This is an enhanced PDF from The Journal of Bone and Joint Surgery

The PDF of the article you requested follows this cover page.

Effects of Posterolateral Reconstructions on External Tibial Rotation

and Forces in a Posterior Cruciate Ligament Graft
Keith L. Markolf, Benjamin R. Graves, Susan M. Sigward, Steven R. Jackson and David R. McAllister
J Bone Joint Surg Am. 2007;89:2351-2358. doi:10.2106/JBJS.F.01086

This information is current as of November 5, 2007

Supplementary material

Commentary and Perspective, data tables, additional images, video clips and/or
translated abstracts are available for this article. This information can be accessed
at http://www.ejbjs.org/cgi/content/full/89/11/2351/DC1

Reprints and Permissions

Click here to order reprints or request permission to use material from this
article, or locate the article citation on jbjs.org and click on the [Reprints and
Permissions] link.

Publisher Information

The Journal of Bone and Joint Surgery

20 Pickering Street, Needham, MA 02492-3157
www.jbjs.org

Markolf.fm Page 2351 Thursday, October 11, 2007 11:53 AM

2351
COPYRIGHT 2007

BY

THE JOURNAL

OF

BONE

AND JOINT

SURGERY, INCORPORATED

Effects of Posterolateral Reconstructions

on External Tibial Rotation and
Forces in a Posterior Cruciate Ligament Graft
By Keith L. Markolf, PhD, Benjamin R. Graves, MD,
Susan M. Sigward, PhD, Steven R. Jackson, and David R. McAllister, MD
Investigation performed at the Biomechanics Research Section, Department of Orthopaedic Surgery,
David Geffen School of Medicine at the University of California at Los Angeles, Los Angeles, California

Background: In patients with a Grade-3 injury, reconstructions of the lateral collateral ligament, popliteus tendon,
and popliteofibular ligament are commonly performed in conjunction with a reconstruction of the posterior cruciate
ligament. The objectives of this study were (1) to compare the abilities of three types of posterolateral graft reconstruction to restrain external tibial rotation and alter forces in a posterior cruciate graft and (2) to compare tibial rotations and posterior cruciate graft forces associated with two levels of initial posterolateral graft tension.
Methods: Forces in the posterior cruciate ligament were recorded as the knee was extended from 120 to 0 and a 5N-m external tibial torque was applied. The posterior cruciate ligament was reconstructed, and external tibial rotation
and the forces in the posterior cruciate graft were recorded. These measurements were again recorded after sectioning
of the posterolateral structures and after reconstruction of the lateral collateral ligament, alone as well as in combination with reconstruction of the popliteus tendon and in combination with reconstruction of the popliteofibular ligament.
Results: With the lateral collateral ligament intact, removal of the popliteus tendon from its femoral origin significantly increased external tibial rotation. Applying tension to a popliteus or popliteofibular graft internally rotated the
tibia, with no significant difference between the rotations caused by the tensioning of the two grafts. Tibial rotation
was significantly greater when graft tensioning had been performed with the tibia free to rotate than it was when the
tensioning had been done with the tibia locked in neutral rotation. With an applied external tibial torque, a reconstruction of the lateral collateral ligament alone was not sufficient to reduce posterior cruciate graft forces to normal. The
addition of a popliteus or popliteofibular reconstruction to the lateral collateral ligament reconstruction significantly
reduced posterior cruciate graft forces to normal (or below normal) levels. The external rotations associated with
these two combined reconstructions were equivalent and significantly less than that in the intact knee. Increasing
tension in either the popliteus or the popliteofibular graft from 10 to 30 N significantly decreased external rotation.
Conclusions: The posterolateral grafts acted to resist applied external torque, thereby off-loading the posterior cruciate graft. Popliteus and popliteofibular grafts were more favorably aligned than a lateral collateral ligament graft to resist external rotation, and they had similar effects.
Clinical Relevance: Holding the tibia in neutral rotation when tensioning a popliteus or popliteofibular graft will help
limit internal tibial rotation. The popliteus and popliteofibular graft tensioning protocols used in this study overly constrained external rotation and failed to produce optimal load-sharing with the posterior cruciate graft.

Disclosure: In support of their research for or preparation of this work, one or more of the authors received, in any one year, outside funding or
grants in excess of $10,000 from NFL Charities. Neither they nor a member of their immediate families received payments or other benefits or a
commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any
benefits to any research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which the authors, or a
member of their immediate families, are affiliated or associated.
A commentary is available with the electronic versions of this article, on our web site (www.jbjs.org) and on our quarterly CD-ROM (call our
subscription department, at 781-449-9780, to order the CD-ROM).

