Journal of

Clinical Medicine

Article
Biomechanical Difference between Conventional Transtibial
Single-Bundle and Anatomical Transportal Double-Bundle
Anterior Cruciate Ligament Reconstruction Using
Three-Dimensional Finite Element Model Analysis
Jae Gyoon Kim 1,† , Kyoung Tak Kang 2,† and Joon Ho Wang 3, *

1 Department of Orthopedic Surgery, Ansan Hospital, Korea University College of Medicine, Ansan-si 15355,
Gyeonggi-do, Korea; [email protected]
2 Department of Mechanical Engineering, Yonsei University, Seoul 03722, Korea; [email protected]
3 Department of Orthopedic Surgery, Samsung Medical Center, Sungkyunkwan University School of Medicine,
Seoul 06351, Korea
* Correspondence: [email protected]; Tel.: +82-2-3410-3507; Fax: +82-2-3410-0061
† Equal contribution.

Abstract: The purpose of our study was to analyze the graft contact stress at the tunnel after
transtibial single-bundle (SB) and transportal double-bundle (DB) anterior cruciate ligament (ACL)



reconstruction. After transtibial SB (20 cases) and transportal DB (29 cases) ACL reconstruction,


the three-dimensional image of each patient made by postoperative computed tomography was
Citation: Kim, J.G.; Kang, K.T.; adjusted to the validation model of a normal knee and simulated SB and DB ACL reconstructions
Wang, J.H. Biomechanical Difference
were created based on the average tunnel position and direction of each group. We also measured
between Conventional Transtibial
graft and contact stresses at the tunnel after a 134 N anterior load from 0◦ to 90◦ flexion. The graft
Single-Bundle and Anatomical
and contact stresses became the greatest at 30◦ and 0◦ flexion, respectively. The total graft and contact
Transportal Double-Bundle Anterior
stresses after DB ACL reconstruction were greater than those after SB ACL reconstruction from 0◦
Cruciate Ligament Reconstruction
Using Three-Dimensional Finite
to 30◦ and 0◦ to 90◦ knee flexion, respectively. However, the graft and contact stresses of each graft
Element Model Analysis. J. Clin. Med. after DB ACL reconstruction were less than those after SB ACL reconstruction. In conclusion, the
2021, 10, 1625. https://doi.org/ total graft and total contact stresses after DB ACL reconstruction are higher than those after SB ACL
10.3390/jcm10081625 reconstruction from 0◦ to 30◦ and 0◦ to 90◦ knee flexion, respectively. However, the stresses of each
graft after DB ACL reconstruction are about half of those after SB ACL reconstruction.
Academic Editors: Yong Seuk Lee
and Enrique Gómez-Barrena Keywords: knee; anterior cruciate ligament; double bundle; single bundle; finite element model;
graft stress; contact stress
Received: 5 March 2021
Accepted: 7 April 2021
Published: 12 April 2021

1. Introduction
Publisher’s Note: MDPI stays neutral
Recently, placing a graft within the anterior cruciate ligament (ACL) footprint has been
with regard to jurisdictional claims in
published maps and institutional affil-
emphasized in anatomical ACL reconstruction [1]. Anatomical ACL reconstruction means
iations.
that tunnels are made in the femoral and tibial ACL footprints regardless of the number of
bundles [2]. In conventional single-bundle (SB) ACL reconstruction, the tunnels are made
in the posterolateral (PL) tibial footprint and the anteromedial (AM) femoral footprint,
resulting in a non-anatomical and more vertical direction than the native ACL, which cannot
restore rotatory laxity [3,4]. Anatomical double-bundle (DB) ACL reconstruction shows
Copyright: © 2021 by the authors.
superior biomechanical results, including both anterior and rotatory stability, compared to
Licensee MDPI, Basel, Switzerland.
conventional non-anatomical SB ACL reconstruction [5–8].
This article is an open access article
distributed under the terms and
Better positioning of a femoral tunnel anatomically would be accomplished using a
conditions of the Creative Commons
femoral tunnel drilling technique independently of the tibial tunnel [9]. The necessity to
Attribution (CC BY) license (https:// make a femoral tunnel independently of a tibial tunnel has drawn interest in independent
creativecommons.org/licenses/by/ techniques such as a transportal (TP) and an outside-in technique [9,10]. We assumed that
4.0/). these changes in the technique of performing anatomical ACL reconstruction would also

J. Clin. Med. 2021, 10, 1625. https://doi.org/10.3390/jcm10081625 https://www.mdpi.com/journal/jcm

Med. 2021, 10, x FOR PEER REVIEW 2 of 13

J. Clin. Med. 2021, 10, 1625 2 of 13

techniques such as a transportal (TP) and an outside-in technique [9,10]. We assumed that
these changes in the technique of performing anatomical ACL reconstruction would also
change the femoral tunnel geometry and the stress patterns of both graft and tunnel. This
changepattern
change in the stress the femoral
wouldtunnel
affect geometry
longer-termand the stressclinical
follow-up patterns of both
results graft the
despite and tunnel. This
changetunnel
anatomical femoral in theposition.
stress pattern would affect
In a previous study,longer-term
the femoral follow-up clinicalangle
graft bending results despite the
was defined as the angle between the femoral tunnel axis and the graft, and this angle wasbending angle
anatomical femoral tunnel position. In a previous study, the femoral graft
was defined
compared between as the angle
conventional SB ACLbetween the femoralusing
reconstruction tunnel axis
the and the graft,
transtibial and this angle was
(TT) tech-
compared between conventional SB ACL reconstruction using the transtibial (TT) technique
nique and anatomical DB ACL reconstruction using the TP technique. The authors found
and anatomical DB ACL reconstruction using the TP technique. The authors found that
that anatomical DB ACL reconstruction showed a more acute angle than conventional SB
anatomical DB ACL reconstruction showed a more acute angle than conventional SB ACL
ACL reconstruction in an extended knee position, which might have increased the stress
reconstruction in an extended knee position, which might have increased the stress in the
in the graft at the femoral tunnel opening [11].
graft at the femoral tunnel opening [11].
Many studies have evaluated the biomechanics of the reconstructed ACL using ca-
Many studies have evaluated the biomechanics of the reconstructed ACL using ca-
daver or finite element model (FEM) analysis [12–18]. Some studies have compared knee
daver or finite element model (FEM) analysis [12–18]. Some studies have compared knee
kinematics and biomechanics between SB and DB ACL reconstructions [14,17]. As far as
kinematics and biomechanics between SB and DB ACL reconstructions [14,17]. As far
we are aware, few studies have evaluated and compared the graft stress and contact stress
as we are aware, few studies have evaluated and compared the graft stress and contact
at the tunnel after conventional SB ACL reconstruction and anatomical DB ACL recon-
stress at the tunnel after conventional SB ACL reconstruction and anatomical DB ACL
struction using FEM analysis on the basis of a surgical simulation model made using com-
reconstruction using FEM analysis on the basis of a surgical simulation model made using
puted tomography (CT) scans
computed of patients.
tomography (CT)This type
scans of of model can
patients. Thisrepresent the real
type of model tunnel
can represent the real
position and direction after ACL reconstructions using both surgical techniques. The
tunnel position and direction after ACL reconstructions using both surgical techniques. pur-
pose of this FEM
Thestudy
purposewasof to this
analyze
FEMthe graftwas
study stress
to and contact
analyze thestress
graft at the tunnel
stress after stress at the
and contact
conventional SB ACL reconstruction and anatomical DB ACL reconstruction at different
tunnel after conventional SB ACL reconstruction and anatomical DB ACL reconstruction at
angles of kneedifferent
flexion. We
angleshypothesized that the
of knee flexion. Wegraft stress andthat
hypothesized contact stressstress
the graft after and
ana-contact stress
tomical DB ACL reconstruction using the TP technique would be greater than those
after anatomical DB ACL reconstruction using the TP technique would be greater than after
conventional SB ACL
those reconstruction
after conventionalusing the reconstruction
SB ACL TT technique. using the TT technique.

