Knee Surg Sports Traumatol Arthrosc (2017) 25:1015–1023

DOI 10.1007/s00167-017-4494-x

KNEE

Anterolateral knee biomechanics

Andrew A. Amis1,2

Received: 7 November 2016 / Accepted: 21 February 2017 / Published online: 15 March 2017
© The Author(s) 2017. This article is an open access publication

Abstract This article reviews the evidence for the roles with the knee held in flexion and the tibia in external rota-
of the anterolateral soft-tissue structures in rotatory stabil- tion—presumably to try to avoid ‘stretching-out’ of the ten-
ity of the knee, including their structural properties, isom- odesis [23]. However, as may be imagined, such treatment
etry, and contributions to resisting tibial internal rotation. could cause stiffness, and there was also a suspicion that
These data then lead to a biomechanical demonstration that it led to degenerative changes of the lateral compartment
the ilio-tibial band is the most important structure for the of the knee. Thus, these procedures fell from use when
restraint of anterolateral rotatory instability. arthroscopy led to better and more accurate intra-articular
Level of evidence V. ACL reconstruction.
Although intra-articular ACL reconstruction led to reli-
Keywords ACL · Anterolateral rotatory instability · able restoration of anterior tibiofemoral joint laxity to nor-
Tibial internal rotation · Ilio-tibial band · Biomechanics mal (or close to normal) in most cases, it was also widely
reported that there remained a small percentage of patients
whose knees continued to feel unstable [4, 15, 19]. The
Introduction symptoms reported were what, in the clinic, matched the
observation of a trace of remaining pivot-shift instability,
In the past, it was common practice to treat the instabil- the so-called ‘pivot-glide’. It was realised that this pivot-
ity that followed anterior cruciate ligament (ACL) injury glide represented a residual rotatory laxity. It is possible to
by means of a lateral extra-articular tenodesis, and proce- measure the small transient instability during intra-opera-
dures such as those described by Lemaire and Combelles tive pivot-shift testing post ACL reconstruction [6]. It has
[30], Ellison [13] and Galway and MacIntosh [17] were also been shown that increasing the ACL graft tension can
used widely. However, the line of action of a lateral con- cause over-constraint of tibial anterior translation laxity,
struct was far away from that of the ruptured ACL, and so yet still leave the rotational laxity greater than in the native
it was not surprising that normal knee mechanics was not knee [31]. Recognition of this shortcoming in some knees
restored. A further aspect is that most authors describing has led to many studies on intra-articular ACL reconstruc-
the extra-articular procedures used a post-operative regime tion in recent years, in which there have been explorations
which included immobilisation in plaster of Paris, usually of factors such as graft tension, femoral tunnel position,
single-bundle versus double-bundle grafts, in attempts to
abolish the troublesome residual rotational laxity.
* Andrew A. Amis There has recently been a renewed realisation that rota-
[email protected] tional laxity may result from injuries to the peripheral
1 structures. This was prompted in part by an anatomy paper
Biomechanics Group, Mechanical Engineering Department,
Imperial College London, London SW7 2AZ, UK describing an anterolateral ligament (ALL), which received
2 much publicity [10], but it has actually led to a return to
Musculoskeletal Surgery Group, Department of Surgery
and Cancer, Imperial College London School of Medicine, consideration of the role of the anterolateral peripheral
London W6 8RF, UK

13
Vol.:(0123456789)
1016 Knee Surg Sports Traumatol Arthrosc (2017) 25:1015–1023

soft-tissue structures and of their repair or reconstruc- the greater mobility and lack of constraint of the lateral
tion; this paper examines the biomechanical role of these compartment, due to both differences of meniscocapsular
structures. attachments—the lateral meniscus being more mobile than
the medial—and of articular geometry—the medial tibial
plateau being concave, which locates the femoral condyle,
Anterior–posterior and internal–external rotation while the lateral plateau is flat or even convex in the sagittal
laxity of the knee plane.

