PETTERI KOUSA
Graft Fixation in ACL Reconstruction
ACADEMIC DISSERTATION
To be presented, with the permission of
the Faculty of Medicine of the University of Tampere,
for public discussion in the small auditorium of Building B,
Medical School of the University of Tampere,
Medisiinarinkatu 3, Tampere, on April 23rd, 2004, at 12 o’clock.
A c t a U n i v e r s i t a t i s T a m p e r e n s i s 1001
ACADEMIC DISSERTATION
University of Tampere, Medical School
Tampere University Hospital, Department of Surgery
Finland
Supervised by
Professor Markku Järvinen
University of Tampere
Docent Teppo Järvinen
University of Tampere
Reviewed by
Docent Arsi Harilainen
University of Helsinki
Docent Andreas Weiler
University of Berlin
Distribution
University of Tampere
Bookshop TAJU
P.O. Box 617
33014 University of Tampere
Finland
Tel. +358 3 215 6055
Fax +358 3 215 7685
[email protected]
http://granum.uta.fi
Cover design by
Juha Siro
Printed dissertation
Acta Universitatis Tamperensis 1001
ISBN 951-44-5944-X
ISSN 1455-1616
Tampereen yliopistopaino Oy Juvenes Print
Tampere 2004
Electronic dissertation
Acta Electronica Universitatis Tamperensis 334
ISBN 951-44-5945-8
ISSN 1456-954X
http://acta.uta.fi
To Nina,
Emilia, Sofia, Maria and Elias
3
CONTENTS
LIST OF ORIGINAL PUBLICATIONS ...................................................... 7
ABBREVIATIONS....................................................................................... 8
ABSTRACT .................................................................................................. 9
INTRODUCTION....................................................................................... 11
REVIEW OF THE LITERATURE............................................................. 13
1. Knee joint motion and stability..........................................................................13
2. Anatomy and function of the ACL ....................................................................14
3. ACL reconstruction............................................................................................16
3.1. Operative technique...............................................................................16
3.2. Graft selection........................................................................................16
3.3. Graft placement .....................................................................................20
3.4. Graft preconditioning and tension .........................................................21
3.5. Graft healing ..........................................................................................22
4. ACL graft fixation..............................................................................................23
4.1. Tibial and femoral fixations ..................................................................24
4.2. Ideal graft fixation .................................................................................24
5. Biomechanics of ACL graft and fixation...........................................................26
5.1. Material properties of the graft..............................................................26
4
5.2. Structural properties of graft fixation complex..................................... 27
5.3. Biomechanical testing of ACL reconstructions .................................... 28
5.4. Factors influencing the biomechanical testing of ACL graft fixation .. 29
5.4.1. Biologic factors......................................................................... 30
5.4.2. Experimental factors................................................................. 31
6. Current ACL graft fixation options ................................................................... 32
6.1. Bone-tendon graft fixation .................................................................... 32
6.1.1. Cortical fixation........................................................................ 32
6.1.2. Interference fixation ................................................................. 33
6.2. Soft tissue graft fixation........................................................................ 36
6.2.1. Cortical fixations ...................................................................... 37
6.2.2. Interference fixation ................................................................. 39
6.2.3. Transfixation............................................................................. 42
AIMS OF THE STUDY..............................................................................44
MATERIALS AND METHODS ................................................................45
1. Specimens.......................................................................................................... 45
2. Study groups and fixation procedures ............................................................... 46
2.2. Bone-tendon graft fixation techniques (studies I, II and III) ................ 46
2.3. Soft tissue graft fixation techniques (studies IV and V) ....................... 50
2.3.1. Study IV.................................................................................... 50
2.3.2. Study V ..................................................................................... 53
3. Biomechanical testing ....................................................................................... 56
3.1. Single load-to-failure test...................................................................... 58
3.2. Cyclic loading test followed by a single load-to-failure test ................ 58
4. Statistical analysis ............................................................................................. 60
RESULTS....................................................................................................62
1. Single load-to-failure tests................................................................................. 62
1.1. Study I ................................................................................................... 62
1.2. Study II.................................................................................................. 62
1.3. Study III ................................................................................................ 63
5
1.4. Study IV.................................................................................................64
1.5. Study V ..................................................................................................64
2. Cyclic loading tests ............................................................................................65
2.1. Study II ..................................................................................................65
2.2. Study III.................................................................................................66
2.3. Study IV.................................................................................................66
2.4. Study V ..................................................................................................68
DISCUSSION ............................................................................................. 70
1. Methodological issues........................................................................................71
2. BPTB graft fixation............................................................................................72
3. Hamstring graft fixation.....................................................................................76
CONCLUSIONS......................................................................................... 81
ACKNOWLEDGMENTS........................................................................... 82
REFERENCES............................................................................................ 84
ORIGINAL PUBLICATIONS.................................................................. 109
6
LIST OF ORIGINAL PUBLICATIONS
This thesis is based on the following original publications, referred to as I-V in the text:
I
Kousa P, Järvinen TLN, Pohjonen T, Kannus P, Kotikoski M, Järvinen M (1995):
Fixation strength of a biodegradable screw in anterior cruciate ligament
reconstruction. J Bone Joint Surg 77B: 901-905.
II
Kousa P, Järvinen TLN, Kannus P, Järvinen M (2001): Initial fixation strength of
bioabsorbable and titanium interference screws in anterior cruciatae ligament
reconstruction. Biomechanical evaluation by single cycle and cyclic loading. Am J
Sports Med 29: 420-425.
III
Kousa P, Järvinen TLN, Kannus P, Ahvenjärvi P, Kaikkonen A, Järvinen M (2001):
A bioabsorbable plug in bone-tendon-bone reconstruction of the anterior cruciate
ligament: Introduction of a novel fixation technique. Arthroscopy 17: 144-150.
IV
Kousa P, Järvinen TLN, Vihavainen M, Kannus P, Järvinen M (2003): The fixation
strength of six hamstring tendon graft fixation devices in anterior cruciate ligament
reconstruction. Part I: Femoral site. Am J Sports Med 31: 174-181.
V
Kousa P, Järvinen TLN, Vihavainen M, Kannus P, Järvinen M (2003): The fixation
strength of six hamstring tendon graft fixation devices in anterior cruciate ligament
reconstruction. Part II: Tibial site. Am J Sports Med 31: 182-188.
7
ABBREVIATIONS
ACL
Anterior cruciate ligament
ANOVA
Analysis of variance
BMD
Bone mineral density
BPTB
Bone-patellar tendon-bone
E
Elastic modulus
LCL
Lateral collateral ligament
MCL
Medial collateral ligament
MRI
Magnetic resonance imaging
N
Newton
PLA
Polylactide
PLA96/4
Poly-L-lactide/D-lactide copolymer
PCL
Posterior cruciate ligament
QT
Quadriceps tendon
SR-PLLA
Self-reinforced poly-L-lactide
STG
Semitendinosus and gracilis tendons
SD
Standard deviation
8
ABSTRACT
Intra-articular reconstruction with a biologic tendon graft is the procedure of choice for
restoring stability of a knee after rupture of the anterior cruciate ligament (ACL). During the
immediate postoperative period after ACL reconstruction the graft fixation site, not the graft
material itself, is considered the weakest link. In this study series, the initial graft fixation
strength of the different bone-patellar tendon-bone (BPTB) and hamstring tendon graft
fixation alternatives were evaluated by single load-to-failure and cyclic loading tests.
In the first part of this study series, it was found that the bioabsorbable fibrillated selfreinforced poly-L-lactide (SR-PLLA) screw provided as good initial fixation strength as the
Kurosaka interference screw and the AO cancellous screw in the fixation of a BPTB graft in
the bovine knee, and thus, could be considered a suitable alternative for the metal screws in
the BPTB graft fixation in the reconstruction of the ACL.
Also to validate the suitability of interference screws “design-wise”, the initial BPTB graft
fixation strength of a bioabsorbable interference screw designed for endoscopic ACL
reconstruction was evaluated in the second part of this study series. The bioabsorbable
interference screw provided as good initial fixation strength as the standard metal
interference screw in the fixation of a BPTB graft in the reconstruction of the ACL.
Although the interference screw fixation of the BPTB graft is safe and reliable in most cases,
there are some pitfalls related to the operative technique, especially in the technically more
demanding femoral site. To avoid these difficulties, a new plugging technique was
developed. In the third part of this study series, the initial fixation strength of the new
plugging and the standard interference screw techniques in the BPTB reconstruction of the
ACL were compared. Considering the biomechanical testing results that we obtained with
9
these two techniques, the plugging technique and standard interference technique provide
similar initial fixation strength in the fixation of a BPTB graft in the femoral tunnel in the
reconstruction of the ACL.
The fixation strength of different hamstring tendon graft fixation devices in the ACL
reconstruction were evaluated in studies four and five. The results show that the cross pin
technique (BoneMulchScrew) provided superior fixation strength in the femoral tunnel and
the central four-quadrant sleeve and screw system (Intrafix) was clearly superior in the tibial
tunnel in securing the quadrupled hamstring graft in the ACL reconstruction. In the femoral
site EndoButton CL, RigidFix, BioScrew and RCI screw and in the tibial site tandem spiked
washers, BioScrew, SoftSilk and SmartScrewACL allowed increased residual displacement
during cyclic loading. Therefore, some caution may be needed in the rehabilitation after ACL
reconstruction when using these implants.
10
INTRODUCTION
The rupture of the anterior cruciate ligament (ACL) produces abnormal kinematics of the
knee, which may lead to symptoms of knee instability particularly during cutting and
pivoting activities, recurrent injury, damage to the menisci and the articular cartilage, and
ultimately, osteoarthrosis (Johnson et al. 1992, Frank and Jackson 1997, Roos and Karlsson
1998). It has been estimated that the annual incidence of anterior cruciate ligament injuries is
about one in 3000 in the United States (Miyasaka et al. 1991). The goal of treatment for an
ACL rupture is to prevent the progression of intra-articular damage by restoring the stability
of the knee (Johnson et al. 1992, Frank and Jackson 1997).
The management of the anterior cruciate ligament deficient knee has developed dramatically
during the past decades. Although it is commonly accepted that not all patients with ACL
deficiency have progressive functional disability, it has been stated that in the majority nonoperative treatment leads to functionally unacceptable outcome (Fu et al. 1999). Various
surgical approaches have been used to treat the ruptured ACL. Despite encouraging short
term results, current evidence indicates that primary repair, primary repair with
augmentation, reconstruction with inadequate graft material, and extra-articular procedures
are technically inferior, fail to restore satisfactory knee stability, and thus have generally
been abandoned (Frank and Jackson 1997, Fu et al. 1999). Consequently, the intra-articular
reconstruction with a biologic graft is currently the procedure of choice for the treatment of a
rupture of the ACL (Frank and Jackson 1997, Fu et al. 1999, Fu et al. 2000).
Many factors influence the clinical success of the ACL reconstruction, including the graft
material itself, the fixation of the graft, the placement of the graft, and the rehabilitation after
the reconstruction (Johnson et al. 1992, Frank and Jackson 1997, Fu et al. 1999, Fu et al.
2000). Today, the two most commonly used ACL substitutes are the central third of the
11
patellar tendon and the hamstring (semitendinosus tendon alone or combined with gracilis
tendon) tendon (Fu et al. 1999, Brand et al. 2000c). In the early postoperative period, the
fixation of the graft is the most critical point of the anterior cruciate reconstruction limiting
vigorous postoperative rehabilitation (Johnson et al. 1992, Frank and Jackson 1997). The
fixation has to be strong enough to avoid failure due to sudden overload or repetitive
submaximal loading and stiff enough to re-establish the stability of the knee and to minimize
graft-tunnel motion. In addition to sufficient strength and stiffness, an ideal graft fixation is
anatomic, biocompatible, safe and reproducible, allows undisturbed postsurgical magnetic
resonance imaging (MRI) of the knee, and does not complicate possible revision surgery (Fu
et al. 1999, Brand et al. 2000c, Fu et al. 2000, Martin et al. 2002).
12
REVIEW OF THE LITERATURE
1. Knee joint motion and stability
Although the principle motion of the knee is flexion and extension, the knee is not a simple
hinge joint. In fact, it allows movement in six degrees of freedom (three translations and
three rotations) (Woo et al. 1999, Dienst et al. 2003). The motion of the knee joint can be
related to three principle axes perpendicular to each other (tibial shaft axis, epicondylar axis
and anteroposterior axis) (Woo et al. 1999). Movements relative to these axes are referred to
as proximal-distal translation, medial-lateral translation, anterior-posterior translation,
internal-external rotation, flexion-extension, and varus-valgus rotation (Woo et al. 1999).
The knee joint is between the long lever arms of the femur and the tibia, and consequently,
according to basic principles of biomechanics, is extremely vulnerable to distortion injuries.
Ligaments and other supporting soft tissue structures (joint capsule, muscles, tendons and
menisci) control the stability of the knee (Swenson and Harner 1995). The role of menisci,
restraining anterior-posterior translation, is particularly important in the ACL deficient knee
(Swenson and Harner 1995). The ACL, in turn, is the primary restraint preventing posterior
displacement of the femur relative to tibia, but also serves as an important secondary restraint
to varus-valgus rotation, as well as internal-external rotation (Dienst et al. 2003). As regards
the other ligaments about the knee and their stabilizing function, the posterior cruciate
ligament (PCL) acts as the primary restraint to posterior translation of the tibia, as well as a
secondary restraint to varus angulation and external tibial rotation, the medial collateral
ligament (MCL) as a primary restraint to valgus stress and the posterolateral corner, being
comprised of three major ligaments -the lateral collateral ligament (LCL), the oblique
13
popliteal ligament, and the arcuate popliteal ligament- restrains mainly varus angulation and
external rotation (Swenson and Harner 1995).
2. Anatomy and function of the ACL
The cruciate ligaments are named for their tibial insertion sites. The tibial origin of the ACL
is in the anterior part of the intercondylar area just posterior to the attachment of the medial
meniscus and anterolateral to the medial intercondylar tubercle. It directs superiorly,
posteriorly and laterally through the intercondylar notch to attach to the posteromedial aspect
of the lateral femoral condyle (Figures 1 and 2). The ACL consists of two bundles grouped
into anteromedial and posteromedial bundles, named according to their attachment site to the
tibia. (Girgis et al. 1975, Dienst et al. 2003)
The two bundles are under variable stress during the whole range of passive flexionextension motion (Swenson and Harner 1995, Fu et al. 1999, Dienst et al. 2003). As a result
of anterior tibial translation, the anteromedial bundle is under relatively constant load
throughout the flexion-extension, while the posterolateral bundle is tight near maximum
extension (Sakane et al. 1997). It has also been stated that the ACL is under variable stress
during rotation of the tibia and abduction of the knee (Swenson and Harner 1995, Frank and
Jackson 1997).
14
ACL
PCL
LCL
MCL
Lateral
meniscus
Medial
meniscus
Figure 1. Anterior view of the right knee.
PCL
Lateral
meniscus
Medial
meniscus
ACL
Patellar
tendon
Figure 2. The tibial plateau, menisci and cruciate ligaments of the right knee, viewed from above.
