Author’s Accepted Manuscript

Optimal graft stiffness and pre-strain restore normal

joint motion and cartilage responses in ACL
reconstructed knee

K.S. Halonen, M.E. Mononen, J. Töyräs, H.

Kröger, A. Joukainen, R.K. Korhonen
www.elsevier.com/locate/jbiomech

PII: S0021-9290(16)30549-8
DOI: http://dx.doi.org/10.1016/j.jbiomech.2016.05.002
Reference: BM7719
To appear in: Journal of Biomechanics
Received date: 8 September 2015
Revised date: 17 April 2016
Accepted date: 2 May 2016
Cite this article as: K.S. Halonen, M.E. Mononen, J. Töyräs, H. Kröger, A.
Joukainen and R.K. Korhonen, Optimal graft stiffness and pre-strain restore
normal joint motion and cartilage responses in ACL reconstructed knee, Journal
of Biomechanics, http://dx.doi.org/10.1016/j.jbiomech.2016.05.002
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.
Optimal graft stiffness and pre-strain restore normal
joint motion and cartilage responses in ACL
reconstructed knee

K S Halonen1, M E Mononen1, J Töyräs1,2, H Kröger3,4, A Joukainen3, R K Korhonen1,2

1
Department of Applied Physics, University of Eastern Finland, POB 1627, FI-70211
Kuopio, Finland
2
Diagnostic Imaging Centre, Kuopio University Hospital, POB 100, FI-70029 KUH, Kuopio,
Finland
3
Department of Orthopaedics, Traumatology and Hand Surgery, Kuopio University Hospital,
POB 100, FI-70029 KUH, Kuopio, Finland
4
Kuopio Musculoskeletal Research Unit (KMRU), University of Eastern Finland, POB 1627,
FI-70211, Kuopio, Finland

Corresponding author:
Kimmo Halonen
Department of Applied Physics, University of Eastern Finland
POB 1627, FI-70211 Kuopio, Finland
Tel. +358 50 3580102
Fax. +358 17 162585
E-mail: kimmo.halonen@uef.fi

Abstract

Anterior cruciate ligament (ACL) rupture leads to abnormal loading of the knee joint and

increases the risk of osteoarthritis. It is unclear how different ACL reconstruction techniques

affect knee joint motion and mechanics. As the in vivo measurement of knee joint loading is

not possible, we used finite element analysis to assess the outcome of ACL reconstruction

techniques.

1
Effects of different ACL reconstruction techniques on knee joint mechanics were studied

using six models during gait; with 1) healthy ACL, 2) ACL rupture, 3) single bundle ACL

reconstruction, 4) double bundle ACL reconstruction, 5) weakened (softer) single bundle

reconstruction and 6) single bundle reconstruction with less pre-strain.

Early in the gait, the ACL rupture caused substantially increased tibial translation in the

anterior direction as well as a smaller but increased lateral translation and internal tibial

rotation. ACL rupture substantially reduced average stresses and strains, while local peak

stresses and strains could be either increased or decreased. Single bundle and double bundle

reconstructions restored joint motion close to normal levels. However, cartilage strains and

stresses were elevated during the entire gait cycle. Models with modulated graft stiffness and

pre-strain restored the joint motion and cartilage stresses and strains close to the normal,

healthy levels.

Results suggest that rather than the choice of reconstruction technique, stiffness and pre-

strain of the ACL reconstruction affect the motion and mechanics of the operated knee. We

suggest that an optimal choice of graft properties might help restore normal knee joint

function and cartilage responses, thus, minimizing the risk of osteoarthritis.

Keywords: Anterior cruciate ligament, reconstruction, graft, knee joint, articular cartilage,
gait, finite element analysis

1. Introduction

Anterior cruciate ligament (ACL) is an essential structure of the knee, providing stability to

the joint during its movement. ACL rupture is a common, often sports-related injury, which if

left untreated, has been shown to greatly increase the risk of osteoarthritis (OA) in the knee

articular cartilage (Barenius et al., 2014; Kessler et al., 2008). Even though the mechanism

2
behind the development of OA in relation to the ACL injury is not fully understood, the

altered motion of the knee and cartilage stresses are believed to be major causes (Kessler et

al., 2008; Tsoukas et al., 2015). In ACL reconstruction, an autograft, harvested from the

patient, or a donor-origin allograft, is used to reconstruct the ruptured ligament (Meredick et

al., 2008). A recent high quality randomized controlled trial, comparing conservative

treatment with early ACL reconstruction, showed similar clinical follow-up results between

the groups at five years. However, 51% of the conservative rehabilitation-only treatment

group had gone through reconstructive surgery during the 5-year follow-up (Frobell et al.,

2013).

Single bundle and double bundle techniques are the most common ACL reconstruction

techniques (Meredick et al., 2008). In the single bundle reconstruction technique, a graft is

harvested either from the central part of the patellar tendon or from gracilis and

semitendinosus tendons (Herrington et al., 2005). Two insertion holes are drilled, one into

femur and the other into tibia. The graft is then tensioned and fastened with special bone

screws (Kim et al., 2013). In the double bundle reconstruction technique, two grafts, usually

folded gracilis and semitendinosus tendons, are inserted into the knee to mimic the anterior-

medial (AM) and posterior-lateral (PL) bundles of a healthy ACL. Though the causality

mechanism of ACL rupture and secondary OA is not known, the ability to preserve knee joint

function seems to be important for minimizing the risk of OA (Barenius et al., 2014; Kessler

et al., 2008). Therefore a means to assess and optimize the outcome of ACL reconstruction is

needed.

