Arch Orthop Trauma Surg (2007) 127:743752

DOI 10.1007/s00402-006-0241-3

A R T HR O SC OP Y A ND S PO RT S M E D I C IN E

Anterolateral rotational knee instability: role of posterolateral

structures
Thore Zantop Tobias Schumacher
Nadine Diermann SteVen Schanz
Michael J. Raschke Wolf Petersen

Received: 2 October 2006 / Published online: 28 October 2006

Springer-Verlag 2006

Abstract
Introduction The aim of this study was to determine
the anterolateral rotational instability (ALRI) of the
human knee after rupture of the anterior cruciate ligament (ACL) and after additional injury of the diVerent
components of the posterolateral structures (PLS). It
was hypothesized that a transsection of the ACL will
signiWcantly increase the ALRI of the knee and
furthermore that sectioning the PLS [lateral collateral
ligament (LCL), popliteus complex (PC)] will additionally signiWcantly increase the ALRI.
Materials and methods Five human cadaveric knees
were used for dissection to study the appearance and
behaviour of the structures of the posterolateral corner
under anterior tibial load. Ten fresh-frozen human
cadaver knees were subjected to anterior tibial load of
134 N and combined rotatory load of 10 Nm valgus and
4 Nm internal tibial torque using a robotic/universal
force moment sensor (UFS) testing system and the
resulting knee kinematics were determined for intact,
ACL-, LCL- and PC-deWcient (popliteus tendon and
popliteoWbular ligament) knee. Statistical analyses
were performed using a two-way ANOVA test with
the level of signiWcance set at P < 0.05.
Results Sectioning the ACL signiWcantly increased
the anterior tibial translation (ATT) and internal tibial

This study is a winner of the AGA DonJoy Award 2006.

T. Zantop (&) T. Schumacher N. Diermann
S. Schanz M. J. Raschke W. Petersen
Department of Trauma, Hand and Reconstructive Surgery,
Westfalian Wilhelms University Muenster,
Waldeyer Strasse 1, 48149 Muenster, Germany
e-mail: thore.zantop@ukmuenster.de

rotation under a combined rotatory load at 0 and 30

Xexion (P < 0.05). Sectioning the LCL further
increased the ALRI signiWcantly at 0, 30 and 60 of
Xexion (P < 0.05). Subsequent cutting of the PC
increased the ATT under anterior tibial load
(P < 0.05), but did not increase the ALRI (P > 0.05).
Conclusion The results of the current study conWrm
the concept that the rupture of the ACL is associated
with ALRI. Current reconstruction techniques should
focus on restoring the anterolateral rotational knee
instability to the intact knee. Additional injury to the
LCL further increases the anterior rotational instability signiWcantly, while the PC is less important. Cautions should be taken when examining a patient with
ACL rupture to diagnose injuries to the primary
restraints of tibial rotation such as the LCL. If an additional extraarticular stabilisation technique is needed
for severe ALRI, the technique should be able to
restore the function of the LCL and not the PC.
Keywords ACL reconstruction Revision
Rotational instability Non-coopers
Robotic/UFS testing system

Introduction
The surgical reconstruction of the ACL is a common
procedure to restore knee stability, and good to excellent clinical results have been reported. However, a critical review of the literature reveals that the success rates
reported for ACL reconstruction after relatively shortterm follow-up are between 69 and 95% [10]. The
causes for the failure after ACL reconstruction are multifactorial (tunnel malplacement, infection, insuYcient

123

744

Wxation, bone tunnel enlargement, postoperative stiVness) [10, 15, 18, 20]. One important factor for the failure of an ACL reconstruction might be an insuYcient
treatment of associated injuries such as the menisci or
injuries of the posteromedial or posterolateral corner
[2, 16, 26, 27, 41]. Especially, associated injuries to the
posterolateral corner may be underestimated and may
play an important role for the surgical outcome [1, 16].
It is well known that the anterior cruciate ligament
(ACL) is the primary restraint to tibial anterior translation. However, in addition to Xexion and extension,
the human knee also allows tibial internalexternal
rotations due to the lack of congruency of the femoral
condyles and tibial plateau [1, 37, 47, 48]. In general,
the lateral compartment is more mobile than the
medial because of the attachment of the medial meniscus to the joint capsule [32, 37]. The mobility of the lateral compartment increases in the weight-bearing
situation because of the anatomical characteristics of
the medial and the lateral tibial plateau. When bearing
weight, the concavity of the medial plateau stabilizes
the medial femoral condyle, whereas the convexity of
the lateral plateau cannot stabilize the lateral femoral
condyle [1, 37]. Thereby, under an anterior tibial load,
anterior tibial translation (ATT) is accompanied by a
tibial internal rotation (coupled tibial rotation,
Fig. 1). In the ACL-injured knee, this coupled tibial
rotation may lead to anterolateral rotatory instability

Fig. 1 Due to the attachment of the medial meniscus to the joint

capsule and the convexity of the lateral tibial plateau, the lateral
compartment of the knee joint is more mobile than the medial
compartment. An anterior tibial translation is therefore associated with an internal tibial rotation. This coupled motion is regarded as coupled tibial rotation

