International Orthopaedics (SICOT) (2015) 39:423–428 DOI 10.1007/s00264-014-2523-7
ORIGINAL PAPER
Reliability of radiographic landmarks in medial patello-femoral ligament reconstruction in relation to the anatomical femoral torsion Martin Kaipel & Sebastian Schützenberger & Sebastian Farr & Istvan Gergely & Alexander Vlcek & Franz Kainberger & Harald Boszotta & Michael Pretterklieber
Received: 7 July 2014 / Accepted: 30 August 2014 / Published online: 24 September 2014 # SICOT aisbl 2014
Abstract Purpose Anatomically correct graft positioning at the femoral insertion site is a key factor in surgical reconstruction of the medial patello-femoral ligament (MPFL). Basically there are two techniques to define this point in fluoroscopy during surgery. The role of the anatomical femoral torsion on the accuracy and reproducibility of both procedures has not been clarified. Methods Twenty human anatomical leg specimens were dissected. The femoral insertion of the MPFL was marked by two K-wires. The position of the ligament insertion was determined fluoroscopically in the true lateral view as used in routine clinical practice. The anatomical MPFL insertion M. Kaipel (*) : H. Boszotta Department of Traumatology, Barmherzige Brüder Hospital, Johannes von Gott Platz 1, 7000 Eisenstadt, Austria e-mail: [email protected] S. Schützenberger Department of Traumatology, Trauma Centre Meidling, Kundratstraße 37, 1120 Vienna, Austria
was compared to the radiographic landmarks which were recommended by two previous studies. The anatomical femoral torsion of the specimens was assessed by computed tomography scans. Results In true lateral view fluoroscopy, the mean distance of the femoral MPFL insertion was −0.2 mm distal to the vertical reference line intersecting the posterior point of Blumensaat’s line. In the anteroposterior direction, the mean distance was −2.0 mm posterior to the femoral cortex reference line. There was no correlation between anatomical femoral torsion and the distance of the femoral MPFL insertion to the posterior cortex. Conclusions The results of this study strongly recommend use of a vertical line intersecting the most posterior point of Blumensaat’s line as a reference to identify the MPFL insertion in the craniocaudal direction. In the anteroposterior direction, the femoral MPFL insertion showed distinctive variation and was found −2.0 mm posterior to the femoral cortex reference line without being influenced by the anatomical femoral torsion.
S. Farr Department of Paediatric Orthopaedics, Orthopaedic Hospital Speising, Speisinger Straße 109, 1130 Vienna, Austria
Keywords Medial patello-femoral ligament . Radiographic landmarks . Anatomical femoral torsion . MPFL reconstruction
I. Gergely Department of Radiology, Barmherzige Brüder Hospital, Esterhazystrasse 26, 7000 Eisenstadt, Austria
Introduction
A. Vlcek : F. Kainberger Department of Neuroradiology and Musculoskeletal Radiology, Medical University Vienna, Währinger Gürtel 18-20, 1090 Vienna, Austria M. Pretterklieber Department for Applied Anatomy, Centre of Anatomy and Cell Biology, Medical University Vienna, Währingerstraße 13, 1090 Vienna, Austria
The medial patello-femoral ligament (MPFL) acts as the most important passive stabiliser of the patella during knee flexion [1–5]. Traumatic patellar dislocations are frequently associated with MPFL lesions [6] leading to persistent instability and recurrent dislocations [7]. Surgical reconstruction of the MPFL has been shown to produce beneficial results [8] and effectively prevents patients from chronic patellofemoral joint
424
instability [9]. As a consequence, the number of MPFL operations has increased and anatomical reconstruction of the MPFL has gained wide acceptance among orthopaedic surgeons. Recent studies have promoted mini-open methods using a tubular graft (e.g. semitendinosus tendon) fixed by anchors or biodegradable screws [10–12]. Anatomically correct graft positioning, especially at the femoral insertion site, has been shown to be crucial to regain physiologic biomechanics within the patello-femoral joint [13, 14]. During surgery, the femoral insertion point of the MPFL is normally identified fluoroscopically in the true lateral view according to radiographic landmarks. Basically there are two various recommendations how to state this point precisely [15, 16]. Both Schöttle et al. and Barnett et al. used the posterior femoral cortex as reference line. They recommended graft insertion 1.3 and 3.8 mm anterior to this radiographic landmark, respectively. Schöttle et al. created a perpendicular reference line intersecting the posterior aspect of the medial condyle and recommended graft insertion 2.5 mm distal to this landmark. Barnett et al. found a perpendicular line intersecting the most posterior point of Blumensaat’s line more reproducible. They recommended graft insertion 0.9 mm distal to this reference. Both data sets were gained by studying a limited number of anatomical specimens. However, none of the previous studies had proven if anatomical femoral torsion, which shows a broad variation among human subjects (−21 to 47.2°) [17, 18], can additionally influence the accuracy of the procedure. It is possible that the position of the femoral MPFL insertion point in the anteroposterior direction shows distinctive variation which might be explained by the pre-existing individual anatomical femoral torsion. As a consequence, extreme internal or external femoral torsion might impair the detection of the anatomically correct femoral MPFL insertion site by using fluoroscopy. Thus, the aim of this study was to evaluate current concepts of detecting the femoral MPFL insertion site by using radiographic landmarks in a larger series of anatomical specimens. Beyond that we assessed the potential influence of the anatomical femoral torsion on the radiographically determined location of the femoral MPFL insertion site.