15
3. ACL reconstruction
3.1. Operative technique
Currently, endoscopic intra-articular reconstruction with a biologic graft is the procedure of
choice for the treatment of an intrasubstance rupture of the ACL. Graft type, fixation
technique and surgeons' attraction influence the surgical technique. Therefore, the following
surgical steps should be considered outlines and vary between surgeons. Diagnostic
arthroscopy is commonly performed initially to confirm associated injuries. The ACL stumps
and intercondylar notch are cleared to ensure visibility, and the stenotic notch may require
notchplasty to prevent graft impingement. After graft harvesting and preparation, the tibial
and femoral tunnels are drilled to the corresponding graft size. In the two-incision technique
both tunnels are drilled outside-in (rear entry). However, during endoscopic technique, the
femoral bone tunnel is commonly drilled either through the tibial tunnel or medial portal in
inside-out fashion. If the medial portal is used, the femoral tunnel is often drilled first to
maintain fluid pressure in the joint. The prepared graft is then pulled through the tibial tunnel
to the femoral tunnel with the aid of passing sutures and the femoral side of the graft is
secured. Thereafter, the graft is preconditioned to tighten the graft and the femoral fixation,
and finally the tibial side is secured while tension is applied to the graft.
3.2. Graft selection
The two types of biologic substitutes used in intra-articular reconstruction for ruptured ACL
are autografts and allografts. Most commonly used autografts are bone-patellar tendon-bone
(BPTB), multiple strand hamstring tendons and quadriceps tendon-bone, whereas the two
most commonly used allografts are BPTB and Achilles tendon-bone (Fu et al. 1999, Brand et
al. 2000c). Several factors are important in the selection of the graft tissue for ACL
reconstruction, such as the initial mechanical properties of the graft tissue, morbidity
resulting from graft harvesting, graft healing, and the initial mechanical properties of the
graft fixation (Miller and Gladstone 2002).
16
The BPTB graft taken from the central third of the patellar tendon with adjacent tibial and
patellar bone blocks has been the most popular autogenous graft for arthroscopic
reconstruction of the ruptured ACL during the past two decades (Fu et al. 1999, Miller and
Gladstone 2002). The popularity of BPTB graft is based on its structural properties, quality
of fixation, excellent long-term clinical success and the fact that it provides bone-to-bone
healing (Fu et al. 1999, Miller and Gladstone 2002). The major concern with the use of
BPTB graft has been the associated donor-site morbidity, including anterior knee pain,
kneeling discomfort, loss of motion, and weakness of the quadriceps muscle (Fu et al. 1999,
Kartus et al. 2001, Miller and Gladstone 2002). Fractures of the patella have also been
reported (Stein et al. 2002).
The improvements in the fixation methods, enabling the use of multiple strand hamstring
grafts and the relatively low donor-site morbidity have resulted in the increased popularity of
the hamstring graft (Fu et al. 1999, Brand et al. 2000c, Miller and Gladstone 2002). Multiply
looped hamstring grafts have been shown to have initial ultimate tensile load and stiffness
similar or higher to that of normal ACL (Woo et al. 1991, Rowden et al. 1997, Hamner et al.
1999, Ferretti et al. 2003). In addition, the hamstring tendon graft has a diameter close to that
of a normal ACL and its multiple bundle construct is thought to mimic the double bundle
construct of an ACL (Brown et al. 1993, Fu et al. 1999, Hamner et al. 1999, Brand et al.
2000c). However, the concerns associated with the use of the hamstring tendon graft include
the failure to achieve rigid initial fixation to bone, the slower bone tunnel incorporation
compared to the BPTB graft, the increased knee laxity, the potential hamstring muscle
weakness, and the discomfort associated with some of the fixation hardware (Fu et al. 1999,
Miller and Gladstone 2002). To date, there are several studies directly comparing patellar
tendon and hamstring tendon autografts used in ACL reconstruction (Marder et al. 1991,
Otero and Hutcheson 1993, Aglietti et al. 1994, O’Neill 1996, Corry et al. 1999, Aune et al.
2001, Beard et al. 2001, Eriksson et al. 2001, Beynnon et al. 2002, Pinczewski et al. 2002,
Shaieb et al. 2002, Ejerhed et al. 2003, Feller and Webster 2003, Gobbi et al. 2003, Jansson
et al. 2003). In general both grafts have provided similar stability and patient satisfaction.
However, in some studies hamstring grafts have shown increased laxity compared with
17
BPTB grafts (Otero and Hutcheson 1993, O’Neil 1996, Corry et al. 1999, Beynnon et al.
2002, Feller and Webster 2003). The increased knee laxity can be, at least partly, explained
by the use of two-strand hamstring tendon grafts (O’Neill 1996, Beynnon et al. 2002), by the
fixation method (Otero and Hutcheson 1993), and the apparent risk for the increased laxity in
the female patients (Corry et al. 1999). In contrast, the use of BPTB grafts appears to cause
more donor-site morbidity compared to the use of hamstring tendon grafts (Corry et al. 1999,
Pinczewski et al. 2002, Shaieb et al. 2002, Ejerhed et al. 2003, Feller and Webster 2003,
Gobbi et al. 2003), although there is also somewhat contrasting evidence, as it has been
shown that the current rehabilitation protocols which allow full weight bearing and full range
of motion immediately after the reconstruction with BPTB graft result in lower prevalence of
anterior knee pain and development of motion loss than reported earlier (Sachs et al. 1989,
Shelbourne and Trumper 1997, Kartus et al. 2001). Both meta-analyses comparing the
outcome using these two graft choices have concluded that patellar tendon autografts provide
better stability than hamstring tendon autografts (Yunes et al. 2001, Freedman et al. 2003).
The quadriceps tendon-bone graft has also been used for ACL reconstruction (Fu et al. 1999).
It has been shown to have sufficient structural properties compared to BPTB and hamstring
grafts (Miller and Gladstone 2002). The quadriceps tendon-bone graft, in turn, is commonly
used for revision ACL reconstructions and for multiple ligament reconstructions (Fu et al.
1999).
Allografts are commonly harvested sterile and preserved by deep freezing, or secondarily
sterilized by low-dose gamma irradiation (Fu et al. 1999). The interest in allografts has
increased because harvesting of autografts increases the operation time and exposes the knee
to donor-site morbidity (Fu et al. 1999, Miller and Gladstone 2002, Allen et al. 2003). The
disadvantages of allograft include the fear for disease transmission and the apparently
delayed graft incorporation (Fu et al. 1999, Miller and Gladstone 2002). In addition the
preservation and sterilization procedures have been shown to have an adverse effect on
structural properties of the allografts (Miller and Gladstone 2002). Accordingly the allografts
are today most commonly proposed for multiple ligament reconstructions and for revision
surgery (Fu et al. 1999, Allen et al. 2003).
18
Table 1. Ultimate tensile load and stiffness of the intact ACL and commonly used autografts.
STG (semitendinosus and gracilis tendons)
Ultimate failure
load (N)
Stiffness
(N/mm)
Reference
ACL
2160 ± 157
242
Woo et al. 1991
ACL
2195 ± 427
306
Rowden et al. 1997
ACL
2362
not reported
Stapelton et al. 1998
Two gracilis strands
tensioned with weights
1550 ± 369
370
Hamner et al. 1999
Two semitendinosus strands
tensioned with weights
2640 ± 320
534
Hamner et al. 1999
Four combined strands
(STG) tensioned with
weights
4090 ± 295
776
Hamner et al. 1999
Four combined strands
(STG) tensioned with
weights
4590 ± 674
861
Hamner et al. 1999
Four combined strands
(STG) tensioned manually
2831 ± 538
456
Hamner et al. 1999
Doubled semitendinosus
and gracilis graft
1709 ± 581
213
Ferretti et al. 2003
Doubled semitendinosus
and gracilis graft (twisted)
2428 ± 475
310
Ferretti et al. 2003
Quadrupled STG
2421 ± 538
238
Wilson et al. 1999
Bone-patellar tendon (10
mm, unconditioned)
1953 ± 325
423
Schatzmann et al. 1998
BPTB 10 mm
1784 ± 580
210
Wilson et al. 1999
Bone-quadriceps tendon (10
mm, unconditioned)
(Mean ± SD)
2172 ± 618
312
Schatzmann et al. 1998
19
3.3. Graft placement
Graft placement has been considered one of the most critical factors in determining the
outcome of the ACL reconstruction (Bealle and Johnson 1999). Incorrect tibial or femoral
tunnel placement results in changes in the graft excursion and either restricts knee motion or
results in joint laxity (Bealle and Johnson 1999). Therefore, a lot of emphasis is in the correct
position of tibial and femoral tunnels. Optimal graft placement, however, differs from patient
to patient because of the variations in the individual anatomy of the knee and therefore the
graft placement has to be customized at surgery.
In the tibia, a tunnel placed too anteriorly causes impingement, increases graft tension at full
extension and full flexion, and is related to early graft failures (Howell and Clark 1992),
whereas a tibial tunnel placed too posteriorly causes excessive laxity during flexion and also
endangers the PCL during tunnel drilling (Jackson and Gasser 1994). Tunnels placed too
medial and lateral are related to graft impingement. It has been reported that the PCL is the
most reproducible landmark for anatomic placement of the tibial tunnel (Hutchinson and Bae
2001).
In the femur, the tunnel is currently placed to the insertion of the anteromedial bundle or
partially anterior and posterior to the isometric point. The femoral tunnel is commonly drilled
to the intercondylar roof at 11 o’clock position for the right knee and at 1 o’clock position for
the left knee, which is believed to reproduce the anteromedial bundle of the ACL. If the graft
is placed toward 12 o’clock position, it is less capable of resisting rotatory loads. Therefore, a
more lateral tunnel position has been advocated. A femoral tunnel placed too posteriorly
causes increased graft tension in flexion, while one placed too anterior is associated with
increase in graft failure rates (Aglietti et al. 1997). Notchplasty is commonly performed to
prevent graft impingement and to improve visualization, although a recent study suggests
that it changes graft placement, excursion and function (Markolf et al. 2002).
Current reconstruction procedures replicating mainly the anteromedial bundle are successful
in limiting anterior tibial load. However, in a recent study, the advantage of replication of
both bundles during ACL reconstruction was clearly demonstrated (Woo et al. 2002). After
20
anatomic reconstruction the both bundles followed their known function. Namely, the level
of the in situ force in response to anterior tibial translation in the posterolateral bundle was
largest in full extension and decreased with increasing flexion angle, whereas the force in the
anterolateral bundle increased during flexion (Woo et al. 2002). The importance of the twobundle reconstruction was more clearly demonstrated when the knee was subjected to
combined rotatory loads simulating the pivot shift test (Yagi et al. 2002).
3.4. Graft preconditioning and tension
The preconditioning of the graft prior to fixation and the initial graft tension at the time of
fixation are considered as important factors determining the long-term function of the
reconstructed knee (Amis and Jakob 1998). However, the need of graft preconditioning has
been questioned recently (Nurmi et al. 2004b). The ACL grafts have time- and historydependent viscoelastic properties resulting in decrease of the initial graft tension over time.
Under a constant load, the intra-articular portion of the graft elongates (that is, creep) over
time, and conversely, when the graft material is kept at fixed elongation the load will
decrease overtime (that is, stress-relaxation). During cyclic loading, the peak load decreases
rapidly in the early phase, but after multiple loading cycles, the curves become relatively
repeatable, indicating that a steady state is reached as no further creep occurs (Schatzmann et
al. 1998). In other words, cyclic loading can prevent stress-relaxation induced loss of initial
graft tension and may therefore help in maintaining the correct graft tension, and ultimately,
knee joint laxity. In addition, a proper graft tension has also been shown to be beneficial for
the ligamentization process of the graft (Frank et al. 1983, Yamakado et al. 2002). The ideal
amount of graft tension during graft fixation is, however, controversial and has to be defined,
as an excessive graft tension might constrain joint motion and consequently result in damage
to the articular cartilage. Excessive graft tension may also have adverse effects on the
postoperative healing process and result in decrease in the biomechanical properties of the
graft (Yoshiya et al. 1987, Amis and Jakob 1998, Numazaki et al. 2002, Yoshiya et al. 2002).
Conversely, inadequate graft tension may possibly fail to re-establish the stability of the
knee. In a prospective randomised study, Yasuda et al. (1997) observed that hamstring
tendon grafts pretensioned to 80 Newtons (N) before fixation resulted in significantly less
21
knee laxity at 2 years compared with those pretensioned at 20 N or 40 N. Zeminski et al.
(2000) used cadaver knees and quadrupled hamstring grafts to study the effect of high initial
graft tension on the biomechanical outcome of an ACL reconstruction. In contrast to Yasuda
et al. (1997), they concluded that the reconstructed knee is sufficiently stabilized by 44 N of
initial tension and doubling the tension did not increase the in situ forces in the graft nor
significantly increase the knee stability. In an ex vivo biomechanical study, Numazaki et al.
(2002) showed that the effect of initial graft tension on the biomechanical properties of the
femur-graft-tibia construct vary between graft tissues and fixation methods.
The position of the tibia relative to the femur during graft fixation has also been shown to
affect knee kinematics and graft tension. It has been shown that in full extension the distance
between the femoral tunnel and the tibial tunnel will be longest. As a consequence, a graft
fixed at full extension will slacken during flexion. Conversely, a graft fixed at 30 degrees of
knee flexion will tighten as the knee is extended. The anterior-posterior position of the tibia
relative to the femur at the time of graft fixation has also major a impact on the graft function
after the ACL reconstruction. In a cadaveric study, Höher et al. (2001) found that a graft
fixed at 30 degrees of flexion with 67 N posterior tibial load most closely re-established the
function of the ACL.
3.5. Graft healing
Several studies have investigated the healing of the biologic grafts histologically and
biomechanically both in animals and in humans during second-look and revision procedures
(Arnoczky et al. 1982, Butler et al. 1989, Rodeo et al. 1993, Grana et al. 1994, Panni et al.
1997, Scranton et al. 1998, Goradia et al. 2000, Yoshiya et al. 2000, Papageorgiou et al.
2001, Weiler et al. 2002a, Weiler et al. 2002b, Nebelung et al. 2003). In animal studies, a
complete incorporation of the BPTB graft has occurred between 6 and 12 weeks and a direct
type of insertion site, similar to that of ACL, has also shown to develop by six months
following ACL reconstruction (Panni et al. 1997, Yoshiya et al. 2000, Papageorgiou et al.
2001).
22
Different time frames for the healing of a soft tissue graft within the bone tunnel have been
described: During the first 4 weeks, a fibrous interface develops between the tendon graft and
the bone tunnel (Rodeo et al. 1993, Grana et al. 1994, Papageorgiou et al. 2001). The earliest
sign of osseus integration is the development of specific collagen fibers, known as Sharpeylike fibers, perpendicular to the bone tunnel bridging the graft and the cancellous bone. The
first Sharpey-like fibers have been found to develop as early as at 3 weeks, but the fibrous
interface is first poorly organized and highly cellular maturing gradually into dense
connective tissue (Rodeo et al. 1993, Grana et al. 1994). After 9 to 12 months, the fibrous
interface between the graft and the bone tunnel becomes less well defined (Rodeo et al. 1993,
Grana et al. 1994). By using interference fixation, a soft tissue graft has been shown to heal
partially by direct contact healing without the development of a fibrous interface (Weiler et
al. 2002a).