2. Materials and Methods and Methods

2. Materials
2.1. Intact Model Intact Model
2.1.
A three-dimensional
A three-dimensional (3D) FEM of the(3D) FEMextremity
lower of the lower
wasextremity
developed was developed
based on com-based on com-
puted tomography (CT) images. A light speed volume
puted tomography (CT) images. A light speed volume computed tomography (VCT, GEcomputed tomography (VCT, GE
Medical Systems, Milwaukee, WI, USA) scanner was used.
Medical Systems, Milwaukee, WI, USA) scanner was used. CT images of a 0.1 mm slice CT images of a 0.1 mm slice
from male
from a 34-year-old a 34-year-old male subject
subject (height 178 cm,(height
weight 178 cm,were
75 kg) weight 75 kg)The
obtained. were
3Dobtained.
sur- The 3D
surface
face of the femur, offibula,
tibia, the femur, tibia, fibula,
and patella at fulland patellawas
extension at full extension
generated by was
Mimicsgenerated
soft- by Mimics
software
ware (Materialise (Materialise
Inc., Leuven, Inc., Leuven,
Belgium) Belgium)
using digital using(Figure
CT data digital1)CT data (Figure 1) [19].
[19].

Figure 1. A three-dimensional
Figure 1. A three-dimensional surface reconstruction
surface reconstruction of the femurofand
the tibia
femur ofand tibia of a 34-year-old
a 34-year-old male
male subject, using Mimics
subject,
(Materialise Inc., usingBelgium).
Leuven, Mimics (Materialise Inc., Leuven, Belgium).

Based on magnetic
Basedresonance
on magneticimaging (MRI),imaging
resonance the femoral cartilage,
(MRI), both menisci,
the femoral pa- both menisci,
cartilage,
tellar tendon, patellar
and four ligaments
tendon, and (anterior cruciate,
four ligaments posterior
(anterior cruciate,
cruciate, medial
posterior collateral,
cruciate, medial collateral,
and lateral collateral ligaments)
and lateral were
collateral segmented
ligaments) weremanually
segmented in 3D reconstruction
manually models.
in 3D reconstruction models.
This segmentation was accurate was
This segmentation to 0.1accurate
mm. Theto Initial
0.1 mm. Graphics Exchange
The Initial Specification
Graphics Exchange Specification
(IGES) files exported from
(IGES) files Mimicsfrom
exported were loadedwere
Mimics intoloaded
Unigraphics (UG) NX (UG)
into Unigraphics 7.0 (Siemens
NX 7.0 (Siemens PLM
Software, Torrance, CA, USA) to make solid models for each femur, tibia, fibula, patella,
in. Med. 2021, 10, x FOR PEER REVIEW 3 of 13
J. Clin. Med. 2021, 10, 1625 3 of 13

PLM Software, Torrance, CA, USA) to make solid models for each femur, tibia, fibula,
patella, and and soft-tissue
soft-tissue segment,
segment, which
which werewere loaded
loaded into into Hypermesh
Hypermesh 8.0 (Altair
8.0 (Altair Engineering, Inc.,
Engineer-
Troy, MI, USA) to make the FE mesh (Figure
ing, Inc., Troy, MI, USA) to make the FE mesh (Figure 2). 2).

Figure 2. (A) AFigure 2. (A) A three-dimensional

three-dimensional soft-tissuebased
soft-tissue reconstruction reconstruction
on magneticbased on magnetic
resonance resonance
imaging imagingcartilage,
(MRI); articular
(MRI); tendon,
both menisci, patellar articularand
cartilage, both menisci,
four ligaments patellar
(anterior tendon,
cruciate, and four
posterior ligaments
cruciate, medial(anterior
collateral,cruciate, pos-collateral
and lateral
ligaments). (B)terior cruciate,
Generation medial
of the finitecollateral, and lateral
element model (FEM)collateral ligaments).
using Hypermesh 8.0(B) Generation
(Altair of theInc.,
Engineering, finiteTroy,
ele-MI, USA).
ment model (FEM) using Hypermesh 8.0 (Altair Engineering, Inc., Troy, MI, USA).
The FE mesh was analyzed using ABAQUS 6.6-1 (Hibbitt, Karlsson and Sorenson,
The FEInc.,
mesh was analyzed
Providence, using The
RI, USA). ABAQUS
cortical6.6-1
bone,(Hibbitt, Karlsson
cancellous bone, and
and Sorenson,
intramedullary canal
Inc., Providence, RI, USA).in
were included The
thecortical bone, cancellous
bone model. bone, were
The bone parts and intramedullary
assumed to be canal
rigid contrary to
were includedsoftin the bone
tissues model.toThe
according bone parts
a previous study were
[20].assumed
Therefore, toabe rigid contrary
primary to
node positioned at the
soft tissues center
according to a previous study [20]. Therefore, a primary node positioned
of rotation at full extension represented each bony structure. The FE models of the at
the center ofsoft
rotation
tissueat full extension
comprised represented
the meniscus and each bony structure.
four ligaments. The FE models
The meniscus of
and cartilage models
the soft tissue
werecomprised the meniscus
also developed based onand four ligaments.
a previous TheThe
study [20]. meniscus and between
attachment cartilagethe cartilage
models wereand also developed
bones based on
was assumed to abeprevious
completely study [20]. The
bonded. [19]attachment
The anteriorbetween the
and posterior horns of
cartilage andboth
bones was assumed
menisci to be completely
were attached to the tibia bonded.
plateau, [19]
and The anteriormeniscus
the medial and posterior
was additionally
horns of both meniscitowere
attached attached
the joint capsuleto the
alongtibiathe
plateau, and the
outer edge, medial meniscus
following Pena et al.was[20]. The liga-
additionallyments
attached to the
model wasjoint capsule
assumed to along the outer edge,
be a hyperelastic, following
rubber-like Pena etwhich
material, al. [20].
represents the
The ligaments model was
nonlinear assumedrelations
stress–strain to be a hyperelastic, rubber-like
[21,22]. A strain material, function
energy potential which rep- characterized
resents the the
nonlinear stress–strain
model [21], relations [21,22].
and the polynomial form of A the
strain energy
strain energypotential
potentialfunction
was selected from
characterizedthethe
ABAQUS
model [21],material library
and the [23]. Soft
polynomial tissues
form arestrain
of the normally exposed
energy to in
potential wasvivo residual
selected fromstresses that theymaterial
the ABAQUS undergo.library
The initial
[23]. strain modelare
Soft tissues of ligaments was also to
normally exposed made
in according
vivo residualto stresses
a previous thatstudy
they[20].
undergo. The initial strain model of ligaments was also
made according to a previous study [20].
2.2. Surgical Procedure
The patients were classified into a conventional TT SB group (20 patients) and an
2.2. Surgical Procedure
The patients were TP
anatomical DB group
classified (29a patients).
into conventionalWe divided the patients
TT SB group according
(20 patients) andto an
the time lapse
anatomical TP DB group (29 patients). We divided the patients according to the time lapse had up to
from the injury to the reconstruction. Group 1 consisted of 20 patients who
a 6-month
from the injury interval between
to the reconstruction. the injury
Group and reconstruction
1 consisted and had
of 20 patients who underwent
up to a 6-conventional
SB ACL reconstruction using the TT technique, because in
month interval between the injury and reconstruction and underwent conventional SB this case, we performed the
remnant preservation technique and the quality of remnant tissue
ACL reconstruction using the TT technique, because in this case, we performed the rem- would be related with
this interval. Group 2 consisted of 29 patients with more than a 6-month interval between
nant preservation technique and the quality of remnant tissue would be related with this
the injury and the reconstruction and underwent anatomical DB ACL reconstruction using
interval. Group 2 consisted of 29 patients with more than a 6-month interval between the
the TP technique with non-remnant preservation.
injury and the reconstruction and underwent anatomical DB ACL reconstruction using
After the usual portal formation and arthroscopic examination, an accessory antero-
the TP technique with non-remnant preservation.
medial (AAM) portal was made. The graft was made using the hamstring tendon. For SB
After the usual portal formation and arthroscopic examination, an accessory antero-
reconstruction, four stranded grafts of semitendinosus and gracilis were made, and for
medial (AAM) portal was made. The graft was made using the hamstring tendon. For SB
DB reconstruction, six stranded grafts (triple-stranded semitendinosus for AM bundles
reconstruction, four stranded grafts of semitendinosus and gracilis were made, and for
and triple-stranded gracilis for PL bundles) were created. We harvested and used both
DB reconstruction, six stranded grafts (triple-stranded semitendinosus for AM bundles
semitendinosus and gracilis in all cases. The mean graft diameter of conventional SB ACL
and triple-stranded gracilis for PL bundles) were created. We harvested and used both
reconstruction was 8.1 ± 0.7 mm (range from 7 to 9 mm), and that of anatomical DB ACL
semitendinosus and gracilis
reconstruction in 7.3
was all cases. The (range
± 0.8 mm mean graft
7 to 8diameter of conventional
mm) for the AM graft andSB ACL
from 5.5 ± 0.6 mm
reconstruction was 8.1 ± 0.7 mm (range from
(range 5 to 6 mm) for the PL graft. 7 to 9 mm), and that of anatomical DB ACL
J. Clin. Med. 2021, 10, 1625 4 of 13