J Bone Joint Surg Am. 2007;89:2351-8 doi:10.2106/JBJS.F.01086

Markolf.fm Page 2352 Thursday, October 11, 2007 11:53 AM

2352
THE JOURNAL OF BONE & JOINT SURGER Y JBJS.ORG
VO L U M E 89-A N U M B E R 11 N O VE M B E R 2007

njuries to the posterolateral structures of the knee are

commonly found in association with a posterior cruciate
ligament injury1-3. Normally, ruptures of the lateral collateral ligament are easily recognized, and an anatomic graft reconstruction is straightforward. However, injuries to the
popliteus tendon and popliteofibular ligament are often unappreciated4, and a number of posterolateral reconstruction
techniques have been described in the literature5-8. The popliteus has an origin on the lateral femoral condyle, and it passes
through the lateral aspect of the capsule to a muscular insertion on the posterior surface of the tibia. The popliteus bypass
reconstruction is typically done with a tendon graft, with an
origin at the popliteus footprint on the lateral femoral condyle, that passes beneath the lateral collateral ligament and
joint capsule and into a tunnel drilled approximately 1 cm beneath the lateral tibial plateau9-11. The popliteofibular ligament has an attachment on the fibular head near the posterior
styloid and a diffuse connection to the posterolateral complex;
it has no direct attachment to the femur12. One method of reconstructing the popliteofibular ligament involves use of a
tendon graft that originates at the center of the popliteus footprint on the femur and passes through a tunnel drilled at the
fibular styloid. This reconstruction has direct bone-to-bone
attachments, unlike the native ligament that it is replacing8,9.
Previous biomechanical studies with application of external tibial torque have shown that external tibial rotation
and forces in the native posterior cruciate ligament (and in a
posterior cruciate ligament graft) increase significantly when
the posterolateral structures are cut13-15. However, there have
been relatively few biomechanical studies on external rotatory
laxity after the reconstruction of the posterolateral structures
in knees with an intact posterior cruciate ligament10,16 and
those in which the posterior cruciate ligament has been reconstructed17,18. The objectives of this study were to measure
changes in tibial rotation and forces in a posterior cruciate ligament graft resulting from applied external tibial torque after
removal of the popliteus tendon, popliteofibular ligament,
and lateral collateral ligament and to repeat the measurements
after reconstruction of the lateral collateral ligament alone, the
lateral collateral ligament and the popliteus tendon, and the
lateral collateral and popliteofibular ligaments. The effects of
varying tension in the posterolateral grafts and of locked compared with unlocked tibial rotation during graft tensioning
were also investigated.
Materials and Methods
ifteen fresh-frozen cadaveric knees were used for this study.
The mean age of the donors at the time of death was 35.1
years (range, seventeen to sixty-five years); thirteen donors
were male, and two were female. A coring cutter was used to
mechanically isolate a cylinder of bone containing the femoral
origin of the posterior cruciate ligament. The bone cylinder
was attached to a custom-designed femoral load cell that recorded force in the ligament as the knee was extended from
120 to 0 and 5 N-m of external tibial torque was applied. A
single-bundle tibial inlay reconstruction of the posterior cruci-

E F F E C T S O F P O S T E RO L A T E R A L R E CO N S T R U C T I O N S O N T I B I A L
R O T A T I O N A N D PO S T E R I O R C R U C I A T E G R A F T F O R C E S

ate ligament was performed on each specimen with use of an

11-mm-wide bone-patellar tendon-bone graft. The femoral
end of the graft was passed through a 10-mm-diameter tunnel
drilled into an acrylic replica of the bone cylinder at the center
of the femoral footprint of the anterolateral bundle of the posterior cruciate ligament. The acrylic replica was connected to
the end of the femoral load cell. The graft was tensioned to restore anteroposterior laxity at 200 N to within 1 mm of that in
the intact knee at 90 of flexion. The popliteus muscle belly was
partially removed during installation of the tibial inlay bone
block. Details of the load-cell installation, graft placement, and
graft tensioning can be found in a prior publication19.
Tibial rotation and the resultant force in the posterior
cruciate ligament graft were recorded continuously over a
120 flexion range with application of 5 N-m of external
torque as described previously20. Testing was repeated with the
popliteus tendon detached from its femoral origin, and then
the popliteofibular ligament was cut at its attachment to the
fibular styloid. A 1 1-cm calcaneal bone block of an Achilles
tendon graft was fixed into a square mortised recess near the
femoral footprint of the popliteus tendon; the tissue was sized
to fit within a 7-mm-diameter tunnel. This graft was used for
both the popliteus and the popliteofibular reconstruction.
When used for the popliteofibular reconstruction, the tendon
entered a tunnel drilled into the styloid of the fibula (Fig. 1). A
high-strength, low-stretch synthetic line (135-lb [61.2-kg] test
Spectra Fiber; Izorline, Gardena, California), sutured into the
free end of the graft tissue with use of a whip stitch, was passed
through a split clamp attached to the tibia anterior to the fibula (Fig. 1). When used to reconstruct the popliteus tendon,
the Achilles tendon passed into a 7-mm-diameter tibial tunnel
1 cm inferior to the lateral tibial plateau. The line attached to
the free end of the graft was passed through a separate split
clamp, which was also located on the anterior aspect of the
tibia (Fig. 1). These particular reconstructions were selected
for experimental study because they represented the most
simple and direct reconstructions of normal posterolateral
anatomy. Since the amount of tension applied to a posterolateral graft and the rotational position of the tibia at the time of
graft tensioning vary in the literature21-23, these variables were
also selected for study.
The precise location of the femoral bone-block site for
the popliteus and popliteofibular grafts was determined after
performance of isometry measurements. This was done by
measuring relative length changes between trial locations on
the lateral femoral condyle and the distal graft tunnel sites. A
suture (fixed at a trial site) was passed through each tunnel
and through a split clamp. A dial caliper was used to measure
the relative length change between the split clamp and a forceps attached to the suture at a fixed distance from the split
clamp. An optimum femoral point for both grafts was selected
on the basis of the minimum relative length change of the suture. The bone block was placed such that the leading edge of
the graft tissue on it was centered over the optimum point. For
the reconstruction of the lateral collateral ligament, the mean
center of the 1 1-cm calcaneal bone block was located 2.4