Strains in the knee joint cartilage have been measured in vivo during static loading (Halonen

et al., 2014; Hosseini et al., 2010). However, during dynamic gait loading, direct assessment

of stresses and strains in articular cartilage is not possible, but they can be simulated with

finite element (FE) modeling (Bae et al., 2012; Peña et al., 2006; Shirazi et al., 2008). In

3
previous FE studies, ACL rupture has either been modeled using a fixed flexion angle

(Huang et al., 2012; Peña et al., 2005) or a simple linear elastic material model for cartilage

(Moglo and Shirazi-Adl, 2003; Yao et al., 2005). Simulation of gait is obviously more

relevant when assessing the risk of OA. In order to study the effect of changes in the ACL on

the knee motion and cartilage responses, implementation of the gait as forces and moments is

needed. Furthermore, cartilage is a poroelastic and fibril-reinforced material, and these

properties have an influence when assessing risk locations and mechanisms of damage in

joints (Brody, 2014; Hosseini et al., 2014). However, for instance changes in collagen fibril

strains (high strains may lead to damage and OA (Butler et al. 1978; Wang 2007)) due to

injury and/or reconstruction have not been shown before. Furthermore, these complex

properties may also have an effect on other parameters, such as fluid pressure and maximum

principal stress, the latter one describing typically tissue failure (Danso et al. 2014; Mononen

et al. 2016).

In this study, we simulated the effect of ACL rupture and reconstruction techniques on the

knee joint motion as well as stresses and strains in the tibial articular cartilage during the

stance phase of gait. A model with ACL deficiency as well as two most common (Herrington

et al., 2005) reconstruction models, single bundle and double bundle ACL reconstructions,

were compared with the healthy knee joint model. We hypothesized that, due to the way the

double bundle technique mimics the anatomy of a healthy ACL, it would preserve the knee

motion better than the single bundle, but cause similar stresses and strains in the tibial

articular cartilage (as a consequence of the graft stiffness). We also hypothesized that the

reduced stiffness and pre-strain of the graft would cause decreased stresses and strains in the

cartilage. The reduced stiffness and pre-strain were tested on both the single bundle and

double bundle grafts.

4
2. Materials and methods

Workflow of the study is presented in Figure 1.

Gait analysis

Analysis of the subject’s gait pattern was conducted in a motion analysis laboratory and is

described in detail in Klodowski et al. (2015) and Halonen et al. (2016). The final gait input

(Fig. 2 a-c) for the FE model (see below) included the quadriceps forces as well as knee joint

moments and translational forces, excluding the extension-flexion angle, which was defined

as rotation. Similar implementation of the gait data into the FE model was done recently

(Tanska et al., 2015).

It has been reported that the passive internal-external moment is approximately 10-20% of

the measured ones (Adouni et al., 2012). Another study by Tanska et al. (2015) found the best

match with experimental studies (Benoit et al., 2006; Kabada et al., 1990; Reinschmidt et al.,

1997) by implementing 10% of the internal-external moment in the knee joint FE model. In

this study, the supporting effect of muscles and other tissues were accounted for by using

50% of the total varus-valgus and internal-external moments within the knee, which was

found to produce knee rotations that matched the observed ones well (Halonen et al., 2016)

(better than when implementing 20% of the measured values (Adouni et al., 2012) – this was

also tested). The difference most likely arises from the use of patella and quadriceps forces in

contrast to the model without patella (such as in the model of Tanska et al. (2015)). In

addition, the resulting amplitudes of the moments were in accordance with experimental

studies of moments in the knee joint (Astephen et al., 2008).

Finite element model

The subject’s knee was imaged with an MR scanner (a clinical 3.0 T scanner, Philips) using

3D fast spin-echo sequence (VISTA) (in-plane resolution = 0.5 mm, slice thickness = 0.5

5
mm, TR = 1300 ms, TE = 32.3 ms). The knee geometry (Fig. 3a) was obtained by

segmentation in Mimics v.15.01 (Materialise). The geometry was imported into Abaqus v.

6.12-3 (Dassault Systèmes Simulia) and meshed. The femoral cartilage and menisci were

meshed with four element layers. By using the ‘subdivide’ option and TIE constraint, tibial

cartilage was meshed more densely at the superficial zone than at the deep zone in order to

improve convergence and accuracy of the implementation of the collagen fibril architecture.

The element type was set as C3D8P (Halonen et al., 2014) and the number of elements for

femoral, tibial and patellar cartilages and menisci was 8980, 20710, 738 and 9300,

respectively. Frictionless surface-to-surface contact was defined for all contacting surfaces.

The master surfaces were determined as a surface, whereas the slave surfaces were defined as

a node surface (CNS). The master-to-slave interactions were: femoral cartilage to tibial

cartilage and menisci, menisci to tibia, patellar cartilage to femoral cartilage. A convergence

test for the model was conducted in our previous work (Halonen et al., 2014).

Fibril-reinforced poroviscoelastic (FRPVE) material (Wilson et al., 2004; Wilson et al., 2005)

was defined for the cartilages and menisci. For cartilages, depth-dependent arcade-like

collagen fibril architecture with split-line patterns was implemented. The FRPVE material

model is well validated vs. experiments in previous studies (Wilson et al., 2004; Wilson et al.,

2005; Julkunen et al., 2007; Halonen et al., 2014), and it allows for instance the evaluation of

fibril strains. Find more details about the material model and its time-dependent response to

compressive and tensile loading from the Supplementary material. Only circumferential

collagen fibrils were implemented into the menisci. See our previous papers for more details

of the materials in the knee joint model (Halonen et al., 2013; Mononen et al., 2012). Split-

line architecture was implemented in the superior-inferior direction in the patellar cartilage

(Bae et al., 2007).

