Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 https://doi.org/10.1186/s40634-019-0200-x Journal of Experimental Orthopaedics RESEARCH Open Access Parametric finite element model of medial patellofemoral ligament reconstruction model development and clinical validation Vicente Sanchis-Alfonso1*, Diego Alastruey-López2, Gerad Ginovart3, Erik Montesinos-Berry4, Fabio García-Castro5, Cristina Ramírez-Fuentes6, Joan Carles Monllau7, Angel Alberich-Bayarri8 and María Angeles Pérez2 Abstract Background: Currently, there is uncertainty regarding the long-term outcome of medial patellofemoral ligament reconstructions (MPFLr). Our objectives were: (1) to develop a parametric model of the patellofemoral joint (PFJ) enabling us to simulate different surgical techniques for MPFLr; (2) to determine the negative effects on the PFJ associated with each technique, which could be related to long-term deterioration of the PFJ. Methods: A finite element model of the PFJ was created based on CT data from 24 knees with chronic lateral patellar instability. Patella contact pressure and maximum MPFL-graft stress at five angles of knee flexion (0, 30, 60, 90 and 120°) were analysed in three types of MPFLr: anatomic, non-anatomic with physiometric behaviour, and non-anatomic with non-physiometric behaviour. Results: An increase in patella contact pressure was observed at 0 and 30° of knee flexion after both anatomic and non-anatomic MPFLr with physiometric behaviour. In both reconstructions, the ligament was tense between 0 and 30° of knee flexion, but at 60, 90 and 120°, it had no tension. In the third reconstruction, the behaviour was completely the opposite. Conclusion: A parametric model of the PFJ enables us to evaluate different types of MPFLr throughout the full range of motion of the knee, regarding the effect on the patellofemoral contact pressure, as well as the kinematic behaviour of the MPFL-graft and the maximum MPFL-graft stress. Keywords: Patellofemoral joint, Medial patellofemoral ligament, MPFL reconstruction, Finite element methodology, Patellofemoral contact pressure, MPFL-stress Background and Torga-Spak 2004; Weinberger et al. 2017). However, Currently, medial patellofemoral ligament reconstruction there is uncertainty regarding the long-term outcome of (MPFLr) is the “gold standard” in chronic lateral patellar these MPFL reconstructions techniques. To classify a instability surgery. It is typically performed whenever surgical technique for MPFLr as being effective, it is not there have been at least two previous episodes of lateral enough for the instability and pain to disappear. For a patellar dislocation (Sanchis-Alfonso 2014, 2016). Differ- surgical technique to be considered effective, new prob- ent surgical techniques with different attachment points, lems like chondropathy or patellofemoral osteoarthritis different types of grafts and different configurations for (PFOA), should never be caused. These problems might the reconstruction have been described for MPFLr. Each be the consequence of the increase in the patellofemoral one has good short-term clinical results (Fink et al. 2014; contact pressure secondary to an inadequate MPFLr (Elias Fulkerson and Edgar 2013; Sanchis-Alfonso 2014; Teitge and Cosgarea 2006; Rood et al. 2015; Stephen et al. 2014, 2016), which is clinically relevant because surgery for lat- * Correspondence: vicente.sanchis.alfonso@gmail.com eral patellar instability is generally performed in young in- 1 Department of Orthopaedic Surgery, Hospital Arnau de Vilanova, C/Sant dividuals. Moreover, the development of symptomatic Climent, 12, 46015 Valencia, Spain PFOA in young persons does not currently have a good Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 2 of 15 solution. In an ideal MPFLr, the graft should be tense at were measured from CT scans both on the axial plane and 0–30° of knee flexion. Beyond 30°, the graft should be by using a multi-planar reformatting (MPR) technique. Pa- loose (Sanchis-Alfonso et al. 2017). All the other parame- tients were pathological. Therefore, the parametric geom- ters should be considered inadequate (Sanchis-Alfonso etry also considered their particular geometry. The main et al. 2017). Given that in the daily clinical practice many parts of the PFJ parametric model were the bones of the MPFL reconstructions with a clearly incorrect femoral fix- femur (femoral condyle) and patella as rigid parts as well as ation point can be seen, we must evaluate not only the the femoral and patellar cartilages as hexahedral deform- correct reconstructions but also the clearly incorrect ones. able components (Fig. 1a-d). As previously stated, each part An effective way to evaluate patellofemoral contact pres- was simplified to obtain nearly anatomical geometry with sure throughout the range of motion of the knee after variable parameters (Elias et al. 2016). The patellar bone MPFLr is by using the Finite element methodology (FEM) was modelled starting from a concave-revolution-solid (DeVries et al. 2015; Elias et al. 2005; Elias and Cosgarea shape, with the parametric radius, height and radius curva- 2006; Shah et al. 2015). Moreover, this technology also en- ture (Fig. 1e). Several revolution cuts were performed on ables us to evaluate the kinematic behaviour of the MPFL- the solid part, and its final geometry was obtained (Fig. 1b). graft and maximum MPFL-graft stress, that is, the tension The patellar cartilage was created following the same pro- that the graft can withstand before breaking, in all knee cedure while maintaining the patellar dimensions (Fig. 1a). flexion-extension positions. The femoral bone was the most complex part of the model. The objective was to create a parametric model of the It was defined as a discrete rigid part that had four main patellofemoral joint (PFJ) where the joint geometry is sim- elements: a revolution shape that defined the bottom plified and can be meshed by means of automatic mesh geometry, with a parametric width and radius (lateral and generation programs with suitable finite element aspect ra- medial); a solid loft for the irregular section, with different tios for all meshes. Additionally, the aim was that the para- width and length parametric sections (width, width 2, width metric model would enable a surgeon to simulate different 3, length, length 2, length 3); a revolution shape in the pos- types of surgical techniques for MPFLr. It is hypothesized terior geometry, where the radius can be modified; and two that this model would allow to evaluate patellofemoral con- revolution shapes (Fig. 1f) that represent the femoral epi- tact pressure and the maximum MPFL-graft stress in each condyles (Fig. 1c). Width and length parameters corre- specific reconstruction at different knee flexion-extension sponded to the maximum distance between both femoral angles. The objective was to determine the negative theor- epicondyles. Width 2 and length 2 were taken at the