It is well-known that the ACL is the primary restraint of Effect of ACL injury in combination with anterolateral
tibial anterior translation [8], where ‘primary restraint’ injury
means that, for a given anteriorly directed force applied
to the proximal tibia, most of the restraint (>50%) is from If the mechanism of ACL rupture has included a large
the ACL. It follows that rupture of the ACL allows a large anterior translation of the lateral aspect of the tibial pla-
increase of tibial anterior translation laxity. That increase of teau, it follows that structures which cross the anterolateral
laxity is greatest around 20°–40° knee flexion, and so ACL joint line and are oriented in a direction that will resist that
injury is best diagnosed near knee extension. motion (that is: slanting from relatively posterior on the
The role of the ACL in resisting tibial internal–exter- femur, across the joint to relatively anterior on the tibia)
nal rotation has been less clear, and there have been papers are likely also to be stretched or ruptured—a ‘combined’
which differ as to whether an isolated ACL rupture leads injury. This has been most clearly demonstrated by the
to any significant increase in tibial rotational laxity: some Segond fracture, when the anterolateral capsular structures
did find a small and significant increase in rotational lax- cause an avulsion of a flake of bone from the rim of the
ity [33, 59], while others did not [11, 28]. This controversy tibial plateau [18]. The resulting increase in laxity of the
probably follows from the ACL being sited centrally over internally rotated knee is diagnostic for anterolateral injury
the tibial plateau, and so it cannot have any large moment [53].
arm around the axis of tibial internal–external rotation, A combined injury of the ACL plus anterolateral struc-
about which to resist any torque when the knee and ACL tures leads to increased tibial anterior translation, and also
are intact or reconstructed. The axis of tibial rotation is increased tibial internal rotation laxity, when compared to
close to the centre of the tibial plateau, usually around the the changes arising from an isolated ACL injury. This is
medial spinous process [24]. It is usually the case, however, reflected in the movements of the bones during the pivot-
that an injury mechanism which ruptures the ACL includes shift examination. Bedi et al. [5] have shown that the move-
at least some rotational component of loading and, hence, ment of the lateral tibial plateau is much greater than that
of bone–bone excursion. That is shown clearly by the pres- of the medial, the difference in motion resulting from the
ence of bone bruises on MRI, when it is common to find internal rotation of the tibia as it subluxes anteriorly. A
a bone bruise near to the centre of the distal aspect of the range of values for the movements of the medial and lateral
lateral femoral condyle, which matches the point where it condyles have been published [32]. This observation has
impacted against the posterior rim of the lateral tibial pla- led to a method to grade the pivot-shift which visualises
teau during the injury [54]. This shows clearly that the lat- the movements of the lateral aspect of the knee [36]. The
eral aspect of the tibia had moved anteriorly from under the overall effect is that the anterolateral aspect of the proximal
femur during the injury. That is not usually found to a simi- tibia moves anteriorly as the internal rotation adds to the
lar degree in the medial compartment of the knee [32], and anterior translation movement during clinical examination,
so it demonstrates that the injury had included a substantial and that is what has long been recognised and described as
component of tibial internal rotation. ‘anterolateral rotatory instability’ (ALRI).
The internal rotation which accompanies ACL injury It should be remembered, however, that the pattern of
mechanisms is a part of the normal behaviour of the intact pathological laxity is extremely variable between knees,
knee. Imposing a force which translates the proximal tibia so that some respond to the pivot-shift test with a large
anteriorly also induces a coupled internal rotation. (A cou- rotational component of motion, while others yield a pre-
pled motion is one which occurs automatically in response dominance of anterior–posterior translation [6]. If there is
to inducing a motion in a different degree of freedom of the a large translation of both compartments, it suggests that
joint.) Data in the literature suggest that the coupled tibial there has been damage also in the medial compartment,
internal rotation is in the range of 3°–10° when examined such as posteromedial menisco-capsular lesion.
by hand [2, 39]; this motion may be larger under func- Experiments in vitro, which allow the loads to be con-
tional loading. It is usually accepted that the internal rota- trolled accurately, have found that not only does a lesion
tion which accompanies anterior translation results from of the anterolateral structures [which includes the capsular

13
Knee Surg Sports Traumatol Arthrosc (2017) 25:1015–1023 1017

structures and the more proximal attachments of the ilio- tensile strength of 628 ± 35 N, reached a failure stress
tibial band (ITB)] lead to increased laxity, it also leads to a of 79 ± 5 MPa, and had a tensile extension stiffness of
persistence of some abnormal laxity after an isolated ACL 614 ± 271 N/mm. Thus, the ITB has been reported to be
reconstruction: both anterior translation and internal rota- stronger and stiffer than any of the measurements reported
tion remain greater than when the knee was intact [21]. for the isolated ALL or the ALL complex which also incor-
This observation provides a logic for using some addi- porates deep capsule-osseous fibres of the ITB. Another
tional anterolateral procedure to restore native knee laxity study [45] found that a 10 mm wide strip of the ITB was
behaviour. 50% stronger than a 20 mm wide area of the anterolateral
joint capsule.
Structural properties of anatomical structures These structural properties along the length of the ITB
at the anterolateral aspect of the knee do not necessarily mean that it controls tibial anterolateral
subluxation when the knee is nearly fully extended, because
Recent interest has focussed on the anterolateral ligament its axis is not aligned to resist that displacement, whereas it
(ALL), which has been suggested to be the structure which is when the knee is in flexion. However, the ITB is tethered
avulses the bone fragment in a Segond fracture [9, 18]. to the lateral aspect of the femur via the Kaplan’s fibres and
However, the literature has included differing descriptions lateral intermuscular septum [58], and also has an anterior
of the ALL, and it has not even been found in all knees by expansion, the lateral anterior aponeurosis, which sweeps
some authors. The implication is that it is a relatively small anteriorly from the ITB, linking it to the lateral aspect of
structure, and so it is unlikely to be either strong or stiff. the patella, and also to the patellar tendon [35]. Thus, the
The result of these differences of opinion is a range of data ITB is set up to resist anterolateral subluxation of the tibia.
on the structural properties of the ALL. The strengths of other stabilising structures at the lat-
Zens et al. [64] found that the isolated ALL had an ulti- eral aspect of the knee have also been reported: Sugita and
mate tensile strength of 50 ± 15 N, at a strain of 36 ± 4%. Amis [57] found tensile strengths for the lateral (fibular)
With a mean cross-sectional area of only 1.54 mm2, the collateral ligament (LCL): 309 ± 91 N, and the popliteofib-
ultimate tensile stress was 33 ± 4 MPa, and the overall ular ligament 186 ± 65 N. LaPrade et al. [29] found similar
stiffness was 4.2 N/mm extension. The specimens failed at tensile strengths for the lateral (fibular) collateral ligament
mid-substance; they did not induce a bone avulsion. (LCL): 295 ± 96 N; popliteofibular ligament 298 ± 144 N;
Other authors have reported greater strength for the and popliteus tendon 700 ± 232 N.
ALL, but this observation relates to the difficulty and dif-
ferences in anatomical interpretation of what, exactly, Length change patterns of anatomical structures
is the ALL? While Zens et al. [64] separated an isolated at the anterolateral aspect of the knee
ALL structure, it is clear that the literature has included
work where the deep capsulo-osseous layer of the ITB and The literature of ACL anatomy and reconstruction has
adjacent capsular tissue have been taken to be part of ‘the shown clearly that, because ligaments attach to bone over
ALL’, and it may be more appropriate to view the entire an area and not just at a point, there will be a spectrum of
anterolateral capsule-ligamentous complex as being the tightening–slackening behaviour across the width of the
structure which should be reconstructed, rather than the structure, due to each fibre having a different moment arm
isolated ALL. Thus, Kennedy et al. [25] reported that the about the axis of flexion–extension. It has also been shown
ALL had a tensile strength of 175 N (139–211 N 95% CI) that the isometry depends principally on the location of the
and stiffness 20 N/mm (16–25) with a more substantial femoral graft attachment, and not the tibial [3]. These prin-
structure than that shown by Zens et al. [64]. ciples also apply at the lateral aspect of the knee. Although
This strength is similar to that of the deep fibres of the posterior part of the lateral femoral condyle is close to
the medial collateral ligament (MCL), reported to be being spherical, that does not mean that the axis of rota-
194 ± 82 N (mean ± SD) [47], which is sheltered by the tion is fixed, because knee function includes components of
superficial MCL, the way that the ALL is protected by the both rolling and sliding of the tibiofemoral joint surfaces
ITB. The deep MCL has a significant role in restraint of during flexion–extension, although a trans-epicondylar axis
tibial external rotation [48], mirroring the role of the ALL is a good approximation.
in restraining internal rotation, and may be injured by a tib- The lateral (fibular) collateral ligament (LCL) attaches
ial external rotation mechanism [38]. close, and just posterior, to the lateral femoral epicondyle.
In comparison, Noyes et al. [41] reported that a graft As the condyle rolls posteriorly, it also moves ‘down-
taken from the distal part of the ITB, 18 ± 2 mm wide, had hill’ on the increasing posterior slope of the lateral tibial
a mean tensile strength of 1068 N, and that a more proxi- plateau and so the LCL slackens significantly with knee
mal strip of the fascia lata 16 ± 1 mm wide had a mean flexion [57]. Taking this as a guide, it may be imagined

