Injury to the PCL has recently become an intriguing topic
of discussion in the sports medicine arena. Advancements
in the understanding of PCL functional anatomy, injury
incidence, natural history of PCL-deficient knees, and
methods of reconstruction have led to more interest in
The complex anatomy of the PCL has initially been described
as consisting of 2 functional primary bundles.15,19,29The
anterolateral bundle is the primary restraint at 90° of flex-
ion, is approximately twice the width of the posteromedial
bundle, is stiffer, and has a higher ultimate load to tensile
testing.19,29Because of these findings, reconstruction of the
PCL has focused on the anterolateral bundle.
Many operative techniques have been described for PCL
reconstruction. Unfortunately, no current surgical proce-
dure has been able to consistently correct abnormal poste-
rior laxity or to provide consistent functional results.1,5,22
Clinical results after PCL reconstruction have not been as
predictable as those after ACL reconstruction.1It is cur-
rently unclear whether PCL reconstruction techniques can
A Biomechanical Comparison of
Posterior Cruciate Ligament
Reconstructions Using Single- and
Double-Bundle Tibial Inlay Techniques
John A. Bergfeld,* MD,Scott M. Graham,†‡MD,Richard D. Parker,* MD,
Antonio D. C. Valdevit,§MSc, andHelen E. Kambic,llPhD
From the *Department of Orthopaedic Surgery, Cleveland Clinic Foundation, Cleveland, Ohio,
the ‡South County Orthopedic Specialists, Laguna Hills, California, the §Department of Surgery,
Lutheran Medical Center, Brooklyn, New York, and the llDepartment of Biomedical Engineering,
Cleveland Clinic Foundation, Cleveland, Ohio
Background: The efficacy of using a double-bundle versus single-bundle graft for posterior cruciate ligament reconstruction has
not been demonstrated.
Hypothesis: A double-bundle graft restores knee kinematics better than a single-bundle graft does in tibial inlay PCL recon-
Study Design: Controlled laboratory study.
Methods: Eight cadaveric knees were subjected to 6 cycles from a 40-N anterior reference point to a 100-N posterior transla-
tional force at 10°, 30°, 60°, and 90° of flexion. Testing was performed for the intact and posterior cruciate deficient knee as well
as for both reconstructed conditions. Achilles tendons, divided into 2 equal sections, were prepared as both single-bundle and
double-bundle grafts. Both grafts were employed in the same knee, and the order of graft reconstruction was randomized.
Results: There were no statistical differences in translation between the intact state and either of the reconstructions (P > .05)
or between either of the reconstructions at any flexion angle (P > .05).
Conclusion: No differences in translation between the 2 graft options were identified.
Clinical Relevance: The use of a double-bundle graft may not offer any advantages over a single-bundle graft for tibial inlay
posterior cruciate reconstructions.
Keywords: posterior cruciate ligament (PCL); double bundle; reconstruction; ligament; knee
†Address correspondence to Scott M. Graham, MD, South County
Orthopedic Specialists, 24331 El Toro Road, Suite 200, Laguna Hills, CA
92653 (e-mail: firstname.lastname@example.org).
Presented at the ORS and AAOS annual meeting, Dallas, Texas,
February 2002, and at the AOSSM annual meeting, Orlando, Florida, July
No potential conflict of interest declared.
The American Journal of Sports Medicine, Vol. 33, No. 7
© 2005 American Orthopaedic Society for Sports Medicine
¶References 3-5, 8-12,16-18, 23, 25, 27, 28, 30.
Vol. 33, No. 7, 2005 Biomechanical Comparison of PCL Reconstructions
restore normal knee laxity in vivo and thus alter the nat-
ural history of the PCL-deficient knee.
The reconstruction of the PCL varies from surgeon to
surgeon with respect to ligament graft selection and tech-
nique of fixation. An important variable is the orientation
of the graft on the tibia (ie, tibial inlay vs tibial tunnel).