Materials and methods Specimens The study protocol and all measurements were approved by the Institutional Review Board of the Medical University of Vienna (No. 1064/2013). A total of 20 formalin-fixed human anatomical leg specimens including the pelvis were taken from six male and 11 female body donors. The study included 16 right and four left specimens. The mean age at death of the body donors was 73.5 years (range 51–101 years). Specimens
International Orthopaedics (SICOT) (2015) 39:423–428
presenting with any form of limb alignment disorder (e.g. fractures, osseous tumours, corrective surgery) were excluded from the study. Computed tomography (CT) imaging All specimens underwent CT scanning by using a 64-row detector scanner (PHILIPS BRILLIANCE 64©, Philips Austria Healthcare, Vienna, Austria). The specimens were placed in supine position on the scanner table with the knees stretched parallel to the z-axis. The scanning parameters were: 140 kV, 100 mA/slice, 64×0.625 mm slice collimation, 512× 512 matrix and a bone algorithm with no gantry tilting. The imaging data were processed using a digital imaging workstation (SYNEDRA VIEW©, Synedra Information Technologies, Innsbruck, Austria). Anatomical torsion of the femora was assessed according to the method of Jend [19] as adapted by Schneider et al. [20]. In short the specimens were placed in supine position with extended legs parallel to the zaxis of the CT scanner. CT sections in line with the femoral neck and through the femoral condyles were preformed. On the proximal femur the reference line connects the centre of the femoral head with the centre of the femoral neck. On the distal femur the reference line is a tangent to the posterior border of the femoral condyles (Fig. 1a, b). Angles between the reference lines and the horizontal plane were measured. Anatomical torsion of the femora was assessed arithmetically. Anatomical preparation and fluoroscopy After completing the CT scans, all specimens were anatomically dissected to expose the MPFL. According to a previous anatomical study [21], the femoral insertion of the MPFL was exposed. The proximal-distal extension of the femoral insertion site of the MPFL was marked by anchoring two K-wires (diameter 0.8 mm, length 3 mm) into the bone (Fig. 2). Following this preparation, radiographs in the true lateral view were performed using a mobile C-arm image intensifier (ARCADIS ORBIC 3D©, Siemens Healthcare, Erlangen, Germany). The anatomical specimens were placed on an examination desk in 30° knee flexion. The position of the image intensifier was modified to gain a true lateral X-ray with superimposed posterior aspects of the medial and lateral condyles of the femur. A radiopaque millimetre scale was placed under the prepared MPFL to enable exact calibration. The obtained X-rays were digitally processed using a commercially available graphic platform (Photoshop© version CS6 extended, Adobe Systems Inc., San Jose, CA, USA). All measurements were performed electronically using the Photoshop© surface. Before starting the measurements, a calibration factor has to be determined in every X-ray by using the radiopaque scale placed underneath the
International Orthopaedics (SICOT) (2015) 39:423–428
425
MPFL. All further measurements were arithmetically corrected by using this specimen-specific calibration factor. In a first step, the centre of the femoral MPFL insertion was defined as the midpoint between the two K-wires. The distance to the posterior femoral cortex and the perpendicular reference line meeting the most posterior point of Blumensaat’s line was measured. In a second step, the proposed centre of the femoral MPFL insertion as defined by Barnett et al. [15] and Schöttle et al. [16] was assessed graphically. The distance between the anatomically dissected centre of the MPFL and the two recommended insertion sites described by Barnett et al. and Schöttle et al. was measured in the anteroposterior and craniocaudal direction (Table 1).
Statistics Statistical analysis was performed using a standard software package (GraphPad Prism©, GraphPad Software Inc., La Jolla, CA, USA). Data were analysed for normal distribution using the Kolmogorov-Smirnov test. Group comparisons were performed using the analysis of variance (ANOVA). Pearson’s correlation coefficient was calculated comparing the anatomical femoral torsion (°) and the distance of the femoral MPFL insertion to the posterior femoral cortex (mm). R < 0.5 indicated weak and r < 0.3 insignificant correlation.
Results Fig. 1 a, b Anatomical femoral torsion was assessed by using CT scans and was defined as the angle between the femoral neck axis and the horizontal reference line minus the angle between the posterior condyles and the horizontal reference line. Positive angles represent anteversion and negative angles retroversion of the femoral neck
Fig. 2 Anatomical dissection of the MPFL. The proximal and distal ends of the femoral insertion were marked by two K-wires. A radiolucent scale was placed underneath the ligament
Anatomical rotation and MPFL insertion at the distal femur Anatomical dissection of the MPFL and identification of the femoral insertion site was achieved in all 20 specimens. The mean anatomical femoral torsion was determined by measurements performed on the CT scans and was 6.8° (minimum −10.3° retroversion and maximum 24.1° anteversion of the femoral neck). On true lateral view fluoroscopy, the mean distance between the femoral MPFL insertion and the posterior femoral cortex was −2.0 mm [minimum −8.0 mm posterior and maximum 3.1 mm anterior of the posterior femoral cortex (Fig. 3 and Table 1)]. The mean distance of the femoral MPFL insertion to the perpendicular reference line intersecting the most posterior point of Blumensaat’s line was −0.2 mm [minimum −8.8 and maximum 7.5 mm (Fig. 3 and Table 1)]. There was no correlation between anatomical femoral rotation and distance of the femoral MPFL insertion to the posterior cortex (Pearson’s correlation coefficient r= 0.082/r2 =0.007/p=0.739).
426
International Orthopaedics (SICOT) (2015) 39:423–428
Table 1 Distances of the anatomically dissected femoral MPFL insertion compared to radiographic landmarks and previously recommended insertion sites No.