Markolf.fm Page 2353 Thursday, October 11, 2007 11:53 AM

2353
THE JOURNAL OF BONE & JOINT SURGER Y JBJS.ORG
VO L U M E 89-A N U M B E R 11 N O VE M B E R 2007

mm anterior to the anatomic center of the femoral footprint

of the lateral collateral ligament. For the popliteus and popliteofibular reconstructions (the same graft was used for both),
the mean bone-block center was 2.7 mm proximal and 11 mm
anterior to the center of the popliteus footprint. The mean relative length changes of in situ lateral collateral, popliteus, and
popliteofibular grafts with these bone-block placements were
<1.5 mm from 0 to 90 of flexion.
An external torque test was performed with the distal
lines of the popliteus or popliteofibular graft unclamped and
the lateral collateral ligament removed (simulating a Grade-3
posterolateral injury). This test was performed first with the
femoral origin of the popliteus intact and then with the popliteus tendon detached from the femur. Next, the lateral collateral ligament was reconstructed with use of a second Achilles
tendon graft. The bone block of this graft was fixed into a
mortised recess centered over the femoral footprint of the lateral collateral ligament. The tissue was sized to fit within a 6mm-diameter tunnel drilled at the fibular insertion of the
lateral collateral ligament (Fig. 1). External torque tests were
performed at two graft-tension levels (10 and 30 N) after reconstruction of the lateral collateral ligament alone, reconstruction of the lateral collateral ligament and the popliteus
tendon, and reconstruction of the lateral collateral and popliteofibular ligaments. The tibia was locked during tensioning.

E F F E C T S O F P O S T E RO L A T E R A L R E CO N S T R U C T I O N S O N T I B I A L
R O T A T I O N A N D PO S T E R I O R C R U C I A T E G R A F T F O R C E S

Prior to external torque testing, the knee was moved

through a passive range of motion (with no tibial torque applied) to record the internal tibial rotations caused by tensioning of the popliteus and popliteofibular grafts. The grafts were
tested in random order. Each reconstruction was tested both
with the graft tensioned to 10 N and with it tensioned to 30 N
at 30 of flexion with the tibia locked in neutral rotation and
with the tibia free to rotate.
A repeated-measures analysis of variance with paired
comparisons was used to analyze differences in mean tibial rotations and graft forces between test conditions. The test conditions included the tibial rotation status (free or locked)
during graft tensioning, the posterolateral corner status
(popliteus attached, popliteus detached, or lateral collateral
ligament removed), and the reconstruction status (reconstruction of the lateral collateral ligament alone, reconstruction of the lateral collateral ligament and the popliteus tendon,
and reconstruction of the lateral collateral and popliteofibular
ligaments), and the amount of graft pretension (10 or 30 N).
The level of significance was p < 0.05.
Results
ith the lateral collateral ligament and the femoral origin
of the popliteus intact, the mean external tibial rotations resulting from the applied external torque were 9.1 at 0

Fig. 1

The combined reconstruction of the lateral collateral ligament (LCL) and the popliteus tendon (POP). The bone block of the
lateral collateral ligament graft is fixed on the lateral femoral condyle at the center of the native footprint of the lateral collateral ligament. The distal end of the graft passes through a tunnel on the fibular head. Lines sutured to the end of the graft
pass through a split clamp (fixed to the tibia) for graft tensioning and fixation. The bone block of the popliteus graft is fixed
near the center of the popliteus footprint on the lateral femoral condyle and passes through a tunnel drilled into the fibular
styloid. When used for a popliteus reconstruction, the same graft passes through a tunnel drilled 1 cm inferior to the lateral
tibial plateau. PFL = popliteofibular ligament. (Reprinted, with permission, from: Markolf KL, Graves BR, Sigward SM, Jackson SR, McAllister DR. How well do anatomical reconstructions of the posterolateral corner restore varus stability to the
posterior cruciate ligament-reconstructed knee? Am J Sports Med. 2007;35:1117-22.)