Jakob and Rüegsegger,20as well as Berg,3have advocated
the tibial inlay technique for its theoretical advantages
regarding graft fiber orientation. Previous research has
shown that at initial fixation and after cyclic loading, the
posterior tibial inlay technique results in decreased poste-
rior translation with less graft degradation when com-
pared with the tunnel technique of PCL reconstruction.5,25
Recent discussions have focused on a more anatomical
PCL reconstruction method using a double-bundle (DB)
graft.Although no clinical outcome studies have been pub-
lished, previous authors have evaluated DB grafts and
have concluded that this method of reconstruction may be
superior.17,23,30Two authors have advocated using a DB
graft with the arthroscopic (tibial tunnel) method of recon-
struction,17,23while other researchers showed that DB
reconstruction using the tibial inlay technique of fixation
provided improved stability over a range of knee motion.30
In these biomechanical studies, the authors added a sec-
ond graft for their DB reconstructions, demonstrating that
2 grafts were an improvement over a single graft. There
are no published mechanical studies evaluating DB PCL
reconstructions that control for the quantity of graft used
for each testing condition.
The purpose of this study was to compare the resultant
laxity provided by the tibial inlay technique using both
single-bundle (SB) and DB grafts for PCL reconstruction
while using the same quantity of graft material. We
hypothesized that with the tibial inlay method of PCL
reconstruction, a DB reconstruction would restore knee
kinematics better than an SB reconstruction would.
MATERIALS AND METHODS
A total of 4 male and 4 female unpaired fresh-frozen
cadaveric knees with a mean age of 48.5 years (range, 37-
63 years) was used in this study. The specimens did not
show signs of injury or degenerative joint disease.
For sample size determination, we assumed a statistical
power of 90% (β = .1) and a significance level (α) of .05,
while using the method described by Lieber.21Standard
deviations from an initial study5were used, and a differ-
ence in the translation data equal to 2 times the standard
deviation was obtained. For a power of 0.9, 6 knees per
group were needed at the various angles. In our study, we
used 8 knees per angle.
Before testing, each specimen was thawed overnight at
room temperature. Each speciman was dissected free of
skin,sparing the joint capsule,ligaments,popliteus tendon
and muscle, and the extensor mechanism. A medial
arthrotomy was performed for joint inspection along with
a vertical posterior arthrotomy to evaluate the intact PCL.
The arthrotomies remained open throughout all testing.
The fibula was resected 4 cm distal to the tibiofibular joint
and stabilized to the tibia with a fully threaded Steinmann
pin. The proximal end of the femur and the distal end of
the tibia were potted in aluminum tubes with polymethyl-
methacrylate. Crossing transfixion pins were used to elim-
inate any rotation between the bones and respective hold-
ing tubes. Specimens were then mounted in a custom test-
ing apparatus described previously,5 with 6 degrees of
adjustability that did not overconstrain the knee while
allowing for testing through flexion and extension.
The custom testing apparatus was similar to that
described by Fleming et al13 and used by Bergfeld et al5
(Figure 1). The potted femoral end of the specimen was
clamped into a yoke attached to the load ram of a servo-
hydraulic universal testing machine (MTS Systems Corp,
Eden Prairie, Minn) that applied an anterior-posterior
(AP) shear force to the knee. The tibial mechanism was
mounted on a translating table that allowed free motion of
the tibia in the coronal plane and that locked in neutral
rotation. The epicondylar axis of the knee joint was placed
at the pivot axis of the femoral yoke.The knee joint flexion
angle was adjusted with the yoke and locked at the desired
flexion angle before each test. A bearing system allowed
free motion of varus-valgus angulation. A posterior-to-
anterior tibial force (simulated anterior drawer) of 40 N
was used to establish an initial reference point from which
posterior translation was measured. Since the ACL is not
involved in the surgery, it is reasonable that the reference
for displacement should be established using an anterior-
ly applied load to the tibia that will engage the ACL. This
position was reproduced at each angle by the MTS
machine and was used as the reference point from which
posterior translation was measured. During loading
cycles, this same point was used as the anterior limit. Six
sequential AP loading cycles from the initial anterior ref-
erence point to 100 N of posterior-directed force were then
applied. These loading cycles were performed at 1 mm/s at
the level of the joint line while AP translation was meas-
ured by the linear variable displacement transducer
(LVDT) of the MTS machine. In all cases, the load-
deformation curve became reproducible after 2 loading
cycles. The testing was conducted at 10°, 30°, 60°, and 90°
of flexion. Each knee served as its own control and was
tested with an intact and transected PCL, followed by both
SB and DB PCL reconstructions with the tibial inlay tech-
nique (Figure 2). The order of graft reconstruction was
randomized throughout the study.