13
1018 Knee Surg Sports Traumatol Arthrosc (2017) 25:1015–1023

that a structure attaching anteriorly will be stretched as and so be unable to resist the anterolateral subluxation of
the knee flexes, and that a more-posterior attachment the pivot-shift.
will cause more slackening. These length change patterns An important practical finding in the work of Kittl et al.
were mapped by Sidles et al. [52]. They showed that, if [26] was that if a lateral tenodesis graft was taken from the
the area on the tibia near to Gerdy’s tubercle was taken area of Gerdy’s tubercle and then routed deep to the LCL
as the datum point (which is appropriate when consider- for attachment to the femur, as described in the Galway and
ing the behaviour of a lateral extra-articular tenodesis), MacIntosh tenodesis [17, 23], then the proximal femoral
then the matching area on the femur closest to isomet- attachment of the LCL acted as a pulley and kept the graft
ric behaviour was along the posterior/lateral edge of the behaviour close to isometric, as described above for the
femoral metaphysis. ALL, with graft elongation with knee extension, a desirable
The isometry of the ALL was measured by Dodds et al. characteristic for stabilising the knee. It was found that it
[12], by threading a suture along the ligament fibres, attach- did not matter where along the lateral condylar ridge the
ing it to the moving tibia and then measuring the changes graft was attached, the isometric behaviour remained very
of the separation distance between the bone attachments similar. This finding allows the surgeon to choose where to
using a transducer. It was shown that, with the tibia follow- place the fixation along the lateral femur, possibly to avoid
ing its ‘neutral’ path of motion (that is: without any control interacting with an ACL graft tunnel, for example.
of tibial internal–external rotation while the knee was flex-
ing/extending) the ALL was close to being isometric from Role of structures to resist ALRI and tibial internal
0 to 60° knee flexion, after which the length reduced, so rotation
the ALL slackened in deeper flexion. If the tibia was held
in internal rotation, the ALL was longer across the arc of Recent work has found that, in the presence of an injury
motion, particularly in the flexed knee, showing that it was which includes damage to both the ACL and the ante-
stretched and, therefore, would be resisting tibial internal rolateral structures, an isolated ACL reconstruction left a
rotation. This ‘close to isometric’ behaviour was associ- residual abnormal rotational laxity of the tibiofemoral joint
ated with the femoral attachment being identified in that [21]. In response, biomechanical studies have examined the
work as proximal and slightly posterior to the lateral epi- effects of cutting individual anatomical structures around
condyle. Although the exact measurement varied between the knee, to demonstrate their roles in constraint of tibial
knees, the mean position was 8 ± 5 mm SD proximal and rotation. When the effect of an isolated ACL deficiency is
4 ± 5 mm posterior to the tip of the epicondyle. A similar studied, the changes in tibial internal rotation laxity may be
femoral attachment position has been reported in another small enough that it is difficult to find a significant change
study [25], at a mean of 7 mm from the lateral epicondyle. in tibial rotation in response to a simulated pivot-shift test,
In contrast to the above studies, Zens et al. [65] reported between intact, ACL-deficient, and ACL-reconstructed
that the ALL attached antero-distal to the lateral epicon- states [11]. Because the pivot-shift test usually involves
dyle, and this difference of anatomical identification led to application of a valgus moment to the knee, there is a ten-
the ALL being reported to be stretched by knee flexion and dency for it to induce a coupled tibial internal rotation, due
slack near extension in that study, a characteristic which to the compressive load in the lateral compartment and
would not help to stabilise the knee when weight-bearing. then the associated tendency of the femoral lateral condyle
Noting the differing findings of isometric behaviour, to slip ‘downhill’ down the posterior slope of the lateral
isometry of anatomical structures and of several of the tibial plateau [42].
points published for lateral extra-articular tenodeses was Changes in rotational laxity are clearer when there is a
measured by Kittl et al. [26], using similar methods to the combination of ACL plus peripheral lesions. Thus, Spencer
work by Dodds et al. [12]. The results confirmed that the et al. [56] showed that the mean coupled internal rotation
ALL was close to isometric, with slight tightening (that is during a simulated pivot-shift test increased from 5° when
lengthening of the distance between the bone attachments) the knee was intact to 7° after isolated ACL transection, to
as the knee was extended. This behaviour was identified 9° when the ALL was cut. Other work [55] reported greater
as being most desirable for a lateral tenodesis or similar coupled internal rotation during the simulated pivot-shift
procedure, because it would need to be tight to resist the test, from 18° intact, to 20° ACL cut, 26° ACL + ALL cut,
pivot-shift subluxation close to knee extension (the weight- to 35° with ACL + ALL + ITB cut. They found greater lax-
bearing posture), yet slacken in knee flexion so as not to ity when an isolated 2 N m tibial internal rotation torque
over-constrain the increased tibial internal–external rota- was applied: 30–32–35–42°, for the same cutting sequence,
tion laxity of the native undamaged knee. Conversely, if a at 20° knee flexion. Another study [46] found that cutting
lateral tenodesis were to be fixed antero-distal to the lateral the ACL led to increased tibial internal rotation from 0 to
femoral epicondyle, it would slacken as the knee extended, 45° knee flexion, but only up to 2° increases. In contrast,