Fem. rot. (°)
A/P post. cortex (mm, +ant/-post)
C/C BL (mm, +prox/-dist)
A/P Barnett et al. (mm, +ant/-post)
C/C Barnett et al. (mm, +prox/-dist)
A/P Schöttle et al. (mm, +ant/-post)
C/C Schöttle et al. (mm, +prox/-dist)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
12.1 24.1 0.0 8.5 1.0 1.0 1.8 6.0 6.0 3.1 20.8 11.9 −10.3 1.2 19.1 −3.4
−2.8 0.4 −2.9 2.9 −6.5 −0.4 −4.1 −1.8 −6.5 −2.1 −8.0 3.1 −1.3 0.0 −2.1 −2.9 −1.7
−2.1 7.5 −1.6 −1.8 −0.7 0.3 −1.0 −6.3 −1.9 −0.4 1.7 −8.8 0.4 2.6 0.5 −0.5 −1.5
−6.5 −2.7 −6.6 0.2 −10.2 −4.0 −8.1 −5.7 −10.3 −5.7 −11.3 −0.4 −4.7 −3.8 −5.9 −6.7 −5.6
1.6 8.6 −0.8 −0.6 0.4 1.9 0.0 −5.4 −0.8 0.6 2.7 −8.0 1.2 3.7 1.2 0.5 −0.7
−4.0 −0.6 −4.1 2.1 −8.1 −1.9 −5.2 −3.1 −7.8 −3.0 −9.1 2.0 −2.1 −1.3 −3.6 −3.9 −3.0
−7.1 2.9 −5.8 −7.8 −3.8 −3.5 −4.5 −10.1 −8.4 −5.5 −3.8 12.3 −3.7 −1.2 −6.4 −5.3 −5.9
18 19 20 Mean SD
8.3 15.3 2.5 6.8 8,8
−1.1 1.0 −2.4 −2.0 2.9
−0.3 −0.5 3.4 −0.6 3.3
−4.9 −2.8 −5.9 −5.6 3.0
0.5 0.4 4.3 0.6 3.3
−2.2 −0.3 0.3 −2.9 3.0
−6.5 −4.1 −3.6 −5.3 3.2
The first column shows the specimen numbers (=No.). The second column presents CT-assessed anatomical femoral torsion (=Fem. rot.) of the specimen (°). The third and fourth columns show the distance (mm) of the anatomically dissected MPFL insertion to the posterior cortex reference line (=A/P post. cortex) and the reference line intersecting the posterior aspect of Blumensaat’s line (=C/C BL) (Fig. 3). The fifth and sixth columns show the distance (mm) of the anatomically dissected MPFL insertion to the insertion site recommended by Barnett et al. [15] in the anteroposterior (=A/P Barnett et al. [15]) and craniocaudal (=C/C Barnett et al. [15] ) direction, respectively. The seventh and eighth columns show the distance (mm) of the anatomically dissected MPFL insertion to the insertion site recommended by Schöttle et al. [16] in the anteroposterior (=A/P Schöttle et al. [16]) and craniocaudal (=C/C Schöttle et al. [16]) direction, respectively. Femoral torsion of the first specimen could not be assessed due to technical reasons and was not included in the analysis of correlation
MPFL insertion compared to Barnett et al./Schöttle et al.
Discussion
The centre of the femoral MPFL insertion verified by anatomical dissection was assessed radiographically and compared to MPFL insertion sites as recommended by Barnett et al. [15] and Schöttle et al. [16] (Fig. 3). In the anteroposterior direction, the femoral MPFL insertion would have been assessed more precisely by the method of Schöttle et al. (in 17 of 20 specimens). Anatomically dissected femoral MPFL insertions in their study show a mean distance of −2.9 mm distal to Schöttle et al.’s recommended insertion site. In the craniocaudal direction, the method of Barnett et al. showed increased precision (17 of 20 specimens). Anatomically dissected femoral MPFL insertions in this study show a mean distance of 0.6 mm to the recommended insertion site (Table 1). Anatomically dissected MPFL insertion sites did not differ significantly compared to the insertion sites recommended by Barnett et al. or Schöttle et al.
The results of our study confirm the accuracy of Barnett et al.’s vertical reference line in the craniocaudal direction. Anatomical MPFL insertion was found within a 2 mm range proximal and distal to this reference line in 13 of 20 specimens (Table 1). Schöttle et al.’s reference line intersecting the medial condyle was shown to be too far proximal (Fig. 3) to correctly detect the femoral MPFL insertion in the majority of specimens. Beyond that in some cases accurate fluoroscopic identification of Schöttle et al.’s vertical reference line was impeded. This was mainly due to the well-known fact that correct definition of the transition point between the posterior femoral condyles and the cortex of the femoral shaft is prone to interobserver variance [15]. As a consequence, the authors of this study strongly recommend use of a vertical line intersecting the most posterior point of Blumensaat’s line as a reference to identify MPFL insertion in the craniocaudal
International Orthopaedics (SICOT) (2015) 39:423–428
Fig. 3 Fluoroscopy of a representative anatomical specimen in true lateral view. The anatomically dissected MPFL insertion is marked by two K-wires (white circles). Reference lines through the posterior femoral cortex, the posterior aspect of the medial condyle and the posterior point of Blumensaat’s line are created according to the method of Schöttle et al./ Barnett et al. Recommended femoral insertion sites of the MPFL according to Schöttle et al. and Barnett et al. are marked by a red xS and xB, respectively. The anatomical dissected femoral MPFL insertions of 20 specimens are marked by yellow x
direction. Thus, results of previous studies [15, 22, 23] which also recommended this vertical reference line due to its high reproducibility during fluoroscopy can be supported. In contrast to these consistent results, in the craniocaudal direction, data of this study showed distinctive variation in the anteroposterior plane. By anatomical dissection, the MPFL insertion site was found within or posterior to the cortical reference line in 16 of 20 specimens (Fig. 3). This is remarkable, as both Schöttle et al. and Barnett et al. originally described that the MPFL insertion should be situated anterior to this reference mark. Further analysis of both studies shows that MPFL insertion posterior to the femoral cortical line also occurred in their specimens. A previous comparative study found the anatomical MPFL insertion 1.0–4.5 mm posterior to Schöttle et al.’s recommended insertion site in seven of eight specimens [22]. In relation to the femoral cortical reference line, MPFL insertion was posterior in six of eight specimens. This indicates that in most cases the femoral MPFL insertion may be far more posterior than expected when using true lateral fluoroscopy during surgery. Despite conflicting results even in our study, the authors recommend MPFL graft insertion 2.1 mm posterior to the femoral cortex in contrast to previously published works. Which factors caused those inconsistent results in this study and previously published studies? Barnett et al. studied
427
the impact of limb rotation during fluoroscopy on the recommended MPFL graft position. As expected, they found a significantly increased anterior position of the radiolucent MPFL insertion markers in external limb rotation. In contrast, internal rotation led to a significantly more posterior position in relation to the femoral cortex. It appeared possible that similar effects might be caused by the anatomical femoral torsion which shows a broad individual variation. In this study, we assessed the anatomical femoral torsion by using CT scans of all specimens. Seventeen specimens exhibited a relatively internal rotation of the distal femur compared to the femoral neck ranging from 0.0 to 24.1° (Table 1). In 15 of 20 specimens, the MPFL insertion was found to be posterior to the femoral cortical reference line. It would have been possible that a relatively internal anatomical rotation of the distal femur results in more posterior MPFL position during lateral view fluoroscopy. However, data of this study did not show any correlation between the anatomical femoral torsion and the MPFL insertion in the anteroposterior direction (r=0.082). The variation of the femoral MPFL insertion site in the anterorposterior direction is not caused by significant differences in the anatomical femoral rotation. Another factor responsible for those inconsistent results could be the local anatomy of the distal femur. During anatomical dissection of the specimens, the femoral MPFL insertion could be constantly found between the medial epicondyle and the adductor’s tubercle. This finding is supported by a previous anatomical study [21]. It has been established that the position of the medial epicondyle shows distinctive variation in relation to the posterior femoral condyles. A previous knee arthroplasty study assessed the preoperative angle between axes intersecting the posterior condyles and the epicondyles [24]. The angles ranged from 2 to 12° and demonstrate the interindividual variation of the distal femur’s anatomy. It is possible that the spreading MPFL insertion sites in the anteroposterior direction are influenced by the variable location of the medial femoral epicondyle in relation to the posterior condyles. A third factor responsible for those inconsistent results could be the limited number of the anatomical specimens studied. Previous studies only used eight to ten specimens [15, 22, 16] due to limited availability. In this study, we assessed 20 specimens. However, a manifold increased sample size could not be achieved for the same reason. In conclusion, the results of this study strongly recommend use of a vertical line intersecting the most posterior point of Blumensaat’s line as a reference to identify MPFL insertion in the craniocaudal direction. In the anteroposterior direction, the femoral MPFL insertion showed distinctive variation and was found −2.0 mm posterior to the femoral cortex reference line without being influenced by the anatomical femoral torsion.