Markolf.fm Page 2354 Thursday, October 11, 2007 11:53 AM

2354
THE JOURNAL OF BONE & JOINT SURGER Y JBJS.ORG
VO L U M E 89-A N U M B E R 11 N O VE M B E R 2007

Fig. 2

Mean curves of tibial rotation versus knee flexion angle produced by

application of 5 N-m of external tibial torque in the fifteen knees, with
(1) the popliteus (pop) insertion intact, (2) the popliteus tendon detached from the lateral femoral condyle, and (3) the lateral collateral
ligament (lcl), popliteus tendon, and popliteofibular ligament removed
(a Grade-3 injury). PCL = posterior cruciate ligament. Sample standard
deviations are indicated by error bars. P < 0.05 indicates a significant
difference between test conditions.

of flexion, 14.1 at 30 of flexion, and 10.3 at 120 of flexion

(Fig. 2). Detachment of the popliteus tendon from the femur
significantly increased the mean external rotation by 1.6 at 0
of flexion to 7.3 at 120 of flexion. The mean external tibial
rotations after all posterolateral structures were removed (a
Grade-3 injury, with the lateral collateral ligament cut and the
popliteus tendon detached) reached a maximum of 29.6 (at
35 of flexion) and were significantly greater than those for the
other test conditions at all flexion angles (Fig. 2).
The mean forces in the posterior cruciate ligament graft
(resulting from the applied external torque) with the lateral
collateral ligament and the origin of the popliteus intact were
significantly greater than the forces in the intact posterior cruciate ligament beyond 50 of flexion (Fig. 3). The mean forces
with the popliteus tendon detached were significantly greater
than those with the popliteus intact beyond 80, and they were
significantly greater than those for the intact posterior cruciate
ligament beyond 55. The mean forces with all posterolateral
structures removed were significantly greater than those with
the popliteus detached beyond 15 and significantly greater
than those in the posterior cruciate ligament beyond 5. The
mean posterior cruciate ligament forces remained near zero up
to approximately 45 of flexion, at which point they began to
increase. The mean posterior cruciate graft forces with all posterolateral structures removed were 119 32 N at 45, and they
increased to a maximum of 197 35 N at 110 (Fig. 3).
With an applied external tibial torque, the mean external rotations with a lateral collateral ligament graft were not

E F F E C T S O F P O S T E RO L A T E R A L R E CO N S T R U C T I O N S O N T I B I A L
R O T A T I O N A N D PO S T E R I O R C R U C I A T E G R A F T F O R C E S

significantly different from those with all posterolateral structures intact beyond 25 of flexion, and they were approximately 3.5 greater between 0 and 25 of flexion (Fig. 4). The
mean rotations with 10 N of tension on the lateral collateral
ligament graft were not significantly different from those with
30 N of tension. The mean rotations with the popliteus and
popliteofibular grafts were significantly less than those with all
posterolateral structures intact, with one exception (the popliteus graft at 0) (Fig. 4). The mean rotations with both grafts
were approximately 2 greater with 10 N of graft tension than
they were with 30 N. The mean rotations with the popliteus
and popliteofibular grafts were not significantly different from
each other at either tension level.
With the tibia free to rotate, applying 30 N of tension to a
popliteus or popliteofibular graft rotated the tibia internally to
5.5 at 0 of flexion and to 15.6 at 120 of flexion. There was
significantly less mean internal rotation when the tibia was
locked during tensioning (1.8 to 9.8, respectively) (Fig. 5).
For these tests, the lateral collateral ligament was intact and the
knee was passively flexed through a 120 range of motion with
no external tibial torque applied. The mean rotation with the
popliteus graft did not differ significantly from that with the
popliteofibular graft at either pretension condition (Fig. 5).
The mean forces in the posterior cruciate graft with a
lateral collateral ligament reconstruction alone were significantly greater than those in the posterior cruciate ligament beyond 80 of flexion (Fig. 6). The mean posterior cruciate graft

Fig. 3

Mean curves of resultant force versus knee flexion angle produced by

application of 5 N-m of external tibial torque (with the lateral collateral
ligament [lcl] intact) in the fifteen knees, with the posterior cruciate ligament (pcl) intact and with a posterior cruciate ligament graft and (1)
the popliteus (pop) insertion intact, (2) the popliteus tendon detached
from the lateral femoral condyle, and (3) the lateral collateral ligament,
popliteus tendon, and popliteofibular ligament removed (a Grade-3 injury). Sample standard deviations are indicated by error bars. ns = no
significant difference between test conditions at the indicated degrees
of flexion.