123

Arch Orthop Trauma Surg (2007) 127:743752

(ALRI) as determined by Hughston et al. [14]. For a

patient, these altered kinematics may cause symptoms
of instability such as giving way phenomenon and can
clinically be assessed by the pivot shift test [37].
While the role of the ACL in controlling ATT is well
understood, there is much more controversy about its
role in controlling tibial internal rotation. Lipke et al.
[25] for example could show that the injury of the ACL
led to a signiWcant increase in internal tibial rotation.
Similar results are reported by Amis and Scammel [2].
In their study, coupled internal rotation increased signiWcantly after ACL rupture. In contrast, biomechanical investigations of Fukubashi et al. [11] showed that
coupled tibial rotation decreased after transsection of
the ACL.
The converse results reported in these studies may
reXect the consequences of isolated ACL injury after
trauma versus isolated transsection of the ACL in a biomechanical experiment. A clinical study reported a high
incidence of associated injuries of the posterolateral
structures (PLS) in patients with an ACL rupture after
rotational trauma [38]. As the axis of rotation of the
tibia plateau is close to the line of action of the ACL,
this ligament seems to be only a secondary restraint
against rotatory loads while the peripheral PLS are better placed for controlling tibial rotation [1, 26].
The PLS of the knee can be divided into two primary components, the lateral collateral ligament
(LCL) and the popliteus complex (PC) (Fig. 2). The
PC consists of the popliteus muscletendon unit and
ligamentous connections between the popliteal tendon
and the Wbula, tibia and meniscus, known as the popliteoWbular ligament and popliteotibial and popliteomeniscal fascicles, respectively [7, 16, 19]. With its
tendinous and ligamentous components, the PC
imparts both static and dynamic restraint to the knee
[7, 16, 40]. The arcuate ligament complex and fabelloWbular ligament are also considered part of the PLS, but
the importance of these structures is believed to be relatively minor [7, 16, 19, 36, 40] (Fig. 2).
The role of the posterolateral corner in rotational
posterolateral instability has been extensively studied
[7, 16, 40]. However, when compared with posterolateral rotational instability, there has been less work on
the role of the structures of the posterolateral corner in
ALRI. Because of this lack of work, present understanding of the functions of these structures in ALRI
remains incomplete, particularly relating to the control
of tibial rotational laxity. Therefore, clinically controversy exists about the management of the resulting
complex instability.
The aim of the present study is to determine the rotational instability of the human knee after the rupture of

Arch Orthop Trauma Surg (2007) 127:743752

745

and described earlier [24, 31, 33, 45] (Fig. 3). This system allows obtaining the knee kinematics in response to
diVerent external loading conditions testing the same
specimen in diVerent conditions: intact, ACL deWcient,
LCL deWcient and with sectioned PLS. The application
of a combined rotatory and valgus load is a biomechanical method to simulate the pivot shift test and to assess
rotational stability in an in vitro setting [17, 42, 46].
It was hypothesized that a transsection of the ACL
will signiWcantly increase the ALRI of the knee and
furthermore that sectioning the PLS (LCL, PC with
popliteus tendon and popliteoWbular ligament) will
additionally signiWcantly increase the ALRI.

Materials and methods

Specimens

Fig. 2 Anatomical preparation of the posterolateral structures,

view from posterolateral. The posterolateral structures of the
knee can be divided into two primary components, the lateral collateral ligament (LCL, 1) and the popliteus complex consisting
out of popliteus tendon (2) and the popliteoWbular ligament (3)

the ACL and after additional injury to the extraarticular primary restraints such as LCL and PC (popliteus
tendon, popliteoWbular ligament) and to elucidate
which structures of the posterolateral corner play an
essential role in controlling the ALRI. To accomplish
this, a robotic/universal force moment sensor (UFS)
testing system will be used as introduced by Woo et al.

In this study 15 fresh-frozen human cadaveric knees

were used (age range: 5978 years). Ten knees were
used for the kinematic study; Wve knees were used for
dissection to study the appearance and behaviour of
the structures of the posterolateral corner. No specimens did show any signs of surgical intervention and
the knees were radiographed and clinically examined
to exclude specimens with bony abnormalities and
osteoarthritis or ligamentous injury.
Prior to testing, the knees were stored at 20 and
thawed for 24 h at room temperature [35]. Femur and
tibia were cut 20 cm from the joint line and the surrounding skin and muscles more than 10 cm away from
the joint line were removed. To maintain the anatomic
position in the proximal tibioWbular joint, the Wbula
was rigidly Wxed to the tibia with a cortical screw.

