (range 20–40 mm) in our study, and present in 100 % of the knees. These results were larger than in the literature, Nomura et al. [17] found a length of 20.3 ± 6.1 mm (range 10–32 mm), and Philippot et al. [18] found a length of 25.7 ± 6.0 mm (range 19–38 mm). Even if our results are not significantly different from the literature, our results insist on the importance of this insertion. The existence of this insertion, even if there is no tight adhesion, could demonstrate the role in the dynamic restraint in lateral displacement of the patella played by the MPFL through the action of the VMO. General dimensions
Patellar insertion Patellar insertion has been reported at the proximal half of the patella [1, 3, 25], proximal two-thirds of the patella [23, 26, 27], at the vastus medialis tendon [8], and to the quadriceps tendon [3, 14]. In our study, we found the length of the patellar insertion was 27 ± 5.9 mm (range 20–39), and a patellar height of 51 ± 2.5 mm (range 47–55). First mentioned by Amis et al. [1], functional bundles have been described by Kang et al. [10].They found two functional bundles of the MPFL on the patellar insertion. The inferior straight bundle runs from the femoral insertion to the medial side of the patella, and is the main static soft structure restraint. The superior oblique bundle runs from the patellar insertion to the upper end of the patella and quadriceps fibers, and is a dynamic soft-tissue restraint to patellar lateral displacement. In our study, we dissected
In our study, the length of the MPFL was 59 ± 6.6 mm (range 50–67). This is similar to the literature (Table 1). Presentation could be described as a hourglass shape [4]. In our study, the width at the middle of the MPFL was greater to the width at the femoral insertion in 11 knees (92 %). In one knee (8 %) the widths at the middle of the MPFL and at its femoral insertion were equal. The width at the patellar insertion was always bigger to the width in the middle of the MPFL. This triangular shape, found in all the knees (100 %), is present in other studies [2, 23]. Implications for knee surgeons Surgical treatment for patellar instability has been related to lower rate of re-dislocation than non-operative management [24]. Some authors [21] proposed to restore the
13
triangular shape of the MPFL, using a double-bundle graft, or to combine reconstruction of MPFL and MPTL [6]. Knowledge of the anatomy of the medial part of the knee and of soft-tissue restraint to lateral displacement of the patella seems to be the key for any surgery of the MPFL. Elias and Cosgarea [7] demonstrated that the femoral insertion of the graft, in MPFL reconstructions, had a significant impact on forces applied on the medial patellofemoral cartilage, which can lead to cartilage degradation, arthritis, and pain. Smirk and Morris [23] also demonstrated that the location of the femoral tunnel had an influence on graft isometry. As a result, we advocate the use of the adductor tubercle as a landmark for surgeries instead of the most commonly used MFE. Lower variations in the distances between the adductor tubercle and the femoral insertion of the MPFL than the MFE could allow avoiding some errors. Moreover, the MFE is usually not well defined: it looks like a C-shaped ridge with a central sulcus for the insertion of the superficial medial collateral ligament [12].
Conclusion The MPFL was always present, running from the proximal and medial part of the patella to a femoral insertion distal to the adductor tubercle and posterior to the MFE. The VMO inserted on the MPFL on its superior edge, whereas the MPFL extended under the VMO to insert on the quadriceps tendon. Our results are mostly similar to the literature, but imply to use the adductor tubercle as a landmark during reconstructions of the MPFL, which would provide a more constant anatomical landmark. Conflict of interest None.
References 1. Amis AA, Firer P, Mountney J, Senavongse W, Thomas NP (2003) Anatomy and biomechanics of the medial patellofemoral ligament. Knee 10(3):215–220 2. Aragao JA, Reis FP, de Vasconcelos DP, Feitosa VL, Nunes MA (2008) Metric measurements and attachment levels of the medial patellofemoral ligament: an anatomical study in cadavers. Clinics (Sao Paulo) 63(4):541–544 3. Conlan T, Garth WP Jr, Lemons JE (1993) Evaluation of the medial soft-tissue restraints of the extensor mechanism of the knee. J Bone Jt Surg Am 75(5):682–693 4. Desio SM, Burks RT, Bachus KN (1998) Soft tissue restraints to lateral patellar translation in the human knee. Am J Sports Med 26(1):59–65 5. Dirim B, Haghighi P, Trudell D, Portes G, Resnick D (2008) Medial patellofemoral ligament: cadaveric investigation of anatomy with MRI, MR arthrography, and histologic correlation. AJR 191(2):490–498. doi:10.2214/AJR.07.3590