In biomechanical studies, the stiffness, strength and modulus of both BPTB and soft tissue
grafts has been shown to decrease during the early postoperative period, but gradually
improving over time (Butler et al. 1989, Rodeo et al. 1993, Weiler et al. 2002b). During
ligamentization both BPTB and soft tissue grafts transform into a structure similar to that of
the ACL between 6 and 12 months postoperatively (Arnoczky et al. 1982, Panni et al. 1997,
Scranton et al. 1998, Goradia et al. 2000).
4. ACL graft fixation
Graft fixation site is the weakest link in the ACL reconstruction during the immediate
postoperative period until biologic fixation occurs within the bone tunnel, after which the
strength of the graft material becomes the primary factor in limiting aggressive rehabilitation
(Rodeo et al. 1993, Weiler et al. 2002b). Normally graft incorporation takes 6 to 12 weeks
after the reconstruction (Rodeo et al. 1993, Grana et al. 1994, Papageorgiou et al. 2001).
Many methods of graft fixation have been used, including staples, sutures over a screw post,
sutures tied to a button, polyester tape-titanium button, screws and washers, transfixations,
23
and interference screws of various designs, materials and sizes. Fixation devices have been
classified as either direct or indirect (Fu et al. 1999). Indirect fixations rely on connecting
material, which is attached to the graft, whereas in direct fixations, the graft is fixed directly
to the bone.
4.1. Tibial and femoral fixations
ACL graft fixation is commonly considered more problematic at the tibia compared to femur,
because forces are subjected to the ACL graft in line with the tibial bone tunnel (as opposed
to the femur, in which the line of force does not come parallel with the femoral bone tunnel
until 100ο of flexion) and the bone quality is worse in the tibial metaphysis than in the femur
(Vuori et al. 1994, Brand et al. 2000c). In addition, the four free tendons of the hamstring
graft are usually fixed in tibia. Several biomechanical studies (Liu et al. 1995, Caborn et al.
1998, Giurea et al. 1999, Magen et al. 1999, Brand et al. 2000b) have raised the current
concern on the suitability of the interference screws in the fixation of the hamstring grafts,
especially in the tibia. To improve the strength of fixation with interference screws in the
biomechanically more demanding tibial tunnel, larger (up to 10x35 mm) screws combined
with back up fixations have been advocated for primary hamstring graft fixation (Martin et
al. 2002).
4.2. Ideal graft fixation
Optimal initial ACL graft fixation requires not only sufficient initial strength to avoid
fixation failure, but also sufficient stiffness to restore the stability of the knee and to
minimize graft-tunnel motion, as well as sufficient resistance to slippage under cyclic loading
conditions to avoid gradual loosening in the early postoperative period after ACL
reconstruction. In addition, an ideal graft fixation is anatomic, biocompatible, safe and
reproducible, allows undisturbed postsurgical MRI evaluation of the knee, and does not
complicate the revision surgery, if required (Fu et al. 1999, Brand et al. 2000c, Fu et al. 2000,
Martin et al. 2002).
24
The fixation has to be both strong enough to avoid failure during unexpected loading events
and also be able to prevent the gradual slippage or deterioration under repetitive loading prior
to biologic healing. The strength requirements of the graft fixation are not clearly known, but
is has been estimated that the graft is loaded to approximately 150-500 N during normal
activities (Noyes et al. 1984, Rupp et al. 1999a, Toutoungi et al. 2000). Noyes et al. (1984)
estimated that the ACL is loaded to approximately 454 N (20 % of its strength in
biomechanical testing) during normal activities. Toutoungi et al. (2000) have recently shown
that an isokinetic/isometric extension of the knee produces peak forces 0.55 x body weight in
the ACL. In addition, Rupp et al. (1999a) showed that quadriceps pull against gravity alone
produces resultant forces up to 247 N (mean peak force 219 ± 25 N) in the native ACL. In
addition, in a cadaver study, Markolf et al. (1996) have demonstrated that after the ACL
reconstruction the BPTB graft experiences higher in situ forces compared to the native ACL
and a 45 N initial tension further increased the graft forces up to 497 N. In a normal knee,
strains of the ACL have been estimated to be approximately one-quarter of their failure
capacity during intensive rehabilitation exercises (Beynnon et al. 1995). It can be speculated
that strains and their resultant forces may be lower in reconstructed knees, which are
protected because of pain.
Currently, the fixation is regarded as the least stiff point in the graft-fixation device-bone
construct. Sufficient stiffness of the graft fixation construct not only restores the normal loaddisplacement response of the knee, but also diminishes graft motion within the bone tunnel.
When using the indirect fixation methods, the low stiffness of the graft and the connecting
materials allow motion of the tendons within the bone tunnel wall and the graft. In addition,
graft fixations at distance from the articular tunnel opening (e.g. extra-articular fixations)
allow the graft to move in sagittal plane during knee motion (Morgan et al. 1995, Fu et al.
1999). Longitudinal and sagittal graft motions within the bone tunnel are also known as the
bungee cord effect and windshield wiper effect, respectively. Excessive graft-tunnel motions
have been accused of impairing the graft incorporation and resulting in enlargement of the
bone tunnels, and altering the biomechanical function of the knee after ACL reconstruction
(Fu et al. 1999, Weiler et al. 2002b). Both BPTB and soft tissue grafts have been shown to
cause tunnel enlargement (L’Inslata et al. 1997).
25
An ideal fixation is at the tunnel opening (aperture fixation) minimizing graft motion relative
to the bone tunnel and preventing synovial fluid access within the graft and the bone tunnel
(Weiler et al. 2002a). In addition, aperture fixation has been shown to increase anterior knee
stability and graft isometry (Morgan et al. 1995, Ishibashi et al. 1997). However, it has been
stated that the improvement in anterior knee stability is actually rather attributable to the
difference in the stiffness of the two fixation methods, not from shortening the graft (Magen
et al. 1999, To et al. 1999). In fact, it has been shown that the stiffness of the graft fixation
bone construct is actually more influenced by the stiffness of the fixation than the length of
the graft (Magen et al.1999, To et al. 1999).
5. Biomechanics of ACL graft and fixation
The biomechanical properties of ACL graft fixation have been evaluated extensively in
numerous laboratory studies. It is somewhat difficult to compare the studies because the
experimental methods of the studies have varied so widely. In principle, the biomechanical
testing can be divided into characterization of the material and structural properties of the
given specimen.
5.1. Material properties of the graft
The material properties of the graft tendon material are represented by a stress-strain curve.
When an external tension load is applied, the graft responds by elongating and resists the
elongation with equal force. The stress of the tendon is determined as the externally applied
load per cross-sectional area of the tendon (N/mm2=MPa). The strain, in turn, characterizes
the stretching of the tendon during externally applied tensile loading and defines the amount
of stretching per unit length of the original tendon. The strain is typically expressed as a
percentage. The elastic (or tangent or Young’s) modulus (E) is determined as the slope of the
stress-strain curve and represents the inherent stiffness of the graft material. Other parameters
that can be obtained from the stress-strain curve include the ultimate stress, ultimate strain
and strain energy. (Woo et al. 1999)
26
5.2. Structural properties of graft fixation complex
When a graft fixation complex is subjected to biomechanical testing, we are determining the
structural properties of the entire specimen. During the uniaxial tensile testing, the response
of the specimen to loading is obtained in the form of force-displacement curve (Figure 3),
which typically consists first of the concave toe region with small increase in load producing
large elongation (low initial stiffness). During this phase of testing the whole graft-fixation
construct undergoes tightening (some of the natural creep of the tendons is eliminated and to
some extent the fixation site and specimen fixation to testing machine are tightened).
Loading beyond the toe region produces nearly linear curve (linear region). Since the
stiffness of any loaded construct is calculated as the ratio of force to displacement, this linear
portion of the curve provides us with the slope of the curve, based on which it is also simple
to evaluate the stiffness: the more vertical the line, the stiffer the construct. The yield/linear
load is defined as the force at which the slope of the force-displacement curve first clearly
decreases, this point having also significance in terms of clinical relevance. As the first
significant slippage of the ACL graft typically occurs at a yield load point, it thus represents
the beginning of abnormal laxity. After the yield point, the specimen undergoes significant
stretching (presumably because both the fixation fails and the graft tissue itself deteriorates),
but the construct often shows a strong resistance to loading and the load values keep
increasing until the maximum load is reached. Beyond the yield point, the force-displacement
curve is usually non-linear, indicating the accumulation of permanent deformation and
stretching on the graft fixation construct. The stiffness of the construct typically decreases
dramatically after the yield point. The other parameters used to characterize the structural
properties of graft fixation construct are the ultimate failure displacement and energy
absorbed at failure. (Woo et al. 1999)
27
1500 N
∆F/∆d=Linear stiffness
Ultimate failure load
Yield load
Energy absorbed
at failure
Force (F)
∆F
Ultimate failure
displacement
∆d
Displacement (d)
50 mm
Figure 3. A typical force-displacement curve obtained from the single load-to-failure testing of a
graft fixation (∆F is the increase in force and ∆d is the corresponding change in displacement).
5.3. Biomechanical testing of ACL reconstructions
It is generally agreed that prior graft incorporation the ACL graft is subjected to periodic
incidental high loads and thousands of loading cycles during currently used rehabilitation
protocols. Therefore, it has been proposed that in addition to the conventional single load-tofailure testing, the testing protocol evaluating the biomechanical characteristics of the graft
fixation should also include cyclic loading tests (Beynnon and Amis 1998). The single loadto-failure test is designed to determine the structural properties of a graft fixation construct
during a single overload mimicking a traumatic incidence. The cyclic loading test, in turn,
attempts to simulate the repetitive loads that the graft fixation construct is believed to be
subjected to during the early postoperative period. Single load-to-failure tests are also
28
performed after the cyclic loading test to determine the cyclic loading-induced deterioration
in fixation strength.
The cyclic loading protocols used to test the ACL graft fixation vary considerably between
studies. Beynnon and Amis (1998) recommended to use a cyclic loading protocol, in which
the graft fixation construct is loaded at low levels over multiple cycles (Dalldorf et al. 1998,
Seil et al. 1998, Giurea et al. 1999, Höher at al. 1999, Yamanaka et al. 1999, Brand et al.
2000a, Höher at al. 2000, Nakano et al. 2000, Nagarkatti et al. 2001, Honl et al. 2002, Miller
et al. 2002, Numazaki et al. 2002, Nurmi et al. 2002, Harvey et al. 2003, Lee et al. 2003,
Musahl et al. 2003, Nurmi et al. 2003, Nurmi et al. 2004a, Nurmi et al. 2004c). However,
several authors have used a cyclic creep test, in which the loading is progressively increased
(Reznik et al. 1990, Liu et al. 1995, Magen et al. 1999, Stadelmaier et al. 1999, Rittmeister et
al. 2001, Weiler et al. 2001, Rittmeister et al. 2002, Scheffler et al. 2002, Starch et al. 2003).
Magen et al. (1999) determined the slippage of the soft tissue graft fixation using a cyclic
loading protocol in which progressively higher loads were applied in 50 N increments under
load control for subsequent cycles until failure. The slippage was compared at load levels of
250 N and 500 N. Magen et al. (1999) concluded that “a fixation method should function to
loads at least 500 N if a reconstructed knee is to be intensively rehabilitated, assuming that
the graft is not overtensioned”. Giurea et al. (1999) have demonstrated the justification of a
more vigorous loading regimen. For example, an eight mm soft titanium interference screw
provided good initial fixation strength (879 ± 74 N) and survived 1100 loading cycles
simulating walking (0 - 150 N), but all specimens failed very rapidly when loaded cyclically
to simulate jogging (0 - 450 N).
5.4. Factors influencing the biomechanical testing of ACL graft fixation
There are several biological and experimental factors that can affect the results and
conclusions of different studies comparing various different graft fixation alternatives in
failure tests, making between-studies comparison often difficult. These include the difference
in bone quality, graft material, orientation of load, number of tested fixation sites, and strain
29
rate. However, proper storage of the specimen has been shown not to have a deleterious
effect on the biomechanical properties (Pelker et al. 1984, Woo et al. 1986).
5.4.1. Biologic factors
Many different tissue sources have been used to evaluate ACL graft fixation, including
human, porcine, bovine and ovine. Human knees are the most commonly used for evaluation
of ACL graft fixation devices (Kurosaka et al. 1987, Shapiro et al. 1992, Brown et al. 1993,
Matthews et al. 1993, Butler et al. 1994, Kohn and Rose 1994, Steiner et al. 1994, Kao et al.
1995, Brown et al. 1996, Johnson et al. 1996, Pena et al. 1996, Caborn et al. 1997, Rowden et
al. 1997, Aune et al. 1998, Caborn et al. 1998, Dalldorf et al. 1998, Matthews et al. 1998,
Stapelton et al. 1998, Höher et al. 1999, Magen et al. 1999, Seitz et al. 1999, Stadelmaier et
al. 1999, To et al. 1999, Brand et al. 2000a, Brand et al. 2000b, Shino and Pflaster 2000,
Nagarkatti et al. 2001, Rittmeister et al. 2001, Selby et al. 2001, Honl et al. 2002, Miller et al.
2002, Rittmeister et al. 2002, Scheffler et al. 2002, Starch et al. 2003, Steenlage et al. 2002,
Nurmi et al. 2003, Aydin et al. 2004, Kocabey et al. 2004, Nurmi et al. 2004a, Nurmi et al.
2004c). Obviously it would be optimal to use human cadaver tissue from young and healthy
donors in evaluating the structural properties of ACL graft fixation, but since human tissues
are difficult to obtain in the required extent, animal tissues have been used widely in
biomechanical experiments. Porcine knees are the most popular animal tissues source
(Reznik et al. 1990, Butler et al. 1994, Paschal et al. 1994, Liu et al. 1995, Pierz et al. 1995,
Rupp et al. 1997, Clark et al. 1998, Rupp et al. 1998, Seil et al. 1998, Magen et al. 1999,
Rupp et al. 1999b, Yamanaka et al. 1999, Black et al. 2000, Höher et al. 2000, Nakano et al.
2000, Nagarkatti et al. 2001, Weiler et al. 2001, Nurmi et al. 2002). They have been
considered to mimic human knee specimens in terms of their size, shape, and bone quality.
However, there is evidence that the structural properties of a fixation method vary between
animal and human tissue (Magen et al. 1999, Nurmi et al. 2004c). Porcine knees have been
shown to have clearly higher (volumetric) bone density compared to human bones (Nurmi et
al. 2002, Nurmi et al. 2004c) and it has been shown that bone mineral density is strongly
related to the strength of the quadrupled hamstring graft interference screw fixation (Brand et
al. 2000b). Bone mineral density might also have an effect on other hamstring tendon graft
30
fixations. In addition, Magen et al. (1999) found substantially lower yield loads and greater
slippage when interference screws were used to fix hamstring tendon grafts in human bone
compared with porcine tendons fixed in porcine bone. They concluded that caution is
warranted when animal tissues are used for predicting the performance of interference screws
in human ACL fixation.
As the quality of human bone specimens often varies considerably, porcine and even bovine
knee specimens with more uniform bone quality offer a reasonable alternative to human
bone. It has been speculated that the increase in bone density mainly affects in fixations that
rely on friction to secure the graft within the bone tunnel (i.e. interference screws). Therefore,
results from animal studies evaluating interference fixation are presumably overly optimistic
in comparison with the situation in humans (Nurmi et al. 2004c).
5.4.2. Experimental factors
The tests evaluating the fixation characteristics are typically conducted by applying a load in
one direction while other motions are constrained. One approach to evaluate graft fixation
characteristics is to perform an ACL reconstruction and test both the fixations (tibia and
femur) simultaneously in the same test and another is to isolate a single attachment site
(Beynnon and Amis 1998).