Fig. 3 Robotic/UFS testing

system (a). The robotic
manipulator can move the
knee in six degrees of freedom, while the universal force
moment sensor (UFS) can
measure three orthogonal
forces and moments (b). The
knee is mounted to the system
with the tibia attached to the
Xunch of the robot via the
UFS while the femur is
mounted to the base

123

746

Femur and tibia were then securely Wxed within thickwalled aluminium cylinders with polymethylmethacrylate bone cement (Palacos, Merk, Darmstadt, Germany). The femoral cylinder was mounted to the base
of the robot (KR 125, KUKA Robots, Augsburg, Germany) with a custom-made clamp while the tibial cylinder was connected through a UFS (FTI Theta 1500240, Schunk, LauVen, Germany) (Fig. 3). The UFS was
Wrmly Wxed to the end eVector of a six degree of freedom robotic manipulator. To prevent exsiccation, specimens were kept moist using saline solution (0.9%).
Anatomical study
To study the anatomy of the posterolateral corner the
skin and subcutaneous fat, tissue was removed in Wve
knees leaving the ligamentous and tendinous structures
intact. The appearance of the PLS was recorded as the
specimens were extended, Xexed and rotated using digital photography. We studied how the ligaments
appeared to act and in what positions diVerent parts of
the complex either tightened or slackened.
Robotic/UFS testing system
To determine knee kinematics, a testing system for
knee kinematics, which combines robotic technology
with a UFS was used. The robot (KR 125, KUKA
Robots, Augsburg, Germany, Fig. 3) is a six-joint, serially articulated manipulator, which allows six degree of
freedom movement of the knee. The system is capable
of highly accurate kinematic measurements, such as
anteroposterior translation, mediallateral translation,
proximaldistal translation, varusvalgus rotation and
internalexternal rotation of joint motion [24, 31, 33,
45, 46]. The repeatability of this system is 0.2 mm and
0.02 for orientation and position of the end eVector,
respectively [24, 31, 33, 45, 46]. The UFS can measure
three forces and three moments along a Cartesian axis
system with repeatability of 0.2 N for forces and
0.01 Nm for moments [24, 31, 33, 45, 46] (Fig. 3).
The robotic manipulator is capable of achieving
positional control of the knee in six degrees of freedom, while the UFS can measure three orthogonal
forces and moments. Simultaneously, this system is
capable of operating in a force-controlled mode via the
force feedback from the UFS to the robot.

Arch Orthop Trauma Surg (2007) 127:743752

the robotic/UFS testing system by targeting force and

moment of zero in all remaining degrees of freedom.
The system found the positions of the knee that minimized all external forces and moments applied to the
joint throughout the range of Xexion from 0 to 90 in
increments of 1. The positions determined by this procedure served as the starting point for application of
external loads. To imitate clinical evaluation for the
knee, an anterior tibial load of 134 N and a combined
rotatory load of 10 Nm valgus and 4 Nm internal tibial
torque was applied at 0, 30, 60 and 90 of knee Xexion. The anterior tibial load was chosen because the
ACL is the primary restraint to ATT and to simulate
clinical tests such as the anterior drawer or Lachman
tests. The force of 134 N was chosen because this is the
force used for instrumented knee laxity measurements
in the KT 1000 [8]. To evaluate the kinematics in
response to a pivot shift test [23, 37], a combined rotatory load of 10 Nm valgus and 4 Nm internal tibial
torque was chosen. The forces have been used previously [33, 46] to perform a simulated pivot shift test
[17]. Using this approach, the same specimen can be
tested in diVerent conditions: intact, ACL-, LCL- and
PC-deWcient (popliteus tendon and popliteoWbular ligament) knee thereby increasing the statistical power.
Next, the ACL was cut through a small lateral parapatellar incision. The external loading conditions were
then reapplied to the knee and the new kinematics
were recorded. Subsequently, the LCL and the PC
(popliteus tendon and popliteoWbular ligament) were
cut (Fig. 2) through a small posterolateral incision and
the resulting kinematics was recorded by the testing
system (Table 1). The order of sectioning the LCL and
the PC was alternated (Table 1).
Statistics
The kinematic data for the intact, ACL-, LCL- and PCdeWcient knees were analysed by using a two-factor
repeated-measures analysis of variance (ANOVA).
Since all tests were performed in the same specimen,
multiple contrasts were performed. The two factors evaluated were the condition of the knee and the knee Xexion angle. The dependent variables evaluated were knee
kinematics. The level of signiWcance was set at P < 0.05.

Results
Testing protocol
Macroscopic observations
The experimental protocol and the data acquired are
displayed in Table 1. The path of passive Xexion
extension of the intact knee joint was determined by

123

In all specimens careful dissections revealed a strong

LCL, popliteus tendon and popliteoWbular ligament

Arch Orthop Trauma Surg (2007) 127:743752

Table 1 The experimental
protocol and the data acquired

747

Loading condition

Data obtained

Intact knee
134 N anterior tibial load (ATL)
10 Nm valgus and 4 Nm internal
tibial torque at 0, 30, 60 and 90

Intact knee kinematics in response to ATL

Intact knee kinematics in response
to combined rotatory load

Transsection of the ACL

134 N anterior tibial load
10 Nm valgus and 4 Nm internal
tibial torque at 0, 30, 60 and 90

ACL-deWcient knee kinematics in response to ATL

ACL-deWcient knee kinematics in response
to combined rotatory load

Transsection of the LCL

134 N anterior tibial load
10 Nm valgus and 4 Nm internal
tibial torque at 0, 30, 60 and 90

LCL-deWcient knee kinematics in response to ATL

LCL-deWcient knee kinematics in response
to combined rotatory load

Transsection of the posterolateral complex

(popliteal tendon, popliteoWbular ligament)
134 N anterior tibial load
10 Nm valgus and 4 Nm internal
tibial torque at 0, 30, 60 and 90