6
The ACL and posterior cruciate ligament (PCL) were segmented from the MR images, and

the exact insertion sites in femur and tibia were determined (Fig. 3b-d). Two experienced

orthopaedic surgeons were consulted in the segmentation as well as the insertion sites for the

single bundle and double bundle ACL reconstruction techniques. The healthy ACL consisted

of AM and PL bundles, while anterior, central, posterior-longitudinal and posterior-oblique

bundles were included in the PCL (Makris et al., 2000). In the single bundle and double

bundle ACL reconstructions, the graft was set as approximately 6-8 mm in diameter. The

exact locations of femoral and tibial graft insertion sites are presented in Fig. 3b-c.

The ligaments were assumed to be in tension at their normal length obtained from MRI, so

that a pre-strain of 5% was applied to the ACL and PCL and 4% to the LCL and MCL

(Gantoi et al., 2013). The pre-strains were implemented using the nonlinear spring response

option in Abaqus. Ligament stiffness values were k=201 Nmm-1 for ACL and k=258 Nmm-1

for PCL. For MCL and LCL, the stiffness values were k=114 Nmm-1 and k=134 Nmm-1,

respectively (Momersteeg et al., 1995). The ACL graft was modeled as a

gracilis/semitendinosus graft with a stiffness of k=715 Nmm-1 (Kim et al., 2003). A review

study by Herrington et al. (2005) showed no significant differences between the outcomes of

reconstructions using patellar tendon grafts and gracilis/semitendinosus grafts. Therefore only

the latter was simulated.

Linear springs without pre-strains were used to represent meniscal attachments and tendons.

The combined total spring constant of each meniscal horn attachment was 350 Nmm-1

(Villegas et al., 2007). The stiffness values used for quadriceps tendon and patellar tendon

were 475 Nmm-1 and 545 Nmm-1, respectively (Schatzmann et al., 1998). In order to omit the

compression resistance of linear springs, the lateral and medial patellofemoral ligaments

(LPFL and MPFL, respectively) were defined as linearly elastic truss elements with no

7
compressive stiffness. The elastic modulus was defined as 17 MPa for LPFL and 19 MPa for

MPFL (Atkinson et al., 2000). The Poisson’s ratio was set to 0.499.

Six models with varying ACL properties were created (see also Fig. 3):

1. Healthy ACL model with anatomically accurate insertion sites and AM and PL

bundles (k=201 Nmm-1, 5% pre-strain).

2. ACL rupture model, with ACL completely removed.

3. Single bundle ACL reconstruction model, consisting of a single bundle (k=715 Nmm-
1
, 5% pre-strain).

4. Double bundle ACL reconstruction model, consisting of AM and PL bundles (k=715

Nmm-1, 5% pre-strain).

5. Weakened (softer) ACL reconstruction model (k=201 Nmm-1 (change in stiffness= -

72%), 5% pre-strain), both single bundle and double bundle.

6. ACL reconstruction model with less pre-strain (k=715 Nmm-1, 2% pre-strain (change

in pre-strain= -3%-points)), both single bundle and double bundle.

Model verification and application

The gait simulation consisted of three steps:

1) In the first step, the femoral cartilage was brought into light contact with the patellar

and tibial cartilages as well as the menisci. The cartilage-bone interface of the tibial

cartilage was fixed.

2) In the second step, the femur was rotated from its position in the MR images (9°

flexion) to the initial position (5.73° flexion) at the beginning of the stance phase of

gait, observed in the motion analysis. Boundary conditions for translations and

internal-external and varus-valgus rotations were removed, and initial forces and

8
 moments (at the beginning of stance phase of gait) were implemented into the model

through the reference point, i.e. the midpoint between the epicondyles of femur,

similarly as before (Cappozzo et al., 1995; Mononen et al., 2013).

3) In the third step, the subject-specific translational forces (inferior-superior, medial-

lateral and anterior-posterior), moments (in the internal-external and valgus-varus

directions) and extension-flexion rotation were implemented (Fig. 2a-c). All femoral

forces, moments and rotations with respect to tibia were implemented through the

reference point (see step 2). The quadriceps force followed the direction of the distal-

proximal axis of femur during the gait.

To assess the effect of the ACL rupture as well as the single bundle and double bundle

reconstruction techniques, tibial rotations with respect to femur were determined from the

reference point described above as a function of the gait. Furthermore, reaction forces

through the joint as well as stresses, strains, contact pressures and fluid pressures of tibial

cartilage in all models were determined and compared with the healthy knee joint model.

Abaqus/standard with soils consolidation (*SOILS) analysis was used in all simulations. This

solution uses implicit time integration.

3. Results

Compared with the healthy ACL model (Fig. 4a), an increase of up to 542% in the anterior

translation of tibia relative to femur was observed in the model with the ACL rupture. The

change was most prominent during the heel strike and lowest between ipsilateral heelrise and

contralateral heelstrike. Lateral tibial translation was also increased in the model with the

ACL rupture (compared with the healthy joint, Fig. 4c), and an increase of up to 261% in the

9
internal rotation was observed (Fig. 4e). Single bundle and double bundle reconstruction

models showed slightly smaller or similar joint motion as simulated for the healthy knee joint

model. The best fit in rotations and translations was observed between the weakened (softer)

single bundle reconstruction model and the healthy knee joint model (Fig. 4). The weakened

double bundle reconstruction model showed a similar response in all observed parameters

with the weakened single bundle model (results not shown). Correspondingly, the double

bundle reconstruction model with reduced pre-strain also showed a similar response in all

observed parameters with the single bundle model with reduced pre-strain (results not

shown).