13
Knee Surg Sports Traumatol Arthrosc (2017) 25:1015–1023 1019

adding the ALL cut led to a further 3°, across 0 to 120° study [43] found that the ACL resisted 35% of a 5 N m
knee flexion. These papers showed clearly that cutting the internal rotation torque near knee extension, falling to
ALL led to significant increases in tibial internal rotation 20% at 60° flexion, with the converse trend for the ALL:
laxity, which means that the ALL must act to resist tibial rising from 5% near extension to 40% at 60° knee flexion.
internal rotation. In contrast to these studies, it has also Unfortunately, this experiment was performed after the
been reported that adding an ALL lesion to an ACL-defi- ITB had been removed, and so the percentage contribu-
cient knee did not lead to measureable increases in either tions would have been over-stated, because the ITB has
anterior translation or internal rotation during clinical test- since been shown to be the primary restraint (see below).
ing of whole cadavers [50]. Thein et al. [61] found that the ALL had resisted <10 N
However, although changes of internal rotational laxity force during a tibial anterior translation test, and <17 N
are what are examined and may be diagnostic for specific during the simulated pivot-shift test; thus, they concluded
ligament injuries during a clinical evaluation of a knee, the that the ALL could only be considered a secondary
resulting data do not tell us how much of the applied load restraint.
(internal rotation torque in the case of testing for ALRI) The study by Kittl et al. [27] measured the reductions
has been resisted by each of the relevant anatomical struc- in tibial internal rotation torque during a sequence of cuts
tures. To do that—to discover which structures are the pri- of the anterolateral structures: the superficial ITB, the
mary restraints—a different type of test has to be used, in deep/capsule-osseous fibres of the ITB; the ALL, then
which the changes of load are measured when a test that the anterolateral joint capsule. This was done for knees
displaces the tibia a constant amount is repeated after cut- with the ACL intact, and also for ACL-deficient knees.
ting a structure. The main finding of this study was that it was the ITB
The method of sequential cutting of anatomical struc- which was the primary restraint to tibial internal rota-
tures was introduced by the work of Butler et al. [8] which tion (Fig. 1). Also, contrary to the recent opinion based
led to identification of the cruciate ligaments as being the on anatomic observations and laxity changes, neither
primary restraints to tibial anterior and posterior transla- the extracapsular ALL nor the anterolateral capsular
tion. They held the femur and tibia in a materials testing structures offered significant resistance to tibial internal
machine and displaced the tibia a known distance and rotation from 0 to 90° knee flexion. At full knee exten-
recorded the force required. They then repeated that ante- sion, where it becomes tight, the ACL was a significant
rior translation after cutting the ACL, and the reduction in restraint in the intact knee, but that contribution fell rap-
force was what had been resisted by the ACL. That simple idly with knee flexion, as the more-posterior fibres of the
method, however, held the bones rigidly and prevented sec- ACL slackened. In an ACL-deficient knee, the resistance
ondary movements. In particular, release of tibial internal
rotation was found to allow 30% more anterior translation
[16]. So test rigs with many sliding and rotating bearings
were developed, to allow multiple degrees of freedom of
motion when the tibia was displaced, and led to detailed
data on the roles of the ligaments (for example, for the PCL
bundles [44]). The problem then was that the repetition of
the drawer force could not ensure that the tibia followed the
same path of motion (the cut structure might have altered
the internal rotation during a draw test, for example), and
so the forces in the remaining ligaments would change,
affecting the data on their contributions. This dilemma was
solved by the introduction of robotic tests of knees, because
the robot would record the path of motion during a test of
the intact knee, and could reproduce it precisely after a lig-
ament had been cut [49].
There have been several in vitro studies of inter-
nal rotation of the knee which used robotic technology.
One study [46] used the robot in order to measure the Fig. 1  Mean contribution (%) of tested structures in restraining a
increases of laxity when the ACL and then ALL were cut, 5 N m internal rotation torque at 0°, 30°, 60°, and 90°. sITB superfi-
cial layer of the iliotibial band, dITB deep and capsulo-osseous layer
across the arc of knee flexion, but did not report using the
of the iliotibial band, ALL anterolateral ligament, ALCap anterolateral
load sensor to also discover how those cuts affected the capsule, ACL anterior cruciate ligament. (Based on data from Kittl
loads on the knee during the movements. Another in-vitro et al. [26])