(Fig. 4). The LCL was tight in full extension and slackened as the knee Xexed beyond 45 of Xexion. ATT in
neutral rotation position tensioned the LCL (Fig. 4).
Internal tibial rotation of the tibia moved the tibial
insertion of the LCL anteriorly thereby tensioning this
structure to a higher degree of Xexion. ATT in tibial
internal rotation did additionally tension the LCL.
When the tibial insertion of the LCL at the Wbular head
was anterior to the origin of the LCL at the lateral
epicondyle, the LCL seemed to be aligned with the
ACL (Fig. 4). In internal rotation, the LCL slackened
after more than 60 of knee Xexion.
The PC with the popliteoWbular ligament, however,
seemed to be tight over the whole range of passive Xexion and extension.
ATT under anterior tibial load
Under the 134 N anterior tibial load, ATT of the intact
knee was a mean of (SD) 2.9 (1.2), 8.4 (1.7), 10.2
(2.2) and 7.8 mm (1.9) at full extension, 30, 60 and
90 of knee Xexion, respectively (Fig. 5). After the
ACL was sectioned, the translations increased signiWcantly at all Xexion angles tested (P < 0.05). The resulting ATT under 134 N anterior tibial load was a mean
of 8.2 (1.5), 14.2 (1.3), 16.7 (2.5) and 13.4 mm
(1.6). After sectioning the LCL, the ATT was a mean
of 12.2 mm (3.1) at full extension, 23.7 mm (1.9) at
30, 18.6 mm (3.6) at 60 and 13.8 mm (2.7) at 90 of
knee Xexion (Fig. 5). This diVerence was statistically
signiWcant when compared to the ACL-deWcient knee
at full extension and 30 of Xexion (P < 0.05). Compared to the LCL-deWcient knee, sectioning the popliteoWbular ligament and the popliteus muscle did

Posterolateral-deWcient knee kinematics

in response to ATL
Posterolateral-deWcient knee kinematics in
response to combined rotatory load

further increase the ATT in response to an anterior tibial load signiWcantly at 90 of knee Xexion up to
18.3 mm (2.9) (P < 0.05).
ATT under combined rotatory load
Anterior tibial translation in response to a combined
rotatory load of 10 Nm valgus and 4 Nm internal rotation was comparable to the ATT under an anterior tibial load (Fig. 6). In response to a combined rotatory
load, the ATT for the intact knee was 2.9 (1.9), 9.9
(3.1), 8.5 (4.1) and 7.8 mm (3.3) for 0, 30, 60
and 90 of knee Xexion, respectively. The values
increased after the sectioning of the ACL up to 7.9 mm
(2.4) at 0, 14.9 mm (2.7) at 30, 9.3 mm (4.2) at
60 and 8.7 mm (3.5) at 90 (Fig. 6). The increase in
ATT at full extension and 30 of knee Xexion was statistically signiWcant (P < 0.05). The LCL was found to
be the primary stabilizer to ATT under combined rotatory load. Sectioning the LCL increased the ATT signiWcantly at full extension, 30 and 60 knee Xexion
(P < 0.05) up to 11.7 (2.1), 20.7 (3.0) and 14.2 mm
(2.5), respectively (Fig. 6). Sectioning of popliteus
tendon and popliteoWbular ligament did not increase
the ATT in response to a combined rotatory load signiWcantly (P > 0.05).
Coupled tibial internal rotation under combined
rotatory load
The internal tibial rotation in response to a combined
rotatory load of 10 Nm valgus and 4 Nm internal
rotation for the intact knee revealed a maximum internal rotation at 30 of knee Xexion (Fig. 7). The mean

123

748

Arch Orthop Trauma Surg (2007) 127:743752

Anterior tibial translation (mm)

intact

ACL def

30

LCL def

PC def

25

*
20

15

*
10
5
0

30

60

90

Knee flexion (degree)

Fig. 5 Anterior tibial translation in mm in response to anterior

tibial load of 134 N (mean SD). Asterisks indicate statistically
signiWcant diVerences (P < 0.05)

Ant erior tibial t ranslation

( mm)

intact

ACL def

30

LCL def

PC def

25
20
15
10

*
*
*

5
0

30

60

90

Knee flexion (degree)

Fig. 6 Coupled anterior tibial translation in mm under combined

10 Nm valgus and 4 Nm internal tibial torque (mean SD).
Asterisks indicate statistically signiWcant diVerences (P < 0.05)

Fig. 4 The role of the LCL in limiting the anterolateral knee

instability. Under combined rotatory load of valgus torque and
tibial internal rotation the tibial insertion of the LCL is displaced
anteriorly (a). This causes the LCL to tension and thereby limiting the anterior tibial translation (b). In ACL deWciency, this
mechanism is even more pronounced (c). The course of the LCL
is aligned with the ACL and the LCL is the primary restraint to
anterior tibial translation (d)

tibial internal rotation for the intact knee was 6.6

(1.0), 21.1 (1.7), 17.1 (1.8) and 23.6 (0.6) at
full extension, 30, 60 and 90 of knee Xexion, respectively. Subsequent sectioning of the ACL increased the
internal tibial rotation at all Xexion angles; however,
most pronounced was this increase at full extension
and 30 of knee Xexion up to 9.5 (1.3) and 24.6
(1.5), respectively (Fig. 7). At these two Xexion
angles, the diVerence between intact and ACL-deWcient knee was statistically signiWcant (P < 0.05). Sectioning the LCL had an additional eVect on the
increase of the internal tibial rotation. The internal
rotation increased up to 12.0 (1.9) at full extension,
27.2 (1.5) at 30, 22.6 (2.1) at 60 and 25.9 (2.2) at