428 Acknowledgments The authors gratefully acknowledge Katrina Sinz for statistical support. Conflict of interest The authors declare that they have no conflict of interest.
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. 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 3. Nomura E, Horiuchi Y, Kihara M (2000) Medial patellofemoral ligament restraint in lateral patellar translation and reconstruction. Knee 7(2):121–127 4. Sandmeier RH, Burks RT, Bachus KN, Billings A (2000) The effect of reconstruction of the medial patellofemoral ligament on patellar tracking. Am J Sports Med 28(3):345–349 5. de Oliveira V, de Souza V, Cury R, Camargo OP, Avanzi O, Severino N, Fucs P (2014) Medial patellofemoral ligament anatomy: is it a predisposing factor for lateral patellar dislocation? Int Orthop 38(8): 1633–1639. doi:10.1007/s00264-014-2357-3 6. 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 7. Hawkins RJ, Bell RH, Anisette G (1986) Acute patellar dislocations. The natural history. Am J Sports Med 14(2):117–120 8. Drez D Jr, Edwards TB, Williams CS (2001) Results of medial patellofemoral ligament reconstruction in the treatment of patellar dislocation. Arthroscopy 17(3):298–306. doi:10.1053/jars.2001. 21490 9. Csintalan RP, Latt LD, Fornalski S, Raiszadeh K, Inacio MC, Fithian DC (2014) Medial patellofemoral ligament (MPFL) reconstruction for the treatment of patellofemoral instability. J Knee Surg 27(2): 139–146. doi:10.1055/s-0033-1360652 10. Fernandez E, Sala D, Castejon M (2005) Reconstruction of the medial patellofemoral ligament for patellar instability using a semitendinosus autograft. Acta Orthop Belg 71(3):303–308 11. Wang CH, Ma LF, Zhou JW, Ji G, Wang HY, Wang F, Wang J (2013) Double-bundle anatomical versus single-bundle isometric medial patellofemoral ligament reconstruction for patellar dislocation. Int Orthop 37(4):617–624. doi:10.1007/s00264-013-1788-6 12. Matthews JJ, Schranz P (2010) Reconstruction of the medial patellofemoral ligament using a longitudinal patellar tunnel
International Orthopaedics (SICOT) (2015) 39:423–428 technique. Int Orthop 34(8):1321–1325. doi:10.1007/s00264-0090918-7 13. Ntagiopoulos PG, Sharma B, Bignozzi S, Lopomo N, Colle F, Zaffagnini S, Dejour D (2013) Are the tubular grafts in the femoral tunnel in an anatomical or isometric position in the reconstruction of medial patellofemoral ligament? Int Orthop 37(10):1933–1941. doi: 10.1007/s00264-013-1938-x 14. Nelitz M, Williams RS, Lippacher S, Reichel H, Dornacher D (2014) Analysis of failure and clinical outcome after unsuccessful medial patellofemoral ligament reconstruction in young patients. Int Orthop. doi:10.1007/s00264-014-2437-4 15. Barnett AJ, Howells NR, Burston BJ, Ansari A, Clark D, Eldridge JD (2012) Radiographic landmarks for tunnel placement in reconstruction of the medial patellofemoral ligament. Knee Surg Sports Traumatol Arthrosc 20(12):2380–2384. doi:10.1007/s00167-011-1871-8 16. Schöttle 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 17. Abadie P, Galaud B, Michaut M, Fallet L, Boisrenoult P, Beaufils P (2009) Distal femur rotational alignment and patellar subluxation: a CT scan in vivo assessment. Orthop Traumatol Surg Res 95(4):267– 271. doi:10.1016/j.otsr.2009.04.004 18. Diederichs G, Köhlitz T, Kornaropoulos E, Heller MO, Vollnberg B, Scheffler S (2013) Magnetic resonance imaging analysis of rotational alignment in patients with patellar dislocations. Am J Sports Med 41(1):51–57. doi:10.1177/0363546512464691 19. Jend HH (1986) Computed tomographic determination of the anteversion angle. Premises and possibilities. Rofo 144(4):447– 452. doi:10.1055/s-2008-1048820 20. Schneider B, Laubenberger J, Jemlich S, Groene K, Weber HM, Langer M (1997) Measurement of femoral antetorsion and tibial torsion by magnetic resonance imaging. Br J Radiol 70(834):575– 579. doi:10.1259/bjr.70.834.9227249 21. Baldwin JL (2009) The anatomy of the medial patellofemoral ligament. Am J Sports Med 37(12):2355–2361. doi:10.1177/ 0363546509339909 22. Redfern J, Kamath G, Burks R (2010) Anatomical confirmation of the use of radiographic landmarks in medial patellofemoral ligament reconstruction. Am J Sports Med 38(2):293–297. doi:10.1177/ 0363546509347602 23. 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 Joint Surg Am 91(3):521– 529. doi:10.2106/JBJS.H.00909 24. Kaipel M, Gergely I, Sinz K, Neumann C, Sinz G (2013) Femoral rotation in ligament balanced knee arthroplasty: a prospective clinical study. J Arthroplasty 28(7):1103–1106. doi:10.1016/j.arth.2013.02.031
ORIGINAL PAPER
Reliability of radiographic landmarks in medial patello-femoral ligament reconstruction in relation to the anatomical femoral torsion Martin Kaipel & Sebastian Schützenberger & Sebastian Farr & Istvan Gergely & Alexander Vlcek & Franz Kainberger & Harald Boszotta & Michael Pretterklieber
Received: 7 July 2014 / Accepted: 30 August 2014 / Published online: 24 September 2014 # SICOT aisbl 2014