2.2.1. Conventional SB Reconstruction Using the TT Technique

To make a tibial tunnel, we used an ACL tibial guide (Linvatec, Largo, FL, USA), and
the tibial guide tip was positioned on the ACL tibial footprint (PL bundle center) at the
point on the extended line from the lateral meniscus anterior horn. The femoral tunnel
was reamed near the AM femoral footprint at the level of the 10:30 (right knee) and 1:30
(left knee) positions, approximately 1–2 mm anterior to the posterior cortex of the femur,
through the tibial tunnel previously reamed.

2.2.2. Anatomical DB Reconstruction Using the TP Technique

The femoral footprints of both bundles were determined using ACL remnants and
bony landmarks [23] and marked using a Steadman awl (ConMed (Linvatec), Largo, FL,
USA). A 2.4 mm guide pin was advanced through the Bullseye® femoral guide (ConMed
(Linvatec), Largo, FL, USA) passed through the AAM portal, with the tip aimed at the
center of the femoral footprints of both bundles, previously determined. The AM bundle
footprint was determined 5 to 6 mm anterior to the posterior cartilage margin and 3 to 4 mm
inferior to the posterolateral corner of the intercondylar notch in 90◦ flexion, and the center
of the PL bundle was determined 5 mm superior to the margin of the articular cartilage
on an imaginary line perpendicular to the tangent line of the lateral femoral condyle in
90◦ flexion. The anatomical center of the tibial footprint was determined using the ACL
remnant. The center of the AM bundle was determined at a point posterior to the anterior
ridge of ACL tibial footprint, and the center of the PL bundle was determined at a point
posterior to the AM bundle footprint. A Sentinel cannulated reamer (ConMed (Linvatec),
Largo, FL, USA) and a 4.5 mm EndoButton drill bit (Smith & Nephew Endoscopy, Andover,
MA, USA) were then used to make femoral tunnels. Next, a tibial tunnel was reamed using
an ACL tibial guide (Linvatec, Largo, FL, USA), and the guide tip was positioned at the
center of the tibial footprints of both bundles.

2.3. Surgical Simulation Model

Three days after ACL reconstruction, CT scans were taken of all knees (49 patients)
with the patients’ consent. A light speed VCT scanner (GE Medical Systems, Milwaukee,
WI, USA) was used in all cases. The CT scans were taken in full extension. The collimation
and tube parameters were 16 × 0.625 mm and 120 kVp/200 mA, respectively. The acqui-
sition matrix and the field of view were 512 × 512 and 140 mm with a slice thickness of
0.625 mm, respectively. CT images of each group were used for 3D reconstruction. This 3D
reconstruction model of each technique group was adjusted to the validation model. The
average position of the AM and PL femoral tunnel in a parallel direction to the Blumensaat
line was 23.7% ± 5.4% and 34.7% ± 6.3%, respectively, after anatomical DB ACL recon-
struction. The average position of the AM and PL femoral tunnel in a vertical direction to it
was 20.5% ± 6.1% and 50.1% ± 6.9%, respectively, after anatomical DB ACL reconstruction.
The average position of the femoral tunnel in a parallel direction to it was 33.3% ± 3.4%
and in a vertical direction to it was 40.8% ± 6.3% after conventional SB ACL reconstruction.
The average AM and PL femoral graft bending angles were 111.5◦ ± 8.8◦ and 118.9◦ ± 9.8◦ ,
respectively, after anatomical DB ACL reconstruction. The mean femoral graft bending
angle after conventional SB ACL reconstruction was 125.3◦ ± 11.1◦ [11]. To simulate con-
ventional SB ACL reconstructions in the 0◦ analytic model, two 8-mm-diameter virtual
tunnels were made at the average femoral and tibial tunnel positions evaluated by the
methods described by Bernard et al., [24] and also made according to the direction of the
tunnel estimated by the average femoral and tibial graft bending angles [11]. To simulate
anatomical DB ACL reconstructions, four 6-mm-diameter virtual tunnels were made using
the same method as for SB reconstruction. The simulation process was performed using
UG NX 7.0. The AM and PL grafts were simulated in each tunnel with 20 N tension in an
extended position. [15] The interface between the grafts and the tunnel was bonded using
mesh tie kinematic constraints. The contacts between bone and ligament and between
ligament and ligament were constructed using the penalty formulation, assuming 0.1 and
Med. 2021, 10, x FOR PEER REVIEW 5 of 13

performed
J. Clin. Med. 2021, 10, 1625using UG NX 7.0. The AM and PL grafts were simulated in each tunnel with 5 of 13
20 N tension in an extended position. [15] The interface between the grafts and the tunnel
was bonded using mesh tie kinematic constraints. The contacts between bone and liga-
ment and between ligament and ligament were constructed using the penalty formula-
tion, assuming0.001 of frictional
0.1 and 0.001 ofcoefficients, respectively.respectively.
frictional coefficients, [25] The validated model
[25] The was combined with
validated
the actual models through surgery (Figure 3).
model was combined with the actual models through surgery (Figure 3).

Figure 3. (A)Figure 3. (A) tomography

Computed Computed tomography
(CT) images(CT) images
of each of each
group weregroup were
used for 3Dused for 3D reconstruction
reconstruction in Mimics. To simulate
single-bundle or double-bundle ACL reconstructions in the 0 analytic model, tunnels wereinreamed
in Mimics. To simulate single-bundle or double-bundle
◦ ACL reconstructions the 0° analytic
at the center of tunnels
model, tunnels were reamed at the center of tunnels of the femur and tibia of the validation model
of the femur and tibia of the validation model using the average femoral tunnel position measured by Bernard’s quadrant
using the average femoral tunnel position measured by Bernard’s quadrant method [24] and
method [24] and drilled according to the femoral tunnel direction, estimated by the femoral graft bending angle described
drilled according to the femoral tunnel direction, estimated by the femoral graft bending angle
by Wang et al. [11] (B)by
described The reconstructed
Wang et al. [11] ACL wasreconstructed
(B) The attached to the
ACLfemoral and tibial
was attached totunnels in theand
the femoral model.
tibial
tunnels in the model.
2.4. Loading and Boundary Conditions and Evaluation
2.4. Loading and Boundary Conditions
The tibial and Evaluation
translations under 0 to 100 N of the anterior and posterior forces working
◦
The tibialon the knee center
translations underat0 0to extension
100 N of thewere compared
anterior and with previous
posterior studies
forces workingthat validated the
intact knee model [26,27]. Second loading conditions included
on the knee center at 0° extension were compared with previous studies that validated the a 134 N anterior load to
the tibia at 0 ◦ extension as well as at 30◦ , 60◦ , and 90◦ flexion. We measured and analyzed
intact knee model [26,27]. Second loading conditions included a 134 N anterior load to the
anterior
tibia at 0° extension astibial
well translation andand
as at 30°, 60°, graft
90°stresses,
flexion. which means the
We measured andtotal stress loaded at the
analyzed
anterior tibial whole graft,and
translation and graft
contact stresses
stresses, (von-Mises
which meansstress), which
the total means
stress loadedtheatstress
the loaded at the
whole graft, andcontact surface
contact between
stresses graft and
(von-Mises tunnel.
stress), which means the stress loaded at the
contact surface between graft and tunnel.
3. Results
3. Results 3.1. Validation