After laxity data had been obtained from the intact knees,
the PCL and meniscofemoral ligaments were resected
using the anterior and posterior arthrotomy previously
made.The meniscofemoral ligaments were removed,as they
Bergfeld et al The American Journal of Sports Medicine
are closely associated with the PCL and have previously
been shown to contribute to posterior knee stability.19
Each specimen was removed from the testing jig between
the loading conditions to ensure removal of the PCL with-
out damaging other structures. The specimen was then
returned to the jig in the same position.The zero-load neu-
tral position was maintained after reloading the speci-
mens. The AP loading was repeated using the previously
described anterior reference point and testing conditions.
Each knee underwent PCL reconstruction with both an SB
and a DB graft using the tibial inlay technique. Both grafts
were employed in the same knee, and the order in which
each graft was used for reconstruction was randomized.From
this design, each knee was able to serve as its own control.
To retain comparable structural and material properties
for each graft, allograft material from 1 specimen was used
for both grafts (Figure 3).Allograft Achilles tendons (MTF,
Edison, NJ), including the os calcis bone block, were sepa-
rated into 2 equal sections (average age for grafts, 48
years). One half was used for the SB graft while the second
half was split again into 2 bundles. The bone block of this
second half was left intact, thereby creating a DB graft.
Special care was taken to tease the tendon apart from dis-
tal to proximal, as the fibers of the Achilles tendon are not
purely longitudinal in direction. The 2 grafts had essen-
tially the same material properties and quantity of colla-
gen, thus the translational differences observed stemmed
from graft geometry and not merely from the addition of a
second graft. A No. 5 Ticron suture (Syneture, Norwalk,
Conn) was woven through the end of each limb in a
Krakow-type fashion so that the suture would extend from
within the bone tunnel and not be intra-articular. This
method was used for graft tensioning and improved inter-
ference graft fixation.
Bone tunnels were created through the medial femoral
condyle (MFC) in the anatomical footprint of the PCL with
an outside-in technique, by use of a PCL femoral drill
guide (Acufex, Mansfield, Mass). The tunnel reproducing
the anterolateral bundle was made anterior and distal in
the footprint of the native PCL, while the tunnel repro-
ducing the posteromedial bundle, when used, was made
posterior and slightly proximal (Figure 4).17,23The direc-
tions used for the guidewire for each tunnel were conver-
gent and directed toward the proximal posterior tibia with
the knee at 90° of flexion.
The previously made posterior capsulotomy was opened
to expose the proximal posterior tibia. A bone trough
measuring 10 × 14 mm was created with osteotomes at the
tibial insertion site of the PCL.The matching bone block of
the graft was secured flush to the posterior surface of the
Figure 1. A custom test jig with 6 degrees of adjustability
was mounted on an MTS testing machine.
Figure 2. Illustrations from a posterior view of single-bundle
(A) and double-bundle (B) reconstructions using the tibial
Figure 3. Separation of an Achilles tendon to generate both
a single-bundle and double-bundle graft of equal amounts of
tissue from the same allograft specimen. One half of the
graft was split sagittally into 2 equal bundles for the double-
Vol. 33, No. 7, 2005 Biomechanical Comparison of PCL Reconstructions
proximal tibia with a 4.5-mm bicortical screw (Synthes,
Paoli, Pa), with a washer and a nut securing the fixation
on the anterior aspect of the tibia.
An anterior force of 40 N was applied to the proximal
tibia,simulating an anterior drawer maneuver while femoral
interference screw(s) (Acufex) were placed to secure the
graft(s). The SB graft, reproducing the anterolateral bun-
dle, was tensioned to 89 N (Chatillon digital tensioner,
Ametek, Largo, Fla) at 90° of knee flexion. For the DB
reconstruction, the anterolateral and posteromedial bun-
dles were each tensioned to 89 N at 90° and 30° of flexion,
respectively. Biomechanical evaluation has shown that at
these angles, the 2 respective portions of the PCL become
taut.19In addition, this method of tensioning closely resem-
bles that which may be used clinically. While under ten-
sion, additional soft tissue clamps were placed across each
graft where the woven suture was placed as it exited the
MFC. This procedure further improved immediate graft
fixation by buttressing the graft against the outer cortex of
the MFC. Laxity testing was then repeated under the same
conditions as previously described. At the conclusion of
mechanical testing, the grafts were removed from the knees
of both groups and inspected for defects. No specimens
from either group showed evidence of mechanical abrasion.