13
1020 Knee Surg Sports Traumatol Arthrosc (2017) 25:1015–1023

which had been taken by the ACL was then taken by an

increased contribution from the deep capsulo-osseous
layer of the ITB. When the contributions of the superfi-
cial and deep layers of the ITB were added together, the
ITB was found to have been the primary restraint (that
is: it resisted more than 50% of the torque) to tibial inter-
nal rotation above 30° knee flexion, and this contribution
increased with knee flexion, to 74% of the total resistance
at 60° knee flexion.
The data from Kittl et al. [27] fits with the operative
observations of Terry et al. [59, 60], who found injury of
the deep capsulo-osseous layer of the ITB in 93% of the
functionally unstable knees that they reconstructed. They
also reported that this ITB damage correlated significantly
with the grade of the pivot-shift, whereas ACL damage
did not. Similarly, it has been reported [37] that the pivot-
shift laxity was increased in knees with anterolateral soft-
tissue injury identified on MRI, in relation to knees with
isolated ACL rupture. Work related to double-bundle ACL
reconstruction has found that the posterolateral bundle of
the ACL has a role in controlling tibial internal rotation
at low flexion angles [63], and this fibre bundle is known
to slacken with knee flexion [1]. It has long been known
Fig. 2  Dissection to show the anterolateral ligament (ALL) as
that the ITB shares the resistance to anterior draw of the described by Dodds et al. [12]. View of the anterolateral aspect of
internally-rotated tibia approximately equally with the a left knee at 90° flexion. The red pin is at the lateral femoral epi-
ACL [40]. It may be concluded that the role of the ITB condyle; the blue pin is at the distal attachment of the lateral (fibu-
as a restraint to ALRI might have been underestimated in lar) collateral ligament (LCL); the green pin is at Gerde’s tubercle.
The ALL passes superficially over the proximal part of the LCL and
the recent literature, which has focussed mostly on intra- attaches mid-way between Gerde’s tubercle and the head of the fibula
articular ACL reconstruction, and then the more recent
attention on the ALL, which is a comparatively flimsy
and compliant structure (Fig. 2). Yamamoto et al. [62] and If an anterolateral procedure is considered in combina-
Hassler and Jakob [20] found that cutting the ITB caused tion with an ACL reconstruction, it is worth noting that
a large increase in internal rotation, and that the reduction the biomechanical evidence derived from the natural soft
of the lateral tibial plateau in a pivot shift test disappeared. tissue structures points towards a tenodesis or reconstruc-
That was presumably caused by the loss of the posteri- tive procedure that aims to recreate the restraint provided
orly directed component of the tension in the ITB, which by the ITB, rather than that of the ALL. In addition to the
increases as knee flexion increases, until it is able to sud- data on percent restraint to tibial internal rotation shown in
denly reduce the tibia beneath the lateral condyle of the Fig. 1 [27], the data on tensile stiffness show that a tibial
femur [7, 34]. displacement which stretches the anterolateral structures
will cause the restraining tension to rise more rapidly in the
Implications for surgical treatment ITB. Based on the data reported above, a 5-mm elongation
would lead to an ITB tension of 3070 N (stiffness 614 N/
The biomechanical evidence reported above suggests that mm reported by Noyes et al. [41]), while the ALL tension
a combined intra-plus extra-articular procedure may be would reach 100 N (stiffness 20 N/mm reported by Ken-
appropriate when damage to both the ACL and peripheral nedy et al. [25]). Furthermore, this disparity in tensions is
structures has been identified. Recent work in vitro [21] has compounded by the directions in which the tensions act, as
found that, in the presence of a combined injury, an isolated they cross the joint line: Gerdy’s tubercle is approximately
ACL reconstruction allowed a residual abnormal laxity in 18 mm anterior to the tibial attachment of the ALL [12],
both anterior translation and internal rotation, and that the and so a tenodesis based there will have a larger ante-
addition of a lateral extra-articular tenodesis abolished that rior–posterior component of the tension acting to resist the
deficit. Engebretsen et al. [14] found that an extra-articular tibial internal rotation than would the tension in an ALL
graft shared the load on an ACL reconstruction, so may reconstruction (Fig. 3). Thus, although the ALL has a role
help to protect it during graft remodelling. in restraining tibial internal rotation, the biomechanical

13
Knee Surg Sports Traumatol Arthrosc (2017) 25:1015–1023 1021

a graft tension of 20 N with the foot held in neutral rota-

tion and the knee at 30° flexion restored native knee lax-
ity, while significant over-constraint and elevated lateral
compartment contact pressures resulted from 80 N graft
tension, particularly if the tibia was left free to move into
external rotation when the graft was fixed.
However, judgement as to which cases may be suitable
for a combined ACL plus lateral extra-articular procedure
is beyond the scope of this paper on the biomechanics of
the natural structures at the anterolateral aspect of the knee;
and then which of the possible lateral procedures will give
the best clinical results, also remains a matter for debate,
which requires collection of more outcomes data.