3.1. Validation The translation results for validation were similar to those of previous studies [26,27].
The tibial anterior and posterior translations for 100 N forces were 2.89 and 4.10 mm,
The translation results (2.43
respectively for validation
and 5.28 were
mm insimilar to those of previous
the experimental studies
study and 2.55[26,27].
and 4.86 mm in the
The tibial anterior
computational study). The element size was decided according to amm,
and posterior translations for 100 N forces were 2.89 and 4.10 re- study [20].
previous
spectively (2.43 and 5.28 mm in the experimental study and 2.55 and 4.86 mm in the com-
putational study). The element
3.2. Anterior Tibialsize was decided
Translation underaccording to a previous
a 134 N Anterior Load study [20].
The anterior tibial translation (ATT) in the TT SB group and the TP DB group ranged
3.2. Anterior Tibial
from 4.2 mm forUnder
Translation 0◦ to a9.1
134 N Anterior
mm Loadfrom 4.6 mm for 0◦ to 8.1 mm for 30◦ (Figure 4).
for 60◦ and
The anterior tibial translation (ATT) in the TT SB group and the TP DB group ranged
from 4.2 mm for 0° to 9.1 mm for 60° and from 4.6 mm for 0° to 8.1 mm for 30° (Figure 4).
J.J.Clin.
Clin.Med.
Med.2021, 10,x1625
2021,10, FOR PEER REVIEW 66ofof13
13

Comparisonof
Figure4.4.Comparison
Figure ofanterior
anteriortibial
tibial translation
translation under
under aa 134
134 N
N anterior
anterior load
loadbetween
betweenconventional
conven-
single-bundle
tional and anatomical
single-bundle double-bundle
and anatomical double-bundleanterior cruciate
anterior ligament
cruciate reconstruction.
ligament Abbrevia-
reconstruction. Ab-
tions: TT, transtibial;
breviations: TP, transportal.
TT, transtibial; TP, transportal.

3.3. Graft Stress

3.3. Graft Stress
The graft stress under an anterior load after ACL reconstruction using the two tech-
The graft stress under an anterior load after ACL reconstruction using the two tech-
niques became the greatest at 30◦ knee flexion (13.1 MPa for SB, 13.3 MPa for DB) (Table 1).
niques became the greatest at 30° knee flexion (13.1 MPa for SB, 13.3 MPa for DB)
(Table 1). a
Table 1. Graft stress during knee flexion combined with a 134 N anterior tibial load.
Table 1. Graft stress during knee
0◦ flexion combined
30◦with a 134 N anterior
60◦ tibial load.
a
90◦
Transtibial SB b 12.0 13.1 0° 30°11.4 60° 10.090°
Transportal DB c 12.3 13.312.0
Transtibial SB b 13.110.3 11.4 9.210.0
AM d 6.2 c 6.4 5.2 4.8
Transportal DB 12.3 13.3 10.3 9.2
PL e 6.1 6.9 5.1 4.4
AM d 6.2 6.4 5.2
a All data are expressed in MPa. b single bundle; c double bundle; d anteromedial; e posterolateral.
4.8
PL e 6.1 6.9 5.1 4.4
a All data are expressed
At lower flexionin MPa. b(0single
angles ◦ andbundle;
30◦ ), thedouble
c bundle;
total graft
d anteromedial; e posterolateral.
stress (the sum of AM and PL graft
stresses) of DB ACL reconstruction was slightly higher than that of SB ACL reconstruction.
At lower
However, at flexion
higher angles
flexion(0° and 30°),
angles (60◦ the
andtotal
90◦ ),graft
the stress (the sum
total graft stressofin
AM DBand
ACLPLrecon-
graft
stresses)
structionofwasDBlower
ACL thanreconstruction was slightly
that in SB ACL higher It
reconstruction. than that of
is worth SB ACL
noting thatreconstruc-
each AM or
tion.
PL graft stress was lower than in conventional SB ACL reconstruction fromin0◦DB
However, at higher flexion angles (60° and 90°), the total graft stress to ACL re-
90◦ knee
construction was lower than that in SB ACL reconstruction. It is worth
flexion. In the simulation of DB ACL reconstruction, the AM graft stress ranged from 6.4 to noting that each
AM or PLatgraft
4.8 MPa stress
30◦ and 90was lower
◦ knee thanrespectively,
flexion, in conventional SB ACL
whereas reconstruction
PL graft stress rangedfrom 0° 6.9
from to
90° knee flexion. In the simulation of DB ACL reconstruction, the AM
to 4.4 MPa at 30 and 90 knee flexion, respectively. At lower flexion angles (0 and 30◦ ),
◦ ◦ graft stress
◦ ranged
from
the PL6.4graft
to 4.8 MPa
stress wasathigher,
30° and 90°
and at knee
higherflexion,
flexion respectively,
angles (60◦ and whereas PLPLgraft
90◦ ), the graftstress
stress
ranged
was lowerfromthan
6.9 to
the4.4
AM MPa at stress
graft 30° and 90° knee
(Figure 5). flexion, respectively. At lower flexion an-
gles (0° and 30°), the PL graft stress was higher, and at higher flexion angles (60° and 90°),
the PL graft stress was lower than the AM graft stress (Figure 5).
J.J.Clin.
Clin.Med.
Med.2021, 10,x1625
2021,10, FOR PEER REVIEW 77ofof13
13

Comparison
Figure5.5.Comparison
Figure of of graft
graft stress
stress under
under a 134
a 134 N anterior
N anterior load load between
between conventional
conventional single-single-
bundleand
bundle andanatomical
anatomicaldouble-bundle
double-bundleanterior
anteriorcruciate
cruciateligament
ligamentreconstruction.
reconstruction. Abbreviations:
Abbreviations: TT,
transtibial;
TT, TP,TP,
transtibial; transportal; AM,
transportal; anteromedial;
AM, PL,PL,
anteromedial; posterolateral.
posterolateral.

3.4.Contact
3.4. ContactStress
StressBetween
BetweenGraft
Graftand
andTunnel
Tunnel
Thepatterns
The patternsofofcontact
contactstress
stressaccording
accordingto
toknee
kneeflexion
flexionare
areshown
shownin
inTables
Tables 22 and
and 3.
3.

Contactstress
Table2.2.Contact
Table stress between
between femoral
femoral tunnel
tunnel and
and graft
graft during
during knee
knee flexion
flexioncombined
combinedwith
withaa134
134N
anterior load.aa
N anterior load.