point etical effects (patellofemoral contact pressure and the max- where the medial epicondyle joins the main femoral bone. imum MPFL-graft stress) on the PFJ in each type of Width 3 and length 3 were measured at the same point as MPFLr. This negative effect could be related to long-term the highest position of the patella (0° knee flexion angle). deterioration of the PFJ. The posterior radius defined the contact region between the patellar and femoral cartilages. The femoral cartilage Methods was defined as deformable, and its generation was based on Parametric finite element model of the patellofemoral femur geometry and consisted of a revolution shape for the joint bottom geometry and a combination of elements that de- From a previous study (Sanchis-Alfonso et al. 2017), high fined the upper region (Fig. 1d). The PFJ parametric model spatial resolution Computerized Tomography (CT) data was developed using the Abaqus/CAE v.6.14 software (Das- were available from 24 knees of patients with chronic sault Systèmes, France). Measuring previous geometrical lateral patellar instability. Images were acquired with a 64- characteristics on the 24 knees, a mean parametric model detector Multi-Detector CT system (Philips Medical was generated (Table 1). Systems, Best, the Netherlands) at the highest spatial reso- As cartilages cannot be reconstructed correctly from a lution, without slice interpolation (0.255 × 0.255 × 0.672 CT, a fixed thickness of 3 mm was assumed (Cohen mm3). An iterative thresholding scheme was used to extract et al. 2003). Tendons and ligaments were also included bones from the imaging data, and triangulated surfaces since they help to stabilize the patella and better distrib- were defined to describe the outer surfaces (MIMICS, Ma- ute patellofemoral pressures (Fig. 2a). The quadriceps terialise NV, Leuven, Belgium). The main characteristics tendon (QT), which consists of the vastus medialis and dimensions considered for the parametric model were (VM), vastus lateralis (VL), vastus intermedius (VI), and measured as a reference (femur and patella bone dimen- the rectus femoris (RF) tendons and the patellar tendon sions) from the 24 knees (Sanchis-Alfonso et al. 2017). (PT) were modelled as a group of four and two truss ele- Knee geometry was simplified to construct a 3D parametric ments, respectively (Fig. 2a) whilst the MPFL and the model that achieved nearly anatomical geometry with vari- lateral retinaculum (LR) were defined as beam elements able parameters (i.e., trochlear dysplasia, patellar width, pa- (B33) (Fig. 2a). The QT was oriented from the insertion tellar diameters, geometry of the patella). The parameters site on the patella to the muscle origin or the most distal Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 3 of 15 Fig. 1 Parametric geometry of the four main parts of the PFJ model: a Patellar cartilage; b patellar bone; c femoral bone; d femoral cartilage; e geometric parameters of the patellar bone; f geometric parameters of the femoral bone wrapping point on the femur. The PT was oriented from elements along their thickness. Finally, the patellar car- the distal patella to the tibia (Elias et al. 2005; Elias and tilage was compounded by 5756 nodes and 4125 ele- Cosgarea 2006; Elias et al. 2006). The tendon and liga- ments, while the femoral cartilage was defined by 24,918 ment properties were taken from previous studies (Cic- nodes and 18,201 elements. The cartilages were mod- cone II et al. 2006; Drez Jr et al. 2001; Elias and elled with an elastic modulus of 10 MPa and Poisson’s Cosgarea 2006) and are summarized in Table 2. A radius ratio of 0.45 (Blankevoort and Huiskes 1991; Fernandez of 1 mm was assumed for the beam elements. A mesh et al. 2012; Shah et al. 2015). convergence analysis was performed for the deformable Bone-cartilage interactions, i.e., femoral bone with fem- parts, which determined that an element size should be oral cartilage and patellar bone with patellar cartilage, 1 mm, so that the cartilages would have at least three were defined as a tie constraint. The contact between both Table 1 Mean values (mm) and standard deviation (±) (mm) of the geometrical parameters defining the parametric PFJ measured on the 24 knees (Sanchis-Alfonso et al. 2017) Femoral geometrical characteristics Width Width 2 Width 3 Length Length 2 Length 3 Medial radius Lateral radius Posterior radius 72.28 ± 8.92 54.00 ± 5.83 39.01 ± 3.85 47.71 ± 5.21 33.99 ± 3.40 28.44 ± 3.00 28.91 ± 4.75 26.09 ± 4.21 13.84 ± 2.71 Patella geometrical characteristics Radius Curvature radius Height 20.26 ± 6.24 45.49 ± 3.31 19.09 ± 2.46 Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 4 of 15 Fig. 2 a Final model reconstruction including the joint ligaments and tendons. b Reconstruction with a patellar bone fixation point and an anatomic femoral fixation point. c Reconstruction with a patellar bone fixation point and a non-anatomic femoral fixation point that has physiometric behaviour. d Reconstruction with a patellar bone fixation point and a femoral fixation point that is too far anterior and without physiometric behaviour cartilage surfaces (femoral cartilage with patellar cartilage) was defined as a surface-to-surface standard contact with a contact adjustment of 0.1, a hard contact for the normal behaviour and a penalty friction formulation with a fric- tion coefficient of 0.02 for the tangential behavior (Besier Table 2 Material properties considered for ligaments and et al. 2008). A sensitivity analysis was performed changing tendons in the FEM simulation the friction coefficient (see Additional file 1). Material Properties Stiffness (N/mm) Poisson Ratio MPFL reconstruction techniques Quadriceps Tendon (QT) 1350 0.3 Three types of MPFL double-bundle semitendinosus re- constructions with patellar and femoral bony attachment Patellar Tendon (PT) 2000 0.3 were simulated based on a previous study (Sanchis-Al- Lateral Retinaculum (LR) 2 0.3 fonso et al. 2017): anatomic reconstruction, meaning a Native Medial Patellofemoral Ligament 12 0.3 reconstruction with a femoral anatomic fixation point (MPFL) (Fig. 2b); non-anatomic but physiometric reconstruction, MPFL Reconstruction (Semitendinosus 100 0.3 meaning the femoral fixation point is not anatomic, but Graft) behaves kinematically like a native MPFL (Fig. 2c); and MPFL Reconstruction (Gracilis Graft) 80 0.3 non-anatomic and non-physiometric reconstruction (Fig. MPFL Reconstruction (Quadriceps 33.6 0.3 2d). For this last type of reconstruction, the femoral fix- Tendon Graft) ation point is too anterior, which means the ligament is Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 5 of 15 too short and that it behaves kinematically, the opposite presented in terms of relative contact pressures, which of a native ligament (Sanchis-Alfonso et al. 2017). The we subsequently refer to as the contact pressure, to length of a normal MPFL increases during flexion from compare the different