When specimens underwent an SB reconstruction first,
a single femoral tunnel was drilled with a 9-mm drill
through the anterodistal aspect of the femoral anatomical
attachment site of the PCL. This tunnel reproduced the
anterolateral bundle of the PCL. The 9-mm graft was then
tensioned as described and secured with a 7-mm interfer-
ence screw.Testing then took place.After completion of the
SB testing, the graft was removed and the knee was pre-
pared for a DB graft. For the following DB reconstruction
of the same knee, a second tunnel was drilled with a 7-mm
drill through the posterior aspect of the anatomical attach-
ment site for the posteromedial bundle. Because both arms
of the DB graft measured 7 mm, a 9-mm interference
screw was placed in the first tunnel (9-mm anterolateral
tunnel), while a 7-mm interference screw was used for the
second tunnel (7-mm posteromedial tunnel). Testing again
When specimens underwent a DB reconstruction first,
the 9-mm anterolateral and the 7-mm posteromedial tun-
nels were initially drilled. The DB graft was then secured
with the same fixation as described above.Testing then took
place. After completion of the DB testing, the graft was
removed and the knee was prepared for an SB graft. This
procedure simply entailed the placement of a 9-mm inter-
ference screw into the posteromedial tunnel to fill it. The
SB graft could then be brought through the previously placed
9-mm anterolateral tunnel and secured in place with a 7-
mm interference screw, as was performed when the order
of testing was reversed.By following this protocol,each test-
ing situation used the same diameter tunnels and inter-
ference screws, regardless of the order of reconstruction.
Each graft was measured with a graft sizer (Acufex) to
within 0.5-mm diameter. The SB grafts averaged 9.0 mm
in diameter, while each arm of the DB grafts averaged 6.5
mm. The total cross-sectional area of each graft used for
any single knee (SB vs DB) was within 5%.
The total AP translation from the final loading cycle of
each specimen was extracted for each flexion angle tested.
Translational measurements (mm) from the same knee
under 4 different conditions (intact, PCL-deficient, SB, and
DB graft reconstructions) and at 4 different angles were
statistically compared. The translational data of the PCL-
deficient and reconstructed groups were compared to those
of the intact knees. The reconstructed groups were also
compared to each other. A repeated-measures analysis of
variance, in conjunction with a Newman-Keuls analysis,
was used for comparison between the groups. Statistical
significance was taken as P = .05.
There was a trend for the DB construct to slightly over-
tighten the knee from 30° to 60°, whereas the SB construct
tended to allow slightly more translation between 10° and
90° (Figure 5). However, no differences in translation
between the intact state and either of the reconstructions
were found to be statistically significant at any flexion
angle tested (P > .05). Additionally, no statistical differ-
ences between the SB and DB reconstructions were found
at any angle (P > .05) (Figure 5).With the tibial inlay tech-
nique, both the SB and DB PCL reconstructions closely
reproduced the stability seen in the intact knee.
In our review of the current literature, few articles have
been published about biomechanical testing of DB PCL
Figure 4. For double-bundle reconstructions, an anterior
and distal (shallow) femoral tunnel reproducing the anatomi-
cal anterolateral bundle was used. A posterior and slightly
proximal second tunnel was added to reproduce the
anatomical posteromedial bundle while staying within the
native PCL footprint. Both images are of a right knee.
Bergfeld et al The American Journal of Sports Medicine
reconstructions,17,23,30although many technical notes and
surgical methods have been described.4,16,26,31This rela-
tively new concept has been initiated because of the varied
clinical outcomes seen after PCL reconstruction.8,22
Studies have documented the importance of femoral tun-
nel placement2,23,30in PCL reconstruction and have shown
that nonisometric positioning of the graft best corrects
abnormal posterior laxity.6,10,14,30For either SB or antero-
lateral bundle reconstruction, Bomberg et al,6Burns et al,7
and Galloway et al14suggested making the femoral tunnel
in the anterior distal footprint of the native PCL.