0◦ ◦
300° 30° 60◦ 60° 90◦90°
SB b
TranstibialTranstibial SB b
Femur Femur 12.0 9.1
12.0 9.1 6.3 6.3 4.14.1
Transportal DB c
Transportal DB12.5
c 9.3
12.5 9.3 6.6 6.6 4.84.8
AM d femur 6.2 4.4 3.2 2.7
PL e femur
AM femur 6.3
d 6.2
4.9
4.4 3.4
3.2 2.1
2.7
a PL e femur b 6.3 4.9 3.4 2.1
All data are expressed in MPa. single bundle; c double bundle; d anteromedial; e posterolateral.
a All data are expressed in MPa. b single bundle; c double bundle; d anteromedial; e posterolateral.
Table 3. Contact stress between tibial tunnel and graft during knee flexion combined with a 134 N
Table 3. Contact stress between tibial tunnel and graft during knee flexion combined with a 134 N
anterior load.a.
anterior load.a.
0◦ 30◦ ◦ 90◦
0° 30° 60 60° 90°
SB b
Transtibial Transtibial SB b
Tibia
Tibia 6.0 5.4
6.05.4 4.1 4.1 3.5
3.5
Transportal DB c 6.2 5.4 4.3 3.6
Transportal DB
AM d tibia 2.9
c 6.2
2.6 5.4 2.2 4.3 1.83.6
PL tibia AM tibia 3.3
e d 2.9
2.8 2.6 2.1 2.2 1.81.8
PL e in
a All data are expressed tibia
MPa. single bundle; double bundle; anteromedial; posterolateral. 1.8
b c 3.3 d 2.8 e 2.1
a All data are expressed in MPa. b single bundle; c double bundle; d anteromedial; e posterolateral.
In the simulation of SB ACL reconstruction, the contact stress in the femoral tunnel
In the
ranged simulation
from 4.1 MPaofatSB 90ACL reconstruction,
◦ to 12.0 MPa at 0◦ the contact
flexion. Instress in the femoral
the simulation of DBtunnel
ACL
ranged from 4.1 MPa at 90° to 12.0 MPa at 0° flexion. In the simulation of DB
reconstruction, the total contact stress in the femoral tunnel ranged from 4.8 MPa at 90◦ to ACL recon-
struction,
12.5 MPa theat 0total
◦ kneecontact stress
flexion. The in thecontact
total femoral tunnel
stress (sumranged
of AM from
and4.8
PLMPa at 90°
bundle to 12.5
stresses) at
MPa at 0° knee flexion. The total contact stress (sum of AM and PL bundle
the femoral tunnel in DB ACL reconstruction was higher than that in SB ACL reconstruction, stresses) at the
femoral tunneleach
even though in DB ACL stress
contact reconstruction
at the AMwas higher
or PL thantunnel
femoral that inofSBDBACL
ACLreconstruction,
reconstruction
even though each contact stress at the AM or PL femoral ◦tunnel
was about half of that of SB ACL reconstruction from 0 to 90 flexion (Figure◦ of DB ACL reconstruction
6).
was about half of that of SB ACL reconstruction from 0° to 90° flexion (Figure 6).
J. Clin. Med. 2021, 10, x FOR PEER REVIEW 8 of 13
J.J. Clin.
Clin. Med.
Med. 2021, 10, x1625
2021, 10, FOR PEER REVIEW 88of
of13
13

Figure
Figure6. Comparison
Comparisonof ofcontact
contact stress
stress at
at the femoral
femoral tunnelunder
under 134NNanterior
anterior tibialload
load be-
Figure 6.6.Comparison of contact stress at the
the femoral tunnel
tunnel under134
134 N anteriortibial
tibial loadbetween
be-
tween conventional single-bundle and anatomical double-bundle anterior cruciate ligament recon-
conventional
tween single-bundle
conventional and anatomical
single-bundle double-bundle
and anatomical anterioranterior
double-bundle cruciatecruciate
ligament reconstruction.
ligament recon-
struction. Abbreviations: TT, transtibial; TP, transportal; AM, anteromedial; PL, posterolateral.
struction. Abbreviations:
Abbreviations: TT, transtibial;
TT, transtibial; TP, transportal;
TP, transportal; AM, anteromedial;
AM, anteromedial; PL, posterolateral.
PL, posterolateral.

The
Thetotal
total contact
contact stress in the
stress in the tibialtunnel
tunnel rangedfrom from 3.5MPa MPa at9090°◦ toto
6.06.0 MPa at at
0◦
The total contact stress in the tibial
tibial tunnelranged
ranged from3.5 3.5 MPaatat 90° to MPa
6.0 MPa at
0° flexion
flexion in
inin SB ACL
SBSBACL reconstruction and 3.6 MPa at 90°
◦ to 6.2 MPa at 0°◦ flexion in DB ACL
0° flexion ACLreconstruction
reconstructionand and3.63.6MPa
MPaatat90 90°to to6.2
6.2 MPa
MPa at at 00° flexion
flexion inin DB
DB ACL
ACL
reconstruction.
reconstruction. The
The total
total contact
contact stress
stress atatthe
the tibial
tibial tunnel
tunnel in
inDB
DB ACL
ACL reconstruction
reconstruction was
was
reconstruction. The total contact stress at the tibial tunnel in DB ACL reconstruction was
higher
higher than
than that
thatin
in SB
SB ACL
ACL reconstruction
reconstruction at
at 0°,
0 ◦ , 60°,
60 ◦ , and
and 90°
90 ◦ knee
knee flexion,
flexion, even
even though
though
higher than that in SB ACL reconstruction at 0°, 60°, and 90° knee flexion, even though
each
eachcontact
contactstress
stressat
atthe
theAMAMor orPL
PLtibial
tibialtunnel
tunnelin inDB DBACL ACLreconstruction
reconstructionwas waslower
lowerthan
than
each contact stress at the AM or PL◦
tibial
◦
tunnel in DB ACL reconstruction was lower than
in SB
inSB ACL
SBACL reconstruction
ACLreconstruction
reconstructionfromfrom 0°
from 0°
0 toto 90°
to90° flexion
90 flexion (Figure
flexion(Figure
(Figure7). 7).
7).
in

Comparisonofofcontact
Figure7.7.Comparison
Figure contactstress
stressatatthe
thetibial
tibialtunnel
tunnelunder
under134
134NNanterior
anteriortibial
tibialload
loadbetween
between
Figure 7. Comparison of contact stress at the tibial tunnel under 134 N anterior tibial load between
conventionalsingle-bundle
conventional single-bundle and
and anatomical
anatomical double-bundle anterior cruciate
cruciate ligament
ligament reconstruction.
reconstruc-
conventional single-bundle and anatomical double-bundle anterior cruciate ligament reconstruc-
tion. Abbreviations:
Abbreviations: TT, transtibial;
TT, transtibial; TP, transportal;
TP, transportal; AM, AM, anteromedial;
anteromedial; PL, posterolateral.
PL, posterolateral.
tion. Abbreviations: TT, transtibial; TP, transportal; AM, anteromedial; PL, posterolateral.
4. Discussion
4. Discussion
4. Discussion
The main findings of our study were that the total graft stress after anatomical DB
The main findings of our study were that the total graft stress after ◦ to anatomical DB
◦ flexion and
ACLThe main findings
reconstruction of our
using the study were that
TP technique the totalhigher
is slightly graft stress
from 0after anatomical
30 DB
ACL reconstruction using the TP technique is slightly higher from 0° to 30° flexion and
ACL from 60◦ to 90using
lowerreconstruction ◦ flexion
the TPthantechnique
SB ACL is slightly higher
reconstruction fromthe
using 0° TT
to 30° flexion and
technique. The
lower from 60° to 90° flexion than SB ACL reconstruction using
◦ flexion the ◦TT technique. The
lower
contact from 60°was
stress to 90° flexion
largest at 0than SB ACL
and reconstruction
decreased fromusing 0◦ tothe
90 TT technique.
flexion The
after ACL
contact stress was largest at 0° flexion and decreased from 0° to 90° flexion after ACL
contact stress was
reconstruction largest
using bothattechniques.
0° flexion and
The decreased
total contactfrom 0° to
stress at 90° flexion after
the femoral and ACL
tibial
reconstruction using both techniques. The total contact stress at the femoral and tibial tun-
reconstruction
tunnels after DB using
ACL both techniques. The
reconstruction wastotal
equal contact stress
or higher at the
than femoral
in SB and tibial tun-
ACL reconstruction
nels after DB ACL reconstruction was equal or higher than in SB ACL reconstruction from
nels 0◦ to
fromafter DB ◦ flexion.
90ACL However,was
reconstruction the equal
graft stress andthan
or higher contact stress
in SB ACLofreconstruction
each AM/PL from graft
0° to 90° flexion. However, the graft stress and contact stress of each AM/PL graft were
were
0° about
to 90° half However,
flexion. of those intheSBgraft
ACLstress
reconstruction
and contact from 0◦ to
stress of 90 ◦ flexion.
each AM/PL There have
graft were
about half of those in SB ACL reconstruction from 0° to 90° flexion. There have been many
been many
about half ofstudies
those inthat have reconstruction
SB ACL evaluated the biomechanics
from 0° to 90°of the reconstructed
flexion. There have been ACLmany
using
J. Clin. Med. 2021, 10, 1625 9 of 13