surgical techniques under the 0 to 30° and decreases from 30 to 120° (Sanchis-Alfonso same conditions. The femur position was fixed once et al. 2017). This pattern is considered as the in vivo every knee flexion position was simulated. MPFL standard dynamic length change. In a normal The data considered for the MPFL inclusion were (anatomic) MPFL reconstruction, the graft is isometric taken from a previous study (Sanchis-Alfonso et al. in all the cases between 0 and 30° of knee flexion (San- 2017). Table 3 summarizes the mean distance between chis-Alfonso et al. 2017). In 83% of cases, it is isometric the patella and femoral insertion points for the different from 0 to 60° of knee flexion (Sanchis-Alfonso et al. MPFL reconstructions. Based on that data, the insertion 2017). Beyond 60° of knee flexion, the MPFL becomes nodes for each technique and the elongation suffered by progressively lax and isometry is lost (Sanchis-Alfonso the ligaments were determined. The reference position, et al. 2017). Regarding isometry, a ligament is considered where the ligaments did not experience any strain was isometric when there is less than 5 mm of length change considered knee flexion at 40°. The average MPFL throughout the range-of-motion (Smirk and Morris lengths were considered, in this part of the study, to 2003). compare the performance of the different surgical tech- niques over the mean parametric FEM of the PFJ. LR Simulation of the different surgical techniques lengths were assumed to be the same as the MPFL The three surgical techniques were analysed for 5 knee length to preserve the equilibrium on both sides of the flexion positions: 0, 30, 60, 90 and 120°, as in a previous joint. dynamic CT scan study (Sanchis-Alfonso et al. 2017). The average length of the MPFL for each surgical Initially, for all of the surgical techniques, the patellar technique was analysed (Table 3), indicating that the dis- group (bone and cartilage) was not in contact with the tance between the femur and patella insertion points femoral group (bone and cartilage) to avoid non- was smaller than the reference distance (40°) in some desirable initial contact problems. The patella was ini- knee flexion positions. That means that the ligament is tially aligned with the trochlear groove using the CT im- not experiencing any type of stress. Thus, analysis of cer- ages. A perpendicular displacement (approximately 0.5 tain positions was not necessary (Table 3, cases indicated mm) to the femoral cartilage surface was imposed upon by a). In the remaining positions, two different types of the patella. Once the contact between both cartilages simulations were performed. First, in certain positions, was generated, initial contact pressures were stored. the MPFL undergoes an elongation, which is simulated Then, the ligaments and tendons were included and the by applying a pretension force, Δl ∗ K, where Δl is the three surgical MPFLr techniques were analysed. The ele- length increment and K is the stiffness of the ligament ments representing the QT and PT were then fixed so (Table 3, cases indicated by b). Second, several positions that the model was in equilibrium and no forces were showed an MPFL length that was only possible if the applied through them. The initial contact pressures were cartilage was compressed because the distance between subtracted from the ones generated with the ligaments the patella and femoral insertion point was further than and tendons inclusions. Therefore, the results are in the reference position (40°). As only-tension elements Table 3 Distance between the patellar and femoral insertion points for the MPFL reconstruction techniques analyzed Flexion Anatomic MPFL Reconstruction Non-anatomic MPFL Reconstruction Non-Anatomic MPFL Reconstruction Angle (°) with Physiometric Behavior but Non-Physiometric Length SD Length SD Length SD (mm) (mm) (mm) (mm) (mm) (mm) 0 60.2b ± 6.1 51.6b ± 4.6 37.5b ± 7.8 b b b 30 57.9 ± 6.8 50.8 ± 5.4 36.5 ± 9.2 40 57.7 ± 6.0 48.8 ± 5.0 36.2c ± 8.1 a a c 60 57.3 ± 6.4 44.9 ± 5.2 35.7 ± 10.1 90 55.6a ± 5.7 38.3a ± 4.9 35.6c ± 7.9 a a c 120 50.7 ± 4.9 33.7 ± 4.8 35.4 ± 5.6 MPFL with non-anatomical femoral attachment point with satisfactory results is always physiometric. MPFL with non-anatomical femoral attachment point with non-satisfactory results is always non-physiometric a No tension b Tension type 1 c Tension type 2 Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 6 of 15 can be compressed, only working under tension, this contact pressures at 60, 90 and 120° were very low com- relative position change was simulated with a pared to the pressures at 0 and 30°. Anatomic recon- temperature reduction equal to Δl=l0MPFL  αMPFL , where struction increased the pressure in all of the knee angles, Δl is the length increment, l0MPFL is the initial length of but the amount of pressure increase was only relevant at the MPFL and αMPFL is the assumed thermal dilatation 0°. In non-anatomic reconstructions with a physiometric coefficient of the MPFL (0.0005 °C− 1). This type of simu- behavior, an increase in all of the positions was found, lation allows cartilages to be modelled in a compressed but the amount of pressure was relevant only at 0 and state. Equilibrium on both sides of the joint was pre- 30°. In non-anatomic reconstruction without physio- served assuming the same Δl for the LR ligament and metric behavior the pressure increased in all the knee with the inclusion of the αLR coefficient for the LR, cal- positions and with a relevant amount of pressure. The culated as Δl=ΔT  l0LR (Table 3, cases indicated by c), maximum patellar cartilage contact pressures are dis- because ΔT was the same for the entire model. This was played in Fig. 3. an iterative process in which ΔT was recalculated until In a native knee, both the MPFL and LR are under the desired length of the MPFL was achieved. tension at 0 and 30° of knee flexion. At 60, 90 and 120°, both the MPFL and LR were loose. In both the anatomic and a non-anatomic MPFLr with physiometric behaviour, Clinical validation of the parametric model the ligament was tense between 0 and 30° of knee flexion, Five patient-specific cases were used for clinical validation but it had no tension at 60, 90 and 120°. In the non- of our parametric model. The geometry of each patient anatomic with non-physiometric behaviour reconstruc- was generated by modifying the main knee parameters of tion, the MPFL was tense at 60, 90 and 120° of knee the parametric model (femur and patella dimensions – see flexion and was completely loose at 0 and 30° of knee parametric finite element model of the patella femoral flexion. The MPFL and LR maximum stresses are dis- joint section-Fig. 1). Patient-specific geometrical data is in- played in Table 6. dicated in Table 4. MPFLr was simulated depending on The following cases demonstrate the sensitivity and patient-specific data. The graft insertion points were based possible clinical