Conclusions on anterolateral knee biomechanics

The biomechanical evidence reported above leads to some

Fig. 3  Left A graft placed at the attachments of the anterolateral liga- observations which may guide the surgeon:
ment (ALL) has a relatively steep orientation as it crosses the tibio-
femoral joint line, thus a graft tension produces a small posterior
component of force to resist tibial internal rotation. Right The more- • Although the ACL is the primary restraint to anterior
anterior attachment of a tenodesis based on Gerdy’s tubercle creates translation of the tibia, its central location means that it
a more efficient orientation to restrain tibial internal rotation than a has a very small moment arm to control tibial internal
procedure based at the tibial attachment of the ALL, particularly if
the graft is passed deep to the LCL, which then acts like a pulley. In rotation.
this diagram, the posterior force vector is four times larger in the right • Tibial internal rotation is resisted primarily by the ITB,
diagram than in the left, yet the graft tensions are the same which is the primary restraint as the knee flexes. The
lateral extra-articular structures have the largest moment
arm to resist tibial internal rotation.
• Internal rotation laxity increases with knee flexion in
the native knee, and is increased further by soft-tissue
injuries. Isolated rupture of the ACL causes a very small
increase of internal rotation, so an obvious increase of
internal rotation laxity implies damage to other struc-
tures.
Compliance with ethical standards

Conflict of interest Prof Amis has received funds into a research

account of Imperial College London from Smith & Nephew Co to un-
dertake research related to the subject field of this article, and has also
been paid for presenting research papers at educational events run by
Fig. 4  This picture of the anterolateral aspect of a left knee held with
the Smith & Nephew Co.
an internal rotation torque applied to the tibia shows clearly that the
more-anterior graft attached at Gerde’s tubercle is better-oriented to
Funding No funding was received for the writing of this article.
resist tibial internal rotation than is the more-posterior arm of the
graft, which simulates the line of action of the anterolateral ligament
Ethical approval This article does not contain any studies with
human participants performed by any of the authors.

data suggest that an ITB tenodesis based on Gerdy’s tuber-

Informed consent For this type of study formal consent is not
cle will be more efficient (Fig. 4).
required.
There has recently been a report that a lateral extra-
articular procedure led to over-constraint of tibial internal
Open Access This article is distributed under the terms of the
rotation laxity across the arc of knee flexion [51]. However,
Creative Commons Attribution 4.0 International License (http://
that study used a graft tension of 88 N. A study of antero- creativecommons.org/licenses/by/4.0/), which permits unrestricted
lateral tenodeses in the author’s laboratory [22] found that use, distribution, and reproduction in any medium, provided you give