cadaver or FEM analysis. [12–18]. However, there has not been any study in which a real
patient image with the real tunnel position and direction after ACL reconstruction using
both techniques was used, nor have there been any FEM studies to determine graft stress
and contact stress after conventional SB and anatomical DB ACL reconstruction.
In our study, the ATT was the largest at 60◦ flexion after SB ACL reconstruction and at
◦
30 flexion after DB ACL reconstruction. Our results also showed that the ATT after DB
ACL reconstruction was slightly higher than that in SB ACL reconstruction at a 0◦ knee
position with a difference of 0.4 mm and was lower from 30◦ to 90◦ knee flexion, although
a maximum 1.5 mm difference in the ATT may not be clinically significant. These results
were similar to another cadaver study that presented that the ATT under a 90 N anterior
load after anatomical DB reconstruction was significantly less than after SB reconstruction
from 30◦ to 90◦ flexion. In the same study, the maximum ATT was observed in the ACL
intact knee at 30◦ flexion and after anatomical DB reconstruction [28]. In some other studies,
DB ACL reconstructions have shown closer biomechanics to an intact knee compared with
SB reconstructions [8,14,29]. Tsai et al. also showed similar result to our study in that
the ATT after DB ACL reconstruction was smaller at a high flexion angle than after SB
ACL reconstruction [14]. This result might be due to the anatomical tunnel positioning
in DB ACL reconstruction. Yasuda et al. suggested that the tunnel location rather than
the number of bundles would be the cause of better results of DB reconstruction than the
conventional SB reconstruction [2].
The two bundles of the ACL function reciprocally in passive flexion and extension,
with the tighter PL bundle in extension and the tighter AM bundle in flexion [30]. The
ligament function evaluated in vitro might be different from that evaluated in vivo, because
the physiological loading conditions would be different [31]. When an anterior load is
applied to the knee, both bundles share the load in near extension, while the majority of
the load is shared by the AM bundle as the knee is flexed [16,32]. Many authors have
reported that AM/PL grafts are the longest in extension and decrease in flexion (from
0◦ to 90◦ ) [15,33]. In many studies, ACL tension has been estimated by measuring the
distance between ACL footprints [31,33,34]. However, the deformation of the graft and
the impingement between the grafts and surrounding bone affect ligament tension [15,35].
Kim et al. suggested that the stress caused by graft impingement between the graft and
surrounding bone according to knee flexion might maintain the tension within the grafts
without actual graft lengthening [15]. Atypical materials such as ligaments lead to large
amounts of stress concentration, accompanied by a relatively small load if there is any
deformation or contact through translation. Therefore, the value of stress was calculated
in this study. In general, the value of the cross-sectional area is required to calculate
stress quantity, and thus, it is much more convenient to conduct FEM rather than cadaver
experiments.
In our study, the total graft stress after DB ACL reconstruction was slightly greater at
0◦ to 30◦ flexion and lesser at 60◦ to 90◦ flexion than after SB ACL reconstruction, although
there was a similar trend in both techniques. In cases of DB ACL reconstruction, an
additional PL tunnel would further enhance the stress concentration around the tunnel at
low knee flexion [17]. In addition, the PL graft tension maximized at a lower flexion angle
and decreased with increased flexion of the knee [16]. This might be the cause of higher
total graft stress after DB ACL reconstruction at a low knee flexion in our study. However,
in this study, the stresses of the AM and PL grafts were about half of those in SB ACL
reconstruction from 0◦ to 90◦ , although the total graft stress in DB ACL reconstruction was
greater than that in SB ACL reconstruction at a low flexion angle. Yasuda et al. described
that in DB reconstruction, forces loaded to the tibia are shared by the two reconstructed
bundles, which can prevent excessive loading to one bundle [7]. This might be one of the
advantages of DB ACL reconstruction.
In our study, the total contact stress after DB ACL reconstruction using the TP tech-
nique was greater than after SB ACL reconstruction using the TT technique at 0◦ to 90◦
knee flexion. As mentioned above, in cases of DB ACL reconstruction, additional PL
J. Clin. Med. 2021, 10, 1625 10 of 13

tunnel creation would further amplify the severity of the stress concentration around the
tunnel [17]. This might be the cause of higher total contact stress after anatomical DB
ACL reconstruction from 0◦ to 90◦ flexion in our study. However, each contact stress at
AM and PL femoral tunnels were less than those in SB ACL reconstruction using the TT
technique. Wang et al. showed that the femoral graft bending angle of the DB ACL using
the TP technique was more acute than that of the SB ACL using the TT technique and
suggested that this acute bending angle might increase the graft stress at low flexion [11].
Our results presented that the graft bending stress at the femoral tunnel after applying both
reconstruction techniques was greatest at 0◦ flexion, which we assumed in our previous
study. However, our results also showed that the contact stress of the AM or PL tunnel
was about half of that of the SB tunnel, regardless of the more acute graft bending angle.
Therefore, if we performed DB ACL reconstruction using the TP technique, the damage
to each AM or PL graft might not be more than that by SB ACL reconstruction using the
TT technique. Even though the anatomical ACL reconstruction technique could increase
the graft stress and contact stress at the tunnel, we would not know clearly its actual
clinical effect. However, we need to study the comparison of the clinical results between
anatomical ACL reconstruction and conventional ACL reconstruction to prove the clinical
effect of this biomechanical difference.
The contact stress in the femoral tunnel was greater than that in the tibial tunnel in our
result. Hirokawa et al. presented in a FEM study that the largest stress was observed near
the femoral insertion, and the area least stretched throughout the whole range of flexion
was the portion near the tibial insertion of an intact ACL [36]. Song et al. also proposed in
their FEM study that the highest stress was focused near the femoral insertion site, and the
least stress was shown near the tibial insertion at full extension under the anterior load [35].
Similar results were observed in our study. Therefore, graft failure might develop more
often near the femoral tunnel in SB or DB ACL reconstruction.
This study had some limitations and simplifications. First, the ligament’s initial ten-
sion value was considered in only one particular case. According to findings from previous
studies, the graft stress and contact stress will be different based on the initial value of
ligament tension [18]. Second, we performed this FEM study under only one loading
condition (134 N anterior tibial translation). If we combined various loading condition
such as varus/valgus and internal/external rotation, it would be more meaningful. How-
ever, the main function of the ACL is limiting the anterior tibial translation. Therefore,
we performed this biomechanical study under a 134 N anterior tibial loading condition.
Actually, there have been many biomechanical studies under similar loading conditions
as our study [13,32,35,37,38]. Li et al. presented in their cadaver study anterior tibial
translation and graft forces under 130 N anterior loading at 0, 15, 30, 60, and 90◦ flexion
after conventional SB ACL reconstruction in comparison to an intact knee [13]. Sakane et al.
also analyzed in situ forces in the ACL and force distribution to the AM and PL bundle
under an anterior load from 22 to 110 N at 0–90◦ flexion using a cadaver [32]. In their FEM
study, Song et al. presented the force and stress distribution within the ACL (AM and PL
bundle) under an anterior load (0–134 N) at full extension. Third, we used just one tendon
diameter regardless of the knee size. The graft diameter can affect the biomechanics after
ACL reconstruction. In some previous studies [39,40], the authors have presented that the
ACL graft size affects the degree of impingement on the intercondylar notch and stresses
occur within the ACL graft during flexion of the knee using Lachman simulation. Proper
determination of the graft size corresponding to the knee size would be included in suture
studies [15,41]. Fourth, we were not able to compare anterior translation and graft stress
patterns between the two ACL reconstruction techniques and a normal ACL. However,
it would have been difficult to conclude whether any ACL reconstruction technique was
superior, since the normal ACL stress pattern was not yet known. Fifth, we performed
the remnant preservation technique only in SB ACL reconstruction. However, the femoral
and tibial tunnel positions were not different from the non-remnant preservation SB ACL
reconstruction. In this FEM study, we used only data of the mean tunnel position and
J. Clin. Med. 2021, 10, 1625 11 of 13