During the first peak force of gait, the ACL rupture shifted the contact pressure and therefore

maximum principal strain areas (peak strain 17.8%) in the posterior direction in both lateral

and medial tibial cartilage (Fig. 5b), compared with the healthy joint model (peak strain

13.8%) (Fig. 5a). The same behavior was observed in fibril strain, fluid pressure and

maximum principal stress (Figs. 6 and 7). The maximum difference between the normal,

healthy joint model and that with the ACL rupture was observed in maximum principal stress

(51%)

Similar maximum principal strains were seen in both single bundle and double bundle ACL

reconstruction models (peak values 14.9% and 14.8%, respectively) (Fig. 5c, d). Compared

with the healthy ACL model, the strains were slightly increased in both reconstruction

models, particularly in the anterior direction of the joint surface. The same behavior was

observed in other parameters (Figs. 6 and 7). The weakened single bundle model and the

single bundle model with less graft pre-strain produced similar maximum principal strains

(peak values 13.8% and 14.6%, respectively) (Fig. 5e, f). Those strains, as well as all the

other analyzed parameters were similar to those in the normal, healthy joint model (Fig. 5, 6

and 7). The maximum difference in parameters between the model with less pre-strain and

10
the healthy joint model was 18% in maximum principal stress. Compared with the healthy

joint model, the maximum difference in the model with the weakened graft was only 10%, in

maximum principal stress.

Same trend was observed in the quantitative analysis of average reaction forces (Fig. 8),

contact pressures, maximum principal strains, fibril strains, fluid pressures and maximum

principal stresses during the entire stance phase of gait (Fig. 9). Lowest values in all

parameters were observed in the ACL rupture model throughout the whole gait apart from the

contact pressure, which showed a small increase during the first 10% of the stance. The

maximum difference between the normal, healthy joint model and that with the ACL rupture

was observed in fibril strains and maximum principal stresses, which were both reduced by

up to 80% at 10% of the stance. Throughout the stance phase, the double bundle

reconstruction model produced stresses and strains very similar to those in the single bundle

model, which were elevated compared with the healthy ACL model with the maximum

difference in fluid pressure (79% at 10% of the stance). Apart from the very end of the

stance, the weakened (softer) single bundle reconstruction showed stresses and strains most

similar to those in the healthy ACL model, with a maximum difference in fibril strains (10%

at 5% of stance). The single bundle reconstruction model with reduced pre-strain slightly

increased contact pressures, strains and stresses during the first 50% of stance compared with

the healthy ACL model. The maximum difference was in contact pressure (37% at 5% of the

stance).

4. Discussion

We investigated the effect of ACL rupture and reconstruction techniques on the femoral

rotations and tibial cartilage response during the stance phase of gait. The healthy knee joint

11
model showed very similar translations and rotations as those reported in a bone-pin study by

Benoit et al., (2007) a dual fluoroscopy study by Kozanek et al. (2009). Ruptured ACL

caused a substantially increased tibial translation in the anterior direction, and a small shift in

the lateral direction. The internal tibial rotation was also increased due to the ACL rupture.

Although knee joint forces were reduced compared with the healthy knee joint, the ACL

rupture shifted the stress and strain areas in the tibial cartilage, especially in the medial tibial

plateau. The single bundle and double bundle ACL reconstruction models were able to

restore the translational and rotational motions of the knee, but they increased the forces,

stresses and strains experienced by the joint and cartilage. The weakened single and double

bundle reconstruction (softer graft) models and those with less graft pre-strain produced the

most similar rotations, translations, reaction forces, stresses and strains compared with the

normal, healthy knee joint model.

It is known that ACL rupture increases anterior tibial translation and internal tibial rotation

(Beynnon et al., 2002; Zhang et al., 2003), the first one being primary and the latter one

secondary restraint (Nigg and Herzog, 2007). This phenomenon was observed in our model

with the ACL rupture. The magnitude of translation in ACL-deficient knees was very similar

to those in the experimental studies (Kvist and Gillquist, 2001; Sonesson and Kvist, 2015).

However, a smaller increase in anterior tibial translation in ACL-deficient knees (compared

with healthy knees) has been reported than what was observed in this study (Kvist and

Gillquist, 2001; Sonesson and Kvist, 2015). The differences arise possibly due to different

testing protocols (e.g. only flexion in Kvist et al. (2001)). In addition, in those studies the

ACL-deficient knee is usually compared with the contralateral healthy knee, whereas in our

model it was the same knee that had its ACL removed. Despite this, the simulated increase in

anterior tibial translation in the ACL-deficient model was similar to that reported by Zhang et

al. (2003). In addition to these increased motions, the ACL rupture model showed slightly

12
increased lateral tibial translation. This might contribute to the increased peak stresses and

strains observed in the medial tibial plateau.

Even though the average stresses, strains, fluid pressures and contact pressures decreased due

to the ACL rupture, the tibiofemoral contact area shifted toward posterior direction. This

caused elevated values in all analyzed parameters in the posterior direction and decreased

values in the anterior direction. This could expose the posterior side of the medial tibial

plateau cartilage to an elevated risk of OA (DeFrate et al., 2006).

It has been suggested that the double bundle reconstruction is able to replicate the function of

the ACL better than the single bundle (Yagi et al., 2002), which mainly mimics the AM

bundle of a healthy ACL (Woo et al., 2002). In our study, both simulated techniques were

able to restore the rotational and translational joint movements close to the normal levels.

This was against our hypothesis, as we expected the double bundle method (mimicking the

AM and PL bundles of a healthy ACL) to perform better than the single bundle. However, the

results are in accordance with the results by Meredick et al. (2008). Further, the differences in

stresses and strains between the single bundle and double bundle techniques were found to be

minimal, which is in accordance with the hypothesis. Though, and in accordance with Yagi et

al. (2002), stresses, strains and fluid pressures experienced by cartilage were slightly lower in

the model with the double bundle than that with the single bundle.