Markolf.fm Page 2355 Thursday, October 11, 2007 11:53 AM

2355
THE JOURNAL OF BONE & JOINT SURGER Y JBJS.ORG
VO L U M E 89-A N U M B E R 11 N O VE M B E R 2007

Fig. 4

Mean curves of tibial rotation versus knee flexion angle produced by

application of 5 N-m of external tibial torque in the fifteen knees, with
(1) all posterolateral structures intact, (2) a lateral collateral ligament
(lcl) graft tensioned to 10 or 30 N, (3) a lateral collateral ligament graft
tensioned to 30 N and a popliteus (pop) graft tensioned to 10 or 30 N,
and (4) a lateral collateral ligament graft tensioned to 30 N and a popliteofibular (pfl) graft tensioned to 10 or 30 N. Sample standard deviations are indicated by error bars. P < 0.05 indicates a significant
difference and ns indicates no significant difference between test
conditions at the indicated degrees of flexion.

forces with reconstruction of the lateral collateral ligament

and the popliteus tendon were not significantly different from
the mean posterior cruciate ligament forces (Fig. 6). The mean
posterior cruciate graft forces with reconstruction of the lateral collateral and popliteofibular ligaments were significantly
less than the posterior cruciate ligament forces beyond 70 of
flexion. The level of graft tension did not significantly affect
the mean posterior cruciate graft forces associated with any
reconstruction (Fig. 6).
Discussion
n this study, we measured the abilities of posterolateral reconstructions to limit external tibial rotation and alter
forces in a posterior cruciate ligament graft in response to an
applied external tibial torque. Variables that were investigated
included the amount of tension applied to the grafts and the
effects of free compared with locked tibial rotation during
graft tensioning. On the basis of preliminary testing with these
graft configurations, we selected 10 N as the minimum tension that could be reliably applied to the graft, whereas 30 N
was thought to be a reasonable amount that would be applied
to a posterolateral graft in clinical practice. With our test apparatus, the tibia could be rigidly locked during graft tensioning to simulate the position during surgery with the tibia and/
or foot held manually. The rotations and graft forces measured in this study would have relevance for the immediate
postoperative period. Relaxation of tension in all grafts after
in vivo cyclic loading would be expected with time, which

E F F E C T S O F P O S T E RO L A T E R A L R E CO N S T R U C T I O N S O N T I B I A L
R O T A T I O N A N D PO S T E R I O R C R U C I A T E G R A F T F O R C E S

could result in greater external rotations than were measured

here24,25.
Protocols for tensioning posterolateral grafts are not well
described in the literature. Lee et al.21 tensioned a split graft
used to reconstruct the lateral collateral and popliteofibular
ligaments with the knee at 30 of flexion with internal rotation. Sekiya and Kurtz22 also utilized a split graft tensioned at
30 with valgus stress and an internal tibial torque. Stannard et
al.23 described a modified two-tailed technique for reconstruction of the popliteus and the lateral collateral ligament: the
graft was tensioned with the knee flexed to 30 and the foot internally rotated. None of these authors mentioned the level of
graft tension that was applied or the amount of internal tibial
rotation when the graft was tensioned.
On the basis of these reports, we chose to tension the
posterolateral grafts at 30 of knee flexion and to measure the
amount of internal rotation as the grafts were tensioned with
the tibia free and locked. We found that the greater the internal rotation after graft tensioning, the lesser the external tibial
rotation with application of external tibial torque. When the
tibia was held in neutral rotation during popliteus and popliteofibular graft tensioning, there was significantly less internal
rotation than there was when the tibia was free to rotate. The
internal rotation that did occur was due to a spring-back
effect produced by the inherent elasticity of the graft tissues,
which pulled the tibia into internal rotation as it was releasedthat is, applying tension to the posterolateral graft
(with the tibia locked) caused it to elongate; when the tibia
was fixed, the graft tissue remained under strain. When the
tibia was unlocked, the tissue strain relaxed as the tibia rotated

Fig. 5

Mean curves of tibial rotation versus knee flexion angle produced by

application of passive knee extension (with the lateral collateral ligament [lcl] intact) in the fifteen knees, with (1) all posterolateral structures intact, (2) a popliteus (pop) graft tensioned to 30 N with free or
locked tibial rotation, and (3) a popliteofibular (pfl) graft tendon tensioned to 30 N with free or locked tibial rotation. Sample standard deviations are indicated by error bars. P < 0.05 indicates a significant
difference between test conditions.

Markolf.fm Page 2356 Thursday, October 11, 2007 11:53 AM

2356
THE JOURNAL OF BONE & JOINT SURGER Y JBJS.ORG
VO L U M E 89-A N U M B E R 11 N O VE M B E R 2007

E F F E C T S O F P O S T E RO L A T E R A L R E CO N S T R U C T I O N S O N T I B I A L
R O T A T I O N A N D PO S T E R I O R C R U C I A T E G R A F T F O R C E S

Fig. 6

Mean curves of resultant force versus knee flexion angle produced by application of 5 N-m of external tibial torque in the fifteen knees, with the posterior cruciate ligament (pcl) intact and with a posterior cruciate ligament graft and (1) all posterolateral structures intact, (2) a lateral collateral ligament (lcl) graft tensioned to 10 or 30 N, (3) a lateral collateral ligament
graft tensioned to 30 N and a popliteus (pop) graft tensioned to 10 or 30 N, and (4) a lateral collateral ligament graft tensioned to 30 N and popliteofibular (pfl) graft tensioned to 10 or 30 N. Sample standard deviations are indicated by error
bars. ns = no significant difference between test conditions at the indicated degrees of flexion.

internally from the position at which the graft was tensioned.