13
1022 Knee Surg Sports Traumatol Arthrosc (2017) 25:1015–1023

appropriate credit to the original author(s) and the source, provide a 18. Goldman AB, Pavlov H, Rubenstein D (1988) The Ségond
link to the Creative Commons license, and indicate if changes were fracture of the proximal tibia: a small avulsion that reflects
made. major ligamentous damage. Am J Roentgenol 151:1163–1167
19. Harner CD, Giffin JR, Dunteman RC, Annunziata CC, Fried-
man MJ (2001) Evaluation and treatment of recurrent instabil-
ity after anterior cruciate ligament reconstruction. Instr Course
Lect 50:463–474
References 20. Hassler H, Jakob RP (1981) Ein beitrag zur ursache der ante-
rolateralen instabilität des kniegelenkes. Arch Orthop Trauma
1. Amis AA, Dawkins GPC (1991) Functional anatomy of the ante- Surg 98:45–50
rior cruciate ligament: fibre bundle actions related to ligament 21. Inderhaug E, Stephen J, Williams A, Amis AA (2017) Bio-
replacements and injuries. J Bone Joint Surg Br 73-B:260–267 mechanical comparison of anterolateral procedures combined
2. Amis AA, Scammell BE (1993) Biomechanics of intraarticular with ACL reconstruction. Am J Sports Med 45:347–354
and extraarticular reconstructions of the anterior cruciate liga- 22. Inderhaug E, Stephen JM, El-Daou H, Williams A, Amis
ment. J Bone Joint Surg Br 75-B:812–817 AA (2017) The effect of graft tension and tibial rotation on
3. Amis AA, Zavras TD (1995) Isometricity and graft placement tibiofemoral joint contact pressures and kinematics following
during anterior cruciate ligament reconstruction. Knee 2:5–17 anterolateral tenodesis. Am J Sports Med (in press)
4. Ardern CL, Webster KE, Taylor NF, Feller JA (2011) Return to 23. Ireland J, Trickey EL (1980) MacIntosh tenodesis for antero-
sport following anterior cruciate ligament reconstruction sur- lateral instability of the knee. J Bone Jt Surg Br 62-B:340–345
gery: a systematic review and meta-analysis of the state of play. 24. Kaneda Y, Moriya H, Takahashi K, Shimada Y, Tamaki T
Br J Sports Med 45:596–606 (1997) Experimental study on external tibial rotation of the
5. Bedi A, Musahl V, Lane C, Citak M, Warren RF, Pearle AD knee. Am J Sports Med 25:796–800
(2010) Lateral compartment translation predicts the grade of 25. Kennedy MI, Claes S, Fuso FAF, Williams BT, Goldsmith
pivot shift: a cadaveric and clinical analysis. Knee Surg Sports MT, Turnbull TL, Wijdicks CA, LaPrade RF (2015) The ante-
Traumatol Arthrosc 18:1269–1276 rolateral ligament. An anatomic, radiographic and biomechani-
6. Bull AMJ, Earnshaw PH, Smith A, Katchburian MV, Hassan cal analysis. Am J Sports Med 43:1606–1615
ANA, Amis AA (2002) Intraoperative measurement of knee 26. Kittl C, Halewood C, Stephen J, Gupte CM, Weiler A, Wil-
kinematics in reconstruction of the anterior cruciate ligament. J liams A, Amis AA (2015) Length change patterns of the lateral
Bone Joint Surg Br 84-B:1075–1081 extra-articular structures of the knee and related reconstruc-
7. Bull AMJ, Andersen HN, Basso O, Targett J, Amis AA (1999) tions. Am J Sports Med 43:354–362
Incidence and mechanism of the pivot shift: an in vitro study. 27. Kittl C, El-Daou H, Athwal KK, Gupte CM, Weiler
Clin Orthops 363: 219–231 A, Williams A, Amis AA (2016) The role of the ante-
8. Butler DL, Noyes FR, Grood ES (1980) Ligamentous restraints rolateral structures and the ACL in controlling laxity of
to anterior-posterior drawer in the human knee. J Bone Joint the intact and ACL-deficient knee. Am J Sports Med.
Surg Am 62-A:259–270 doi:10.1177/0363546516638070
9. Campos JC, Chung CB, Lektrakul N, Pedowitz R, Trudell D, Yu 28. Lane JG, Irby SE, Kaufman K, Rangger C, Daniel DM (1994)
J, Resnick D (2001) Pathogenesis of the Ségond fracture: ana- The anterior cruciate ligament in controlling axial rotation an
tomic and MR imaging evidence of an iliotibial tract or anterior evaluation of its effect. Am J Sports Med 22:289–293
oblique band avulsion. Radiology 219:381–386 29. LaPrade RF, Bollom TS, Wentorf FA, Wills NJ, Meister K
10. Claes S, Vereecke E, Maes M, Victor J, Verdonk P, Bellemans J (2005) Mechanical properties of the posterolateral structures of
(2013) Anatomy of the anterolateral ligament of the knee. J Anat the knee. Am J Sports Med 33:1386–1391
223:321–328 30. Lemaire M, Combelles F (1980) Technique actuelle de plastie
11. Diermann N, Schumacher T, Schanz S, Raschke MJ, Petersen W, ligamentaire pour rupture ancienne du ligament croisé antérieur.
Zantop T (2009) Rotational instability of the knee: internal tibial Rev Chir Orthop 66:523–525
rotation under a simulated pivot shift test. Arch Orthop Trauma 31. Lie DT, Bull AM, Amis AA (2007) Persistence of the mini pivot
Surg 129:353–358 shift after anatomically placed anterior cruciate ligament recon-
12. Dodds AL, Halewood C, Gupte CM, Williams A, Amis AA struction. Clin Orthop Relat Res 457:203–209
(2014) The anterolateral ligament: anatomy, length changes, and 32. Lopomo N, Zaffagnini S, Amis AA (2013) Quantifying the pivot
association with the Ségond fracture. Bone Joint J 96-B:325–331 shift test: a systematic review. Knee Surg Sports Traumatol
13. Ellison AE (1979) Distal ilio-tibial band transfer for anterolateral Arthrosc 21:767–783
rotatory instability of the knee. J Bone Jt Surg Am 61-A:330–337 33. Markolf KL, Park S, Jackson SR, McAllister DR (2009) Ante-
14. Engebretsen L, Lew WD, Lewis JL, Hunter RE (1990) The effect rior–posterior and rotatory stability of single and double-bundle
of an iliotibial tenodesis on intraarticular graft forces and knee anterior cruciate ligament reconstructions. J Bone Joint Surg Am
joint motion. Am J Sports Med 18:169–176 91-A:107–118
15. Freedman KB, D’Amato MJ, Nedeff DD, Kaz A, Bach BR 34. Matsumoto H (1990) Mechanism of the pivot shift. J Bone Joint
(2003) Arthroscopic anterior cruciate ligament reconstruction: Surg Br 72-B:816–821
a metaanalysis comparing patellar tendon and hamstring tendon 35. Merican AM, Amis AA (2008) Anatomy of the lateral retinacu-
autografts. Am J Sports Med 31:2–11 lum of the knee. J Bone Joint Surg (Br) 90-B:527–534
16. Fukubayashi T, Torzilli PA, Sherman MF et al (1982) An in vitro 36. Muller B, Hofbauer M, Rahnemai-Azar AA, Wolf M, Araki D
biomechanical evaluation of anterior–posterior motion of the et al (2016) Development of computer tablet software for clini-
knee. Tibial displacement, rotation, and torque. J Bone Joint cal quantification of lateral knee compartment translation dur-
Surg Am 64-A:258–264 ing the pivot-shift. Comput Methods Biomech Biomed Engin
17. Galway H, MacIntosh D (1980) The lateral pivot shift: a symp- 19:217–228
tom and sign of anterior cruciate ligament insufficiency. Clin 37. Musahl V, Rahnemai-Azar AA, Costello J, Arner JW, Fu FH,
Orthops Relat Res 147: 45–50 Hoshino Y, Lopomo N, Samuelsson K (2016) The influence
of meniscal and anterolateral capsular injury on knee laxity in