implications of the use of a parametric on each patient’s geometry with the help of the corre- model of the PFJ using FEM to evaluate MPFL sponding CT data. Each patient underwent a different reconstructions. type of MPFLr. Each specific case was simulated bearing the surgeon’s MPFL measurements in mind, as indicated Case # 1 (Fig. 4 and Table 7) in Table 5. Moreover, all five cases were clinically evalu- A 17-year-old man was operated on for lateral patellar in- ated by one of the authors (V S-A). stability using a single semitendinosus bundle MPFL graft. The patient expressed persistent lateral patellar instability Results and severe pain. The simulation predicted a contact pres- In a knee with a virtual intact MPFL, which was used as sure on the patellar cartilage of 1.19 MPa for 60° of knee a reference for the comparison among different recon- flexion, 2.25 MPa for the 90° position and an important struction techniques, the maximum patellar cartilage contact pressure of 5.84 MPa for 120° of knee flexion (Fig. 4). The maximum MPFL stress at 60° was 59.03 MPa. At Table 4 Patient-specific geometrical data (Fig. 1). Measurements 90°, it was 119.2 MPa and 252 MPa at 120°. At 0 and 30°, in mm the MPFL was loose. The maximum lateral retinaculum Case 1 Case 2 Case 3 Case 4 Case 5 (LR) stress at 60° was 1.62 MPa, 5.38 MPa at 90° and 7.06 Width 71.6 89.55 66 65.03 70.5 MPa at 120°. At 0 and 30°, the LR was loose. From that Width 2 49.71 60.13 50.2 46 53.48 data, we predicted that the patient would develop patellar Width 3 41.38 41.97 38.78 31.9 40.72 chondropathy, which was in fact seen during the arthros- Length 52.74 48.28 51.25 39.09 47.01 copy performed during the MPFL revision surgery (Fig. Length 2 32.42 40.16 31.3 30.95 33.05 4d). The tension pattern of the MPFL graft is typically seen in a non-anatomic femoral fixation point that is too Length 3 28.75 33.47 25.35 28.66 25.4 far anterior in which the graft exhibits non-physiometric Medial radius 27.43 36.94 29.56 27.57 22.28 behaviour. This can very clearly be seen in the last pre- Lateral radius 27.12 33.47 22.09 23.79 23.18 operative 3D CT scan (Fig. 4c). Posterior radius 14.07 11.2 12.43 14.73 12.09 Radius 19.01 20.32 17.12 16.54 32.23 Case # 2 (Fig. 5 and Table 7) Curvarture radius 44.94 51.24 42.41 43.51 43.17 A 28-year-old woman operated on for lateral patellar in- stability with a double-bundle MPFL plasty, using the Height 19.61 23.02 17.94 20.72 16.22 semitendinosus. The patient complained of severe pain Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 7 of 15 Table 5 Patient-specific data for the model validation Case Graft Material Measured length for each position (mm) Configuration 0° 30° 40° 60° 90° 120° 1 Semitendinosus 36.3a 35.9a 36.83 38.7b 43.7b 46.3b Non-Anatomic Femoral Attachment point with SB Non-Satisfactory Result 2 Semitendinosus 23.1a 33.3a 36.33b 42.4b 46.6b 48.6b Non-Anatomic Femoral Attachment Point with DB (Proximal) Non-Satisfactory Result Semitendinosus 25.4a 39.7a 42.77b 48.9b 54.3b 54.8b DB (Distal) 3 Quadriceps 56.2b 46.8b 43.03 35.5a 24.2a 22.4a Non-Anatomic Femoral Attachment Point with Tendon SB Non-Satisfactory Result 4 Semitendinosus 52.2b 51.1b 50.17 48.3a 41.3a 35a Anatomic Femoral DB (Proximal) Attachment Point with Satisfactory Result Semitendinosus 49.9b 49.7b 48.37 45.7a 39.7a 35.1a DB (Distal) 5 Semitendinosus 56.4b 57b 55.07 51.2a 46.9a 42.3a Anatomic Femoral Attachment Point with Sastisfactory Result DB (Proximal) Semitendinosus 55.1b 56b 54.17 50.5a 45.8a 41.9a DB (Distal) Cases # 1, 2 and 3 are non-anatomic and non-physiometric SB Single bundle, DB Double bundle a No tension b Tension type 1 and incapacitating lateral patellar instability. The simula- 30° of knee flexion, with greater stress than a native tion predicted higher contact pressures than in the pre- MPFL, and the patellofemoral pressure was below the vious simulation: 6.17 MPa for the 60° knee flexion normal values that could cause symptomatic PFOA. In position, 5.18 MPa for the 90° knee flexion position and fact, no chondropathy was seen in this patient during the 7.13 MPa for the 120° knee flexion position (Fig. 5). The arthroscopy performed in the revision surgery (Fig. 6d). maximum MPFL stress at 60° was 19.51 MPa, 29.52 MPa at 90° and 34.7 MPa at 120°. At 0 and 30°, the MPFL was Case # 4 (Fig. 7 and Table 7) loose. The maximum LR stress at 60° was 4.56 MPa, An 18-year-old woman was operated on for lateral patel- 7.54 MPa at 90° and 8.37 MPa at 120°. At 0 and 30°, the lar instability with an anatomic MPFL reconstruction LR was loose. The MPFL was tense at 60, 90 and 120° of using a double-bundle semitendinosus autograft, with an knee flexion and was completely loose at 0 and 30° of excellent clinical result at 5 years of follow up. The simula- knee flexion. Clinically, this tension pattern will lead to tion predicted a contact pressure of 0.2 MPa at 0° of knee PFOA, which was in fact seen during surgery (Fig. 5d). flexion and 0.91 MPa at 30° of knee flexion. The max- This tension pattern is typical of a non-anatomic fem- imum MPFL and LR stresses predicted for the 30° of knee oral fixation point that is far too anterior, as clearly seen flexion position were 29.47 MPa and 0.79 MPa, respect- in the 3D CT scan in which the graft exhibits non- ively. For 0° of knee flexion, they were 60.02 MPa and physiometric behaviour (Fig. 5c). 1.15 MPa, respectively. The prediction fulfils the require- ments for an ideal MPFLr; a tense graft at 0 and 30° of Case # 3 (Fig. 6 and Table 7) knee flexion with far greater stress to failure than a native A 38-year-old woman was operated on for lateral patellar ligament. The patellofemoral pressures were kept below instability with an MPFL single-bundle reconstruction the values that could cause symptomatic PFOA. This ten- using the quadriceps tendon. The patient complained of sion pattern is typical of an anatomic femoral fixation severe pain and incapacitating lateral patellar instability. point as is clearly seen in the 3D CT scan (Fig. 7c). The simulation performed with our FEM showed patello- femoral contact pressures far below those found in a na- Case # 5 (Fig. 8 and Table 7) tive knee (Fig. 6a). The maximum MPFL and LR stresses A 15-year-old woman was operated on for lateral patel- predicted for the 0° knee flexion position were 12.28 MPa lar instability with an anatomic MPFL reconstruction and 8.22 MPa, respectively. They were 3.93 MPa and 2.68 using a double-bundle semitendinosus autograft, with an MPa for 30°, respectively. The prediction fulfils the re- excellent clinical result at 5 years of follow up. The simu- quirements for an effective MPFLr: a tense graft at 0 and lation predicted a contact pressure of 1.57 MPa for 0° of Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 8 of 15 Fig. 3 Patellar cartilage contact pressure (MPa): a Native knee. b Anatomic MPFL reconstruction. c Non-anatomic MPFL reconstruction with physiometric behaviour. d