The results suggest that in order to restore the forces within the joint and stresses and strains

of cartilage, the stiffness of the graft plays a crucial role. Throughout all analyses, the models

with modulated properties of the single bundle reconstruction (softer graft and less pre-strain)

produced the best match with the normal, healthy joint model. This is consistent with our

hypothesis. With the chosen material parameters, the graft stiffness rather than the graft pre-

strain had more influence on the restoration of the normal, healthy knee joint motion.

13
However, it should be noted that this conclusion might change with other choices of the

material parameters. Nonetheless, this result suggests that with optimal stiffness and pre-

strain of the graft, it should be possible to fully restore the normal, healthy joint motion and

mechanics. In order to prevent OA, it would be important to be able to plan the operation and

the properties of the graft in a patient-specific manner.

We also tested if the double bundle technique could achieve a better match with the healthy

ACL by creating a model with weakened double bundle graft. The model showed a similarly

good match with the normal, healthy knee joint model as the weakened single bundle model

(results not shown).

It has been shown that the graft loses some of its tension immediately after the operation,

resulting in increased laxity in the anterior-posterior direction (Beynnon et al., 2001). A

cadaveric study reported a loss of 46% of initial pretension at the end of 1500 extension-

flexion cycles (Arnold et al., 2005). The results of this study emphasize the importance of the

final tension the graft ends up with, as it has an effect on the strains and stresses the articular

cartilage undergoes during daily activities. It is beneficial for the function of the knee that the

graft is initially over-tensioned as it relaxes soon after the operation.

This study has a few limitations. Only one subject was included in the study. However, we

emphasize that this is a methodological study, in which the model is first validated and then

applied for further clinical studies with more subjects. Furthermore, these mechanisms we

present here should not depend on the joint geometry or other variables. Obviously, the exact

values of strains and stresses would vary from patient to patient. However, the conclusions

would not change (Räsänen et al., 2013). Location of the ACL reconstruction insertion sites

was determined by orthopaedic clinicians using only MR images, without a possibility to

confirm the real anatomic location of the insertion sites. In the future, the effect of the

14
variation in the locations of the insertion sites of the graft on knee joint motion and

mechanics could be an interesting topic (Nishizawa et al., 2015). It has been shown that

patients with a ruptured ACL exhibit reduced range of motion in flexion (DeVita et al.,

1998). Since this was a simulation study and we obviously did not have information of the

possible change in the subject’s knee motion due to ACL rupture, we decided to keep all

boundary and loading conditions the same in all models.

The results may help explain why the risk of post-traumatic OA is or is not increased in

ACL-deficient (or reconstructed) knees. The model shows that in ACL deficient knees,

despite the lack of tension by the ligament, local stresses and strains can be increased due to

changes in knee joint motion. On the other hand, the high stiffness of ACL graft increases the

stresses and strains in the tibial cartilage, therefore likely increasing the risk of post-traumatic

OA. The present study further suggests that, for the outcome of an ACL repair operation, the

stiffness and pre-strain of the reconstructed ACL, rather than the choice of the technique, play

a crucial role in the restoration of the knee joint function and cartilage responses.

Conflicts of interest

None.

Acknowledgements

Financial support from International Doctoral Programme in Biomedical Engineering and

Medical Physics (iBioMEP), Jenny & Antti Wihuri Foundation, Academy of Finland

(138574, 286526), Sigrid Juselius Foundation, Kuopio University Hospital (VTR, project

5041752), the strategic funding of the University of Eastern Finland and European Research

Council under the European Union’s Seventh Framework Programme (FP/2007-2013), ERC

15
Grant Agreement no. 281180 is acknowledged. CSC-IT Center for Science, Finland, is

acknowledged for computing resources.

References

Adouni, M., Shirazi-Adl, A., Shirazi, R., 2012. Computational biodynamics of human knee
joint in gait: From muscle forces to cartilage stresses. Journal of Biomechanics 45, 2149-
2156.
Arnold, M. P., Lie, D. T. T., Verdonschot, N., de Graaf, R., Amis, A. A., van Kampen, A.,
2005. The Remains of Anterior Cruciate Ligament Graft Tension After Cyclic Knee Motion.
The American Journal of Sports Medicine 33, 536-542.
Astephen, J. L., Deluzio, K. J., Caldwell, G. E., Dunbar, M. J., 2008. Biomechanical Changes
at the Hip, Knee, and Ankle Joints during Gait Are Associated with Knee Osteoarthritis
Severity. Journal of Orthopaedic Research 26 (3), 332-341.
Atkinson, P., Atkinson, T., Huang, C., Doane, R., 2000. A comparison of the mechanical and
dimensional properties of the human medial and lateral patellofemoral ligaments.
Transactions for 46th Annual Meeting, Orthopaedic Research Society 46, 776.
Bae, J., Park, K., Seon, J., Kwak, D., Jeon, I., Song, E., 2012. Biomechanical analysis of the
effects of medial meniscectomy on degenerative osteoarthritis. Medical and Biological
Engineering and Computing 50, 53-60.
Bae, W. C., Wong, V. W., Hwang, J., Antonacci, J. M., 2007. Wear-lines and split-lines of
human patellar cartilage: relation to tensile biomechanical properties. Osteoarthritis and
Cartilage 16, 841-845.
Barenius, B., Ponzer, S., Shalabi, A., Bujak, R., Norlén, L., Eriksson, K., 2014. Increased
Risk of Osteoarthritis After Anterior Cruciate Ligament Reconstruction: A 14-Year Follow-
up Study of a Randomized Controlled Trial. The American Journal of Sports Medicine 42
(5), 1049-1057.
Benoit, D. L., Ramsey, D. K., Lamontagne, M., Xu, L., Wretenberg, P., Renström, P., 2007.
In Vivo Knee Kinematics during Gait Reveals New Rotation Profiles and Smaller
Translations. Clinical Orthopaedics and Related Research 454, 81-88.
Benoit, D. L., Ramsey, D. K., Lamontagne, M., Xu, L., Wretenberg, P., Renström, P., 2006.
Effect of skin movement artifact on knee kinematics during gait and cutting motions
measured in vivo. Gait & Posture 24, 152-164.
Beynnon, B. D., Fleming, B. C., Labovitch, R., Parsons, B., 2002. Chronic anterior cruciate
ligament deficiency is associated with increased anterior translation of the tibia during
transition from non-weightbearing to weightbearing. Journal of Orthopaedic Research 20,
332-337.
Beynnon, B. D., Uh, B. S., Johnson, R. J., Fleming, B. C., Renström, P. A., Nichols, C. E.,
2001. The Elongation Behavior of the Anterior Cruciate Ligament Graft In Vivo: A Long-
Term Follow-Up Study. The American Journal of Sports Medicine 29, 161-166.