123

In t ern al t ib ial ro t at io n ( d eg ree)

intact

ACL def

LCL def

PC def

30

**
*

25

**

20
15
10

**
*

5
0

30

60

90

Knee flexion (degree)

Fig. 7 Internal tibial rotation in degree under combined 10 Nm

valgus and 4 Nm internal tibial torque (mean SD). Single asterisk indicates statistically signiWcant diVerent when compared to
the intact knee (P < 0.05). Double asterisks indicate statistically
signiWcant diVerent when compared to the intact and the ACLdeWcient knee

90 of knee Xexion (Fig. 7). At full extension, 30 and

60 of knee Xexion this increase was statistically signiWcant (P < 0.05). Sectioning of the popliteus tendon and
the popliteoWbular ligament had no increasing eVect on
the internal tibial rotation of the knee in response to a
combined rotatory load.

Arch Orthop Trauma Surg (2007) 127:743752

Discussion
The aim of the current study was to evaluate the rotational instability of the ACL-deWcient and the posterolateral-deWcient knee under combined rotatory load.
The result supports our hypothesis that the transsection of the PLS results in increased ATT under combined rotatory load. Furthermore, the results suggest
that the LCL is the primary restraint to limit ATT
under a combined rotatory load of valgus and internal
tibial rotation. Injury to the LCL statistically increases
the rotational instability of the ACL-deWcient knee.
Subsequent sectioning of the popliteus tendon and the
popliteoWbular ligament had no signiWcant eVect on the
anterior tibial rotation or the internal tibial rotation
under combined rotatory load. However, sectioning
the PC increased ATT under an anterior load.
The role of the ACL for knee joint stability has been
described as the primary restraint to tibial anterior
translation is well understood [5, 8, 11]. However, there
is some controversy about the role of the ACL to control tibial internal rotation [14, 12, 46]. Some biomechanical studies showed that the ACL does not
contribute in controlling tibial internal rotation [11,
22]. Others could demonstrate that the cutting of the
ACL leads to an increase in internal tibial rotation [2,
3, 29, 30, 46]. These diVerences may be caused by methods used in these studies. With the UFS/robotic system
the present study shows that transsection of the ACL
leads to increased ATT under a combined load with
4 Nm internal tibial and 10 Nm valgus torque.
A study investigating three-dimensional knee kinematics in patients after ACL rupture showed that
patients with ACL injury present with signiWcant rotational instability at higher demanding activities [13].
Interestingly, this altered knee kinematics could not be
restored with a single bundle ACL reconstruction [13,
34]. It has been hypothesised that the single bundle
reconstruction cannot restore rotational instability
because the posterolateral bundle is not restored with
this technique [34, 42]. This data are in accordance with
the results presented by Tashman et al. [39]. Using a
250 frames/s stereoradiographic system, these authors
investigated the three-dimensional kinematics of
patients after ACL reconstruction during simulated
downhill running on a treadmill. Even though the ATT
was similar for the reconstructed and uninjured limbs,
the tibial rotation of the reconstructed knees was not
restored to normal [39].
Several biomechanical studies have shown that the
posterolateral bundle plays a role in controlling rotational stability [12, 43, 46]. Therefore, several authors
recommend double bundle ACL reconstruction to

749

achieve better rotational stability [6, 12, 42, 44, 4648].

The results of the present study underline the role of
the ACL in controlling rotational stability, but they
also show the important role of the LCL in rotational
instability. At 60 of Xexion, the LCL was the primary
restraint in limiting ATT against a combined rotatory
and valgus load (simulated pivot shift test). We
hypothesize that not only the technique of ACL reconstruction should be considered to be a cause for residual rotational instability after single bundle ACL
reconstruction. It seems likely that untreated injuries
of the PLS may also responsible for residual rotational
instability. It has been shown that the mechanism of
many ACL injuries has a rotational component [30,
36]. It is hard to believe that rotational forces are able
to stretch the ACL to failure, leaving the PLS intact.
This theory is underlined by the Wndings of Stubli and
Birrer [38], which show a high incidence of associated
injuries of the PLS in patients with ACL rupture.
To assess ALRI clinically, the pivot shift test can be
used [23, 37]. This test has been reported to correlate
with instability symptoms, reduced sports activity and
meniscal damage [18, 21]. Recent clinical studies have
documented a correlation between surgical outcome
after ACL reconstruction and the presence of pivot
shift test [15, 20]. In the current study a simulated pivot
shift was applied by the robotic/UFS testing system. A
load application of 10 Nm valgus and 4 Nm internal
tibial torque has been used to simulate a pivot shift
test. This load has been used previously for the evaluation of the knee kinematics in response to a pivot shift
test [18, 33, 42, 46].
It has been shown that the pivot shift test is extremely
variable, both between examiners and between patients
[1, 4, 30]. The results of the present study suggest that
additional injuries to the LCL might be a factor, which
explains the diVerences in pivot shift between patients. It
seems to be necessary to develop a standardized method
to assess rotational instability of the ACL-injured knee.
This method would help to detect possible associated
injuries of the peripheral ligamentous structures. An
objective method for grading or to classify the ALRI
could be of clinical relevance. A rough classiWcation
would be ALRI with or without involvement of the PLS.
The observations of the dissection study explain the
results of the biomechanical experiments. In internal
tibial rotation, the tibial insertion of the LCL is moved
anteriorly and thereby tightened. The course of the
LCL is aligned with the ACL and primary restraint to
withstand ATT under combined rotatory load. With an
injury to the LCL this mechanism may be insuYcient
thereby further increasing the rotational tibial laxity
(Fig. 4).