Abstract Purpose Anatomically correct graft positioning at the femoral insertion site is a key factor in surgical reconstruction of the medial patello-femoral ligament (MPFL). Basically there are two techniques to define this point in fluoroscopy during surgery. The role of the anatomical femoral torsion on the accuracy and reproducibility of both procedures has not been clarified. Methods Twenty human anatomical leg specimens were dissected. The femoral insertion of the MPFL was marked by two K-wires. The position of the ligament insertion was determined fluoroscopically in the true lateral view as used in routine clinical practice. The anatomical MPFL insertion M. Kaipel (*) : H. Boszotta Department of Traumatology, Barmherzige Brüder Hospital, Johannes von Gott Platz 1, 7000 Eisenstadt, Austria e-mail: [email protected] S. Schützenberger Department of Traumatology, Trauma Centre Meidling, Kundratstraße 37, 1120 Vienna, Austria
was compared to the radiographic landmarks which were recommended by two previous studies. The anatomical femoral torsion of the specimens was assessed by computed tomography scans. Results In true lateral view fluoroscopy, the mean distance of the femoral MPFL insertion was −0.2 mm distal to the vertical reference line intersecting the posterior point of Blumensaat’s line. In the anteroposterior direction, the mean distance was −2.0 mm posterior to the femoral cortex reference line. There was no correlation between anatomical femoral torsion and the distance of the femoral MPFL insertion to the posterior cortex. Conclusions The results of this study strongly recommend use of a vertical line intersecting the most posterior point of Blumensaat’s line as a reference to identify the MPFL insertion in the craniocaudal direction. In the anteroposterior direction, the femoral MPFL insertion showed distinctive variation and was found −2.0 mm posterior to the femoral cortex reference line without being influenced by the anatomical femoral torsion.
S. Farr Department of Paediatric Orthopaedics, Orthopaedic Hospital Speising, Speisinger Straße 109, 1130 Vienna, Austria
Keywords Medial patello-femoral ligament . Radiographic landmarks . Anatomical femoral torsion . MPFL reconstruction
I. Gergely Department of Radiology, Barmherzige Brüder Hospital, Esterhazystrasse 26, 7000 Eisenstadt, Austria
Introduction
A. Vlcek : F. Kainberger Department of Neuroradiology and Musculoskeletal Radiology, Medical University Vienna, Währinger Gürtel 18-20, 1090 Vienna, Austria M. Pretterklieber Department for Applied Anatomy, Centre of Anatomy and Cell Biology, Medical University Vienna, Währingerstraße 13, 1090 Vienna, Austria
The medial patello-femoral ligament (MPFL) acts as the most important passive stabiliser of the patella during knee flexion [1–5]. Traumatic patellar dislocations are frequently associated with MPFL lesions [6] leading to persistent instability and recurrent dislocations [7]. Surgical reconstruction of the MPFL has been shown to produce beneficial results [8] and effectively prevents patients from chronic patellofemoral joint
424
instability [9]. As a consequence, the number of MPFL operations has increased and anatomical reconstruction of the MPFL has gained wide acceptance among orthopaedic surgeons. Recent studies have promoted mini-open methods using a tubular graft (e.g. semitendinosus tendon) fixed by anchors or biodegradable screws [10–12]. Anatomically correct graft positioning, especially at the femoral insertion site, has been shown to be crucial to regain physiologic biomechanics within the patello-femoral joint [13, 14]. During surgery, the femoral insertion point of the MPFL is normally identified fluoroscopically in the true lateral view according to radiographic landmarks. Basically there are two various recommendations how to state this point precisely [15, 16]. Both Schöttle et al. and Barnett et al. used the posterior femoral cortex as reference line. They recommended graft insertion 1.3 and 3.8 mm anterior to this radiographic landmark, respectively. Schöttle et al. created a perpendicular reference line intersecting the posterior aspect of the medial condyle and recommended graft insertion 2.5 mm distal to this landmark. Barnett et al. found a perpendicular line intersecting the most posterior point of Blumensaat’s line more reproducible. They recommended graft insertion 0.9 mm distal to this reference. Both data sets were gained by studying a limited number of anatomical specimens. However, none of the previous studies had proven if anatomical femoral torsion, which shows a broad variation among human subjects (−21 to 47.2°) [17, 18], can additionally influence the accuracy of the procedure. It is possible that the position of the femoral MPFL insertion point in the anteroposterior direction shows distinctive variation which might be explained by the pre-existing individual anatomical femoral torsion. As a consequence, extreme internal or external femoral torsion might impair the detection of the anatomically correct femoral MPFL insertion site by using fluoroscopy. Thus, the aim of this study was to evaluate current concepts of detecting the femoral MPFL insertion site by using radiographic landmarks in a larger series of anatomical specimens. Beyond that we assessed the potential influence of the anatomical femoral torsion on the radiographically determined location of the femoral MPFL insertion site.