The orientation of the applied load is also expected to affect the fixation characteristics. The
more physiologically applied directions enhance the fixation strength by developing shear
forces typically at the screw-graft-tunnel interface. Application of force parallel to the bone
tunnel transfers the highest forces directly to the fixation site, representing the worst-case
scenario for a fixation technique, and thus, providing the purest test for pullout strength.
The influence of strain rate to failure characteristics of graft fixation has been speculated.
High strain rates (100%) in a single load-to-failure test have been advocated, as such protocol
is believed to better mimic a traumatic incidence. However, it has been shown that strain
rates have no apparent effect on the tensile strength, as no difference was observed in the
31
modulus or failure mode of the BPTB grafts tested with a strain rate of 10 versus 100 percent
(Blevins et al. 1994).
6. Current ACL graft fixation options
6.1. Bone-tendon graft fixation
In the interference technique introduced by Lambert (1983), where the graft is secured in
bone tunnels with metallic interference-fit screws placed parallel to the osseous part of the
BPTB graft has been shown to provide superior biomechanical properties compared to other
methods. Consequently, metal interference screw fixation of the BPTB graft has become the
most widely used operative procedure for the fixation of the intra-articular replacement of the
ruptured ACL. In addition to superior fixation strength, the BPTB graft fixation using
interference technique has been shown to provide good and reproducible clinical outcome.
The major concern with the use of interference technique has been the damage of the bone
block, tendon or the passing sutures during the screw insertion. The fixation may also fail
due to divergent screw placement or the breakage of a bioabsorbable screw (Matthews et al.
1989, Barber et al. 1995, Pierz et al. 1995, Safran and Harner 1996, Marti et al. 1997, Brown
and Carson 1999).
6.1.1. Cortical fixation
Steiner et al. (1994) used a human cadaveric model to evaluate four different BPTB graft
fixation options. They found that suture fixation of BPTB grafts produced lower stiffness
compared to interference screw fixation. Combination of interference screw and suture
techniques provided highest failure load and stiffness (Steiner et al. 1994). Recently, Honl et
al. (2002) compared three different BPTB graft fixation methods (EndoButton, interference
screw and sutures tied over a screw post) using a cyclic loading test. They found that
32
interference fixation had age dependency and for patients over 40 to 45 years, EndoButton
may provide a better fixation. They also concluded that sutures tied over a post technique is
not suitable for graft fixation under clinical circumstances. In addition to inferior stiffness of
the sutures, the prominent screw post on the anterior surface of the tibia may irritate and
require later removal, and at femoral side sutures tied over a screw post technique requires a
lateral thigh incision.
6.1.2. Interference fixation
Various factors affect the initial graft fixation strength of the interference screw, including
the quality of the bone, the shape of the bone block, the gap between the screw and the bone
block, the magnitude of the insertion torque, the divergence of the screw, and the design,
material and size of the screw (Brand et al. 2000c). To date, there are several studies showing
improved strength of fixation of the BPTB tissue graft with increased screw diameter (Brown
et al. 1993, Butler et al. 1994, Kohn and Rose 1994, Gerich et al. 1997, Pomeroy et al. 1998),
while Hulstyn et al. (1993) and Shapiro et al. (1995) did not find any extra benefit in
increasing the screw diameter. Matthews et al. (1998) compared the fixation strength of
large-diameter (9, 11, 13, and 15 mm) metal interference screws. The so-called interference
(screw diameter-gap size) was kept at 4.5 mm. They did not find any differences between the
fixation strengths, and therefore the “extra”-large interference screws were advocated only
for revision reconstructions where the diameter of the tunnel has widened for one reason or
another. The effect of the interference screw length on the fixation properties of the BPTB
graft is also controversial (Brown et al. 1993, Hulstyn et al. 1993, Kao et al. 1995, Gerich et
al. 1997, Black et al. 2000). Gerich et al. (1997) found that the strength of the fixation
increases with increase in the screw length. Hulstyn et al. (1993) had also similar findings
with 5.5 mm diameter screws, although, the fixation strength did not improve with increase
in the screw length when 7 mm or 9 mm diameter screws were used.
The gap between the bone tunnel wall and the bone block also affects the initial fixation
strength. For example, Butler et al. (1994) has demonstrated that when the gap size is 3 or 4
mm the ultimate failure load is significantly increased when a 9 mm diameter screw is
33
substituted for 7 mm diameter screw. Reznik et al. (1990) found also correlation between gap
size and failure loads. In contrast, Brown et al. (1996) did not find significant correlation
between gap size and failure load. Several studies have found correlation between the
magnitude of screw insertion torque and initial fixation strength of the interference screw
fixation (Brown et al. 1993, Kohn and Rose 1994, Brown et al. 1996, Black et al. 2000).
However, excessive insertion torque may predict damage to the graft (Kohn and Rose 1994).
Rupp et al. (1998) used a porcine model to compare the influence of the position of the screw
relative to cancellous or cortical surface of the bone block to the initial fixation strength of a
metal interference screw. Cortical fixation resulted in more tendon ruptures, but there was no
effect on the initial fixation strength.
The influence of inside-out and outside-in screw placements has been evaluated in several
studies (Brown et al. 1993, Pierz et al. 1995, Bryan et al. 1996, Dalldorf et al. 1998,
Stapelton et al. 1998). When the screw is placed parallel to the bone block both inside-out
and outside-in screw placements provide similar results (Brown et al. 1993, Pierz et al. 1995,
Bryan et al. 1996, Dalldorf et al. 1998). Stapelton et al. (1998) used cadaveric knees to
compare two-incision technique with one incision-technique. After the ACL reconstruction
the knees were tested to failure by placing the tibia anteriorly with the knee at 30 degrees of
flexion and 30 degrees of internal rotation. They concluded that one-incision technique is
significantly stronger than two-incision technique. However, the tibial sides of the fixations
were done similarly in both techniques and eleven fixations out of twelve tested knees failed
on the tibial side.
Interference screw divergence, the angle of the interference screw with respect to the bone
block, is a common clinical concern, and consequently, its effect on the strength of fixation
of BPTB graft has been evaluated biomechanically. It has been shown that an increase in the
divergence angle decreases fixation strength at angles greater than 20 degrees (Jomha et al.
1993, Pierz et al. 1995). Further, a divergent screw placement has more influence on the
outside-in screw placement (i.e. rear entry technique) than the inside-out (Pierz et al. 1995).
Lemos et al. (1995) found no difference in fixation strength when the divergence angle was
15.
34
During single incision endoscopic ACL reconstruction using the BPTB graft, the surgeon
may encounter problems associated with graft-tunnel mismatch, the length of the BPTB graft
exceeding the combined tunnel and intra-articular distance with the consequent protrusion of
the bone block out of the tibial tunnel. The incidence of graft-tunnel mismatch has been
reported to be as high as 28 % (Shaffer et al. 1993), however, it can be decreased by
increasing the length of the tibial tunnel. Alternative techniques have been introduced to deal
with the graft-tunnel mismatch, such as advancing the femoral bone block further within the
femoral socket, which has been associated with an increase in the risk of graft laceration,
though. Also, after the femoral fixation, the BPTB graft can be shortened by flipping the
patellar tendon or by recessing the tibial bone block (Pomeroy et al. 1998, Hoffmann et al.
1999, Black et al. 2000). Based on biomechanical studies, Pomeroy et al. (1998) have stated
that the bone block should be at least 1 cm long to provide sufficient interference screw
fixation strength. In addition, Black et al. (2000) have found similar fixation properties with
12.5 mm and 20 mm long interference screws. Other suggested fixation options for the BPTB
bone block protruding distally outside the tibial tunnel are sutures tied over a screw post,
staples, and free bone block interference screw fixations (Novak et al. 1996, Gerich et al.
1997).
Bioabsorbable interference screws may offer many advantages over their metal counterparts,
including the decreased risk for graft laceration, decreased stress protection, undistorted
postsurgical MRI, no need for hardware removal, and lack of corrosion (Black 1988,
Shellock et al. 1992, Safran and Harner 1996, Jacobs et al. 1998, Suh et al. 1998, Brown and
Carson 1999). The increased popularity of the ACL reconstruction surgery has also increased
the number of ACL revisions (Jaureguito and Paulos 1996, Safran and Harner 1996). In
revision surgery, the metal screws that have been inserted in the primary reconstruction have
to be normally removed to achieve a proper graft replacement (Safran and Harner 1996,
Brown and Carson 1999).
Kurzweil et al. (1995) have suggested that metallic tibial interference screw may cause pain.
Twenty-three knees of the 458 ACL reconstruction using BPTB autografts and allografts
35
underwent tibial interference screw removal because of persistent pain. In three knees the
metallic interference screw protruded above the tibial tunnel. Radiographs failed to show any
cyst formation and the bone blocks appeared to have incorporated. However, a bursal sac was
developed over the metal screw in three cases. At follow-up evaluation 2 years after
hardware removal, the pain and tenderness over the tibial site were resolved in 21 of the 23
patients. In a follow-up of 604 patients (Kartus et al. 1998), 39 metal implants used for BPTB
autograft fixation were removed on the tibial site due to pain. The retained metallic devices
may cause pain due to the obvious mismatch of the elastic modulus between the bone
(cancellous bone 0.2-0.7 GPa and cortical bone 9-20 GPa) and the metallic device (for
example, titanium alloy 110 GPa) (Ruluff and McIntyre 1982, Anderson et al. 1992,
McCalden et al. 1993). The less rigid bioabsorbable implants (PLA96/4 4.6-7.5 GPa) may
decrease the incidence of this problem (Pohjonen and Törmälä 1996).
Although Pena et al. (1996) reported significantly lower fixation strength using
bioabsorbable than standard metal interference screws, bioabsorbable and metal screws have
generally provided comparable initial fixation strengths in single load-to-failure and cyclic
loading tests (Abate et al. 1998, Caborn et al. 1997, Johnson and van Dyk 1996, Rupp et al.
1997, Seil et al. 1998, Weiler et al. 1998b).
Although bioabsorbable interference screws have demonstrated good short- and mid-term
results in clinical studies, some complications have been reported. Namely, the breakage of
the screw during insertion, loose intra-articullar bodies, mild inflammatory reactions and
osteolytic cyst formation (Barber et al. 1995, Imhoff et al. 1998, Martinek et al. 1999,
Macdonald and Arneja 2003).
6.2. Soft tissue graft fixation
Various graft fixation devices have been developed to be used with the soft tissue grafts.
Despite the reduced donor site complications and sufficient initial mechanical properties of
the hamstring tendon graft, there remains concern for sufficient fixation of the graft among
the ACL surgeons. Free tendon grafts are generally fixed far away from the joint using either
36
staples, screws and washers, so-called endobuttons, cross pins or tying sutures over buttons
or screw posts (Fu et al. 1999, Brand et al. 2000c, Martin et al. 2002). Currently, many of
these hamstring graft fixation techniques provide high fixation strength, even higher than
interference fixation for BPTB graft (Steiner et al. 1994, Clark et al. 1998, Magen et al. 1999,
To et al. 1999). However, the increased length and the possible connecting materials of the
graft-fixation construct are believed to allow graft motion within the bone tunnel wall (Fu et
al. 1999). Currently, common concern about soft tissue aperture fixation methods (e.g.
interference screw fixation) is their low resistance to slippage under repetitive loads during
early postoperative period, as the strength of the fixation relies on friction to secure the
tendons in the tunnel drilled in bone.
6.2.1. Cortical fixations
A soft tissue fixation using a single staple has provided poor results in biomechanical studies
(Robertson et al. 1986, Kurosaka et al. 1987). However, the results have improved
considerably when two staples have been used (Magen et al. 1999, Yamanaka et al. 1999).
Magen et al. (1999) fixed the porcine extensor tendon to the tibial cortex using a “beltbuckle” technique. The failure load of the graft fixation bone construct was 704 ± 174 N and
the stiffness was 118 ± 47 N/mm. Staple fixation is, however, at distance from the joint
surface and the prominence of the staples may cause irritation at the fixation site and require
later removal. Staple is currently commonly recommended to back-up other fixations of the
soft tissue graft (Martin et al. 2002).
Indirect soft tissue graft fixation options rely on connecting materials (sutures, and polyester
tapes), which connect the graft to the point of actual fixation. Tying sutures over a screw post
and washer or button placed at the cortex outside the bone tunnel has been used both on the
femoral and tibial side. Yamanaka et al. (1999) and Nakano et al. (2000) reported the same
results of single load-to-failure and cyclic loading studies evaluating the fixation strength of
the sutures-tied over a button technique in a doubled flexor graft in the porcine knee. The
ultimate failure load was 458 ± 72 N and the stiffness was only 19.8 ± 1.4 N/mm. The
37
disadvantages of these techniques are the lateral thigh incision on the femoral side and the
strength and stiffness of the sutures connecting the graft and the fixation device. Steiner et al.
(1994) evaluated the fixation strength of four hamstring fixation techniques using gracilis and
semitendinosus grafts in human cadaver knees. They showed that the fixation of a
quadrupled hamstring graft with sutures tied over a screw post provided inferior fixation
properties compared to fixation with soft tissue washers.
Tandem washers are commonly used at the tibial side. In addition to good initial failure
strength and stiffness, they have been shown to provide good resistance to slippage (Magen
et al. 1999). Using bovine bone and bovine extensor tendons, Beynnon et al. (1998) found a
significant correlation between the ultimate failure load and insertion torque magnitude.
WasherLoc is designed to secure the tibial end of the quadrupled hamstring graft within the
tunnel, at the distal tunnel opening, whereas cortical fixations (tandem spiked washers,
sutures over a screw post and washer or buttons) have longer working length, as they are
placed completely outside the tibial tunnel. Cortical fixations are also prone to cause
irritation at the fixation site, and thus, may require hardware removal. Magen et al. (1999)
used human tendons and bones to show that tandem washers and WasherLoc provided
significantly higher yield load and significantly less slippage compared with titanium
interference screw in resisting rapidly increasing progressive cyclic loading protocol. All
three graft fixation constructs provided stiffness values similar to that of native ACL.
Brown and Sklar (1998) have stated that the endoscopic EndoButton technique provides a
quick, simple, accurate, reproducible and strong method for the femoral site hamstring graft
fixation. EndoButton consists of a titanium button and a connecting material. The button is
placed on the lateral femoral cortex of the femur and the connecting material is attached to
the button creating a loop for the graft. The disadvantage of the endoscopic EndoButton
technique is the mobile fixation allowing both longitudinal and sagittal graft motions relative
to the bone tunnel. Höher et al. (1999) showed that titanium button/polyester tape fixation
allowed 1 to 3 mm graft tunnel motion when loading to 100 and 300 N, respectively. Tsuda
et al. (2002) have also demonstrated that EndoButton fixation of a soft tissue graft allows
both increased sagittal and longitudinal motion, and greater anterior knee laxity compared to
38
interference screw fixation. EndoButton technique has been associated with tunnel
enlargement in clinical studies (L’Insalata et al. 1997, Nebelung et al. 1998). The knots of the
connecting material have been the weakest link of the fixation. As a result, a continuous loop
EndoButton (EndoButton CL) was developed, providing a factory-attached loop eliminating
the need for knot tying of the connection material.