Non-anatomic MPFL reconstruction with a femoral fixation that is too far anterior and without physiometric behaviour (M-medial; L-Lateral) knee flexion position and 1.63 MPa for 30° of knee tense graft at 0 and 30° of knee flexion with a far higher flexion position. The maximum MPFL and LR stresses stress to failure than a native ligament. The patellofe- predicted for the 30° knee flexion position were 70.3 moral pressure values were below those thought to cause MPa and 1.27 MPa, respectively. At 0° of knee flexion, a symptomatic PFOA. This tension pattern is typical of they were 40.24 MPa and 0.53 MPa, respectively. The an anatomic femoral fixation point as is clearly seen in prediction fulfils the requirements for an ideal MPFLr; a the 3D CT scan (Fig. 8c). Table 6 MPFL and LR stress Maximum MPFL Stress (MPa) Maximun LR Stress (MPa) Model’s accuracy FEM was very accurate in cases 1, 2, 4 and 5, but not in 0° 30° 60° 90° 120° 0° 30° 60° 90° 120° case 3. Case 3 fulfilled the requirement for a correct A 8.85 0.78 0 0 0 1.52 0.15 0 0 0 plasty relative to the maximum stress and patellofemoral B 74.72 6.55 0 0 0 1.51 0.14 0 0 0 pressure. However, the patient had pain and instability C 97.02 69.60 0 0 0 1.66 1.10 0 0 0 after surgery. The instability could be explained by the D 63.44 14.74 46.71 77.57 92.70 0.78 0.17 1.24 2.09 2.51 single-bundle configuration of the graft, the vertical dir- A) Native knee ection of the graft because of the non-anatomic femoral B) Anatomic MPFL reconstruction with semitendinosus fixation point (Fig. 6b) and the patella alta. All of them C) Non-anatomic MPFL reconstruction with a physiometric behavior D) Non-anatomic MPFL reconstruction with a femoral fixation that is too make this graft non-functional. All these factors can anterior and without a physiometric behavior contribute to instability and therefore to pain. Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 9 of 15 Fig. 4 Case # 1 Surgical failure: a Contact pressure (MPa) on the patellar cartilage L=lateral, M= Medial. b Parametric model of patient # 1. c Femoral attachment point is too far anterior (black arrow). d Visible patellar chondropathy during arthroscopy Discussion This model is the first parametric 3D FEM of the PFJ that analyses the effect of different MPFL reconstructions on the patella contact pressure as well as on the kinematic Table 7 MPFL and LR ligaments stress obtained for each behaviour of the MPFL-graft and MPFL-graft stress along reconstruction and position analyzed the total range-of-motion of the knee. Case Flexion Angle (°) MPFL stress (MPa) LR stress (MPa) Generation of a patient-specific FEM (i.e., a real 1 60 59.03 1.62 FEM) of the PFJ requires CT images to be processed, 90 119.20 5.38 segmented and then converted into a 3D finite elem- 120 252.00 7.06 ent model. This process is complex, expensive and 2 60 19.51 4.56 very time-consuming. However, the parametric model 90 29.52 7.54 is the opposite. Segmentation is a process that re- quires manual correction to eliminate undesired tis- 120 34.7 8.37 sues, and the computational burden makes the real 3 0 12.28 8.22 model unsuitable for clinical integration as a tool for 30 3.93 2.68 MPFLr planning. A parametric model is a generic 4 0 60.02 1.15 model, that is a simplified model valid for any knee 30 29.47 0.79 that could have direct clinical application. 5 0 40.24 0.53 The difference between the current work and previous ones is that in this paper the contact pressures for all 30 70.30 1.27 the angles of knee flexion (from 0 to 120°) in both Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 10 of 15 Fig. 5 Case # 2 Surgical failure: a Contact pressure (MPa) on the patellar cartilage L = Lateral M = Medial. b Parametric model of patient # 2. Trochlear dysplasia type D (red arrow). c Superior femoral attachment point is too far anterior (black arrow) Trochlear dysplasia (red arrow). d Visible patellofemoral osteoarthritis anatomical and non-anatomical (physiometric and non- errors during MPFLr estimating contact pressures between physiometric) MPFL reconstructions are analysed. Since 3 and 6 MPa (Elias and Cosgarea 2006; Elias et al. 2016). this is a novel method, we focused on clinical validation. Shah et al. also obtained very similar values to previous In this way, five clinical cases are presented to demon- computational studies. Various authors have demonstrated strate the accuracy of the model and to show its versatility that the changes in the length of a ligament that occur dur- for predicting challenging clinical cases. An extrapolation ing joint flexion-extension show changes in the tension of of the computational results was performed to provide a that ligament (Good 1995; Moritomo et al. 2009; Sanchis- qualitative comparison to the clinical outcomes. The con- Alfonso et al. 2017; Seo et al. 2014; Tan et al. 2011). Based tribution of these results is the introduction of FEM in on this observation, in a previous study using a dynamic daily clinical practice to optimize surgical procedures by CT scan, it has been concluded that the native MPFL was using personalized treatments. tense during the first 30° of knee flexion in all cases and Findings using the FEM are in agreement with those re- progressively loosened after 30° (Sanchis-Alfonso et al. ported in previous computational studies (Elias et al. 2005; 2017). The explanation behind this conclusion lies in the Elias and Cosgarea 2006; Shah et al. 2015) and could have fact that the attachment points of the MPFL are separated meaningful potential implications for clinicians performing further during the first 30° of knee flexion and become MPFLr surgery (Conlan et al. 1993; Desio et al. 1998; Elias progressively closer from 30° onwards. It is called the phy- and Cosgarea 2006; Hautamaa et al. 1998; Sanchis-Alfonso siometric behaviour of the ligament. The current study en- et al. 2017; Servien et al. 2011). Elias et al. evaluated medial abled us to directly confirm these findings. The ligament is patellofemoral cartilage overload in cases with technical tense between 0 and 30° of knee flexion, but at 60, 90 and Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 11 of 15 Fig. 6 Case # 3 Surgical failure: a Contact pressure (MPa) on the patellar cartilage L = Lateral M = Medial. b One can see that the graft is preserved; however, the orientation is too oblique and extremely vertical. c Parametric model of patient # 3. d Arthroscopy at the time of the revision surgery shows normal patellofemoral cartilage 120°, it has no tension. This fact has clinical relevance as The ideal MPFLr technique must combine a precise the MPFL is a structure that is only involved in the lateral balance between an optimal patellofemoral pressure with stability of the patella during the first 30° of knee flexion. maximum graft stress. It makes a new tear less likely. After 30°, the ligament loosens and the patellofemoral con- The patellofemoral contact pressure of