Materials and methods Specimens The study protocol and all measurements were approved by the Institutional Review Board of the Medical University of Vienna (No. 1064/2013). A total of 20 formalin-fixed human anatomical leg specimens including the pelvis were taken from six male and 11 female body donors. The study included 16 right and four left specimens. The mean age at death of the body donors was 73.5 years (range 51–101 years). Specimens
International Orthopaedics (SICOT) (2015) 39:423–428
presenting with any form of limb alignment disorder (e.g. fractures, osseous tumours, corrective surgery) were excluded from the study. Computed tomography (CT) imaging All specimens underwent CT scanning by using a 64-row detector scanner (PHILIPS BRILLIANCE 64©, Philips Austria Healthcare, Vienna, Austria). The specimens were placed in supine position on the scanner table with the knees stretched parallel to the z-axis. The scanning parameters were: 140 kV, 100 mA/slice, 64×0.625 mm slice collimation, 512× 512 matrix and a bone algorithm with no gantry tilting. The imaging data were processed using a digital imaging workstation (SYNEDRA VIEW©, Synedra Information Technologies, Innsbruck, Austria). Anatomical torsion of the femora was assessed according to the method of Jend [19] as adapted by Schneider et al. [20]. In short the specimens were placed in supine position with extended legs parallel to the zaxis of the CT scanner. CT sections in line with the femoral neck and through the femoral condyles were preformed. On the proximal femur the reference line connects the centre of the femoral head with the centre of the femoral neck. On the distal femur the reference line is a tangent to the posterior border of the femoral condyles (Fig. 1a, b). Angles between the reference lines and the horizontal plane were measured. Anatomical torsion of the femora was assessed arithmetically. Anatomical preparation and fluoroscopy After completing the CT scans, all specimens were anatomically dissected to expose the MPFL. According to a previous anatomical study [21], the femoral insertion of the MPFL was exposed. The proximal-distal extension of the femoral insertion site of the MPFL was marked by anchoring two K-wires (diameter 0.8 mm, length 3 mm) into the bone (Fig. 2). Following this preparation, radiographs in the true lateral view were performed using a mobile C-arm image intensifier (ARCADIS ORBIC 3D©, Siemens Healthcare, Erlangen, Germany). The anatomical specimens were placed on an examination desk in 30° knee flexion. The position of the image intensifier was modified to gain a true lateral X-ray with superimposed posterior aspects of the medial and lateral condyles of the femur. A radiopaque millimetre scale was placed under the prepared MPFL to enable exact calibration. The obtained X-rays were digitally processed using a commercially available graphic platform (Photoshop© version CS6 extended, Adobe Systems Inc., San Jose, CA, USA). All measurements were performed electronically using the Photoshop© surface. Before starting the measurements, a calibration factor has to be determined in every X-ray by using the radiopaque scale placed underneath the
International Orthopaedics (SICOT) (2015) 39:423–428
425
MPFL. All further measurements were arithmetically corrected by using this specimen-specific calibration factor. In a first step, the centre of the femoral MPFL insertion was defined as the midpoint between the two K-wires. The distance to the posterior femoral cortex and the perpendicular reference line meeting the most posterior point of Blumensaat’s line was measured. In a second step, the proposed centre of the femoral MPFL insertion as defined by Barnett et al. [15] and Schöttle et al. [16] was assessed graphically. The distance between the anatomically dissected centre of the MPFL and the two recommended insertion sites described by Barnett et al. and Schöttle et al. was measured in the anteroposterior and craniocaudal direction (Table 1).
Statistics Statistical analysis was performed using a standard software package (GraphPad Prism©, GraphPad Software Inc., La Jolla, CA, USA). Data were analysed for normal distribution using the Kolmogorov-Smirnov test. Group comparisons were performed using the analysis of variance (ANOVA). Pearson’s correlation coefficient was calculated comparing the anatomical femoral torsion (°) and the distance of the femoral MPFL insertion to the posterior femoral cortex (mm). R < 0.5 indicated weak and r < 0.3 insignificant correlation.
Results Fig. 1 a, b Anatomical femoral torsion was assessed by using CT scans and was defined as the angle between the femoral neck axis and the horizontal reference line minus the angle between the posterior condyles and the horizontal reference line. Positive angles represent anteversion and negative angles retroversion of the femoral neck
Fig. 2 Anatomical dissection of the MPFL. The proximal and distal ends of the femoral insertion were marked by two K-wires. A radiolucent scale was placed underneath the ligament
Anatomical rotation and MPFL insertion at the distal femur Anatomical dissection of the MPFL and identification of the femoral insertion site was achieved in all 20 specimens. The mean anatomical femoral torsion was determined by measurements performed on the CT scans and was 6.8° (minimum −10.3° retroversion and maximum 24.1° anteversion of the femoral neck). On true lateral view fluoroscopy, the mean distance between the femoral MPFL insertion and the posterior femoral cortex was −2.0 mm [minimum −8.0 mm posterior and maximum 3.1 mm anterior of the posterior femoral cortex (Fig. 3 and Table 1)]. The mean distance of the femoral MPFL insertion to the perpendicular reference line intersecting the most posterior point of Blumensaat’s line was −0.2 mm [minimum −8.8 and maximum 7.5 mm (Fig. 3 and Table 1)]. There was no correlation between anatomical femoral rotation and distance of the femoral MPFL insertion to the posterior cortex (Pearson’s correlation coefficient r= 0.082/r2 =0.007/p=0.739).
426
International Orthopaedics (SICOT) (2015) 39:423–428
Table 1 Distances of the anatomically dissected femoral MPFL insertion compared to radiographic landmarks and previously recommended insertion sites No.