13
Surg Radiol Anat 6. Ebied AM, El-Kholy W (2012) Reconstruction of the medial patello-femoral and patello-tibial ligaments for treatment of patellar instability. Knee Surg Sports Traumatol Arthrosc 20(5):926–932. doi:10.1007/s00167-011-1659-x 7. Elias JJ, Cosgarea AJ (2006) Technical errors during medial patellofemoral ligament reconstruction could overload medial patellofemoral cartilage: a computational analysis. Am J Sports Med 34(9):1478–1485. doi:10.1177/0363546506287486 8. Feller JA, Feagin JA Jr, Garrett WE Jr (1993) The medial patellofemoral ligament revisited: an anatomical study. Knee Surg Sports Traumatol Arthrosc 1(3–4):184–186 9. Hautamaa PV, Fithian DC, Kaufman KR, Daniel DM, Pohlmeyer AM (1998) Medial soft tissue restraints in lateral patellar instability and repair. Clin Orthop Relat Res 349:174–182 10. Kang HJ, Wang F, Chen BC, Su YL, Zhang ZC, Yan CB (2010) Functional bundles of the medial patellofemoral ligament. Knee Surg Sports Traumatol Arthrosc 18(11):1511–1516. doi:10.1007/ s00167-010-1090-8 11. Kaplan EB (1957) Factors responsible for the stability of the knee joint. Bull Hosp Jt Dis 18(1):51–59 12. LaPrade RF, Engebretsen AH, Ly TV, Johansen S, Wentorf FA, Engebretsen L (2007) The anatomy of the medial part of the knee. J Bone Jt Surg Am 89(9):2000–2010. doi:10.2106/JBJ S.F.01176 13. Lind M, Jakobsen BW, Lund B, Christiansen SE (2008) Reconstruction of the medial patellofemoral ligament for treatment of patellar instability. Acta Orthop 79(3):354–360. doi:10.1080/17453670710015256 14. Mochizuki T, Nimura A, Tateishi T, Yamaguchi K, Muneta T, Akita K (2013) Anatomic study of the attachment of the medial patellofemoral ligament and its characteristic relationships to the vastus intermedius. Knee Surg Sports Traumatol Arthrosc 21(2):305–310. doi:10.1007/s00167-012-1993-7 15. Nomura E (1999) Classification of lesions of the medial patello-femoral ligament in patellar dislocation. Int Orthop 23(5):260–263 16. Nomura E, Horiuchi Y, Inoue M (2002) Correlation of MR imaging findings and open exploration of medial patellofemoral ligament injuries in acute patellar dislocations. Knee 9(2):139–143 17. Nomura E, Inoue M, Osada N (2005) Anatomical analysis of the medial patellofemoral ligament of the knee, especially the femoral attachment. Knee Surg Sports Traumatol Arthrosc 13(7):510– 515. doi:10.1007/s00167-004-0607-4 18. Philippot R, Chouteau J, Wegrzyn J, Testa R, Fessy MH, Moyen B (2009) Medial patellofemoral ligament anatomy: implications for its surgical reconstruction. Knee Surg Sports Traumatol Arthrosc 17(5):475–479. doi:10.1007/s00167-009-0722-3 19. Reider B, Marshall JL, Koslin B, Ring B, Girgis FG (1981) The anterior aspect of the knee joint. J Bone Jt Surg Am 63(3):351–356 20. Sallay PI, Poggi J, Speer KP, Garrett WE (1996) Acute dislocation of the patella. A correlative pathoanatomic study. Am J Sports Med 24(1):52–60 21. Schottle PB, Hensler D, Imhoff AB (2010) Anatomical doublebundle MPFL reconstruction with an aperture fixation. Knee Surg Sports Traumatol Arthrosc 18(2):147–151. doi:10.1007/ s00167-009-0868-z 22. Schottle PB, Schmeling A, Rosenstiel N, Weiler A (2007) Radiographic landmarks for femoral tunnel placement in medial patellofemoral ligament reconstruction. Am J Sports Med 35(5):801– 804. doi:10.1177/0363546506296415 23. Smirk C, Morris H (2003) The anatomy and reconstruction of the medial patellofemoral ligament. Knee 10(3):221–227 24. Smith TO, Song F, Donell ST, Hing CB (2011) Operative versus non-operative management of patellar dislocation. A metaanalysis. Knee Surg Sports Traumatol Arthrosc 19(6):988–998. doi:10.1007/s00167-010-1355-2
Surg Radiol Anat 25. Steensen RN, Dopirak RM, McDonald WG 3rd (2004) The anatomy and isometry of the medial patellofemoral ligament: implications for reconstruction. Am J Sports Med 32(6):1509–1513. doi:10.1177/0363546503261505 26. Tuxoe JI, Teir M, Winge S, Nielsen PL (2002) The medial patellofemoral ligament: a dissection study. Knee Surg Sports Traumatol Arthrosc 10(3):138–140. doi:10.1007/s00167-001-0261-z 27. Warren LF, Marshall JL (1979) The supporting structures and layers on the medial side of the knee: an anatomical analysis. J Bone Jt Surg Am 61(1):56–62
28. Wijdicks CA, Griffith CJ, LaPrade RF, Johansen S, Sunderland A, Arendt EA, Engebretsen L (2009) Radiographic identification of the primary medial knee structures. J Bone Jt Surg Am 91(3):521–529. doi:10.2106/JBJS.H.00909 29. Zaffagnini S, Colle F, Lopomo N, Sharma B, Bignozzi S, Dejour D, Marcacci M (2012) The influence of medial patellofemoral ligament on patellofemoral joint kinematics and patellar stability. Knee Surg Sports Traumatol Arthrosc. doi:10.1007/ s00167-012-2307-9
13