a virgin knee tact pressure, which also contributes somewhat to patello- must be reproduced, and a maximum MPFL-graft stress femoral stability and is already low during the first 30° greater than that of the native MPFL must be created (0.23 MPa), decrease considerably (0.0046 MPa). This find- with the intention to compensate for the anatomic fac- ing is in agreement with several anatomic and biomechan- tors (increased tibial tuberosity – trochlear groove (TT- ical studies that show that the MPFL is the most important TG) distance, patella alta and trochlear dysplasia) that restraint to lateral patellar displacement from 0 to 30° of predispose to lateral patellar dislocation (Sanchis-Al- knee flexion (Conlan et al. 1993; Desio et al. 1998; Hauta- fonso 2014). In fact, the maximum MPFL-graft stress in maa et al. 1998). After 30° of knee flexion, lateral patella both anatomic and non-anatomic but physiometric re- stability depends on the femoral trochlea. Additionally, this constructions is much greater than that of a native study confirms previous findings that show that the loca- MPFL. However, it is very important not to increase tion of the femoral attachment point is of utmost import- maximum MPFL-graft stress with a subsequent increase ance to obtain satisfactory clinical results (Sanchis-Alfonso in the patellofemoral pressure because the technique will et al. 2017). The femoral attachment point is related to the have a suitable result in the short term but will have a patellofemoral contact pressure, tension of the MPFL-graft deleterious effect and will lead to degenerative changes and physiometry of the reconstruction. in the long term. MPFLr evaluation by means of the Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 12 of 15 Fig. 7 Case # 4 Primary surgery with an excellent result: a Contact pressure (MPa) on the patellar cartilage L = Lateral M = Medial. b Parametric model of patient # 4. c Anatomic femoral attachment point (black arrow) FEM is more sensitive than evaluations using only clin- increase. In this way Stephen et al., in 2014 and 2016, ical or radiological tests. The FEM can demonstrate the found that an anatomic MPFLr with a tension of 2 N and validity of a surgical technique in the long term since it fixed at 0, 30 or 60° of knee flexion, regardless of the type enables one to determine whether a specific technique of graft used, restores PFJ contact pressures to the intact will lead to an increase in the patellofemoral pressure, state. However, graft overtensioning and/or non-anatomic which is closely related to future development of PFOA. positioning of the femoral attachment increases PFJ con- The elevated MPFL graft tension or an incorrect femoral tact pressures (Elias et al. 2016; Stephen et al. 2014, 2016). tunnel position will increase the pressure applied to patel- A broad variability in patellofemoral anatomy, graft ten- lofemoral cartilage (Stephen et al. 2014, 2016), and this in- sion and non-anatomic femoral attachment could explain crease in PFJ contact pressure might result in joint these different findings. degeneration (Rood et al. 2015; Stephen et al. 2014). Rood The current tendency is to perform MPFL reconstruc- et al. in 2015 have shown that static MPFL reconstruc- tions with an anatomic femoral bone attachment and pa- tions (i.e., reconstruction with both femoral and patellar tellar bone attachment. In our study, we observed an osseous attachments) result in higher patellofemoral pres- increase in the patellofemoral contact pressures at 0 and sures compared with those in the intact situation and thus 30° of knee flexion after an MPFL reconstruction (2.17 increase the chance of PFOA in the long term. While MPa at 0° and 0.14 MPa at 30° when using the semitendi- Rood et al. showed elevated contact pressures with nosus as a graft) compared to the pressure found in a nor- MPFLr, Stephen et al. in 2014 did not show a pressure mal non-operated knee (0.18 MPa at 0° and 0.016 MPa at Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 13 of 15 Fig. 8 Case # 5 Primary surgery with an excellent result: a Contact pressure (MPa) on the patella cartilage L = Lateral M = Medial. b Parametric model of patient # 5. c Anatomic femoral attachment point (black arrow) 30°). This leads us to consider the possible long-term ef- clinical effects after 2 years when using these reconstruc- fects from a slightly greater patellofemoral contact pres- tions, which could be due to the short follow-up in both sures. However, in theory, the patellofemoral contact cases. In this type of reconstruction, the FEM shows an pressures found in the anatomic reconstructions are not increase in patellofemoral contact pressure at 0 and 30° great enough to cause symptomatic PFOA since they are of knee flexion in comparison to these pressures in the lower than those causing knee osteoarthritis (Segal et al. native knee (2.77 MPa at 0° and 1.91 MPa at 30° vs 0.18 2009). The objective would be not to exceed safe levels of MPa at 0° and 0.016 MPa at 30°). This pressure increase patellofemoral pressure to induce patellofemoral chondro- mainly occurs on the medial patellar facet. According to pathy and ultimately PFOA. It should also be remembered Jones et al. (2016), the average contact stress at 30° is that the increase in patellofemoral contact pressures 1.7 ± 0.6 MPa, with a peak of 3.2 ± 0.6 on the surface of helps to stabilize the PFJ. Therefore, this factor would the patellar cartilage and of 2.8 ± 0.7 MPa at the deepest be beneficial in the classical anatomic reconstruction. point. The differences found between this study and the Thus, a discrete increase in contact pressure, as we one by Jones et al. in 2016 can be explained by the fact have observed, is desirable. that Jones uses a laboratory controlled study with ca- Currently, what is being discussed is the precise conse- daver knees using a different method than us. What is quences of the clinical results of the non-anatomical not known is whether this pressure increase will result techniques for the MPFLr in which the MPFL-graft be- in chondropathy in the long-term and ultimately result haves like a native MPFL (physiometric behaviour) from in symptomatic PFOA. As far as we know, there is no the physiological point of view. Servien et al. in 2011 study of the PFJ that has determined the contact stress and Sanchis-Alfonso et al. in 2017 found no negative threshold that is predictive of symptomatic PFOA. Segal Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 14 of 15 et al. in 2009 observed that a threshold of 3.42 to 3.61 MPFL. Another important limitation of this study is the MPa had a 73.3% sensitivity with specificity ranging fact that the patellofemoral pressure values that predict from 46.7% to 66.7% for the prediction of symptomatic the development of a symptomatic PFOA are not known. knee osteoarthritis. Obviously, these values cannot be ex- We have extrapolated the well-known values that