13
Knee Surg Sports Traumatol Arthrosc (2017) 25:1015–1023 1023

patients with anterior cruciate ligament injuries. Am J Sports 52. Sidles JA, Larson RV, Garbini JL, Downey DJ, Matsen FA
Med 44:3126–3131 (1988) Ligament length relationships in the moving knee. J
38. Narvani A, Mahmud T, Lavelle J, Williams A (2010) Injury to Orthop Res 6:593–610
the proximal deep medial collateral ligament—a problematical 53. Slocum DB, Larson RL (1968) Rotatory instability of the knee –
subgroup of injuries. J Bone Jt Surg (Br) 92-B: 949–953 its pathogenesis and a clinical test to demonstrate its presence. J
39. Nordt WE, Lotfi P, Plotkin E et al (1999) The in vivo assessment Bone Joint Surg Am 50-A:211–225
of tibial motion in the transverse plane in anterior cruciate liga- 54. Song GY, Zhang H, Wang QQ, Zhang J, Li Y, Feng H (2016)
ment reconstructed knees. Am J Sports Med 27:611–616 Bone contusions after acute noncontact anterior cruciate liga-
40. Noyes FR, Grood ES, Butler DL, Paulos LE (1980) Clinical bio- ment injury are associated with knee joint laxity, concomi-
mechanics of the knee: ligament restraints and functional stabil- tant maniscal lesions, and anterolateral ligament abnormality.
ity. In: AAOS symposium on the athlete’s knee. CV Mosby, St. Arthrosc 32: 2331–2341
Louis, p 1–35 55. Sonnery-Cottet B, Lutz C, Daggett M, Freychet B, Niglis L,
41. Noyes FR, Butler DL, Grood ES, Zernicke RF, Hefzy MS (1984) Imbert P (2016) The involvement of the anterolateral ligament in
Biomechanical analysis of human ligament grafts used in knee rotational control of the knee. Am J Sports Med 44:1209–1214
ligament repairs and reconstructions. J Bone Joint Surg (Am) 56. Spencer L, Burkhart TA, Tran MN, Rezansoff AJ, Deo S, Cater-
66-A:344–352 ine S, Getgood AM (2015) Biomechanical analysis of simulated
42. Ostgaard SE, Helmig P, Nielsen S, Hvid I (1991) Anterolateral clinical testing and reconstruction of the anterolateral ligament
instability in the anterior cruciate ligament deficient knee. A of the knee. Am J Sports Med 43:2189–2197
cadaver study. Acta Orthop Scand 62:4–8 57. Sugita T, Amis AA (2001) Anatomic and biomechanical study of
43. Parsons EM, Gee AO, Spiekerman C, Cavanagh PR (2015) The the lateral collateral and popliteofibular ligaments. Am J Sports
biomechanical function of the anterolateral ligament of the knee. Med 29:466–472
Am J Sports Med. doi:10.1177/0363546514562751 58. Terry GC, Hughston JC, Norwood LA (1986) The anatomy
44. Race A, Amis AA (1996) Loading of the two bundles of the pos- of the iliopatellar band and iliotibial tract. Am J Sports Med
terior cruciate ligament: an analysis of bundle function in A-P 14:39–45
drawer. J Biomechs 29:873–879 59. Terry GC, Norwood LA, Hughston JC, Caldwell KM (1993)
45. Rahnemai-Azar AA, Miller M, Guenther D, Fu FH, Lesniak BP, How iliotibial tract injuries of the knee combine with acute ante-
Musahl V, Debski RE (2016) Structural properties of the antero- rior cruciate ligament tears to influence abnormal anterior tibial
lateral capsule and iliotibial band of the knee. Am J Sports Med displacement. Am J Sports Med 21:55–60
44:892–897 60. Terry GC, LaPrade RF (1996) The biceps femoris muscle com-
46. Rasmussen MT, Nitri M, Williams BT, Moulton SG, Cruz RS, plex at the knee Its anatomy and injury patterns associated with
Dornan GJ, Goldsmith MT, LaPrade RF (2016) An in vitro acute anterolateral-anteromedial rotatory instability. Am J Sports
robotic assessment of the anterolateral ligament in the set- Med 24:2–8
ting of an anterior cruciate ligament injury. Am J Sports Med 61. Thein R, Boorman-Padgett J, Stone K, Wickiewicz TL, Imhauser
44:585–592 CW, Pearle AD (2016) Biomechanical assessment of the antero-
47. Robinson JR, Bull AMJ, Amis AA (2005) Structural properties lateral ligament of the knee – a secondary restraint in simulated
of the medial collateral ligament complex of the human knee. J tests of the pivot-shift and of anterior stability. J Bone Joint Surg
Biomech 38:1067–1074 (Am) 98-A: 937–943
48. Robinson JR, Bull AMJ, Thomas R deW, Amis AA (2006) The 62. Yamamoto Y, Hsu WH, Fisk JA, Van Scyoc AH, Miura K, Woo
role of the medial collateral ligament and posteromedial capsule SL (2006) Effect of the iliotibial band on knee biomechanics
in controlling knee laxity. Am J Sports Med 34:1815–1823 during a simulated pivot shift test. J Orthop Res 24:967–973
49. Rudy TW, Livesay GA, Woo SLY, Fu FH (1996) A combined 63. Zantop T, Herbort M, Raschke MJ, Fu FH, Petersen W (2007)
roobotic/universal force sensor approach to determine in situ The role of the anteromedial and posterolateral bundles of the
forces of knee ligaments. J Biomech 29:1357–1360 anterior cruciate ligament in anterior tibial translation and inter-
50. Saiegh YA, Suero EM, Guenther D, Hawi N, Decker S, Krettek nal rotation. Am J Sports Med 35:223–227
C, Citak M, Omar M (2015) Sectioning the anterolateral liga- 64. Zens M, Feucht MJ, Ruhhammer J, Bernstein A, Mayr HO, Sud-
ment did not increase tibiofemoral translation or rotation in an kamp NP, Woias P, Niemeyer P (2015) Mechanical tensile prop-
ACL-deficient cadaveric model. Knee Surg Sports Traumatol erties of the anterolateral ligament. J Exp Orthop. doi:10.1186/
Arthrosc. doi:10.1007/s00167-015-3787-1 s40634-015-0023-3
51. Schon JM, Moatshe G, Brady AW, Cruz RS, Chahla J, Dornan 65. Zens M, Niemeyer P, Ruhhammer J, Bernstein A, Woias P,
GJ, Turnbull TL, Engebretsen L, LaPrade RF (2016) Anatomic Mayr HO, Sudkamp NP, Feucht MJ (2015) Length changes of
anterolateral ligament reconstruction leads to overconstraint at the anterolateral ligament during passive knee motion: a human
any fixation angle. Am J Sports Med 44:2546–2556 cadaveric study. Am J Sports Med 43:2545–2552

13