the tunnel direction after the ACL reconstruction. Therefore, the remnant preservation
technique would not affect the result of this FEM study. Finally, we couldn’t compare the
direct effect of only the femoral reaming technique (TT vs. TP or the number of grafts (SB
vs. DB), because these two variables were different in each two groups. However, in a
previous study, ref. [11] we found that the femoral graft bending angle was significantly
different between anatomical DB ACL reconstruction with the TP technique and conven-
tional SB ACL reconstruction with the TT technique and hypothesized that this more acute
graft bending angle after the anatomical reconstruction technique would increase the graft
stress than the conventional technique. We wanted to evaluate the effect of the anatomical
femoral reaming technique (TP) compared to the conventional technique (TT) following
our previous study that compared the femoral graft bending angle. So, we performed this
FEM study.

5. Conclusions
The total graft stress after anatomical DB ACL reconstruction using the TP technique
was greater from 0◦ to 30◦ flexion and was lesser from 60◦ to 90◦ flexion than after con-
ventional SB ACL reconstruction using the TT technique. The total contact stress at the
tibial and femoral tunnels was greatest at 0◦ knee flexion and decreased from 0◦ to 90◦
flexion after ACL reconstruction using both techniques. The total contact stresses at the
femoral and tibial tunnels after anatomical DB ACL reconstruction were greater than those
after conventional SB ACL reconstruction using the TT technique at 0◦ to 90◦ knee flexion,
respectively. However, the graft and contact stresses of each AM/PL femoral and tibial
tunnel after anatomical DB ACL reconstruction were less than those after conventional SB
ACL reconstruction.

Author Contributions: Conceptualization, J.H.W.; data curation, J.G.K. and K.T.K.; formal analysis,
J.G.K. and K.T.K.; resources, J.H.W.; software, K.T.K.; supervision, J.H.W.; writing—original draft,
J.G.K., K.T.K., and J.H.W.; writing—review and editing, J.G.K., K.T.K., and J.H.W. All authors have
read and agreed to the published version of the manuscript.
Funding: There was no funding in this study.
Institutional Review Board Statement: The study was conducted according to the guidelines of the
Declaration of Helsinki, and approved by the Institutional Review Board of Samsung Medical Center
(2012-07-071-001 and 2012.08.03).
Informed Consent Statement: Patient consent was waived, because in this study we used only CT
data of patients taken previously.
Data Availability Statement: Data available in a publicly accessible repository
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Iriuchishima, T.; Horaguchi, T.; Kubomura, T.; Morimoto, Y.; Fu, F.H. Evaluation of the intercondylar roof impingement after
anatomical double-bundle anterior cruciate ligament reconstruction using 3D-CT. Knee Surg. Sports Traumatol. Arthrosc. 2011, 19,
674–679. [CrossRef]
2. Yasuda, K.; van Eck, C.F.; Hoshino, Y.; Fu, F.H.; Tashman, S. Anatomic single- and double-bundle anterior cruciate ligament
reconstruction, part 1: Basic science. Am. J. Sports Med. 2011, 39, 1789–1799. [CrossRef] [PubMed]
3. Woo, S.L.; Kanamori, A.; Zeminski, J.; Yagi, M.; Papageorgiou, C.; Fu, F.H. The effectiveness of reconstruction of the anterior
cruciate ligament with hamstrings and patellar tendon. A cadaveric study comparing anterior tibial and rotational loads. J. Bone
Joint Surg. Am. Vol. 2002, 84, 907–914. [CrossRef] [PubMed]
4. Brophy, R.H.; Selby, R.M.; Altchek, D.W. Anterior cruciate ligament revision: Double-bundle augmentation of primary vertical
graft. Arthrosc. J. Arthrosc. Related Surg. 2006, 22, e681–e685. [CrossRef]
5. Robinson, J.; Stanford, F.C.; Kendoff, D.; Stuber, V.; Pearle, A.D. Replication of the range of native anterior cruciate ligament fiber
length change behavior achieved by different grafts: Measurement using computer-assisted navigation. Am. J. Sports Med. 2009,
37, 1406–1411. [CrossRef]
6. Fu, F.H.; Shen, W.; Starman, J.S.; Okeke, N.; Irrgang, J.J. Primary anatomic double-bundle anterior cruciate ligament reconstruction:
A preliminary 2-year prospective study. Am. J. Sports Med. 2008, 36, 1263–1274. [CrossRef]
J. Clin. Med. 2021, 10, 1625 12 of 13