would trapolated to the PFJ, which is the joint with the thickest lead to the development of a symptomatic tibiofemoral cartilage in the human body. It is logical to think that the osteoarthrosis. It has also been hypothesized that the pressures causing symptomatic PFOA would be greater. In values necessary to develop a symptomatic PFOA should non-anatomical MPFL reconstructions, the maximum be higher than those for a symptomatic tibiofemoral patellofemoral contact pressures are on the order of 2.77 osteoarthritis because the patellar cartilage is much MPa, values that are considerably below the cut-off point thicker than that found on the tibia or in the femur (Segal mentioned above. Therefore, it is likely that a non- et al. 2009). For that reason, we speculate that a higher anatomical but physiometric reconstruction would not have pressure would be necessary to cause damage. Using the long-term negative effects on the PFJ. Consequently, it FEM allows to reliably predict the clinical evolution of an would seem more important for the ligament to be “physio- MPFL-graft. Logically, in a condition with multifactorial metric” rather than perfectly anatomical. etiopathogeny such as lateral patellar instability, the model With the FEM, it is possible to predict which MPFLr fails in some cases because there are additional factors have an increased risk of severe patellofemoral chondro- (e.g., patella alta, increased tibial tubercle-trochlear groove pathy resulting in symptomatic PFOA and requiring ac- distance and trochlear dysplasia) other than the tension of tive treatment. In the cases in which PFOA occurred, it the MPFL-graft and patellofemoral contact pressures that was because the MPFL-graft was loose, with knee flexion could be responsible for the failed surgery. This is a major from 0 to 30°, and was tense from 60° onward. In these limitation of this study. The abovementioned anatomic cases, the patellofemoral contact pressures were over 5 additional factors are often associated in patients requiring MPa from 60° onward, the femoral attachment point be- MPFLr and can change the pressures at the PFJ and lead ing extremely non-anatomical (too far anterior) and the to different outcomes. Although it has not been addressed MPFLr was not physiometric. The predictive value of in the present work, the conditions in which the graft the parametric model of the PFJ has made its clinical would not prevent post-operative instability could be in- validation possible. corporated (Farahmand et al. 1998; Hautamaa et al. 1998; A limitation of this study is that the patellar and femoral Sanchis-Alfonso 2014). cartilages had a constant thickness of approximately 3 mm. The PFJ was reconstructed from CT data in which Conclusion soft tissues are not clearly distinguished. However, the gap The main finding of this study is that the use of a para- between both bones was approximately 6 mm. Accord- metric 3D finite element model of the PFJ allows the ingly, the same thickness for both cartilages was assumed. evaluation of different types of surgical techniques for Small differences would have been predicted if other MPFLr with regard to the effect on the patellofemoral thickness values had been considered. Additionally, the contact pressure. That also goes for the kinematic be- ligament material properties were taken from the litera- haviour of the MPFL-graft with flexion-extension of the ture (Ciccone II et al. 2006; Drez Jr et al. 2001; Elias and knee and the maximum MPFL-graft stress based on a Cosgarea 2006). In the future, patient-specific material previous study which has shown that the graft length properties should be considered. The inclusion of mag- variation differs in each type of MPFLr. In this way, from netic resonance (MR) data from the same patients and the diagnostic images like a CT, for example, it is possible to use of image registration techniques might combine MR simulate different surgical treatments and customize the and CT data. It which would not only make it possible to treatment for individual patients. extract cartilage thickness accurately but also to determine patient-specific multi-variate matrix properties, such as Additional file the T1 or T2 relaxation times, which are related to proteo- glycan and collagen matrix integrity, respectively (Martí- Additional file 1: Figure S1. Patellar cartilage contact pressure (MPa) Bonmatí et al. 2008). Another limitation is that there was obtained for different values of friction coefficient (µ) (0.01, 0.02, 0.025, 0.03) in the 30º knee flexion angle for the native knee model. Figure S2. no estimation of the amount of error in the patient- Patellar cartilage contact pressure (MPa) obtained for different patellar- specific shape when creating the patient-specific model. femoral cartilages combinations in the 30º knee flexion angle for the There was only qualitative assessment of the global intact knee model. PCT=Patellar Cartilage Thickness, FCT=Femoral Cartilage Thickness. (DOCX 387 kb) patient-specific shape. Additionally, to preserve equilib- rium, the elements representing the QT and PT were fixed Abbreviations and no forces were applied through them. Furthermore, CT: Computerized Tomography; FEM: Finite element methodology; the same LR length changes were assumed as for the LR: Lateral retinaculum; MPa: Megapascal; MPFL: Medial patellofemoral Sanchis-Alfonso et al. Journal of Experimental Orthopaedics (2019) 6:32 Page 15 of 15 ligament; MPFLr: Medial patellofemoral ligament reconstruction; Farahmand F, Senevongse W, Amis A (1998) Quantitative study of the quadriceps MR: Magnetic resonance; PFJ: Patelofemoral joint; PFOA: Patellofemoral muscles and trochlear groove geometry related to instability of the osteoarthritis; PT: Patellar tendon; QT: Quadriceps tendon; RF: Rectus femoris; patellofemoral joint. J Orthop Res 16:136–143 VI: Vastus intermedius; VL: Vastus lateralis; VM: Vastus medialis Fernandez J, Hunter P, Shim V, Mithraratne K (2012) A subject-specific framework to inform musculoskeletal modeling: outcomes from the IUPS physiome Authors’ contributions project. In: Calvo Lopez B, Peña E (eds) Patient-specific computational (I) Conception and design: V.S-A; (II) Provision of patients: V.S-A; (III) modeling. Lecture notes in computational vision and biomechanics, vol 5. Adquisition of the images to reconstruct the FE models: C.R-F and F.G-C; (IV) Springer, Dordrecht, pp 39–60 Creation of the parametric model and FE simulations: D.A-L; (V) Data analysis Fink C, Veselko M, Herbort M, Hoser C (2014) MPFL reconstruction using a and interpretation: All authors; (VI) Manuscript writing: All authors. All authors quadriceps tendon graft: part 2: operative technique and short term clinical read and approved the final manuscript. results. Knee 21:1175–9117 Fulkerson JP, Edgar C (2013) Medial quadriceps tendon-femoral ligament: surgical Funding anatomy and reconstruction technique to prevent patella instability. Arthrosc This project has been partially founded by a grant from the program RETOS Tech 12:e125–e128 (DPI2016-80283-C2-2-R). Good L (1995) In vitro correlation between tension and length change in an anterior cruciate ligament substitute. Clin Biomech (Bristol, Avon) 10:200–207 Hautamaa PV, Fithian DC, Kaufman KR, Daniel DM, Pohlmeyer AM (1998) Medial Competing interests soft tissue restraints in lateral patellar instability and repair. Clin Orthop Relat The authors declare that they have no competing interests. Res 349:174–182 Jones B, Hung CT, Ateshian G (2016) Biphasic analysis of cartilage stresses in the Author details 1 patellofemoral joint. J Knee Surg 29:92–98 Department of Orthopaedic Surgery, Hospital Arnau de Vilanova, C/Sant Martí-Bonmatí L, Sanz-Requena R, Alberich-Bayarri A (2008) Pharmacokinetic MR Climent, 12, 46015 Valencia, Spain. 