6.2.2. Interference fixation
Extra-articular soft tissue graft fixation constructs have been accused of allowing motion of
the tendons and the access of synovial fluid within the bone tunnel, both of which may
impair graft healing and ultimately result in increased knee laxity. In contrast the resulting,
short graft fixation construct in interference fixation has been shown to increase anterior knee
stability. The use of interference screws has also become safer because of blunt threaded and
round headed interference screws, reducing the risk for graft laceration. Weiler et al. (2002a)
recently demonstrated that interference fixation may improve the incorporation of the tendon
within the bone tunnel (tendon-to-bone healing). However, there has been concern of the
aperture fixation methods due to their low resistance to slippage under repetitive loads during
immediate postoperative period.
Recent reports comparing the interference fixation with that of other fixations in soft tissue
graft have been controversial (Giurea et al. 1999, Magen et. al 1999, Miyata et al. 2000,
Scheffler et al. 2002). Several biomechanical studies have evaluated the fixation properties of
the soft tissue interference screw fixation. Recently, Magen et al. (1999) evaluated
biomechanically the fixation properties of six tibial soft tissue fixation methods using a twophase study design. They first compared six different fixation devices using porcine bone and
porcine extensor tendons, after which the three best fixation techniques were compared again
using human cadaveric tissue. Although interference screw performed well in the porcine
specimens (first part of the study), in the second part using the human tibia and the
quadrupled hamstring grafts and progressively increasing loading both the tandem washers
and WasherLoc provided significantly higher yield load and significantly less slippage
compared with the interference screws The yield load value provided by interference fixation
39
was only 350 ± 134 N. In addition, four of seven interference fixations failed at 500 N or a
lower load. In the report by Giurea et al (1999), four different hamstring graft fixation
devices (stirrup, clawed washer, RCI screw, and “soft” and round-headed interference
screws) were compared in pull out and cyclic loading tests. The interference screws
specifically designed for the fixation of soft-tissue graft in the femoral tunnel and stirrup
anchoring the looped end of the quadrupled hamstring tendon graft into the tibial tunnel
showed superior resistance to slippage at low loading levels (0-150 N) of constant cyclic
loading compared to clawed washer and normal round-headed cannulated interference (RCI)
screw.
The BPTB graft interference screw fixation in young human bone is commonly considered
the golden standard fixation. Liu et al. (1995) used porcine bone to show that bonehamstring-bone graft had inferior initial fixation strength compared to BPTB grafts.
Similarly, in cyclic relaxation tests, Yamanaka et al. (1999) and Nakano et al. (2000) found
that unloading was higher and failure strength lower in soft tissue graft interference screw
fixation compared to BPTB graft interference screw fixation.
Although the results from biomechanical studies of hamstring graft interference screw
fixation in human bone have not been very encouraging (Aune et al. 1998, Caborn et al.
1998, Magen et al. 1999, Stadelmaier et al. 1999), the method has gained wide acceptance,
because clinical outcome reports have been promising (Corry et al. 1999, Piczewski et al.
2002).
Many factors contribute to the strength of the initial fixation of the soft tissue graft including
the density of the bone, the insertion torque of the screw and the diameter, length, design and
material of the screw (Weiler et al. 1998a, Brand et al. 2000b, Weiler et al. 2000). Weiler et
al. (2000) and Selby et al. (2001) found that the strength of the fixation of the soft tissue graft
improves with increase in the length of the screw, but the opposite has also been described
(Stadelmeier et al. 1999, Harvey et al. 2003). Weiler et al. (2000) also found that increasing
the screw diameter improves the fixation strength.
40
Brand et al. (2000b) have evaluated the effect of bone mineral density and insertion torque on
the biomechanical properties of hamstring graft interference fixation in ACL reconstruction.
They concluded that the interference screw fixation strength is directly related to the
metaphyseal bone mineral density in both the femur and tibia. Furthermore, femoral
metaphysis has a greater bone mineral density compared to tibia, and insertion torque is
directly correlated to maximum failure load. Weiler et al. (2000) have also found that the
magnitude of insertion torque correlates to the maximum failure load. Harvey et al. (2003)
recently showed that there is less slippage in cortico-cancellous than in cancellous only
fixation.
There is controversy over the effects of the screw placement relative to individual hamstring
tendon strands (eccentric vs. concentric). Shino and Pflaster (2000) recently showed that
concentric placement of the screw could improve the fixation of hamstring graft in tibial
tunnel, although the opposite was shown by Simonian et al. (1998) when hamstring tendons
were fixed with interference screws in polyurethane foam blocks. Starch et al. (2003)
showed that "central four-quadrant sleeve and screw system" (Intrafix) allows less hamstring
graft slippage compared to interference screw fixation.
According to Weiler et al. (1998a) bioabsorbable screws provide a better fixation strength
than their metal counterparts. They speculated that the bioabsorbable material itself was
responsible for the enhanced strength. Caborn et al. (1998) reported that the strength
provided by bioabsorbable interference screw was 40 % higher than that with RCI screw.
However, the difference was not statistically significant. In addition, Giurea et al. (1999)
recently showed that interference screw design affects the possibility for slippage of the soft
tissue fixation when tested under cyclic loading conditions. After 1100 loading cycles at 150
N, the RCI screw allowed a mean displacement of 6.8 ± 5.6 mm, which was over three times
higher than that observed with the interference screws specifically designed for the fixation
of soft tissue grafts.
Compaction drilling or compaction with serial dilators is a common procedure to create a
denser bone tunnel wall and to provide a snug fit for the graft. The denser bone tunnel wall is
believed to enhance the strength of interference screw fixation of the soft tissue grafts.
41
However, neither dilation using serial dilators nor compaction drilling with a stepped router
has been shown to improve hamstring graft fixation in tibia (Rittmeister et al. 2001, Nurmi et
al. 2002, Nurmi et al. 2003, Nurmi et al. 2004a).
For the interference screw soft tissue fixation in tibia, a back-up fixation is often
recommended, especially when the bone is soft, the bone tunnel large, and the patient either
noncompliant or heavy (Martin et al. 2002, Scheffler et al. 2002). It has been shown that free
bone block increases initial interference screw fixation strength and limits graft slippage
(Weiler et al. 1998a, Scheffler et al. 2002).
6.2.3. Transfixation
In transfixation technique the implant is placed transversely through the femoral or tibial
bone tunnel to secure the graft. Various transfixation devices (TransFix, RigidFix,
BoneMulchScrew) have been introduced. Most of them can be considered toggle fixation
where the tendon is looped around a part of the implant to suspend the graft within the bone
tunnel. RigidFix, is a cross pin similar to TransFix, however, it actually should be considered
a “semitoggle” fixation since it is not possible to “wrap” the loops of the tendons around the
cross pins. The fixation relies on the pins crossing the tightly sutured tendon loop bundle in
the femoral tunnel. For the toggle fixation, the tendon is looped over a crossbar inserted into
the metaphyseal bone of lateral femoral condyle allowing tensioning of the individual strands
which has been shown to result in significant increase in graft strength and stiffness (Hamner
et al. 1999). Transfixation seems to approach our current need in terms of initial fixation
strength and stiffness. To et al. (1999) compared the strength of fixation of a
BoneMulchScrew with that of EndoButton and Mitek Anchor in a double looped
semitendinosus and gracilis graft in the young human femur. They found that the ultimate
failure load provided by BoneMulchScrew averaged 1126 ± 80 N and was significantly
higher than the ultimate failure load provided by the EndoButton (430 ± 27 N) and Mitek
Anchor (312 ± 35 N). The stiffness of the femur-BoneMulchScrew-graft was also
significantly higher compared to femur-EndoButton-graft and femur-Mitek Anchor-graft
constructs. Clark et al. (1998) evaluated the fixation strength of two metal cross pins using
42
ovine tendons and porcine femora. The ultimate failure load and yield load ranged from 1003
± 33 N to 1604 ± 134 N and 725.5 ± 87 N to 1363 ± 166 N, respectively.
43
AIMS OF THE STUDY
1. To validate biomechanically the suitability of the bioabsorbable fibrillated SR-PLLA
screw in the BPTB reconstruction of the ACL in terms of its initial fixation strength and
to compare it with the Kurosaka interference screw and the AO cancellous screw.
2. To compare the initial fixation strength of a bioabsorbable interference screw with that of
a standard titanium interference screw in the fixation of a BPTB graft in the femoral
tunnel in the reconstruction of the ACL.
3. To compare the initial fixation strength of the new plugging and the standard interference
screw techniques in the BPTB reconstruction of the ACL.
4. To evaluate the fixation strength of commonly used femoral and tibial fixation devices of
a quadrupled hamstring tendon graft in the ACL reconstruction.
44
MATERIALS AND METHODS
1. Specimens
A total of 353 graft fixation constructs were evaluated in this thesis (Table 2). Knee
specimens (bovine and porcine) were obtained from local slaughterhouses, fresh-frozen at 20-25 oC and thawed for 12 hours at room temperature before testing. In studies I to III, the
BPTB grafts were used. The patella, the patellar tendon, and the triangular bone block were
removed as one block from each knee specimen. A 9 mm wide section from the middle third
of the patellar tendon was then sharply dissected, leaving the patella intact. The bone blocks
were further trimmed into a cylindrical (I) or rectangular shape (II, III). In studies IV and V,
240 freshly harvested human cadaveric semitendinosus and gracilis tendons were used as
graft material. The tendons were cleared from adherent muscle and soft tissue, wrapped in
saline soaked gauze and stored frozen (-25oC) in small plastic bags. The National Authority
for Medicolegal Affairs in Finland had approved the use of human cadaver tissue.
The tibial tunnels (I, II, V) were directed from the original tibial insertion site of the ACL to
the anteromedial wall of the proximal tibia. The femoral tunnels (III, IV) were drilled from
the original femoral insertion site of the ACL towards the lateral wall of the lateral femoral
condyle of the knee (at 1 o’clock for the left and at 11 o’clock for the right knee).
45
Table 2. Specimens.
Study
I (n = 33)
Bone
Graft
Bovine tibia
BPTB (Cylindrical (9 mm) tibial bone block with middle
third of the patellar tendon and the whole patella)
II (n = 40)
Porcine tibia
(20 pairs)
BPTB (Rectangular tibial bone block measuring 7 x 9 x 20
mm (thickness x width x length) with middle third of patella
tendon and the whole patella)
III (n = 40)
Porcine femur
(20 pairs)
BPTB (Rectangular tibial bone block measuring 7 x 9 x 25
mm (thickness x width x length) with middle third of patella
tendon and the whole patella)
IV (n = 120)
Porcine femur
120 quadrupled human cadaveric STG
V (n = 120)
Porcine tibia
120 quadrupled human cadaveric STG
2. Study groups and fixation procedures
In studies I, IV and V, the specimens were assigned into study groups of 11 (I) or 20 (IV,V)
specimens in each. In studies II and III, the 20 pairs of knee specimens were divided into two
treatment groups so that the left and right knee of each animal were randomly assigned into
different groups.
2.2. Bone-tendon graft fixation techniques (studies I, II and III)
In study I, the trimmed tibial bone block was inserted into a 10 mm tibial tunnel. A 1.5 mm
tunnel was then drilled along the cortex of the bone block parallel to the long axis of the
tibial bone tunnel. That tunnel was further enlarged to 2.5 mm for Kurosaka and AOcancellous screws, and to 4.5 mm for SR-PLLA screws. Kurosaka and AO-cancellous screws
were inserted into these 2.5 mm holes, and for the SR-PLLA screw a tapping device was
46
used prior to the screw insertion. All screws were inserted with an outside-in technique
(Figure 4).
Figure 4. The implants used in study I from left to right: Kurosaka interference screw; Acufex
Microsurcical Inc., Mansfield, Massachusetts (7 mm diameter, 25 mm length), bioabsorbable SRPLLA screw; Biofix, Bioscience, Tampere, Finland (6.3 mm in diameter, 25 mm thread length) and
AO cancellous screw; Stratec, Hegendorf, Switzerland (6.5 mm diameter and 25 mm thread length).
In study II, the rectangular tibial bone blocks were fixed into a 9 mm tibial tunnel. Before the
insertion of the bioabsorbable screw, a 4.5 mm drill and dilator provided by the manufacturer
were used to enhance starting. Both screws were inserted over a guide pin using an outside-in
technique into the 3 mm gap that was available between the graft and the tibial bone tunnel.
47
Figure 5. The implants used in study II from left to right: Titanium interference screw (SoftSilk);
Acufex Microsurgical Inc, Mansfield, Massachusetts (7 mm diameter, 25 mm length) and
bioabsorbable poly-L-lactide/D-lactide (PLA96/4) copolymer interference screw; Bionx Implants Ltd,
Tampere, Finland (7 mm diameter, 25 mm length).
Figure 6. The tibial BPTB graft fixation with an interference screw.
48
In study III, the tibial bone block of the graft was first pulled into a 9 mm femoral tunnel,
placing the cancellous side anteriorly. The plug was placed at the tunnel opening anterior to
the graft, so that the groove was next to the tendon. Thereafter, the plug was tapped into the
tunnel. The interference screws were inserted with the inside-out technique, placing the
screw into the 2 to 3 mm gap between the tibial bone tunnel wall and the bone block. A Kwire was used to prevent screw-graft divergence.
Figure 7. The implants used in study III from left to right: Titanium SoftSilk interference screw;
Acufex Microsurgical Inc, Mansfield, Massachusetts (7 mm diameter, 25 mm length) and
bioabsorbable (PLA96/4) copolymer plug; Bionx Implants Ltd, Tampere, Finland.
A
B
Figure 8. The femoral BPTB graft fixation techniques used in study III. (A) Plugging technique and
(B) interference technique.
49
2.3. Soft tissue graft fixation techniques (studies IV and V)
In studies IV and V, the quadrupled semitendinosus and gracilis (STG) grafts were prepared
and fixed according to the specific instruction provided by the manufacturer of the implants
and using device specific instrumentation. The looped end of the quadrupled STG graft was
fixed in the femur (study IV) and the four free strands of the STG graft were fixed in the
tibial tunnel (study V).
2.3.1. Study IV
EndoButton CL: The diameter of the graft was measured in 0.5 mm increments by using
Graft Sizing Tubes (Acufex Microsurgical, Inc.). A femoral socket equal to the graft
diameter was drilled with a compaction drill (Acufex Microsurgical, Inc.). The tendons were
doubled over the loop of the EndoButton CL device and then the button was pulled through a
4.5 mm passing channel and deployed on the lateral cortex. The length of the continuous
loop was chosen so that 20-25 mm of the graft was in the femoral socket.
BoneMulchScrew: The diameter of the prepared graft was measured by using Arthrotek
sizing sleeve (Arthrotek, Inc., Warsaw, Indiana) in 1 mm increments. A femoral socket
matching the graft diameter was drilled with a cannulated reamer. The 10.5 x 25 mm
BoneMulchScrew was inserted into an 8 mm tunnel in the lateral femoral cortex
perpendicular to the femoral socket. The graft was then placed around the tip of the screw.
Thereafter, a compacting rod was used to compact the cacellous bone through the
BoneMulchScrew into the femoral socket.