7. Yasuda, K.; Kondo, E.; Ichiyama, H.; Tanabe, Y.; Tohyama, H. Clinical evaluation of anatomic double-bundle anterior cruciate
ligament reconstruction procedure using hamstring tendon grafts: Comparisons among 3 different procedures. Arthrosc. J.
Arthrosc. Related Surg. 2006, 22, 240–251. [CrossRef]
8. Oh, J.Y.; Kim, K.T.; Park, Y.J.; Won, H.C.; Yoo, J.I.; Moon, D.K.; Cho, S.H.; Hwang, S.C. Biomechanical comparison of single-bundle
versus double-bundle anterior cruciate ligament reconstruction: A meta-analysis. Knee Rurg. Related Res. 2020, 32, 14. [CrossRef]
9. Lubowitz, J.H.; Konicek, J. Anterior cruciate ligament femoral tunnel length: Cadaveric analysis comparing anteromedial portal
versus outside-in technique. Arthrosc. J. Arthrosc. Related Surg. 2010, 26, 1357–1362. [CrossRef] [PubMed]
10. Lubowitz, J.H. Anteromedial portal technique for the anterior cruciate ligament femoral socket: Pitfalls and solutions. Arthrosc. J.
Arthrosc. Related Surg. 2009, 25, 95–101. [CrossRef] [PubMed]
11. Wang, J.H.; Kim, J.G.; Lee do, K.; Lim, H.C.; Ahn, J.H. Comparison of femoral graft bending angle and tunnel length between
transtibial technique and transportal technique in anterior cruciate ligament reconstruction. Knee Surg. Sports Traumatol. Arthrosc.
2012, 20, 1584–1593. [CrossRef]
12. Vercillo, F.; Woo, S.L.; Noorani, S.Y.; Dede, O. Determination of a safe range of knee flexion angles for fixation of the grafts in
double-bundle anterior cruciate ligament reconstruction: A human cadaveric study. Am. J. Sports Med. 2007, 35, 1513–1520.
[CrossRef]
13. Li, G.; Papannagari, R.; DeFrate, L.E.; Yoo, J.D.; Park, S.E.; Gill, T.J. Comparison of the ACL and ACL graft forces before and after
ACL reconstruction: An in-vitro robotic investigation. Acta Orthopaedica 2006, 77, 267–274. [CrossRef]
14. Tsai, A.G.; Wijdicks, C.A.; Walsh, M.P.; Laprade, R.F. Comparative kinematic evaluation of all-inside single-bundle and double-
bundle anterior cruciate ligament reconstruction: A biomechanical study. Am. J. Sports Med. 2010, 38, 263–272. [CrossRef]
15. Kim, H.Y.; Seo, Y.J.; Kim, H.J.; Nguyenn, T.; Shetty, N.S.; Yoo, Y.S. Tension changes within the bundles of anatomic double-bundle
anterior cruciate ligament reconstruction at different knee flexion angles: A study using a 3-dimensional finite element model.
Arthrosc. J. Arthrosc. Related Surg. 2011, 27, 1400–1408. [CrossRef] [PubMed]
16. Wu, J.L.; Seon, J.K.; Gadikota, H.R.; Hosseini, A.; Sutton, K.M.; Gill, T.J.; Li, G. In situ forces in the anteromedial and posterolateral
bundles of the anterior cruciate ligament under simulated functional loading conditions. Am. J. Sports Med. 2010, 38, 558–563.
[CrossRef]
17. Yao, J.; Wen, C.; Cheung, J.T.; Zhang, M.; Hu, Y.; Yan, C.; Chiu, K.Y.; Lu, W.W.; Fan, Y. Deterioration of stress distribution due
to tunnel creation in single-bundle and double-bundle anterior cruciate ligament reconstructions. Ann. Biomed. Eng. 2012, 40,
1554–1567. [CrossRef]
18. Yasuda, K.; Ichiyama, H.; Kondo, E.; Miyatake, S.; Inoue, M.; Tanabe, Y. An in vivo biomechanical study on the tension-versus-
knee flexion angle curves of 2 grafts in anatomic double-bundle anterior cruciate ligament reconstruction: Effects of initial tension
and internal tibial rotation. Arthrosc. J. Arthrosc. Related Surg. 2008, 24, 276–284. [CrossRef]
19. Kwon, O.R.; Kang, K.T.; Son, J.; Kwon, S.K.; Jo, S.B.; Suh, D.S.; Choi, Y.J.; Kim, H.J.; Koh, Y.G. Biomechanical comparison of fixed-
and mobile-bearing for unicomparmental knee arthroplasty using finite element analysis. J. Orthopaedic Res. 2014, 32, 338–345.
[CrossRef] [PubMed]
20. Pena, E.; Calvo, B.; Martinez, M.A.; Doblare, M. A three-dimensional finite element analysis of the combined behavior of
ligaments and menisci in the healthy human knee joint. J. Biomech. 2006, 39, 1686–1701. [CrossRef] [PubMed]
21. Takeda, Y.; Xerogeanes, J.W.; Livesay, G.A.; Fu, F.H.; Woo, S.L. Biomechanical function of the human anterior cruciate ligament.
Arthrosc. J. Arthrosc. Related Surg. 1994, 10, 140–147. [CrossRef]
22. Mesfar, W.; Shirazi-Adl, A. Biomechanics of the knee joint in flexion under various quadriceps forces. Knee 2005, 12, 424–434.
[CrossRef]
23. ABAQUS, version 6.11, Simulia: Providence, RI, USA, 2011.
24. Forsythe, B.; Kopf, S.; Wong, A.K.; Martins, C.A.; Anderst, W.; Tashman, S.; Fu, F.H. The location of femoral and tibial tunnels in
anatomic double-bundle anterior cruciate ligament reconstruction analyzed by three-dimensional computed tomography models.
J. Bone Joint Surg. Am. Vol. 2010, 92, 1418–1426. [CrossRef]
25. Chizari, M.; Wang, B. 3D numerical analysis of an ACL reconstructed knee. SIMULIA Customer Conf. 2009, 2009, 1–13.
26. Li, G.; Gil, J.; Kanamori, A.; Woo, S.L. A validated three-dimensional computational model of a human knee joint. J. Biomech. Eng.
1999, 121, 657–662. [CrossRef] [PubMed]
27. Li, G.; Suggs, J.; Gill, T. The effect of anterior cruciate ligament injury on knee joint function under a simulated muscle load: A
three-dimensional computational simulation. Ann. Biomed. Eng. 2002, 30, 713–720. [CrossRef]
28. Kondo, E.; Merican, A.M.; Yasuda, K.; Amis, A.A. Biomechanical comparisons of knee stability after anterior cruciate ligament
reconstruction between 2 clinically available transtibial procedures: Anatomic double bundle versus single bundle. Am. J. Sports
Med. 2010, 38, 1349–1358. [CrossRef]
29. Yagi, M.; Wong, E.K.; Kanamori, A.; Debski, R.E.; Fu, F.H.; Woo, S.L. Biomechanical analysis of an anatomic anterior cruciate
ligament reconstruction. Am. J. Sports Med. 2002, 30, 660–666. [CrossRef] [PubMed]
30. Amis, A.A.; Dawkins, G.P. Functional anatomy of the anterior cruciate ligament. Fibre bundle actions related to ligament
replacements and injuries. J. Bone Joint Surg. Br. Vol. 1991, 73, 260–267. [CrossRef]
31. Li, G.; DeFrate, L.E.; Sun, H.; Gill, T.J. In vivo elongation of the anterior cruciate ligament and posterior cruciate ligament during
knee flexion. Am. J. Sports Med. 2004, 32, 1415–1420. [CrossRef]
J. Clin. Med. 2021, 10, 1625 13 of 13

32. Sakane, M.; Fox, R.J.; Woo, S.L.; Livesay, G.A.; Li, G.; Fu, F.H. In situ forces in the anterior cruciate ligament and its bundles in
response to anterior tibial loads. J. Orthopaedic Res. 1997, 15, 285–293. [CrossRef] [PubMed]
33. Yoo, Y.S.; Jeong, W.S.; Shetty, N.S.; Ingham, S.J.; Smolinski, P.; Fu, F. Changes in ACL length at different knee flexion angles: An
in vivo biomechanical study. Knee Surg. Sports Traumatol. Arthrosc. 2010, 18, 292–297. [CrossRef]
34. Jordan, S.S.; DeFrate, L.E.; Nha, K.W.; Papannagari, R.; Gill, T.J.; Li, G. The in vivo kinematics of the anteromedial and
posterolateral bundles of the anterior cruciate ligament during weightbearing knee flexion. Am. J. Sports Med. 2007, 35, 547–554.
[CrossRef]
35. Song, Y.; Debski, R.E.; Musahl, V.; Thomas, M.; Woo, S.L. A three-dimensional finite element model of the human anterior cruciate
ligament: A computational analysis with experimental validation. J. Biomech. 2004, 37, 383–390. [CrossRef]
36. Hirokawa, S.; Tsuruno, R. Three-dimensional deformation and stress distribution in an analytical/computational model of the
anterior cruciate ligament. J. Biomech. 2000, 33, 1069–1077. [CrossRef]
37. Grontvedt, T.; Pena, F.; Engebretsen, L. Accuracy of femoral tunnel placement and resulting graft force using one- or two-incision
drill guides. A cadaver study on ten paired knees. Arthrosc. J. Arthrosc. Related Surg. 1996, 12, 187–192. [CrossRef]
38. Hoher, J.; Kanamori, A.; Zeminski, J.; Fu, F.H.; Woo, S.L. The position of the tibia during graft fixation affects knee kinematics and
graft forces for anterior cruciate ligament reconstruction. Am. J. Sports Med. 2001, 29, 771–776. [CrossRef] [PubMed]
39. Orsi, A.D.; Canavan, P.K.; Vaziri, A.; Goebel, R.; Kapasi, O.A.; Nayeb-Hashemi, H. The effects of graft size and insertion site
location during anterior cruciate ligament reconstruction on intercondylar notch impingement. Knee 2017, 24, 525–535. [CrossRef]
[PubMed]
40. Westermann, R.W.; Wolf, B.R.; Elkins, J.M. Effect of ACL reconstruction graft size on simulated Lachman testing: A finite element
analysis. Iowa Orthop. J. 2013, 33, 70–77.
41. Yoon, K.H.; Kim, Y.H.; Ha, J.H.; Kim, K.; Park, W.M. Biomechanical evaluation of double bundle augmentation of posterior
cruciate ligament using finite element analysis. Clin. Biomech. 2010, 25, 1042–1046. [CrossRef]