16
Brody, L. T., 2014. Knee osteoarthritis: Clinical connections to articular cartilage structure
and funcion. Physical Therapy in Sport 16 (4), 1-16.
Butler, D.L., Grood, E.S., Noyes, F.R. 1978. Biomechanics of ligaments and tendons. Exercis
Sport Science Reviews 6, 125-81.
Cappozzo, A., Catani, F., Della Croce, U., Leardini, A., 1995. Position and orientation in
space of bones during movement: anatomical frame definition and determination. Clinical
Biomechanics 10, 171-178.
Danso, E.K., Honkanen, J.T.J., Saarakkala, S., Korhonen, R.K. 2014. Comparison of
nonlinear mechanical properties of bovine articular cartilage and meniscus. Journal of
Biomechanics 47, 200-206.
DeFrate, L. E., Papannagari, R., Gill, T. J., Moses, J. M., Pathare, N. P., Li, G., 2006. The 6
Degrees of Freedom Kinematics of the Knee After Anterior Cruciate Ligament Deficiency:
An In Vivo Imaging Analysis. The American Journal of Sports Medicine 34, 1240-1246.
DeVita, P., Hortobagyi, T., Barrier, J., 1998. Gait biomechanics are not normal after anterior
cruciate ligament reconstruction and accelerated rehabilitation. Medicine & Science in Sports
& Exercise 30 (10), 1481-1488.
Frobell, R. B., Roos, H. P., Roos, E. M., Roemer, F. W., Ranstam, J., Lohmander, L. S.,
2013. Treatment for acute anterior cruciate ligament tear: five year outcome of randomised
trial. BMJ 346, 1442-12.
Gantoi, F. M., Brown, M. A., Shabana, A. A., 2013. Finite Element Modeling of the Contact
Geometry and Deformation in Biomechanics Applications. Journal of Computational and
Nonlinear Dynamics 8, 1-11.
Halonen, K. S., Mononen, M. E., Jurvelin, J. S., Töyräs, J., Klodowski, A., Kulmala, J.-P.,
Korhonen, R. K., 2016. Importance of Patella, Quadriceps Forces and Depth-Wise Cartilage
Structure on Knee Joint Motion and Cartilage Response During Gait. Journal of
Biomechanical Engineering (in press), XX-XX.
Halonen, K. S., Mononen, M. E., Jurvelin, J. S., Töyräs, J., Korhonen, R. K., 2013.
Importance of depth-wise distribution of collagen and proteoglycans in articular cartilage—A
3D finite element study of stresses and strains in human knee joint. Journal of Biomechanics
46, 1184-1192.
Halonen, K. S., Mononen, M. E., Jurvelin, J. S., Töyräs, J., Salo, J., Korhonen, R. K., 2014.
Deformation of articular cartilage during static loading of a knee joint – Experimental and
finite element analysis. Journal of Biomechanics 47, 2467-2474.
Herrington, L., Wrapson, C., Matthews, M., Matthews, H., 2005. Anterior Cruciate Ligament
reconstruction, hamstring versus bone–patella tendon–bone grafts: a systematic literature
review of outcome from surgery. The Knee 12, 41-50.
Hosseini, S. M., Wilson, W., Ito, K., Donkelaar, v. C. C., 2014. A numerical model to study
mechanically induced initiation and progression of damage in articular cartilage.
Osteoarthritis and Cartilage 22, 95-103.