RigidFix: The diameter of the quadrupled graft was measured with a Graft Sizing Tube in
0.5 mm increments. A cannulated reamer corresponding to the diameter of the graft was
selected to drill a 25 mm femoral socket. Two parallel drill tunnels were drilled for
bioabsorbable (PLA) RigidFix crosspins (Mitek Products, Norwood, Massachusetts) from the
lateral femoral condyle perpendicular to the femoral socket using a RigidFix crosspin guide
50
(Mitek Products, Norwood, Massachusetts). The graft was then pulled into the socket and
two cross pins were inserted across the graft.
Interference screws: The prepared graft was measured in 0.5 mm increments (Graft Sizing
Tubes, Acufex Microsurgical, Inc.) and a corresponding compaction drill was selected to
drill the femoral socket. The interference screws were inserted in inside-out fashion to fix the
graft into the femoral socket. For bioabsorbable interference screws a notcher provided by
the manufacturer was used to enhance starting and to accommodate the screw heads. A
conventional RCI (Acufex Microsurgical Inc.) titanium screw was used in the left knees and
a reverse-threaded screw was used in the right knees.
Figure 9. The implants used in study IV from left to right: EndoButton CL; Acufex Microsurgical,
Inc., Mansfield, Massachusetts, BoneMulchScrew; Arthrotek, Inc., Warsaw, Indiana (10.5x25 mm),
two bioabsorbable (PLA) RigidFix cross pins; Mitek Products, Norwood, Massachusetts,
bioabsorbable (PLA96/4) copolymer SmartScrewACL interference screw; Bionx Implants Inc., Blue
Bell, Pennsylvania (7 mm diameter, 25 mm length), bioabsorbable (PLA) interference screw
BioScrew; Linvatec, Inc., Largo, Florida (7 mm diameter, 25 mm length) and titanium RCI
interference screw; Acufex Microsurgical Inc, Mansfield, Massachusetts (7 mm diameter, 25 mm
length).
51
A
B
D
C
Figure 10. The femoral hamstring graft fixations used in study IV. (A) EndoButton CL, (B)
BoneMulchScrew, (C) RigidFix and (D) interference screw (BioScrew, RCI screw and
SmartScrewACL).
52
2.3.2. Study V
WasherLoc: The diameter of the prepared graft was measured in 1 mm increments and the
appropriate size tibial tunnel drilled using a cannulated reamer. The 20 mm WasherLoc was
impacted at the distal tunnel opening perpendicular to the tibial tunnel and fixed with a selftapping WasherLoc compression screw.
Tandem spiked washers: Two 14 x 1.3 mm spiked metal washers and two 4.5 mm
bicortical screws were used in tandem fashion using a figure-of-eight technique to fix the
graft to the anteromedial tibial cortex distal to the tunnel opening. The diameter of the graft
was sized in 0.5 mm increments and a corresponding compaction drill was used to drill the
tibial tunnel.
Intrafix: The prepared graft was measured in 0.5 mm increments and a compaction drill 0.5
mm greater than the graft was selected to drill the tibial tunnel. During graft fixation the
tendons were first separated with a trial instrument. Thereafter, the Intrafix Sheath was
inserted into the tibial tunnel and fixed with the Intrafix Tapered screw.
Interference screws: The graft diameter was measured in 0.5 mm increments using Graft
Sizing Tubes and the tibial tunnel equal to the graft diameter was drilled using a compaction
drill. As with the bioabsorbable screws in the femur, a notcher provided by the manufacturer
was used to ease the insertion and to accommodate the screw heads of the bioabsorbable
interference screw. All interference screws were inserted in outside-in fashion over a guide
pin flush with the intra-articular tunnel opening.
53
Figure 11. The implants used in study IV from left to right: WasherLoc; Arthrotek, Inc., Warsaw,
Indiana (20 mm diameter), tandem spiked washers; Linvatec Inc., Largo, Florida (two 14x1.3 mm
washers and two 4.5 mm bicortical screws), Intrafix; Innovasive Devices Inc., Marlborough,
Massachusetts, bioabsorbable (PLA) BioScrew interference screw; Linvatec, Inc., Largo, Florida (8
mm diameter, 25 mm length), titanium SoftSilk interference screw; Acufex Microsurgical, Inc. (8 mm
diameter, 25 mm length), bioabsorbable (PLA96/4) copolymer SmartScrew ACL interference screw;
Bionx Implants Inc., Blue Bell, Pennsylvania (8 mm diameter, 25 mm length).
54
A
C
B
D
Figure 12. The tibial hamstring graft fixations used in study V. (A) WasherLoc, (B) tandem spiked
washers, (C) Intrafix and (D) interference screw (BioScrew, SoftSilk and SmartScrewACL).
55
3. Biomechanical testing
The biomechanical testing protocols are summarized in Table 3. Biomechanical testing was
performed using Lloyd 6000R (I) and Lloyd LR 5K (II-V) materials testing machines (J.J.
Lloyd Instruments, Southampton, UK). The tibia (I, II, V) and femur (III, IV) were securely
mounted to the testing machine by specially designed clamps. The patella (I-III) was secured
to the testing machine with an 8 mm bar inserted through a drill hole made perpendicular to
the long axis of the patella. The STG graft (IV, V) was prepared for biomechanical testing by
suturing the tendons together in whipstitch fashion (No. 2-0 Vicryl suture, Ethicon, Johnson
& Johnson, Arlington, Texas), while maintaining constant tension. The graft was then
secured to a specially designed soft tissue clamp, leaving 25 mm length of graft between the
clamp and the intra-articular tunnel opening (Figure 13). The biomechanical testing consisted
of single load-to-failure tests (I-V), cyclic loading tests (II-V) and single load-to-failure tests
after cyclic loading (II-V).
Figure 13. The specimen in study V is mounted in the materials testing machine so that the
orientation of the load is parallel to the long axis of the tibial tunnel.
56
Table 3. Biomechanical testing protocols used in this study series.
Biomechanical testing
Single load-to-failure test
Study
Cyclic loading test
I
(n=33)
Vertical loading parallel to the long
axis of the tibia at a rate of 50
mm/min.
10 pairs (n=20)
10 pairs (n=20)
Vertical loading in line with the tibial
bone tunnel at a rate of 50 mm/min.
50 N preload.
II
(n=40)
Vertical loading in line with the tibial bone tunnel.
Cyclic loading started with 100 loading cycles between
50 and 150 N (1 cycle in 2 seconds). The load was
progressively increased in 50 N increments after each set
of 100 cycles. After 100 cycles at 850 N the specimens
were loaded to failure at a rate of 50 mm/min.
10 pairs (n=20)
10 pairs (n=20)
Vertical loading in line with the tibial
bone tunnel at a rate of 50 mm/min.
50 N preload for 10s.
III
(n=40)
Vertical loading in line with the tibial bone tunnel.
1500 loading cycles between 50 and 200 N (1 cycle in 2
seconds). After 1500 loading cycles the surviving
specimens were loaded to failure at a rate of 50 mm/min.
(n=60)
(n=60)
Vertical loading in line with the tibial
bone tunnel at a rate of 50 mm/min.
50 N preload for 10s.
IV
(n=120)
Vertical loading in line with the tibial bone tunnel.
1500 loading cycles between 50 and 200 N (1 cycle in 2
seconds). After 1500 loading cycles the surviving
specimens were loaded to failure at a rate of 50 mm/min.
(n=60)
(n=60)
Vertical loading in line with the tibial
bone tunnel at a rate of 50 mm/min
50 N preload for 10s
V
(n=120)
Vertical loading in line with the tibial bone tunnel
1500 loading cycles between 50 and 200 N (1 cycle in 2
seconds). After 1500 loading cycles the surviving
specimens were loaded to failure at a rate of 50 mm/min
57
3.1. Single load-to-failure test
A vertical load was carried out at a rate of 50 mm/min until failure. The specimen’s response
to the loading was automatically obtained in the form of a force-displacement curve (Figure
3), and the ultimate failure load (I-III), yield load (linear load) (I-V), elastic modulus (I),
stiffness (II-V) and mode of failure were determined (I-V). The yield load was described as
the point at which the force-displacement curve entered into the non-linear region. Stiffness
was determined as the slope of the linear region of the force-displacement curve. In study I,
the apparent elastic modulus of the graft was calculated from the force displacement curve in
the lower tensile load range between 0 and 500 N (equation 1) as well as in the upper load
range between 500 N and linear load (equation 2).
Equation 1:
500N x Initial length of the graft (mm)
Displacement at 500 N load (mm) x Cross-section of the graft (mm2)
Equation 2:
(Linear load (N) – 500 N) x Initial length of the graft (mm)
(Displacement at linear load (mm) – Displacement at 500 N load (mm))
x Cross section of the graft (mm2)
3.2. Cyclic loading test followed by a single load-to-failure test
In study II, the specimens were first subjected to a 50 N preload. Thereafter, the graftconstruct was loaded cyclically at a frequency of one cycle in two seconds. The cyclic
loading started with 100 load cycles between 50 and 150 N. The load was then progressively
increased in increments of 50 N after each set of 100 cycles. After 100 cycles at 850 N, the
surviving specimens were pulled to failure at a rate of 50 mm/min. Similar to the single loadto-failure test, the specimen’s response to loading was obtained in the form of a forcedisplacement curve, and the ultimate failure load, yield load and the failure mode were
determined (Figure 14). The yield load was determined as the load at which the displacement
58
first increased significantly. The graft displacement at the first peak of load at each force
level was also measured to estimate the resistance of fixation to slipping.
Yield load
Yield load
Force (F)
Displacement (d)
Displacement (d)
Figure 14. An example of a force-displacement curve of the cyclic loading test in study II. The cyclic
loading started with 100 loading cycles between 50 and 150 N. The loading was then increased in 50
N increments after each set of 100 cycles. At yield load the displacement first increased significantly.
In studies III to V, a 50 N preload was first applied to the specimens for 10 seconds, after
which the graft-fixation constructs underwent 1500 loading cycles between 50 and 200 N at
one cycle every two seconds parallel with the long axis of the bone tunnels. Similar to the
single load-to-failure test, the force-displacement curve was recorded (Figure 15). The
rigidity of the fixation was evaluated by determining the loading-induced increase in the
displacement from the preload level after 1, 10, 50, 100, 250, 500, 1000, and 1500 loading
cycles. After 1500 loading cycles, the surviving specimens were loaded to failure at a rate of
50 mm/min, and the ultimate failure load (III) and yield load (III-V) were determined.
59
Force (F)
d0 d1
d1500
Displacement (d)
Figure 15. An example of a force-displacement curve of the cyclic loading test in studies III-V (d0 =
displacement after preload, d1 = displacement after first load cycle and d1500 = displacement after 1500
load cycles).
4. Statistical analysis
The results of the continuous variables were reported as mean ± standard deviation (SD). A p
value less than 0.05 was considered statistically significant throughout this study series.
60
In studies I, IV and V, one-way analysis of variance (ANOVA) with the Tukey’s test as a
post hoc test was used to test the difference between the study groups. In studies II and III,
the differences between the groups were determined using paired sample t-test. The
difference in yield load values between single load-to-failure and cyclic loading tests were
further compared using unpaired t-test (II, IV and V). A McNemar test was performed in
study II to test the relationship between the screw type and the number of bone block
fractures.
61
RESULTS
1. Single load-to-failure tests
1.1. Study I
The results are summarized in Table 4. The mean linear loads for the interference screw
(1231 ± 420 N), bioabsorbable SR-PLLA screw (1174 ± 392 N) and AO cancellous screw
(951.3 ± 269 N) were not significantly different, nor were the mean ultimate failure loads or
the elastic modulus of the fixation.
Table 4. Single load-to-failure test in study I.
Fixation
Interference
AO-cancellous
SR-PLLA
(Mean ± SD)
Ultimate failure Yield load (N)
load (N)
1358 ± 348
1231 ± 420
1081 ± 331
951.3 ± 269
1211 ± 362
1174 ± 392
Elastic modulus (N/mm2)
E1
E2
186 ± 34
275 ± 83
201 ± 18
307 ± 116
189 ± 47
304 ± 72
1.2. Study II
Table 5 gives the results of the single load-to-failure test. The yield load was 621 ± 139 N in
the bioabsorbable group and 774 ± 154 N in the titanium group (p=0.07). Significant group
62
differences were found neither in the ultimate failure loads nor in the stiffness of the
fixations.
Table 5. Single load-to-failure test in study II.
Fixation
Ultimate failure
Yield load (N)
Stiffness (N/mm)
load (N)
Bioabsorbable
837 ± 260
621 ± 139
76 ± 20
Titanium
863 ± 192
774 ± 154
80 ± 15
(Mean ± SD)
1.3. Study III
The results of the single load-to-failure tests are presented in Table 6. The mean yield load
was 821 ± 281 N for the plugging technique and 791 ± 166 N for the interference technique
(p=0.82). There was no significant difference in the groups with regard to the ultimate failure
load or the stiffness of the fixation technique.
Table 6. Single load-to-failure test in study III.
Fixation
Ultimate failure
Yield load (N)
Stiffness (N/mm)
load (N)
Plugging
1061 ± 342
821 ± 281
75 ± 17
Interference
971 ± 260
791 ± 166
83 ± 11
(Mean ± SD)
63
1.4. Study IV
The results of the single load-to-failure test are summarized in Table 7. The mean yield loads
for the BoneMulchScrew and EndoButton CL groups were significantly greater than those in
the interference screw fixations. The mean yield load of the RigidFix group was also
significantly higher than in the BioScrew and the RCI screw groups. The highest stiffness
was found in the BoneMulchScrew group (115 ± 28 N/mm), which was significantly higher
than that in the EndoButton CL, RigidFix, RCI screw and BioScrew groups, but no
significant difference was observed between the BoneMulchScrew and SmartScrewACL
groups. The stiffness of the SmartScrewACL group was also significantly higher than in the
BioScrew and RCI screw groups.
Table 7. Single load-to-failure test in study IV.
Fixation
Yield load (N)
Stiffness (N/mm)
EndoButton CL
1086 ± 185
79 ± 7.2
BoneMulchScrew
1112 ± 295
115 ± 28
RigidFix
868 ± 171
77 ± 17
BioScrew
589 ± 204
66 ± 28
RCI screw
546 ± 174
68 ± 15
SmartScrewACL
794 ± 152
96 ± 20
(Mean ± SD)
1.5. Study V
The results of the single load-to-failure test are summarized in Table 8. The mean yield load
for the Intrafix group was significantly higher than those for the other groups and the mean
yield load for the WasherLoc group was significantly greater than that for the interference
screws. The stiffness of the Intrafix (223 ± 62 N/mm) was significantly higher than in the
other fixations. The stiffness of the SmartScrewACL group was also significantly higher than
that for the SoftSilk group.
64
Table 8. Single load-to-failure test in study V.
Fixation
Yield load (N)
Stiffness (N/mm)
WasherLoc
975 ± 232
87 ± 23
Tandem Spiked Washers
769 ± 141
69 ± 14
Intrafix
1332 ± 304
223 ± 62
SoftSilk
612 ± 176
91 ± 34
BioScrew
471 ± 107
61 ± 12
SmartScrewACL
665 ± 201
115 ± 34
(Mean ± SD)
2. Cyclic loading tests
2.1. Study II
The cyclic loading induced graft displacement at each force level was not significantly
different between bioabsorbable or titanium interference screws (Figure 16). After cyclic
loading, the mean linear loads for the bioabsorbable interference screw (605 ± 142 N) and
titanium interference screw (585 ± 103 N) were not significantly different, nor were the mean
ultimate failure loads 708 ± 115 N and 683 ± 136 N, respectively. The number of bone block
fractures was neither significantly different between the screws. The decrease in yield load
values between the single load-to-failure test and the single load-to-failure test subsequent to
cyclic loading was significantly greater in the titanium group (p=0.03).