Mannor et al23performed an in vitro analysis of PCL
graft placement and tension. The authors evaluated the
relationship between SB and DB reconstructions, with
varying positions of the femoral tunnels for each.Although
no conclusion was drawn regarding the superiority of SB
versus DB grafts, the authors did conclude that femoral
tunnel position affected translation and bundle tension. A
shallow SB reconstruction did restore posterior transla-
tion to within 2 mm of that of the intact knee over the
entire range of knee flexion. The DB reconstruction was
evaluated with a high and shallow first bundle and then
either a low and shallow or a deep second bundle. While
both reconstructions controlled posterior translation, each
did so with different load-sharing characteristics. Our
reconstructions were similar to those of Mannor et al.23An
anterior and distal (S1) femoral tunnel reproducing the
anatomical anterolateral bundle was used for SB recon-
struction. For our DB reconstructions, a posterior and
proximal (S2) second tunnel was added to reproduce the
anatomical posteromedial bundle while staying within the
native PCL footprint. However, unlike Mannor et al,23who
tensioned both of their grafts at 90° of flexion, our second
bundle was tensioned at 30° of knee flexion.
Harner et al17found that a DB reconstruction better
reapproximated both normal knee laxity and PCL forces
throughout the range of knee flexion using a robotic
manipulator and the principle of superposition. However,
like Race and Amis,30they also simply added a second
graft to simulate a DB reconstruction. A 10-mm Achilles
tendon graft was used for the SB reconstruction, while a
second 8-mm doubled semitendinosus graft (≈18 mm total)
was added for the DB reconstruction. This procedure may
not have tested the benefit of a DB graft but that of a larger
graft with a greater cross-sectional area and origin foot-
print. The authors used an arthroscopic tibial tunnel
method, which may have introduced bias, as the simple
addition of more graft material is likely to provide
improved stability. Markolf et al25revealed that the
arthroscopic tunnel method of PCL reconstruction places
grafts at risk of degradation and loosening with cycling.
Harner et al17also tensioned the grafts from the tibial
side. Tensioning on the femoral side has been shown to
result in higher tension on the intra-articular portion of
the graft when compared to tensioning being performed on
the tibial side of the graft.24Final tensioning is per-
formed on the femoral side using the tibial inlay method,
as in our study. This practice may be because of the
decreased turn the graft must make with the tibial inlay
method and the reduced resistance to tension as it
Race and Amis30reported that only a DB reconstruction
restored normal knee laxity across the full range of knee
motion when compared to an anatomical SB and an “iso-
metric” SB PCL reconstruction. Their isometric point did
not appear to fall within the femoral anatomical attach-
ment site, which may explain the overconstraint in exten-
sion and the laxity in flexion that they found. We agree
with Ortiz et al27that no true isometric point exists, as the
point shifts under different loading conditions. In compar-
ing their anatomical SB to a DB graft reconstruction, Race
and Amis30used the femoral footprints of the native PCL
while securing the tibial bone block with a posterior tibial
inlay technique.Like Harner et al,17the authors used a 10-
mm graft for the SB and then added a second 8-mm graft
for the DB (≈18 mm total). This method may also have
introduced bias by simply adding a second graft. We con-
trolled for this discrepancy by using the same quantity of
graft for each reconstruction (ie, SB vs DB). Race and
Amis30tensioned the posteromedial bundle at 130°, whereas
we tensioned it at 30°, closer to extension, where the pos-
teromedial bundle acts as a restraint to posterior transla-
tion in a more functional manner.15,19
There were 3 reasons we chose to use Achilles tendon
grafts. Harner et al19felt Achilles tendons have an advan-
tage, as they are more rounded and provide good repro-
duction of the native PCL.Secondly, a single Achilles graft
is large enough to be formed into both an SB and a DB
graft while maintaining similar material properties.
Lastly, the use of Achilles tendon allografts for PCL recon-
struction is a common clinical practice.