17
Hosseini, A., Van de Velde, S. K., Kozanek, M., Gill, T. J., Grodzinsky, A. J., Rubash, H. E.,
Li, G., 2010. In-vivo time-dependent articular cartilage contact behavior of the tibiofemoral
joint. Osteoarthritis and Cartilage 18, 909-916.
Huang, R., Zheng, H., Xu, Q., 2012. Biomechanical evaluation of different techniques in
double bundle anterior cruciate ligament reconstruction using finite element analysis. Journal
of Biomimetics, Biomaterials and Tissue Engineering 13, 55-68.
Julkunen, P., Kiviranta, P., Wilson, W., Jurvelin, J.S., Korhonen, R.K. 2007. Characterization
of articular cartilage by combining microscopic analysis with a fibril-reinforced finite-
element model. J. Biomech. 40, 1862-70.
Kabada, M. P., Ramakrishnan, H. K., Wootten, M. E., 1990. Measurement of Lower
Extremity Kinematics During Level Walking. Journal of Orthopaedic Research 8, 383-392.
Kessler, M. A., Behrend, H., Henz, S., Stutz, G., Rukavina, A., Kuster, M. S., 2008.
Function, osteoarthritis and activity after ACL-rupture: 11 years follow-up results of
conservative versus reconstructive treatment. Knee Surgery, Sports Traumatology,
Arthroscopy 16 (5), 442-448.
Kim, D. H., Wilson, D. R., Hecker, A. T., Jung, T. M., Brown, C. H., 2003. Twisting and
Braiding Reduces the Tensile Strength and Stiffness of Human Hamstring Tendon Grafts
Used for Anterior Cruciate Ligament Reconstruction. The American Journal of Sports
Medicine 31 (6), 861-867.
Kim, H. S., Seon, J. K., Jo, A. R., 2013. Current Trends in Anterior Cruciate Ligament
Reconstruction. Knee Surgery & Related Research 25 (4), 165-173.
Klodowski, A., Mononen, M. E., Kulmala, J. -., Valkeapää, A., Korhonen, R. K., Avela, J.,
Kiviranta, I., Jurvelin, J. S., Mikkola, A., 2015. Merge of motion analysis, multibody
dynamics and finite element method for the subject-specific analysis of cartilage loading
patterns during gait: Differences between rotation and moment-driven models of human knee
joint. Multibody System Dynamics July 29 2015. pp. 1-20. DOI10.1007/s11044-015-9470-y.
Kozanek, M., Hosseini, A., Liu, F., Van de Velde, S. K., Gill, T. J., Rubash, H. E., Li, G.,
2009. Tibiofemoral kinematics and condylar motion during the stance phase of gait. Journal
of Biomechanics 42, 1877-1884.
Kvist, J., Gillquist, J., 2001. Sagittal Plane Knee Translation and Electromyographic Activity
During Closed and Open Kinetic Chain Exercises in Anterior Cruciate Ligament-Deficient
Patients and Control Subjects. The American Journal of Sports Medicine 29 (1), 72-82.
Makris, C. A., Georgoulis, A. D., Papageorgiou, C. D., Moevius, U. G., Soucacos, P. N.,
2000. Posterior Cruciate Ligament Architecture: Evaluation Under Microsurgical Dissection.
Arthroscopy: The Journal of Arthroscopic & Related Surgery 16 (6), 627-632.
Meredick, R. B., Vance, K. J., Appleby, D., Lubowitz, J. H., 2008. Outcome of Single-
Bundle Versus Double-Bundle Reconstruction of the Anterior Cruciate Ligament. The
American Journal of Sports Medicine 36 (7), 1414-1421.
Moglo, K. E., Shirazi-Adl, A., 2003. Biomechanics of passive knee joint in drawer: load
transmission in intact and ACL-deficient joints. The Knee 10, 265-276.

18
Momersteeg, T. J. A., Blankevoort, L., Huiskes, R., Kooloos, J. G. M., Kauer, J. M. G.,
Hendriks, J. C. M., 1995. The effect of variable relative insertion orientation of human knee
bone-ligament-bone complexes on the tensile stiffness. Journal of Biomechanics 28, 745-752.
Mononen, M. E., Jurvelin, J. S., Korhonen, R. K., 2013. Effects of Radial Tears and Partial
Meniscectomy of Lateral Meniscus on the Knee Joint Mechanics during the Stance Phase of
the Gait Cycle - A 3D Finite Element Study. Journal of Orthopaedic Research 31 (8), 1208-
1217.
Mononen, M. E., Mikkola, M. T., Julkunen, P., Ojala, R., Nieminen, M. T., Jurvelin, J. S.,
Korhonen, R. K., 2012. Effect of superficial collagen patterns and fibrillation of femoral
articular cartilage on knee joint mechanics—A 3D finite element analysis. Journal of
Biomechanics 45, 579-587.
Mononen, M.E., Tanska, P., Isaksson, H., Korhonen, R.K., 2016. A Novel Method to
Simulate the Progression of Collagen Degeneration of Cartilage in the Knee: Data from the
Osteoarthritis Initiative. Scientific Reports 6, Feb, 1-14.
Nigg, B. M., Herzog, W., 2007. Biomechanics of the musculo-skeletal system. Third edition.
John Wiley & Sons, UK, pp. 130.
Nishizawa, Y., Thorhauer, E., Irrgang, J. J., Fu, F. H., Tashman, S., 2015. Relationships
Between Femoral Tunnel Position and Dynamic Knee Function After ACL Reconstruction.
In Orthopaedic Research Society 2015. US, Las Vegas, Nevada.
Peña, E., Calvo, B., Martínez, M. A., Doblaré, M., 2006. A three-dimensional finite element
analysis of the combined behavior of ligaments and menisci in the healthy human knee joint.
Journal of Biomechanics 39, 1686-1701.
Peña, E., Martínez, M. A., Calvo, B., Palanca, D., Doblaré, M., 2005. A finite element
simulation of the effect of graft stiffness and graft tensioning in ACL reconstruction. Clinical
Biomechanics 20, 636-644.
Räsänen, L. P., Mononen, M. E., Nieminen, M. T., Lammentausta, E., Jurvelin, J. S.,
Korhonen, R. K., OAI Investigators, 2013. Implementation of subject-specific collagen
architecture of cartilage into a 2D computational model of a knee joint - data from the
osteoarthritis initiative (OAI). Journal of Orthopaedic Research 31, 10-22.
Reinschmidt, C., van den Bogert, A. J., Nigg, B. M., Lundberg, A., Murphy, N., 1997. Effect
of skin movement on the analysis of skeletal knee joint motion during running. Journal of
Biomechanics 30 (7), 729-732.
Schatzmann, L., Brunner, P., Stäubli, H. U., 1998. Effect of cyclic preconditioning on the
tensile properties of human quadriceps tendons and patelar ligaments. Knee Surgery, Sports
Traumatology, Arthroscopy 6, 56-61.
Shirazi, R., Shirazi-Adl, A., Hurtig, M., 2008. Role of cartilage collagen fibrils networks in
knee joint biomechanics under compression. Journal of Biomechanics 41, 3340-3348.
Sonesson, S., Kvist, J., 2015. Dynamic and static tibial translation in patients with anterior
cruciate ligament deficiency initially treated with a structured rehabilitation protocol. Knee
Surgery, Sports Traumatology, Arthroscopy 1, 1-10.