Fem. rot. (°)
A/P post. cortex (mm, +ant/-post)
C/C BL (mm, +prox/-dist)
A/P Barnett et al. (mm, +ant/-post)
C/C Barnett et al. (mm, +prox/-dist)
A/P Schöttle et al. (mm, +ant/-post)
C/C Schöttle et al. (mm, +prox/-dist)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
12.1 24.1 0.0 8.5 1.0 1.0 1.8 6.0 6.0 3.1 20.8 11.9 −10.3 1.2 19.1 −3.4
−2.8 0.4 −2.9 2.9 −6.5 −0.4 −4.1 −1.8 −6.5 −2.1 −8.0 3.1 −1.3 0.0 −2.1 −2.9 −1.7
−2.1 7.5 −1.6 −1.8 −0.7 0.3 −1.0 −6.3 −1.9 −0.4 1.7 −8.8 0.4 2.6 0.5 −0.5 −1.5
−6.5 −2.7 −6.6 0.2 −10.2 −4.0 −8.1 −5.7 −10.3 −5.7 −11.3 −0.4 −4.7 −3.8 −5.9 −6.7 −5.6
1.6 8.6 −0.8 −0.6 0.4 1.9 0.0 −5.4 −0.8 0.6 2.7 −8.0 1.2 3.7 1.2 0.5 −0.7
−4.0 −0.6 −4.1 2.1 −8.1 −1.9 −5.2 −3.1 −7.8 −3.0 −9.1 2.0 −2.1 −1.3 −3.6 −3.9 −3.0
−7.1 2.9 −5.8 −7.8 −3.8 −3.5 −4.5 −10.1 −8.4 −5.5 −3.8 12.3 −3.7 −1.2 −6.4 −5.3 −5.9
18 19 20 Mean SD
8.3 15.3 2.5 6.8 8,8
−1.1 1.0 −2.4 −2.0 2.9
−0.3 −0.5 3.4 −0.6 3.3
−4.9 −2.8 −5.9 −5.6 3.0
0.5 0.4 4.3 0.6 3.3
−2.2 −0.3 0.3 −2.9 3.0
−6.5 −4.1 −3.6 −5.3 3.2
The first column shows the specimen numbers (=No.). The second column presents CT-assessed anatomical femoral torsion (=Fem. rot.) of the specimen (°). The third and fourth columns show the distance (mm) of the anatomically dissected MPFL insertion to the posterior cortex reference line (=A/P post. cortex) and the reference line intersecting the posterior aspect of Blumensaat’s line (=C/C BL) (Fig. 3). The fifth and sixth columns show the distance (mm) of the anatomically dissected MPFL insertion to the insertion site recommended by Barnett et al. [15] in the anteroposterior (=A/P Barnett et al. [15]) and craniocaudal (=C/C Barnett et al. [15] ) direction, respectively. The seventh and eighth columns show the distance (mm) of the anatomically dissected MPFL insertion to the insertion site recommended by Schöttle et al. [16] in the anteroposterior (=A/P Schöttle et al. [16]) and craniocaudal (=C/C Schöttle et al. [16]) direction, respectively. Femoral torsion of the first specimen could not be assessed due to technical reasons and was not included in the analysis of correlation
MPFL insertion compared to Barnett et al./Schöttle et al.
Discussion
The centre of the femoral MPFL insertion verified by anatomical dissection was assessed radiographically and compared to MPFL insertion sites as recommended by Barnett et al. [15] and Schöttle et al. [16] (Fig. 3). In the anteroposterior direction, the femoral MPFL insertion would have been assessed more precisely by the method of Schöttle et al. (in 17 of 20 specimens). Anatomically dissected femoral MPFL insertions in their study show a mean distance of −2.9 mm distal to Schöttle et al.’s recommended insertion site. In the craniocaudal direction, the method of Barnett et al. showed increased precision (17 of 20 specimens). Anatomically dissected femoral MPFL insertions in this study show a mean distance of 0.6 mm to the recommended insertion site (Table 1). Anatomically dissected MPFL insertion sites did not differ significantly compared to the insertion sites recommended by Barnett et al. or Schöttle et al.
The results of our study confirm the accuracy of Barnett et al.’s vertical reference line in the craniocaudal direction. Anatomical MPFL insertion was found within a 2 mm range proximal and distal to this reference line in 13 of 20 specimens (Table 1). Schöttle et al.’s reference line intersecting the medial condyle was shown to be too far proximal (Fig. 3) to correctly detect the femoral MPFL insertion in the majority of specimens. Beyond that in some cases accurate fluoroscopic identification of Schöttle et al.’s vertical reference line was impeded. This was mainly due to the well-known fact that correct definition of the transition point between the posterior femoral condyles and the cortex of the femoral shaft is prone to interobserver variance [15]. As a consequence, the authors of this study strongly recommend use of a vertical line intersecting the most posterior point of Blumensaat’s line as a reference to identify MPFL insertion in the craniocaudal
International Orthopaedics (SICOT) (2015) 39:423–428
Fig. 3 Fluoroscopy of a representative anatomical specimen in true lateral view. The anatomically dissected MPFL insertion is marked by two K-wires (white circles). Reference lines through the posterior femoral cortex, the posterior aspect of the medial condyle and the posterior point of Blumensaat’s line are created according to the method of Schöttle et al./ Barnett et al. Recommended femoral insertion sites of the MPFL according to Schöttle et al. and Barnett et al. are marked by a red xS and xB, respectively. The anatomical dissected femoral MPFL insertions of 20 specimens are marked by yellow x
direction. Thus, results of previous studies [15, 22, 23] which also recommended this vertical reference line due to its high reproducibility during fluoroscopy can be supported. In contrast to these consistent results, in the craniocaudal direction, data of this study showed distinctive variation in the anteroposterior plane. By anatomical dissection, the MPFL insertion site was found within or posterior to the cortical reference line in 16 of 20 specimens (Fig. 3). This is remarkable, as both Schöttle et al. and Barnett et al. originally described that the MPFL insertion should be situated anterior to this reference mark. Further analysis of both studies shows that MPFL insertion posterior to the femoral cortical line also occurred in their specimens. A previous comparative study found the anatomical MPFL insertion 1.0–4.5 mm posterior to Schöttle et al.’s recommended insertion site in seven of eight specimens [22]. In relation to the femoral cortical reference line, MPFL insertion was posterior in six of eight specimens. This indicates that in most cases the femoral MPFL insertion may be far more posterior than expected when using true lateral fluoroscopy during surgery. Despite conflicting results even in our study, the authors recommend MPFL graft insertion 2.1 mm posterior to the femoral cortex in contrast to previously published works. Which factors caused those inconsistent results in this study and previously published studies? Barnett et al. studied
427
the impact of limb rotation during fluoroscopy on the recommended MPFL graft position. As expected, they found a significantly increased anterior position of the radiolucent MPFL insertion markers in external limb rotation. In contrast, internal rotation led to a significantly more posterior position in relation to the femoral cortex. It appeared possible that similar effects might be caused by the anatomical femoral torsion which shows a broad individual variation. In this study, we assessed the anatomical femoral torsion by using CT scans of all specimens. Seventeen specimens exhibited a relatively internal rotation of the distal femur compared to the femoral neck ranging from 0.0 to 24.1° (Table 1). In 15 of 20 specimens, the MPFL insertion was found to be posterior to the femoral cortical reference line. It would have been possible that a relatively internal anatomical rotation of the distal femur results in more posterior MPFL position during lateral view fluoroscopy. However, data of this study did not show any correlation between the anatomical femoral torsion and the MPFL insertion in the anteroposterior direction (r=0.082). The variation of the femoral MPFL insertion site in the anterorposterior direction is not caused by significant differences in the anatomical femoral rotation. Another factor responsible for those inconsistent results could be the local anatomy of the distal femur. During anatomical dissection of the specimens, the femoral MPFL insertion could be constantly found between the medial epicondyle and the adductor’s tubercle. This finding is supported by a previous anatomical study [21]. It has been established that the position of the medial epicondyle shows distinctive variation in relation to the posterior femoral condyles. A previous knee arthroplasty study assessed the preoperative angle between axes intersecting the posterior condyles and the epicondyles [24]. The angles ranged from 2 to 12° and demonstrate the interindividual variation of the distal femur’s anatomy. It is possible that the spreading MPFL insertion sites in the anteroposterior direction are influenced by the variable location of the medial femoral epicondyle in relation to the posterior condyles. A third factor responsible for those inconsistent results could be the limited number of the anatomical specimens studied. Previous studies only used eight to ten specimens [15, 22, 16] due to limited availability. In this study, we assessed 20 specimens. However, a manifold increased sample size could not be achieved for the same reason. In conclusion, the results of this study strongly recommend use of a vertical line intersecting the most posterior point of Blumensaat’s line as a reference to identify MPFL insertion in the craniocaudal direction. In the anteroposterior direction, the femoral MPFL insertion showed distinctive variation and was found −2.0 mm posterior to the femoral cortex reference line without being influenced by the anatomical femoral torsion.