It should be noted that our cadaveric model did not include the inherent passive tension in the knee muscles that
would be present in vivo at the time of surgery. The amount of
internal tibial rotation produced by graft tensioning in vivo
could be less than what was measured in this study, as a result
of passive resistance from the knee muscles. We believe that
our results indicate the maximum amount of tibial rotation
that could occur.
The normal knee has an inherent external rotatory laxity at 30 due to slackening of the native posterolateral structures with knee flexion16,17. Most surgeons test for external
rotatory laxity at the time of graft tensioning. However, the
correct amount of laxity is difficult to judge, and it is possible
that the natural rotatory laxity of the knee may be reduced or
eliminated with the graft in place. It is possible that physiologic external rotatory laxity may return as the posterolateral
grafts stretch out during the healing process, as a result of repeated cyclic loading. However, the clinical effects of excessive
internal prerotation of the tibia at the time of graft tensioning
are unknown.
In vivo, control of external tibial rotation is provided by
the popliteus muscle acting at its tendinous insertion on the
femur. In our cadaveric model, some of the muscle attachment of the popliteus to the tibia was cut while the tibial inlay
was placed. This left the femoral insertion of the popliteus
tendon as well as some of its fascial attachments intact. Detaching the popliteus tendon from the femur significantly
increased external tibial rotation. This was likely due to its at-

tachment to other structures, including its connection to the

fibula through the popliteofibular ligament. Therefore the
popliteus complex provides both static and dynamic restraint
to external tibial rotation. Clinically, detachment of the popliteus tendon from the lateral femoral condyle is commonly observed with a Grade-3 injury to the posterolateral corner26,27.
At the time of surgery, a ruptured lateral collateral ligament is normally recognized and reconstructed to restore
varus laxity. Detachment of the popliteus tendon from the lateral femoral condyle and a tear of the popliteofibular ligament
can be more difficult to appreciate. When all three structures
were absent in conjunction with a posterior cruciate ligament
injury, reconstruction of the posterior cruciate ligament alone
was not sufficient to restore posterior cruciate graft forces to
normal levels or to restore normal external tibial rotations between 0 and 25 of knee flexion. Theoretically, a popliteofibular graft should be more effective in controlling external
rotation than a popliteus graft because it is fixed at a greater
distance from the axis of tibial rotation, thus giving it a greater
mechanical advantage. This was not observed in our study:
both grafts had equivalent effects at the same level of pretension. However, the mean posterior cruciate graft forces associated with a popliteofibular reconstruction were approximately
10 to 20 N less than those associated with a popliteus reconstruction beyond 60 of flexion.
As a result of differences in graft-tensioning protocols
and the combinations of posterolateral graft reconstructions
tested, our results differ somewhat from those in prior studies.
Kanamori et al.8 found that knees subjected to 10 N-m of ex-

Markolf.fm Page 2357 Thursday, October 11, 2007 11:53 AM

2357
THE JOURNAL OF BONE & JOINT SURGER Y JBJS.ORG
VO L U M E 89-A N U M B E R 11 N O VE M B E R 2007

ternal tibial torque after reconstruction of the popliteofibular

ligament had external rotation values that were not significantly different from those in the intact knee. An unspecified
level of tension was applied to the popliteofibular graft at 30
of knee flexion, and the posterior cruciate ligament and lateral
collateral ligament were not reconstructed. Nau et al.17 found
that reconstructions of the posterior cruciate ligament and the
popliteus tendon as well as reconstructions of the posterior
cruciate ligament, popliteus tendon, and popliteofibular ligament restored external rotations to normal at 30 and 90 of
knee flexion with 5 N-m of external torque. The posterolateral
grafts were tensioned to 10 N with the knee at 90, and the
tibia internally rotated 10. The posterior cruciate graft was
tensioned to 70 N at 90. Nau et al.16 performed external
torque tests with reconstructions of the lateral collateral and
popliteofibular ligaments as well as reconstructions of the lateral collateral ligament, popliteofibular ligament, and popliteus tendon; the posterior cruciate ligament was intact. All
posterolateral graft combinations restored normal external
tibial rotations at 30 and 90 of knee flexion. The popliteofibular and popliteus grafts were tensioned to 10 N at 30 of flexion, with the tibia in neutral rotation. The lateral collateral
ligament graft was tensioned to 10 N at 90, with neutral tibial
rotation. Sekiya et al.18 reconstructed the popliteus tendon and
the popliteofibular ligament in knees with a double-bundle
reconstruction of the posterior cruciate ligament. External
tibial rotations (with application of 5 N-m of torque) were not
significantly different from those of the intact knee. Both posterolateral graft reconstructions were tensioned to 67 N at 30
of flexion. Addition of the posterolateral reconstructions to an
isolated double-bundle reconstruction of the posterior cruciate ligament significantly decreased external tibial rotation at
0, 30, and 90 of knee flexion but decreased force in the posterior cruciate ligament reconstruction at 90 only.
The present in vitro study had several limitations. The
tibial rotations and graft forces measured in this study would
be comparable with those present at the time of surgery. Missing from our experimental protocols were the effects of
weight-bearing and muscle forces. With the graft-tensioning
protocols used in this study, reconstructions of the lateral collateral ligament and the popliteus tendon and those of the
lateral collateral and popliteofibular ligaments resulted in
marked overconstraint of external rotation compared with
that of the intact knee. For example, the mean rotations with
those reconstructions tensioned to 30 N were 3 to 10 less
than the mean rotations with all posterolateral structures in-