19
Tanska, P., Mononen, M. E., Korhonen, R. K., 2015. A multi-scale finite element model for
investigation of chondrocyte mechanics in normal and medial meniscectomy human knee
joint during walking. Journal of Biomechanics 48 (8), 1397-1406.
Tsoukas, D., Fotopoulos, V., Basdekis, G., Makridis, K. G., 2015. No difference in
osteoarthritis after surgical and non-surgical treatment of ACL-injured knees after 10 years.
Knee Surgery, Sports Traumatology, Arthroscopy (in print).
Villegas, D. F., Maes, J. A., Magee, S. D., Haut Donahue, T. L., 2007. Failure properties and
strain distribution analysis of meniscal attachments. Journal of Biomechanics 40, 2655-2662.
Wang, J.H.-C. Review mechanobiology of tendon. 2006. Journal of Biomechanics 39, 1563-
82.
Wilson, W., van Donkelaar, C. C., van Rietbergen, B., Ito, K., Huiskes, R., 2004. Stresses in
the local collagen network of articular cartilage: a poroviscoelastic fibril-reinforced finite
element study. Journal of Biomechanics 37, 357-366.
Wilson, W., van Donkelaar, C. C., van Rietbergen, B., Ito, K., Huiskes, R., 2005. Erratum to
“Stresses in the local collagen network of articular cartilage: a poroviscoelastic fibril-
reinforced finite element study” [Journal of Biomechanics 37 (2004) 357–366] and “A fibril-
reinforced poroviscoelastic swelling model for articular cartilage” [Journal of Biomechanics
38 (2005) 1195–1204]. Journal of Biomechanics 38, 2138-2140.
Woo, S. L., Kanamori, A., Zeminski, J., Yagi, M., Papageorgiou, C., Fu, F. H., 2002. The
Effectiveness of Reconstruction of the Anterior Cruciate Ligament with Hamstrings and
Patellar Tendon. The Journal of Bone & Joint Surgery 84, 907-914.
Yagi, M., Wong, E. K., Kanamori, A., Debski, R. E., Fu, F. H., Woo, S. L., 2002.
Biomechanical Analysis of an Anatomic Anterior Cruciate Ligament Reconstruction. The
American Journal of Sports Medicine 30, 660-666.
Yao, J., Snibbe, J., Maloney, M., Lerner, A. L., 2005. Stresses and Strains in the Medial
Meniscus of an ACL Deficient Knee under Anterior Loading: A Finite Element Analysis
with Image-Based Experimental Validation. Journal of Biomechanical Engineering 128, 135-
141.
Zhang, L. -., Shiavi, R. G., Limbird, T. J., Minorik, J. M., 2003. Six degrees-of-freedom
kinematics of ACL deficient knees during locomotion - compensatory mechanism. Gait and
Posture 17, 34-42.

Figure legends

Fig 1. Study workflow.

Fig 2. Gait input for the FE model. (a) Internal-external and varus-valgus moments as well as

extension-flexion rotation. The moments are scaled to 50% of the ones obtained from gait

analysis (see text for details). (b) Total quadriceps force and its components. (c) Translational

20
forces. The dashed lines represent ipsilateral heelrise and contralateral heelstrike,

respectively.

Fig 3. (a) Finite element model, sagittal view. (b) Axial view of the tibial attachments of

healthy ACL and PCL (red and blue areas, respectively) as well as the tibial insertions of

simulated single bundle and double bundle ACL reconstruction techniques. (c) Sagittal view

of the femoral attachment of a healthy ACL (red area) and the insertions of simulated single

bundle and double bundle ACL reconstruction techniques. (d) Reference point (see text) and

the attachment sites of a healthy PCL (blue area).

Fig 4. Tibial translations (a)-(c) and rotations (d)-(e) with respect to femur during the stance

phase of gait. (a) Posterior-anterior translation. (b) Inferior-superior translation. (c) Medial-

lateral translation. (d) Valgus-varus rotation. (e) External-internal rotation. (f) Flexion-

extension rotation. Extension-flexion rotations are identical in all models because they were

in the model input as such (see Materials and methods).

Fig 5. Maximum principal strains in tibial cartilage surface during the first peak force of gait

(20% of stance). (a) Healthy ACL. (b) ACL rupture. (c) Single bundle ACL reconstruction.

(d) Double bundle ACL reconstruction. (e) Weakened single bundle ACL reconstruction. (e)

Single bundle ACL reconstruction with less prestrain.

Fig 6. Differences in maximum principal strains (a)-(e) and fibril strains (f)-(j) in the tibial

cartilage surface. Values of the healthy joint model are subtracted from the values of other

models (differences are in %-points).

Fig 7. Differences in fluid pressures (a)-(e) and maximum principal stresses (f)-(j) in the

tibial cartilage surface. Values of the healthy joint model are subtracted from the values of

other models.

21
Fig 8. Tibial reaction forces through the joint surfaces during the stance phase of gait. (a)

Lateral tibial plateau. (b) Medial tibial plateau.

Fig 9. Average values of the analyzed parameters in the contact area of the medial tibial

cartilage surface during the stance phase of gait. (a) Contact pressures, (b) maximum

principal strains, (c) fibril strains (d) fluid pressures and (e) maximum principal stresses.

22
Figure 1
Figure 2 REVISED
Figure 3 REVISED
Figure 4 REVISED
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9