123

750

In the current study, we did also evaluate the eVect

of cutting the PC (popliteus tendon and popliteoWbular
ligament) on the resulting knee kinematics. The unexpected results show a signiWcant increase in ATT at 90
under anterior tibial load (P < 0.05). In the posterior
aspect of the knee, the popliteus tendon wraps around
the lateral femoral condyle. At a Xexion angle of 90,
the femoral origin of the popliteus tendon is rolled
anterior and proximal to the lateral epicondyle,
thereby tensioning the Wbres of the popliteus tendon.
In an ACL-deWcient knee, this tensioning of the tendon wrapped around the femoral condyle may contribute to restrain ATT under anterior tibial load.
Using a robotic/UFS testing system, Kanamori et al.
[17] investigated the forces in the MCL and PLS in the
intact and the ACL-deWcient knee. In the ACL-deWcient knee, the in situ forces of the PLS under 134 N
anterior tibial load were reported to be Wve times as
high as in the intact knee condition. The authors concluded that, although both the MCL and PLS play only
a minor role in resisting anterior tibial loads in the
intact knee, they become signiWcant after ACL injury
[17]. The results of the current study strongly resemble
these Wndings and provide the kinematic data for the
previously published in situ forces [17].
During the last decades there has been a tremendous eVort to improve arthroscopic techniques for
ACL reconstruction. The latest development is the
double bundle ACL reconstruction technique, which
aims to restore rotational knee stability. The results of
the present study suggest that this technique might be
able to restore knee kinematics only in knees with isolated ACL injury or in knees with minor injury of the
PLS. We hypothesize that in patients with higher
degree of ALRI an additional extraarticular procedure
might be necessary. In the literature diVerent reconstruction procedures for the PLS can be found such as
the Larson sling (reconstruction of the LCL and the
popliteoWbular ligament) [7], augmentation of the LCL
with a strip of the biceps tendon [9] or popliteus
bypasses (graft between tibia and femur) [28]. Based
on the results of the present study, the aim of these
reconstruction techniques should be a technique that
primarily restores the LCL. However, more biomechanical research is needed to evaluate the best peripheral reconstruction technique for ALRI.
Some limitations apply to the current study. First,
we investigated the resulting knee kinematics under
anterior tibial load and coupled rotatory load. Theoretically, the application of an axial compression force
may have additional eVects on the ATT. However,
adding another testing condition would have signiWcantly increased the duration of testing thereby causing

123

Arch Orthop Trauma Surg (2007) 127:743752

the tissue to exsiccate. Secondly, the age of the human

cadaver knees as used in this study may not represent
the typical age for patients suVering ACL rupture. The
scarcity of human donors makes it impossible to test
young human knees in numbers to provide statistically
signiWcant conclusions. Additionally, the test set-up did
not incorporate muscle activity. Theoretically, contraction of muscles such as the hamstring tendons or the
popliteus would have an important eVect on the resulting knee kinematics. However, a study evaluating
EMG signals of intact and ACL-deWcient knees
observed only minor popliteus EMG signal diVerences
after ACL rupture [41]. The authors concluded that
the popliteus muscle does not actively contribute to
rotational instability such as the pivot shift.
In conclusion, the current study shows the importance of the LCL for the resulting knee kinematics
under combined rotatory load. The LCL was found to
be the primary stabilizer at 60 under a combined rotatory load in limiting the ATT. When the results of the
current in vitro study are transferred to the clinical setting, they may suggest that ACL rupture results in an
anterior tibial instability as well as an ALRI. However,
injury to the LCL signiWcantly increases ALRI. It
would be of clinical relevance to distinguish ALRI with
or without the involvement of the PLS. In patients with
severe rotational instability and rupture of the ACL
and LCL, the necessity of additional extraarticular stabilization procedures needs to be discussed. Further
research is needed to elucidate the eVect of combined
ACL and posterolateral reconstruction when the primary restraints to internal tibial rotation are torn.
Acknowledgments Funding of the robotic/UFS testing system
was received by a grant from the German Speaking Association
for Arthroscopy (AGA) and is deeply appreciated by the authors.