E F F E C T S O F P O S T E RO L A T E R A L R E CO N S T R U C T I O N S O N T I B I A L
R O T A T I O N A N D PO S T E R I O R C R U C I A T E G R A F T F O R C E S

tact. Clearly, the graft-tensioning protocols used in the present

study were unable to restore both graft forces and external tibial rotations to normal. Our study was designed to directly
compare reconstruction of the popliteus tendon with that of
the popliteofibular ligament by utilizing the same graft for
both reconstructions. It was not possible to test both reconstructions simultaneously. It is likely that reconstruction of all
three posterolateral structures would overconstrain external
tibial rotation to an even greater degree. More studies are required to develop other graft-tensioning protocols that more
closely replicate normal external rotatory laxity.
In conclusion, our study demonstrated that the knee is
highly unstable with all posterolateral structures removed: the
mean maximum tibial rotations and posterior cruciate ligament graft forces were nearly 30 and 200 N, respectively.
Reconstruction of the lateral collateral ligament alone was insufficient to restore normal posterior cruciate force levels.
Addition of a reconstruction of the popliteus tendon or popliteofibular ligament to a reconstruction of the lateral collateral
ligament significantly reduced forces in the posterior cruciate
ligament graft to normal (or below-normal) levels and markedly reduced external tibial rotation by approximately 7 (in 0
of flexion) to 13 (in 90 of flexion) (p < 0.05). External tibial
rotations associated with reconstructions of the lateral collateral ligament and popliteus tendon were equivalent to those
associated with reconstructions of the lateral collateral and
popliteofibular ligaments, and both were significantly less
than those in the intact knee (i.e., the reconstructed knee was
overconstrained). Increasing tension in either a popliteus or a
popliteofibular graft from 10 to 30 N significantly decreased
mean tibial rotations approximately by 2 to 3. Further research in this important area is warranted to determine optimum graft tensioning that will better restore normal graft
forces and knee kinematics.
NOTE: Human tissues utilized for this study were provided by the Musculoskeletal Transplant
Foundation.

Keith L. Markolf, PhD

Benjamin R. Graves, MD
Susan M. Sigward, PhD
Steven R. Jackson
David R. McAllister, MD
Biomechanics Research Section, Department of Orthopaedic Surgery,
University of California at Los Angeles Rehabilitation Center, 1000 Veteran Avenue, Room 21-67, Los Angeles, CA 90095-1759. E-mail address
for K.L. Markolf: kmarkolf@mednet.ucla.edu

References
1. Clancy WG, Sutherland TB. Combined posterior cruciate ligament injuries. Clin
Sports Med. 1994;13:629-47.
2. Fanelli GC, Edson CJ. Posterior cruciate ligament injuries in trauma patients:
part II. Arthroscopy. 1995;11:526-9.
3. Harner CD, Hher J. Evaluation and treatment of posterior cruciate ligament
injuries. Am J Sports Med. 1998;26:471-82.
4. LaPrade RF, Terry GC. Injuries to the posterolateral aspect of the knee. Association of anatomic injury patterns with clinical instability. Am J Sports Med.
1997;25:433-8.

5. LaPrade RF, Johansen S, Wentorf FA, Engebretsen L, Esterberg JL, Tso A.

An analysis of an anatomical posterolateral knee reconstruction: an in vitro biomechanical study and development of a surgical technique. Am J Sports Med.
2004;32:1405-14.
6. Larsen MW, Moinfar AR, Moorman CT. Posterolateral corner reconstruction:
fibular-based technique. J Knee Surg. 2005;18:163-6.
7. Khanduja V, Somayaji HS, Harnett P, Utukuri M, Dowd GS. Combined reconstruction of chronic posterior cruciate ligament and posterolateral corner deficiency. A
two- to nine-year follow-up study. J Bone Joint Surg Br. 2006;88:1169-72.