Double-bundle PCL reconstruction may not be neces-
sary when using the tibial inlay method.Previous research
has shown that graft abrasion and compromise occur with
repetitive cycling when using the arthroscopic tunnel tech-
nique. These problems are not seen in grafts placed with
the tibial inlay technique.5,25These issues might be more
responsible for the variable clinical results that have been
reported rather than the use of SB grafts. The tibial inlay
method negates the killer turn and provides secure bone-
to-bone fixation of the graft to the tibia.
Figure 5. Posterior translation at the different flexion angles
for each of the testing conditions. No statistically significant
differences were noted.
Vol. 33, No. 7, 2005 Biomechanical Comparison of PCL Reconstructions Download full-text
Using the tibial inlay technique, we evaluated whether
a DB graft would differ from an SB graft in reproducing
stability to a PCL-deficient knee while controlling for the
amount of graft used under each testing condition. Our
results did not find any significant difference in transla-
tion between either of the reconstruction methods (ie, SB
vs DB) and the intact knee (P > .05). In addition, no sta-
tistical differences between the SB and DB reconstruc-
tions were found at any angle (P > .05). The DB construct
appeared to allow less posterior translation than seen with
the SB construct across all angles tested. This restraint to
posterior translation, however, was greater than that of
the intact knee from 30° to 60°.
There are limitations to our study. We did not perform
testing at flexion angles greater than 90°. Testing at flex-
ion angles greater than 90° could show differences that
were not revealed in our study. Additionally, our protocol
did not evaluate how the different grafts behaved over
time with extensive cycling. Such differences, if noted, may
be clinically relevant.
In conclusion, we did not identify any biomechanical
superiority in performing a DB reconstruction compared to
the SB technique and therefore did not support our hypoth-
esis. Although authors have reported advantages with the
DB method,we believe that by using the posterior tibial inlay
technique, the more technically challenging DB PCL recon-
struction may not offer any advantages over an SB graft.
The authors acknowledge the Musculoskeletal Transplant
Foundation for its generous donation of tissues used in
this research project. This study was supported by a grant
from the Cleveland Clinic Foundation Research Program
Committee and the Department of Orthopaedic Surgery,
Section of Sports Medicine Education and Research Fund.
1. Bach BR. Graft selection for posterior cruciate ligament surgery. Oper
Tech Sports Med. 1993;1:104-109.
2. Bach BR, Daluga DJ, Mikosz R, Andriacchi TP, Seidl R. Force dis-
placement characteristics of the posterior cruciate ligament. Am J
Sports Med. 1992;20:67-72.
3. Berg EE. Posterior cruciate ligament tibial inlay reconstruction.
4. Bergfeld JA, Graham SM. Tibial inlay procedure for PCL reconstruc-
tion: one tunnel and two tunnels. Oper Tech Sports Med. 2001;9:69-75.
5. Bergfeld JA, McAllister DR, Parker RD, Valdevit AD, Kambic HE. A
biomechanical comparison of posterior cruciate ligament reconstruc-
tion techniques. Am J Sports Med. 2001;29:129-136.
6. Bomberg BC, Acker JH, Boyle J, et al. The effect of posterior cruci-
ate ligament loss and reconstruction of the knee. Am J Knee Surg.
7. Burns WC, Draganich LF, Pyevich M, Reider B. The effect of femoral
tunnel position and graft tensioning technique on laxity of the poste-
rior cruciate ligament reconstructed knee. Am J Sports Med.
8. Clancy WG Jr, Shelbourne KD, Zoellner GB, Keene JS, Reider B,
Rosenberg TD. Treatment of knee joint instability secondary to rup-
ture of the posterior cruciate ligament: report of a new procedure. J
Bone Joint Surg Am. 1983;65:310-322.
9. Clendenin MB, DeLee JC, Heckman JD. Interstitial tears of the pos-
terior cruciate ligament of the knee. Orthopedics. 1980;3:764-772.
10. Covey DC, Sapega AA, Sherman GM. Testing for isometry during
reconstruction of the posterior cruciate ligament: anatomic and bio-
mechanical considerations. Am J Sports Med. 1996;24:740-746.