References
1. Amis A, Bull AMJ, Lie DT (2005) Biomechanics of rotational instability and anatomic anterior cruciate ligament
reconstruction. Oper Tech Orthop 15:2935
2. Amis AA, Scammell BE (1993) Biomechanics of intra-articular and extra-articular reconstruction of the anterior cruciate ligament. J Bone Joint Surg Br 75:812817
3. Andersen HN, Dyhre-Poulsen P (1997) The anterior cruciate
ligament does play a role in controlling axial rotation of the
knee. Knee Surg Sports Traumatol Arthrosc 5:145149
4. Bull AMJ, Amis AA (1998) The pivot shift phenomenon: a
clinical and biomechanical perspective. Knee 5:141158
5. Butler DL, Noyes FR, Grood ES (1980) Ligamentous restraints to anteriorposterior drawer in the human knee. J
Bone Jt Surg Am 62:259270
6. Cha PS, Brucker PU, West RV, Fu FH (2005) Arthroscopic
double bundle anterior cruciate ligament reconstruction: an
anatomic approach. Arthroscopy 21:12751281

Arch Orthop Trauma Surg (2007) 127:743752

7. Covey DC (2001) Current concepts review: injuries of the
posterolateral corner of the knee. J Bone Joint Surg 83:106
118
8. Daniel DM, Stone ML, Sachs R, Malcom L (1985) Instrumented measurement of anterior knee laxity in patients with
acute anterior cruciate ligament disruption. Am J Sports Med
13:401407
9. Fanelli GC (2005) Surgical reconstruction for acute posterolateral injury of the knee. J Knee Surg 18:157162
10. Freedman KB, DAmato, NedeV DD (2003) Arthroscopic
anterior ligament reconstruction: a metaanalysis comparing
patellar tendon and hamstring tendon autografts. Am J
Sports Med 31:211
11. Fukubayashi T, Torzilli PA, Sherman MF, Warren RF (1981)
An in vitro biomechanical evaluation of anterior-posterior
motion of the knee Tibial displacement, rotation, and torque.
J Bone Joint Surg Am 64:258264
12. Gabriel MT, Wong EK, Woo SL, Yagi M, Debski RE (2004)
Distribution of in situ forces in the anterior cruciate ligament
in response to rotatory loads. J Orthop Res 22:8589
13. Georgoulis AD, Papadonikolakis A, Papageorgiou CD, Mitsou A, Stergiou N (2003) Three-dimensional tibiofemoral
kinematics of the anterior cruciate ligament-deWcient and
reconstructed knee during walking. Am J Sports Med 31:75
79
14. Hughston JC, Andrews JR, Cross MJ, Moschi A (1976) ClassiWcation of knee ligament instabilities. Part I. The medial
compartment and cruciate ligaments. J Bone Joint Surg Am
58:159172
15. Jonsson H, Riklund-Ahlstrom K, Lind J (2002) Positive pivot
shift after ACL reconstruction predicts later osteoarthrosis:
63 patients followed 59 years after surgery. Acta Orthop
Scand 75:594599
16. Kanamori A, Sakane M, Zeminski J, Rudy TW, Woo SL
(2000) In-situ force in the medial and lateral structures of intact and ACL-deWcient knees. J Orthop Sci 5:567571
17. Kanamori A, Zeminski J, Rudy TW, Li G, Fu FH, Woo SL
(2002) The eVect of axial tibial torque on the function of the
anterior cruciate ligament: a biomechanical study of a simulated pivot shift test. Arthroscopy 18:394398
18. Kaplan N, Wickiewicz TL, Warren RF (1990) Primary surgical treatment of anterior cruciate ligament ruptures. A longterm follow-up study. Am J Sports Med 18:35488
19. Kim YC, Chung IH, Yoo WK, Shu JS, Kim SJ, Park CI (1997)
Anatomic and magnetic resonance imaging of the posterolateral structures of the knee. Clin Anat 10:397404
20. Kocher MS, Steadman JR, Briggs KK, Sterett WI, Hawkins
RJ (2004) Relationships between objective assessment of ligament stability and subjective assessment of symptoms and
function after anterior cruciate ligament reconstruction. Am
J Sports Med 32:629634
21. Kujala UM, Nelimarkka O, Koskinen SK (1992) Relationship between the pivot shift and the conWguration of the lateral tibial plateau. Arch Orthop Trauma Surg 111:228229
22. Lane JG, Daniel D (1994) The anterior cruciate ligament in
controlling axial rotation. Am J Spots Med 22:289293
23. Larson RL (1983) Physical examination in the diagnosis of
rotatory instability. Clin Orthop Relat Res 172:3844
24. Lenschow S, Zantop T, Weimann A, Lemburg T, Raschke M,
Strobel M, Petersen W (2006) Joint kinematics and in situ
forces after single bundle PCL reconstruction: a graft placed
at the center of the femoral attachment does not restore normal posterior laxity. Arch Orthop Trauma Surg 126:253259
25. Lipke JM, Janecki CJ, Nelson CL, McLeod P, Thompson C,
Thompson J, Haynes DW (1981) The role of incompetence of
the anterior cruciate and lateral ligaments in anterolateral

751

26.

27.

28.

29.

30.

31.