2Multiscale in Mechanical and Biological analysis of the cartilage is influenced by field strength. Eur J Radiol 67:448–452 Engineering (M2BE), Aragón Institute of Engineering Research (I3A), Moritomo H, Noda K, Goto A, Murase T, Yoshikawa H, Sugamoto K (2009) Department of Mechanical Engineering, University of Zaragoza, Zaragoza, Interosseous membrane of the forearm: length change of ligaments during Spain. 3Department of Orthopaedic Surgery, Hospital Terres de l’Ebre, forearm rotation. J Hand Surg Am 34:685–691 Tortosa, Spain. 4Agoriaz Orthopaedic Center, Riaz & Clinique CIC, Montreux, Rood A, Hannink G, Lenting A, Groenen K, Koëter S, Verdonschot N, van Kampen Switzerland. 5QUIBIM SL, Quantitative Imaging Biomarkers in Medicine, A (2015) Patellofemoral pressure changes after static and dynamic medial Valencia, Spain. 6Hospital Universitario y Politécnico La Fe and Biomedical patellofemoral ligament reconstructions. Am J Sports Med 43:2538–2544 Imaging Research Group (GIBI230), IIS La Fe Research Group, Valencia, Spain. 7 Sanchis-Alfonso V (2014) Guidelines for medial patellofemoral ligament reconstruction Department of Orthopaedic Surgery and Traumatology, Hospital del Mar, in chronic lateral patellar instability. J Am Acad Orthop Surg 22:175–182 Universitat Autònoma de Barcelona, Barcelona, Spain. 8QUIBIM SL, Sanchis-Alfonso V (2016) How to deal with chronic patellar instability: what does Quantitative Imaging Biomarkers in Medicine, GIBI230, Biomedical Imaging the literature tell us? Sports Health 8:86–90 Research Group, La Fe Health Research Institute, Valencia, Spain. Sanchis-Alfonso V, Ramirez-Fuentes C, Montesinos-Berry E, Domenech J, Martí- Bonmati L (2017) Femoral insertion site of the graft used to replace the Received: 19 March 2019 Accepted: 21 June 2019 medial patellofemoral ligament influences the ligament dynamic changes during knee flexion and the clinical outcome. Knee Surg Sports Traumatol Arthrosc 25:2433–2441 References Segal NA, Anderson DD, Iyer KS (2009) Baseline articular contact stress levels Besier TF, Gold GE, Delp SL, Fredericson M, Beaupre GS (2008) The influence of predict incident symptomatic knee osteoarthritis development in the MOST femoral internal and external rotation on cartilage stresses within the cohort. J Orthop Res 27:1562–1568 patellofemoral joint. J Orthop Res 26:1627–1635 Seo YJ, Song SY, Kim IS, Seo MJ, Kim YS, Yoo YS (2014) Graft tension of the Blankevoort L, Huiskes R (1991) Ligament-bone interaction in a three dimensional posterior cruciate ligament using a finite element model. Knee Surg Sports model of the knee. J Biomech Eng 113:263–269 Traumatol Arthrosc 22:2057–2063 Ciccone WJ II, Bratton DR, Weinstein DM, Elias JJ (2006) Viscoelasticity and Servien E, Fritsch B, Lustig S (2011) In vivo positioning analysis of medial temperature variations decrease tension and stiffness of hamstring tendon patellofemoral ligament reconstruction. Am J Sports Med 39:134–139 grafts following anterior cruciate ligament reconstruction. J Bone Joint Surg Shah KS, Saranathan A, Koya B, Elias JJ (2015) Finite element analysis to characterize Am 88:1071–1078 how varying patellar loading influences pressure applied to cartilage: model Cohen ZA, Mow VC, Henry JH, Levine WN, Ateshian GA (2003) Templates of the evaluation. Comput Methods Biomech Biomed Engin 18:1509–1515 cartilage layers of the patellofemoral joint and their use in the assessment of Smirk C, Morris H (2003) The anatomy and reconstruction of the medial osteoarthritic cartilage damage. Osteoarthr Cartil 11:569–579 patellofemoral ligament. Knee 10(3):221–227 Conlan T, Garth WP Jr, Lemons JE (1993) Evaluation of the medial soft-tissue Stephen JM, Kaider D, Lumpaopong P, Deehan DJ, Amis AA (2014) The effect of restraint of the extensor mechanism of the knee. J Bone Joint Surg Am 75: femoral tunnel position and graft tension on patellar contact mechanics and 682–693 kinematics after medial patellofemoral ligament reconstruction. Am J Sports Desio SM, Burks RT, Bachus KN (1998) Soft tissue restraints to lateral patellar Med 42:364–372 translation in the human knee. Am J Sports Med 26:59–65 Stephen JM, Kittl C, Williams A, Zaffagnini S, Marcheggiani GM, Fink C, Amis AA (2016) DeVries WNA, Duchman KR, Bollier MJ (2015) A finite element analysis of medial Effect of medial patellofemoral ligament reconstruction method on patellofemoral patellofemoral ligament reconstruction. Iowa Orthop J 35:13–19 contact pressures and kinematics. Am J Sports Med 44:1186–1194 Drez D Jr, Edwards TB, Williams CS (2001) Results of medial patellofemoral ligament Tan J, Xu J, Xie RG, Deng AD, Tang JB (2011) In vivo length and changes of ligaments reconstruction in the treatment of patellar dislocation. Arthroscopy 17:298–306 stabilizing the thumb carpometacarpal joint. J Hand Surg Am 36:420–427 Elias JJ, Bratton DR, Weinstein DM, Cosgarea AJ (2006) Comparing two Teitge RA, Torga-Spak R (2004) Medial patellofemoral ligament reconstruction. estimations of the quadriceps force distribution for use during patellofemoral Orthopedics 27:1037–1040 simulation. J Biomech 39:865–872 Weinberger JM, Fabricant PD, Taylor SA, Mei JY, Jones KJ (2017) Influence of graft Elias JJ, Cech JA, Weinstein DM, Cosgarea AJ (2005) Reducing the lateral force source and configuration on revision rate and patient-reported outcomes acting on the patella does not consistently decrease patellofemoral after MPFL reconstruction: a systematic review and meta-analysis. Knee Surg pressures. Am J Sports Med 32:1202–1208 Sports Traumatol Arthrosc 25:2511–2519 Elias JJ, Cosgarea AJ (2006) Technical errors during medial patellofemoral ligament reconstruction could overload medial patellofemoral cartilage: a computational analysis. Am J Sports Med 34:1478–1485 Publisher’s Note Elias JJ, Kelly MJ, Smith KE, Gall KA, Farr J (2016) Dynamic simulation of the Springer Nature remains neutral with regard to jurisdictional claims in effects of graft fixation errors during medial patellofemoral ligament published maps and institutional affiliations. reconstruction. Orthop J Sports Med 4(9):2325967116665080