11. Cross MJ, Powell JF. Long-term follow-up of posterior cruciate liga-
ment rupture: a study of 116 cases. Am J Sports Med. 1984;12:292-297.
12. Dandy DJ, Pusey RJ. The long-term results of unrepaired tears of the
posterior cruciate ligament. J Bone Joint Surg Br. 1982;64:92-94.
13. Fleming B, Beynnon B, Howe J, McLeod W, Pope M. Effect of ten-
sion and placement of a prosthetic anterior cruciate ligament on the
anteroposterior laxity of the knee. J Orthop Res. 1992;10:177-186.
14. Galloway MT, Grood ES, Mehalic JN, Levy M, Saddler SC, Noyes FR.
Posterior cruciate ligament reconstruction: an in vitro study of femoral
and tibial graft placement. Am J Sports Med. 1996;24:437-445.
15. Girgis FG, Marshall JL, Monajem A. The cruciate ligaments of the
knee joint: anatomical, functional and experimental analysis. Clin
16. Graham SM, Parker RD, McAllister DR, Calabrese GJ. Tibial inlay
technique for posterior cruciate ligament reconstruction. Tech
17. Harner CD, Janaushek MA, Kanamori A, Yagi M, Vogrin TM, Woo SL.
Biomechanical analysis of a double-bundle posterior cruciate liga-
ment reconstruction. Am J Sports Med. 2000;28:144-151.
18. Harner CD, Janaushek MA, Ma CB, Kanamori A, Vogrin TM, Woo SL.
The effect of knee flexion angle and application of an anterior tibial
load at the time of graft fixation on the biomechanics of a posterior
cruciate ligament–reconstructed knee. Am J Sports Med.
19. Harner CD, Xerogeanes JW, Livesay GA, et al. The human posterior
cruciate ligament complex: an interdisciplinary study. Ligament mor-
phology and biomechanical evaluation. Am J Sports Med.
20. Jakob RP, Rüegsegger M. Therapy of posterior and posterolateral
knee instability. Orthopade. 1993;22:405-413.
21. Lieber RL. Experimental design and statistical analysis. In: Simon SR,
ed. Orthopaedic Basic Science. Rosemont, Ill: American Academy of
Orthopaedic Surgeons; 1994:623-665.
22. Lipscomb AB Jr, Anderson AF, Norwig ED, Hovis WD, Brown DL.
Isolated posterior cruciate ligament reconstruction: long-term results.
Am J Sports Med. 1993;21:490-496.
23. Mannor DA, Shearn JT, Grood ES, Noyes FR, Levy MS. Two-bundle
posterior cruciate ligament reconstruction: an in vitro analysis of graft
placement and tension. Am J Sports Med. 2000;28:833-845.
24. Markolf KL, Slauterbeck JR, Armstrong KL, Shapiro MS, Finerman
GA. A biomechanical study of replacement of the posterior cruciate
ligament with a graft, part I: isometry, pre-tension of the graft, and
anterior-posterior laxity. J Bone Joint Surg Am. 1997;79:375-380.
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;84:518-524.
26. Noyes FR, Medvecky MJ, Bhargava M. Arthroscopically assisted
quadriceps double-bundle tibial inlay posterior cruciate ligament
reconstruction: an analysis of techniques and a safe operative
approach to the popliteal fossa. Arthroscopy. 2003;19:894-905.
27. Ortiz GJ, Schmotzer H, Bernbeck J, Graham S, Tibone JE,
Vangsness CT Jr. Isometry of the posterior cruciate ligament: effects
of functional load and muscle force application. Am J Sports Med.
28. Parolie JM, Bergfeld JA. Long-term results of nonoperative treatment
of isolated posterior cruciate ligament injuries in the athlete. Am J
Sports Med. 1986;14:35-38.
29. Race A, Amis AA. The mechanical properties of the two bundles of
the human posterior cruciate ligament. J Biomech. 1994;27:13-24.
30. Race A, Amis AA. PCL reconstruction: in vitro biomechanical com-
parison of “isometric” versus single- and double-bundled “anatomic”
grafts. J Bone Joint Surg Br. 1998;80:173-179.
31. Richards RS, Moorman CT. Use of autograft quadriceps for double-
bundle posterior cruciate ligament reconstruction. Arthroscopy.