32.
33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

and anteromedial instability. A biomechanical study of cadaver knees. J Bone Joint Surg Am 63:954960
Mannel H, Marin F, Claes L, Durselen L (2004) Anterior cruciate ligament rupture translates the axes of motion within
the knee. Clin Biomech 19:130135
Markolf KL, Mensch JS, Amstutz HC (1976) StiVness and
laxity of the kneethe contributions of the supporting structures. A quantitative in vitro study. J Bone Jt Surg Am
58A:583594
Mller W (1982) Das Knie. Form, Funktion und ligamentre
Wiederherstellungschirurgie. Springer, Berlin Heidelberg
New York
Nordt WE 3rd, LotW P, Plotkin E, Williamson B (1999) The
in vivo assessment of tibial motion in the transverse plane in
anterior cruciate ligament-reconstructed knees. Am J Sports
Med 27:611616
Olson EO, Myklebust G, Engebretsen L, Bahr R (2004) Injury mechanisms for anterior cruciate ligament injuries in team
handball. A systematic video analysis. Am J Sports Med
32:10021012
Petersen W, Lenschow S, Weimann A, Strobel MJ, Raschke
MJ, Zantop T (2006) Importance of femoral tunnel placement in double-bundle posterior cruciate ligament reconstruction: biomechanical analysis using a robotic/universal
force-moment sensor testing system. Am J Sports Med
34:456463
Petersen W, Tillmann B (1998) Collagenous Wbril texture of
the human knee joint menisci. Anat Embryol 197:317324
Petersen W, Tretow H, Weimann A, Herbort M, Fu FH, Raschke MJ, Zantop T (2006) Biomechanical evaluation of two
techniques for double bundle acl reconstruction using a robotic/UFS testing system. Am J Sports Med (in press)
Ristanis S, Stergiou N, Patras K, Vasiliadis HS, Giakas G,
Georgoulis AD (2005) Excessive tibial rotation during highdemand activities is not restored by anterior cruciate ligemant reconstruction. Arthroscopy 21:13231329
Rudy TW, Livesay GA, Xerogeanes JW, Woo SLY (1994) A
combined robotics/UFS approach to measure knee kinematics and determine in-situ ACL forces. ASME Adv Bioeng
28:287288
Seering WP, Piziali RL, Nagel DA, Schurman DJ (1980) The
function of the primary ligaments of the knee in varusvalgus
and axial rotation. J Biomech 13:785794
Slocum DB, Larson RL (1968) Rotatory instability of the
knee. Its pathogenesisand a clinical testto demonstrate its
presence. J Bone Joint Surg 50:211225
Staubli HU, Birrer S (1990) The popliteus tendon and its fascicles at the popliteal hiatus: gross anatomy and functional
arthroscopic evaluation with and without anterior cruciate
ligament deWciency. Arthroscopy 6:209220
Tashman S, Collon D, Anderson K, Kolowich P, Anderst W
(2004) Abnormal rotational knee motion during running after anterior cruciate ligament reconstruction. Am J Sports
Med 32:974983
Vogrin TM, Hoher J, Aroen A, Woo SL, Harner CD (2000)
EVects of sectioning the posterolateral structures on knee
kinematics and in situ forces in the posterior cruciate ligament. Knee Surg Sports Traumatol Arthrosc 8:9398
Weresh MJ, Gabel RH, Brand RA, Tearse DS (1997) Popliteus function in ACL-deWcient patients. Scand J Med Sci
Sports 7:1419
Woo SL, Kanamori A, Zeminski J, Yagi M, Papageorgiou C,
Fu FH (2002) The eVectiveness 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 84:907914

123

752
43. Yagi M, Wong EK, Kanamori A, Debski RE, Fu FH, Woo
SL (2002) Biomechanical analysis of an anatomic anterior
cruciate ligament reconstruction. Am J Sports Med 30:660
666
44. Yasuda K, Kondo E, Ichiyama H, Tanabe Y, Tohyama H
(2006) Clinical evaluation of anatomic double-bundle anterior cruciate ligament reconstruction procedure using hamstring tendon grafts: comparisons among 3 diVerent
procedures. Arthroscopy 22(3):240251
45. Zantop T, Lenschow S, Lemburg T, Weimann A, Petersen W
(2004) Soft-tissue graft Wxation in posterior cruciate ligament
reconstruction: evaluation of the eVect of tibial insertion site

123

Arch Orthop Trauma Surg (2007) 127:743752

on joint kinematics and in situ forces using a robotic/UFS
testing system. Arch Orthop Trauma Surg 124:614620
46. Zantop T, Herbort M, Weimann A, Fu FH, Petersen W
(2006) The role of anteromedial and posterolateral bundle of
the ACL in anterior tibial translation and internal rotation.
Am J Sports Med (in press)
47. Zantop T, Petersen W, Fu FH (2005) Anatomy of the anterior cruciate ligament. Oper Tech Orthop 15:2028
48. Zantop T, Petersen W, Sekiya JK, Musahl V, Fu FH (2006)
Anterior cruciate ligament anatomy and function relating to
anatomical reconstruction. Knee Surg Sports Traumatol
Arthrosc. 2006 Aug 5; [Epub ahead of print]