Markolf.fm Page 2358 Thursday, October 11, 2007 11:53 AM

2358
THE JOURNAL OF BONE & JOINT SURGER Y JBJS.ORG
VO L U M E 89-A N U M B E R 11 N O VE M B E R 2007

E F F E C T S O F P O S T E RO L A T E R A L R E CO N S T R U C T I O N S O N T I B I A L
R O T A T I O N A N D PO S T E R I O R C R U C I A T E G R A F T F O R C E S

8. Kanamori A, Lee JM, Haemmerle MJ, Vogrin TM, Harner CD. A biomechanical
analysis of two reconstructive approaches to the posterolateral corner of the
knee. Knee Surg Sports Traumatol Arthrosc. 2003;11:312-7.

18. Sekiya JK, Haemmerle MJ, Stabile KJ, Vogrin TM, Harner CD. Biomechanical
analysis of a combined double-bundle posterior cruciate ligament and posterolateral corner reconstruction. Am J Sports Med. 2005;33:360-9.

9. Veltri DM, Warren RF. Operative treatment of posterolateral instability of the

knee. Clin Sports Med. 1994;13:615-27.

19. Markolf KL, Feeley BT, Jackson SR, McAllister DR. Biomechanical studies of
double-bundle posterior cruciate ligament reconstructions. J Bone Joint Surg Am.
2006;88:1788-94.

10. Stannard JP, Brown SL, Farris RC, McGwin G Jr, Volgas DA. The posterolateral corner of the knee: repair versus reconstruction. Am J Sports Med.
2005;33:881-8.
11. Mller W. The knee: form, function, and ligament reconstruction. Berlin,
Heidelberg: Springer-Verlag; 1983.
12. LaPrade RF, Ly TV, Wentorf FA, Engebretsen L. The posterolateral attachments
of the knee: a qualitative and quantitative morphologic analysis of the fibular collateral ligament, popliteus tendon, popliteofibular ligament, and lateral gastrocnemius tendon. Am J Sports Med. 2003;31:854-60.
13. Vogrin TM, Hher J, Aren A, Woo SL, Harner CD. Effects of sectioning the
posterolateral structures on knee kinematics and in situ forces in the posterior
cruciate ligament. Knee Surg Sports Traumatol Arthrosc. 2000;8:93-8.
14. LaPrade RF, Muench C, Wentorf FA, Lewis JL. The effect of injury to the posterolateral structures of the knee on force in a posterior cruciate ligament graft: a
biomechanical study. Am J Sports Med. 2002;30:233-8.
15. Markolf KL, Wascher DC, Finerman GA. Direct in vitro measurement of forces
in the cruciate ligaments. Part II: the effect of section of the posterolateral structures. J Bone Joint Surg Am. 1993;75:387-94.
16. Nau T, Chevalier Y, Hagemeister N, Deguise JA, Duval N. Comparison of 2 surgical techniques of posterolateral corner reconstruction of the knee. Am J Sports
Med. 2005;33:1838-45.
17. Nau T, Chevalier Y, Hagemeister N, Duval N, deGuise JA. 3D kinematic in-vitro
comparison of posterolateral corner reconstruction techniques in a combined injury model. Knee Surg Sports Traumatol Arthrosc. 2005;13:572-80.

20. Markolf KL, Burchfield DM, Shapiro MM, Cha CW, Finerman GA, Slauterbeck
JL. Biomechanical consequences of replacement of the anterior cruciate ligament
with a patellar ligament allograft. Part II: forces in the graft compared with forces
in the intact ligament. J Bone Joint Surg Am. 1996;78:1728-34.
21. Lee MC, Park YK, Lee SH, Jo H, Seong SC. Posterolateral reconstruction using split Achilles tendon allograft. Arthroscopy. 2003;19:1043-9.
22. Sekiya JK, Kurtz CA. Posterolateral corner reconstruction of the knee: surgical technique utilizing a bifid Achilles tendon allograft and a double femoral tunnel. Arthroscopy. 2005;21:1400.
23. Stannard JP, Brown SL, Robinson JT, McGwin G, Volgas DA. Reconstruction of
the posterolateral corner of the knee. Arthroscopy. 2005;21:1051-9.
24. Markolf K, Davies M, Zoric B, McAllister D. Effects of bone block position and
orientation within the tibial tunnel for posterior cruciate ligament graft reconstructions: a cyclic loading study of bone-patellar tendon-bone allografts. Am J Sports
Med. 2003;31:673-9.
25. Markolf KL, Zemanovic JR, McAllister DR. Cyclic loading of posterior cruciate
ligament replacements fixed with tibial tunnel and tibial inlay methods. J Bone
Joint Surg Am. 2002;84:518-24.
26. Twaddle BC, Bidwell TA, Chapman JR. Knee dislocations: where are the
lesions? A prospective evaluation of surgical findings in 63 cases. J Orthop
Trauma. 2003;17:198-202.
27. Baker CL, Norwood LA, Hughston JC. Acute posterolateral rotatory instability
of the knee. J Bone Joint Surg Am. 1983;65:614-8.