65
Displacement (mm)
10
8
6
bioabsorbable
titanium
4
2
0
150 200 250 300 350 400 450 500 550 600
Force Level (N)
Figure 16. The mean displacement of the fixation at the first peak of load at each force level up to the
average yield point in the cyclic loading test in study II. There was no significant group difference
between bioabsorbable and titaniun interference screws. The values are given as the mean and
standard deviation.
2.2. Study III
The mean residual displacements of the fixations were not significantly different between the
interference technique and the plugging technique (Figure 17). The mean yield was 576 ±
189 N for the plugging technique and 679 ± 286 N for the interference technique (p=0.28),
and the mean ultimate failure loads were 994 ± 376 N and 1001 ± 343 (p=0.97), respectively.
2.3. Study IV
The only failures occurring during the cyclic loading test were one in the BioScrew group
and one in the RCI screw group. After 1500 cycles the BoneMulchScrew group showed the
lowest cyclic loading induced residual displacement (2.2 ± 0.7 mm) followed by the
66
SmartScrewACL (3.2 ± 0.9 mm), RigidFix (3.7 ± 1.0 mm), EndoButton CL (3.9 ± 0.7 mm),
RCI screw (3.9 ± 1.4 mm) and the BioScrew (4.0 ± 1.4 mm) groups (Figures 17 and 18). The
residual displacement in the EndoButton CL, RigidFix, BioScrew and RCI screw groups was
significantly greater than that in the BoneMulchScrew group. However, the residual
displacements between the BoneMulchScrew and SmartScrewACL groups did not differ
statistically significantly from one another.
Displacement (mm)
BoneMulchScrew/hamstring
SmartScrew ACL/hamstring
Femoral graft fixation options
RigidFix/hamstring
RCI screw/hamstring
EndoButton CL/hamstring
BioScrew/hamstring
interference screw/BPTB
Plug/BPTB
8
7
6
5
4
3
2
1
0
10
250
500
1500
Number of cycles
Figure 17. The mean displacement of the femoral graft fixation options used in this study series after
10, 250, 500 and 1500 cycles.
Table 9 gives the results of the single load-to-failure test after the cyclic loading. The mean
yield load for the BoneMulchScrew group was significantly higher than those for the
BioScrew and RCI screw groups and the mean yield load for the SmartScrewACL group was
significantly greater than that for the RCI screw. The stiffness of the BoneMulchScrew group
(189 ± 38 N/mm) was significantly higher than that in the EndoButton CL, RigidFix,
BioScrew or RCI screw groups. The stiffness of the RigidFix group was also significantly
67
higher than that for the EndoButton CL group, and the stiffness of the SmartScrewACL
group was significantly higher than that in the EndoButton CL and BioScrew groups. The
decrease in yield load values between the single load-to-failure test and the single load-tofailure test subsequent to cyclic loading was significant in the BoneMulchScrew and
EndoButton CL groups.
Table 9. Single load-to-failure test after cyclic loading test in study IV.
Fixation
Yield load (N)
Stiffness (N/mm)
EndoButton CL
781 ± 252
105 ± 13
BoneMulchScrew
925 ± 280
189 ± 38
RigidFix
768 ± 253
136 ± 13
BioScrew
565 ± 137
113 ± 15
RCI screw
534 ± 129
134 ± 23
SmartScrewACL
842 ± 201
162 ± 28
(Mean ± SD)
2.4. Study V
In the cyclic loading test the only failure of fixation occurred in the SoftSilk group (one
specimen failing). After the cyclic loading tests the Intrafix showed the lowest residual
displacement (1.5 ± 0.3 mm) followed by the WasherLoc (3.2 ± 1.5 mm), SmartScrewACL
(3.8 ± 1.5 mm), BioScrew (4.1 ± 1.2 mm), tandem spiked washers (4.2 ± 2.6 mm) and the
SoftSilk (4.7 ± 1.5 mm) (Figure 18). The mean residual displacement after the cyclic loading
test in the tandem spiked washer, BioScrew, SoftSilk and SmartScrewACL groups was
significantly greater than that in the Intrafix group, however, there was no significant
difference in the Intrafix and WasherLoc groups.
68
Displacement (mm)
Intrafix/tibia
SmartScrew ACL/tibia
Spiked washers/tibia
BoneMulchScrew/femur
RigidFix/femur
EndoButton CL/femur
WasherLoc/tibia
BioScrew/tibia
SoftSilk/tibia
SmartScrew ACL/femur
RCI screw/femur
BioScrew/femur
7
6
5
4
3
2
1
0
10
250
500
1500
Number of cycles
Figure 18. The mean displacement of the hamstring graft fixation options used in this study series
after 10, 250, 500 and 1500 cycles.
In the single load-to-failure test after the cyclic loading test, the Intrafix was again the
strongest (1309 ± 302 N) followed by WasherLoc (917 ± 234 N), SmartScrewACL (694 ±
173 N), tandem spiked washers (675 ± 190 N), BioScrew (567 ± 156 N) and SoftSilk (423 ±
75 N). Identically to the single load-to-failure test alone, these differences were highly
significant (p<0.001) for the Intrafix over all the other devices, except of WasherLoc
(p<0.01). The difference was also significant for the WasherLoc over the SoftSilk (p<0.001),
BioScrew (p<0.01) and tandem spiked washer (p<0.05). Similarly to the single load-tofailure test, the highest stiffness was found in the Intrafix group (267 ± 36 N/mm), being
significantly higher than that in the other fixation groups. The stiffness of the
SmartScrewACL group (159 ± 25) was significantly higher than that in the tandem spiked
washer (108 ± 26), BioScrew (125 ± 23), and SoftSilk (120 ± 18) groups. The decrease in
yield load values was not significant after cyclic loading.
69
DISCUSSION
The graft fixation site, not the graft material itself, is commonly considered the weakest link
during the early postoperative period after ACL reconstruction until the biologic fixation to
the host bone and the graft degradation begins, after which the strength of the graft material
becomes the major factor in limiting the rehabilitation (Johnson et al. 1992, Fu et al. 1999,
Brand et al. 2000c). Current grafts are initially stronger than the native ACL and the
postoperative rehabilitation protocols allow full weight bearing and full range of motion
immediately after the reconstruction, thus subjecting the graft to increased loading and
placing a special emphasis on the importance of secure initial graft fixation (Fu et al. 1999,
Brand et al. 2000c). The primary concern has been the fixation of soft tissue grafts, especially
their ability to resist repetitive in vivo forces during the rehabilitation period. Consequently,
the implant manufacturers have been seeking stronger and stiffer fixations and have
introduced various new graft fixation devices every year. Subsequently ACL graft fixation
research has become increasingly popular. A high number of these published ACL graft
fixation studies have certainly helped the orthopaedic surgeons to make better decisions
regarding the ACL reconstruction. However, the apparent influence of the implant business
casts a shadow on many of the previously published ACL graft fixation studies, which makes
the abundant information somewhat controversial and difficult to analyse and compile.
Although the strength requirements of the graft fixation have not been clearly demonstrated,
it has been shown that peak ACL forces may reach 0.55 x body weight during
isokinetic/isometric extension of the knee (Toutoungi et al. 2000). In a cadaver study, Rupp
et al. (1999a) demonstrated that extension of the knee against gravity produces 219 ± 25 N
resultant forces in the ACL. However, much lower fixation strength may be sufficient,
because excellent long-term outcomes have been reported for both BPTB and hamstring
tendon graft fixations, which have been found to provide poor stiffness and fixation strength
70
in biomechanical studies (e.g., sutures tied over a button and RCI) (Steiner et al. 1994,
Shelbourne and Nitz 1990, Pinczewski et al. 2002). In addition, the graft material itself
becomes considerably weaker than the initial fixation during the early postoperative period
(Butler et al. 1989, Rodeo et al. 1993, Weiler et al. 2002b).
1. Methodological issues
To compare the data between biomechanical studies, one has to keep in mind that practically
all biomechanical test modes are extremely case-sensitive i.e., small differences in study
design [e.g., specimen (animal vs. human tissue, young vs. old donors), biomechanical
testing protocol (loading rate, orientation of the applied load and number of loading cycles)]
can seriously influence the results of one test, and thus, comparisons between studies should
be done with extreme caution.
On the basis of the current literature, it is impossible to determine the testing protocol that
would mimic the loading experienced by the ACL substitute during the rehabilitation.
Beynnon and Amis (1998) have described guidelines for the biomechanical evaluation of the
ACL graft fixation. They emphasized the importance of the cyclic loading test and
recommended to evaluate anterior-posterior laxity with constant forces ranging between 150
(anterior) and -150 N (posterior). This method allows to mimic the Lachman and anterior
drawer tests by applying anteriorly directed load to the tibia in different knee flexion angles,
and therefore the results can be considered to be more relevant clinically. They also
suggested an alternative approach for testing ACL graft fixation constructs, in which one
bone (tibia or femur) and the other end of the graft are secured to the testing machine. The
latter approach was used in this study series and it allows to monitor a specific fixation at a
specific location, which gives more precise information about the specific fixation technique.
Despite these guidelines, the cyclic loading protocols have varied considerably.
The possible failure of the fixation in the bone tunnel during cyclic loading was not directly
analysed in this series of five successive experiments. The displacement values of the graft-
71
fixation construct observed during the cyclic loading test could be assumed to result from
either the creep of the tendon or the gradual slippage/failure of the graft fixation. Considering
the carefully standardized testing protocols, the displacement seen in the cyclic loading tests
was apparently mostly a result of the creep of the tendons and the increase in the
displacement values was most likely due to the gradual failure of the fixation. In fact,
Rittmeister et al. (2002) recently showed that over 90% of the total displacement seen in the
biomechanical testing of ACL reconstructs is attributable to soft tissue graft slippage past the
interference screw.
Porcine knees are the most commonly used sources of bone for testing various ACL fixation
methods despite having clearly higher (volumetric) bone density than human bones (Nurmi et
al. 2002, Nurmi et al. 2004c). As the quality of human bone specimens often varies
considerably, porcine and even bovine knee specimens with more uniform bone quality can
be considered to offer a reasonable alternative to the human bone. In studies using porcine or
bovine bone the results are presumably over optimistic compared with human bone, which
probably more influences the evaluation of fixations that rely on the friction to secure the
graft, such as interference screws and Intrafix in comparison with extra-articular techniques
(EndoButton CL, BoneMulchScrew, RigidFix, WasherLoc and tandem spiked washers).
2. BPTB graft fixation
At the time this series of experiments was carried out, the BPTB graft fixation with metal
interference screws was clearly the procedure of choice (“golden standard”), and
accordingly, the basis of comparison in ACL surgery. Bioabsorbable implants were
developed to offer benefits compared to their metal counterparts. They were introduced
clinically in the treatment of fractures and osteotomies in 1984 (Rokkanen et al. 1985,
Böstman 1991). Along with the popularity of ACL reconstruction surgery, the number of
ACL revisions also increased dramatically (Jaureguito and Paulos 1996, Safran and Harner
1996, Allen et al. 2003). As the retained metal screws may severely complicate the revision
72
ACL surgery and disturb MRI (Shellock et al. 1992, Suh et al. 1998), there was an apparent
benefit associated with the use of fixation implants made of bioabsorbable material, but their
suitability had to be properly validated.
On the basis of our biomechanical testing (study I), the bioabsorbable fibrillated SR-PLLA
screw provided as good initial fixation strength as the Kurosaka interference screw and the
AO cancellous screw in the fixation of a BPTB graft in the bovine knee, and thus, could be
considered a suitable alternative for the metal screws in the BPTB graft fixation in the
reconstruction of ACL. These pullout test results were similar to those of other reports using
the BPTB graft fixation and bovine knee specimens (Hulstyn et al. 1993, Lemos et al. 1995,
Shapiro et al. 1995, Brown et al. 1996, Bryan et al. 1996, Abate et al. 1998, Pomeroy et al.
1998, Weiler et al. 1998b).
Also to validate the suitability design-wise, the results of study II showed that both the
bioabsorbable and the titanium interference screws provided similar fixation strength in the
single load-to-failure and cyclic loading tests. The single load-to-failure test results of the
titanium interference screw were also very comparable to those obtained by other
investigators using either young human cadaveric or porcine knee specimens and metal
interference screws for the BPTB graft fixation (Liu et al. 1995, Pierz et al. 1995, Brown et
al. 1996, Pena et al. 1996, Rupp et al. 1997, Rupp et al. 1998, Seil et al. 1998, Stapelton et al.
1998, Rupp et al. 1999b, Honl et al. 2002).
Bioabsorbable and metal interference screws have provided comparable initial fixation
strength in the single load-to-failure testing of BPTB graft fixations (Johnson and vanDyk
1996, Caborn et al. 1997, Rupp et al. 1997, Abate et al. 1998, Weiler et al. 1998b). However,
Pena et al. (1996) found that the metal interference screw provided superior fixation strength
compared to the bioabsorbable interference screw. Beevers (2003) speculated that the quality
of the bones between the two study groups explained their finding and that bioabsorbable and
metal interference screw groups were similar when a correction for different bone mineral
density was done.
73
Seil et al. (1998) were the first to compare under cyclic loading conditions the initial fixation
strength of bioabsorbable and metal interference screws in the fixation of BPTB graft in ACL
reconstructions. The cyclic loading test consisted of 500 loading cycles between 60 to 250 N
at a rate of 300 mm/min, after which the specimens were loaded to failure. With regard to the
ultimate failure load and bone block slippage observed during cyclic loading, no significant
difference was found between the bioabsorbable and titanium interference screws. In our
study, the displacement values were similar at each loading level between the two
interference screws, confirming the previous findings by Seil et al. (1998).
A number of studies have shown a consistent phenomenon of more bone block fractures with
metal interference screws than bioabsorbable interference screws in the single load-to-failure
test (Pena et al. 1996, Caborn et al. 1997, Rupp et al. 1997, Weiler et al. 1998b). Similarly,
there were more bone block fractures in the specimens fixed with a titanium interference
screw compared to bioabsorbable interference screws in our study (II), although the
difference was not statistically significant. The difference in failure mode can be partly
explained by the difference in the elastic modulus between the bone (cancellous bone 0.2-0.7
GPa and cortical bone 9-20 GPa), and the interference screws (titanium alloy 110 GPa,
PLA96/4 4.6-7.5 GPa) (Ruluff and McIntyre 1982, Anderson et al. 1992, McGalden et al.
1993, Pohjonen and Törmälä 1996). It appears, however, that the failure of bone blocks in
the metal interference screw fixation is not of clinical significance.
Although the interference screw fixation of the BPTB graft is safe and reliable in most cases,
there are some pitfalls related to the operative technique, especially in the technically more
demanding femoral site. Damage to the bone block, the tendinous part of the graft or passing
sutures during screw insertion, and divergent screw insertion are some of the reported
problems associated with the femoral interference screw fixation of the BPTB graft in ACL
reconstruction. Furthermore, if the single-incision technique is used, surgeons occasionally
face problems with graft-tunnel mismatch, that is, the graft “being too long” and protruding
out of the tibial drill tunnel. Then alternative fixation techniques are often required (Taylor et
al. 1996, Denti et al. 1998, Fowler et al. 1998, Olszewski et al. 1998, Hoffmann et al. 1999).