428 Acknowledgments The authors gratefully acknowledge Katrina Sinz for statistical support. Conflict of interest The authors declare that they have no conflict of interest.
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. 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 3. Nomura E, Horiuchi Y, Kihara M (2000) Medial patellofemoral ligament restraint in lateral patellar translation and reconstruction. Knee 7(2):121–127 4. Sandmeier RH, Burks RT, Bachus KN, Billings A (2000) The effect of reconstruction of the medial patellofemoral ligament on patellar tracking. Am J Sports Med 28(3):345–349 5. de Oliveira V, de Souza V, Cury R, Camargo OP, Avanzi O, Severino N, Fucs P (2014) Medial patellofemoral ligament anatomy: is it a predisposing factor for lateral patellar dislocation? Int Orthop 38(8): 1633–1639. doi:10.1007/s00264-014-2357-3 6. 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 7. Hawkins RJ, Bell RH, Anisette G (1986) Acute patellar dislocations. The natural history. Am J Sports Med 14(2):117–120 8. Drez D Jr, Edwards TB, Williams CS (2001) Results of medial patellofemoral ligament reconstruction in the treatment of patellar dislocation. Arthroscopy 17(3):298–306. doi:10.1053/jars.2001. 21490 9. Csintalan RP, Latt LD, Fornalski S, Raiszadeh K, Inacio MC, Fithian DC (2014) Medial patellofemoral ligament (MPFL) reconstruction for the treatment of patellofemoral instability. J Knee Surg 27(2): 139–146. doi:10.1055/s-0033-1360652 10. Fernandez E, Sala D, Castejon M (2005) Reconstruction of the medial patellofemoral ligament for patellar instability using a semitendinosus autograft. Acta Orthop Belg 71(3):303–308 11. Wang CH, Ma LF, Zhou JW, Ji G, Wang HY, Wang F, Wang J (2013) Double-bundle anatomical versus single-bundle isometric medial patellofemoral ligament reconstruction for patellar dislocation. Int Orthop 37(4):617–624. doi:10.1007/s00264-013-1788-6 12. Matthews JJ, Schranz P (2010) Reconstruction of the medial patellofemoral ligament using a longitudinal patellar tunnel
International Orthopaedics (SICOT) (2015) 39:423–428 technique. Int Orthop 34(8):1321–1325. doi:10.1007/s00264-0090918-7 13. Ntagiopoulos PG, Sharma B, Bignozzi S, Lopomo N, Colle F, Zaffagnini S, Dejour D (2013) Are the tubular grafts in the femoral tunnel in an anatomical or isometric position in the reconstruction of medial patellofemoral ligament? Int Orthop 37(10):1933–1941. doi: 10.1007/s00264-013-1938-x 14. Nelitz M, Williams RS, Lippacher S, Reichel H, Dornacher D (2014) Analysis of failure and clinical outcome after unsuccessful medial patellofemoral ligament reconstruction in young patients. Int Orthop. doi:10.1007/s00264-014-2437-4 15. Barnett AJ, Howells NR, Burston BJ, Ansari A, Clark D, Eldridge JD (2012) Radiographic landmarks for tunnel placement in reconstruction of the medial patellofemoral ligament. Knee Surg Sports Traumatol Arthrosc 20(12):2380–2384. doi:10.1007/s00167-011-1871-8 16. Schöttle 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 17. Abadie P, Galaud B, Michaut M, Fallet L, Boisrenoult P, Beaufils P (2009) Distal femur rotational alignment and patellar subluxation: a CT scan in vivo assessment. Orthop Traumatol Surg Res 95(4):267– 271. doi:10.1016/j.otsr.2009.04.004 18. Diederichs G, Köhlitz T, Kornaropoulos E, Heller MO, Vollnberg B, Scheffler S (2013) Magnetic resonance imaging analysis of rotational alignment in patients with patellar dislocations. Am J Sports Med 41(1):51–57. doi:10.1177/0363546512464691 19. Jend HH (1986) Computed tomographic determination of the anteversion angle. Premises and possibilities. Rofo 144(4):447– 452. doi:10.1055/s-2008-1048820 20. Schneider B, Laubenberger J, Jemlich S, Groene K, Weber HM, Langer M (1997) Measurement of femoral antetorsion and tibial torsion by magnetic resonance imaging. Br J Radiol 70(834):575– 579. doi:10.1259/bjr.70.834.9227249 21. Baldwin JL (2009) The anatomy of the medial patellofemoral ligament. Am J Sports Med 37(12):2355–2361. doi:10.1177/ 0363546509339909 22. Redfern J, Kamath G, Burks R (2010) Anatomical confirmation of the use of radiographic landmarks in medial patellofemoral ligament reconstruction. Am J Sports Med 38(2):293–297. doi:10.1177/ 0363546509347602 23. 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 Joint Surg Am 91(3):521– 529. doi:10.2106/JBJS.H.00909 24. Kaipel M, Gergely I, Sinz K, Neumann C, Sinz G (2013) Femoral rotation in ligament balanced knee arthroplasty: a prospective clinical study. J Arthroplasty 28(7):1103–1106. doi:10.1016/j.arth.2013.02.031