To avoid these problems, a new plugging technique was developed. With this technique the
74
groove of the implant (plug) protects the graft during insertion (graft fixation), while the plug
is tapped behind the bone block instead of being forced into the gap between the bone tunnel
wall and the bone block (Figure 8). In addition, the plug insertion (tapping) advances the
femoral bone block deeper into the femoral tunnel, subsequently pulling the tibial bone block
into the tibial tunnel diminishing the possibility for mismatch. In addition, as the implant
obstructs the femoral tunnel opening, a smaller bone block is needed, and the risk of a
patellar fracture in the BPTB graft harvest can be reduced (Moholkar et al. 2002, Stein et al.
2002). The plug is also made of bioabsorbable material providing the previously mentioned
material advantages.
In the biomechanical validation of the novel BPTB graft fixation technique (study III), no
significant difference was found in failure loads between a metal interference screw and a
bioabsorbable plug in the single load-to-failure and cyclic loading tests. The load-induced
increase in the displacement was similar in both study groups indicating that both fixations
failed equally. The failure loads of the interference technique were also similar to those of the
previous studies using young human or porcine knee specimens and the BPTB graft fixation
with metal interference screws (Reznik et al. 1990, Matthews et al. 1993, Butler et al. 1994,
Paschal et al. 1994, Pierz et al. 1995, Brown et al. 1996, Pena et al. 1996, Rupp et al. 1997,
Pomeroy et al. 1998). However, as mentioned before, mechanically weaker graft fixations
could be strong enough to prevent a fixation failure and slippage during rehabilitation. This
may particularly concern the femur where shear forces enhance the fixation, as the line of
force does not come parallel with the femoral bone tunnel until 100ο of flexion. It is also
commonly considered that the highest forces are acting near the extension of the knee.
Further, although the press-fit fixations of the BPTB graft (femoral bone plug) have yielded
relatively low values in pull out testing and poor resistance to repetitive loading, the clinical
outcome has proved highly satisfactory (Boszotta 1997, Rupp et al. 1997, Seil et al. 1998).
75
3. Hamstring graft fixation
Many variables in graft fixation seem to influence the overall long-term outcome after the
reconstruction of the ACL, particularly when the soft tissue grafts (hamstring tendon) are
used. The donor-site morbidity associated with the use of BPTB grafts is the most persuasive
argument favouring the use of hamstring grafts. Studies have shown that the initial structural
properties (failure strength and stiffness) of the four strand hamstring grafts are comparable
or even higher than those of the BPTB graft and the hamstring graft fixation techniques have
considerably improved, resulting in a steady increase in the use of the hamstring tendon
grafts during the past decade. However, concerns with the hamstring grafts still include the
initial strength of graft fixation, possibly increased knee laxity and graft-tunnel motion,
slower biological fixation to the bone tunnel compared to the BPTB graft, possibly increased
graft tunnel motion, hamstring muscle weakness, and discomfort associated with cortical
hardware (Fu et al. 1999, Miller and Gladstone 2002). Whether the increased knee laxity
observed in some clinical studies comparing hamstring tendon and BPTB autografts in ACL
reconstruction is due to the method of fixation or the difference in graft material itself is
unclear (Otero and Hutcheson 1993, O’Neil 1996, Corry et al. 1999, Beynnon et al. 2002,
Feller and Webster 2003).
Interference screw fixation has become popular also with hamstring grafts. Anatomic
(aperture) fixation, in which the fixation is close to the original ACL insertion site e.g., an
interference screw placed in the drill tunnel adjacent to the joint surface, has been shown to
improve knee stability and graft isometry (Morgan et al. 1995, Ishibashi et al. 1997).
Anatomic fixation also potentially minimizes graft motion relative to the bone tunnel wall
and prevents the access of synovial fluid between the graft and the bone tunnel wall
decreasing the risk for tunnel widening and enhancing graft healing at the articular tunnel
aperture (Weiler et al. 2002b). However, concern has been raised of the soft tissue graft
interference screw fixation with regard to the possible slippage of the graft past the screw
under repetitive loads during the immediate postoperative period prior to graft incorporation,
as the fixation depends on the friction between the tendons and walls of the bone tunnels
(Giurea et al. 1999, Magen et al. 1999, Scheffler et al. 2002). In contrast, the extra-articular
76
methods, in which the fixation site is at a distance from the joint line, have been implicated to
cause graft elongation, graft-tunnel motion and enlargement of the bone tunnels and
ultimately increase in knee laxity (Jansson et al. 1999, Scheffler et al. 2003). Unexpectedly,
Singhatat et al. (2002) recently demonstrated in an ovine model after 4 weeks of extraarticular fixation technique (WasherLoc) improved graft healing to the bone tunnel in
comparison with interference screw fixation. The extra-articular WasherLoc maintained the
strength and improved the stiffness of the graft-fixation complex, whereas the strength and
stiffness of the fixation with bioabsorbable interference screws decreased considerably. The
major weakness of the study was the absence of intra-articular healing environment. The
possible synovial fluid access between the bone tunnel and the healing tendon combined with
potential graft-tunnel motion could have a considerable effect on the results. It has also been
shown that the improvement in knee stability is strongly influenced by the rigidity and
stiffness of the fixation, not solely by the length of the graft fixation construct (Magen et al.
1999, To et al. 1999).
In studies IV and V, the fixation strength of the commonly used femoral (EndoButton CL,
BoneMulchScrew, RigidFix and three different interference screws) and tibial (WasherLoc,
tandem spiked washers, Intrafix and three interference screws) hamstring ACL graft fixation
methods were compared biomechanically. The extra-articular hamstring graft femoral
fixation methods (study IV) differed from each other by their principle of fixation:
EndoButton CL and BoneMulchScrew can be considered “true” toggle fixations, which
suspend the graft within the tunnel, and the tendon is looped around a part of the implant. In
the EndoButton CL, the tendon is fixed to the cortex of the femur via continuous polyester
loop (indirect fixation) and in the BoneMulchScrew, the tendon is looped over a crossbar
inserted into the metaphyseal bone of the lateral femoral condyle. RigidFix, in turn, should
be considered a “semi-toggle” fixation because it is not possible to wrap the tendon loops
around the cross pins. Therefore the strength of fixation depends on the pins engaging the
tight tendon loop against the walls of the tunnel.
In study IV, the extra-articular techniques generally provided better pullout strength of the
fixation than the apertural (interference screws) with no significant difference between the
77
three extra-articular techniques. As the BoneMulchScrew was found superior to EndoButton
CL and RigidFix, it appears that the rigidity of the device itself improves the fixation
strength. In the cyclic loading test, in turn, the EndoButton CL and RigidFix groups showed
more residual displacement compared to the BoneMulchScrew, although the single cycle
loading results were comparable. Because there were no visible changes in the titanium
button of the EndoButton CL during the cyclic loading test, changes in the continuous loop
or tendon-loop interface may have been the reason for the increased residual slippage. In the
RigidFix group, the increased residual displacement was most likely attributable to the
gradual bending or fracture of the bioabsorbable pins. Of the interference screws, the
SmartScrewACL was found to be comparable to the BoneMulchScrew, whereas the
BioScrew and RCI screws allowed significantly higher residual displacement than the
BoneMulchScrew.
Although all the extra-articular methods gave significantly better strength of fixation than the
RCI screw and BioScrew, the difference in the yield load values between the RigidFix and
SmartScrewACL was not statistically significant, indicating that some difference in the
characteristics of the three interference screws had a considerable effect on the strength of
fixation. While the bioabsorbable material used in manufacturing interference screws and the
surface finish have been speculated to enhance the strength of fixation of interference screws
(Weiler et al. 1998a), our results rather suggest that the design (geometry, core diameter,
pitch/thread thickness, etc.) of certain bioabsorbable screws is more likely the reason for the
improved fixation strength of soft tissue grafts in the ACL reconstruction. In accordance with
our findings, Giurea et al. (1999) recently showed that the interference screw design has an
effect on the slippage of the soft tissue fixation under cyclic loading conditions. After 1100
loading cycles at 150 N, the titanium RCI screw allowed a mean displacement of 6.8 ± 5.6
mm, which was over three times higher than that observed with the titanium interference
screws specifically designed for the soft tissue graft fixation.
In study V, the two extra-articular fixation methods differed in their effective graft lengths.
The WasherLoc secures the graft within the bone tunnel opening, whereas the tandem spiked
washers fixes the graft completely outside the tibial tunnel, which results in an even longer
78
effective graft length. The Intrafix, in turn, secures the graft inside the distal tunnel leaving
approximately 5 to 15 mm of free graft within the proximal opening (aperture) of the drill
tunnel. Finally, the interference screws can be considered truly apertural fixations, as the
screws can be driven flush with the intra-articular tunnel opening. The Intrafix provided
superior pullout strength and stiffness in single load-to-failure tests and it was also superior
in resisting slippage under cyclic loading when compared to other implants. In the cyclic
loading test, although the stiffness values were comparable, the yield load for WasherLoc
was significantly higher than for the interference screws. Although SmartScrewACL was
consistently better than BioScrew and SoftSilk, the only statistically significant difference
was found in stiffness of the fixation.
Bioabsorbable interference screws have been in clinical use for the fixation of ACL
replacements at least since 1992 (McGuire et al. 1999, Stähelin 1997). Some complications
have been reported, such as breakage of the screw during insertion, inflammatory reactions
and osteolytic cyst formation (Barber et al. 1995, Imhoff et al. 1998, Martinek et al. 1999).
The short- and mid-term clinical studies have suggested that bioabsorbable interference
screws provide a reasonable alternative to metal interference screws in the fixation of both
BPTB and hamstring grafts (Barber et al. 1995, Marti et al. 1997, McGuire et al. 1999,
Charlton et al. 2003).
To summarize the results, the concern about the use of interference screws in securing soft
tissue grafts was supported by the results of our studies on the femoral and tibial fixation
devices (studies IV and V). The only complete failures of fixation observed during the cyclic
loading testing of the specimens were one in the BioScrew group, one in the RCI screw
group and one in the SoftSilk group. However, in the femur (study IV), SmartScrewACL was
the second best of all the implants tested in the cyclic loading test. The clinical correlation
can be found from Corry et al. (1999) and Pinczewski et al. (2002), who have shown that
hamstring graft fixation with metal interference screws provides good clinical outcome and
restores satisfactory knee stability. Our results indicate that both BPTB and hamstring tendon
graft fixation methods can provide rigid fixation without concern about increased knee laxity
due to loosening of the fixation under repetitive loading conditions. However, it seems that
79
the design and material characteristics of the implant have a major impact, especially on the
strength of hamstring graft fixation.
This study series evaluated biomechanically the initial fixation strength of various femoral
and tibial fixation devices of both BPTB and quadrupled hamstring tendon grafts in ACL
reconstruction. It is important to be critical and cautious when interpreting these test results.
Also it is essential to understand that the highest fixation strength, superior stiffness and the
best resistance to repetitive loading in biomechanical studies do not necessarily mean that the
fixation is superior to the long-term survival of the graft. For example, a stronger and stiffer
graft fixation construct is less forgiving in case of incorrect tunnel placement and may restrict
knee motion and ultimately jeopardize grafts survival (biologic incorporation and
ligamentization) in long-term. As mentioned earlier, time zero biomechanical ACL graft
fixation studies have been abundant. Whether we like it or not, in retrospect it appears safe to
conclude that the importance of the strength of the ACL graft fixation has been exaggerated,
apparently at least partly due to commercial reasons. We must remember that the goal of an
ACL reconstruction is not the immediate postoperative period, but rather to re-establish the
normal function and stability of the knee for decades. Therefore, in the future, to better
validate the suitability of various implants or fixation techniques in the reconstruction of the
ACL, it is essential to focus on in vivo research.
80
CONCLUSIONS
1. On the basis of the biomechanical testing results that we obtained with the three different
screws, the bioabsorbable fibrillated SR-PLLA screw is suitable for the fixation of a
BPTB graft in the reconstruction of the ACL.
2. Considering the biomechanical testing results that we obtained with the two interference
screws in this study, the bioabsorbable and standard titanium interference screws provide
similar initial fixation strength in the fixation of a BPTB graft in the ACL reconstruction.
3. The results show that in terms of initial fixation strength the new plugging technique is as
good as the standard interference screw technique in the fixation of a BPTB graft in the
femoral tunnel in the reconstruction of the ACL.
4. On the basis of the results, the cross pin technique (BoneMulchScrew) provided superior
fixation strength in the femoral tunnel and the central four-quadrant sleeve and screw
system (Intrafix) was clearly superior in the tibial tunnel in securing the quadrupled
hamstring graft in the ACL reconstruction. In the femoral site, EndoButton CL, RigidFix,
BioScrew and RCI screw and in the tibial site, tandem spiked washers, BioScrew,
SoftSilk and SmartScrew ACL allowed increased residual displacement during cyclic
loading. Therefore, some caution may be warranted in the rehabilitation after the ACL
reconstruction when using these implants. Soft tissue specific titanium interference
screws (RCI screw in femur and SoftSilk screw in tibia) showed inferior pullout strength
in single load-to-failure tests.
81
ACKNOWLEDGMENTS
The present study was carried out at the Department of Surgery, Medical School, University
of Tampere, Finland, from 1993 to 2003.
My deepest gratitude is extended to my supervisors, Professor Markku Järvinen, MD, Head
of the Department of Surgery, University of Tampere and Docent Teppo Järvinen MD,
Department of Surgery, University of Tampere, for giving me the opportunity to do research
and providing me with the excellent facilities to perform this study. Your continuous
encouragement, constructive guidance and unfailing support have enabled me to complete
the thesis. Above all, it has been a great privilege to enjoy your friendship.
I am grateful to the official reviewers of this study, Docent Arsi Harilainen, MD, University
of Helsinki, and Docent Andreas Weiler, MD, University of Berlin, for their careful review,
valuable suggestions and constructive criticism during the final preparation of this thesis.
I sincerely thank Rauno Piirtola, FM, MA (Mich.), for his careful revision of the English
language of the manuscript.
I am grateful to the co-authors of my original studies, Professor Pekka Kannus, MD, Timo
Pohjonen, Licensed Technician, Auvo Kaikkonen, MD, PhD, Mikko Kotikoski, MD, PhD,
Pia Ahvenjärvi, M.Sc. (Eng.), and Mika Vihavainen, MD, for their important contribution.
Also I would like to thank Matti Jäntti, MD, for our fruitful collaboration.
I want to thank my colleagues Minna Kääriäinen, MD, PhD, and Mika Paavola, MD, PhD,
with whom I have shared my office, for their friendship and support. Warm gratitude is also
due to Docent Antero Natri, MD.
82
I wish to express my gratitude to Docent Jukka-Pekka Mecklin, MD, Surgeon in Chief,
Department of Surgery, Jyväskylä Central Hospital, and Docent Ilkka Kiviranta, MD, Chief
in Orthopaedics and Traumatology, Jyväskylä Central Hospital, for providing me the
opportunity to finish this study. Collectively I also want to express my special thanks to my
colleagues at the Department of Surgery in Jyväskylä Central Hospital, for friendship and
support.
My warmest thanks are due to my mother and father, for their encouragement and support
throughout my life. Unfortunately, my father had no opportunity to see me completing this
thesis. I want specially thank my grandmother, Saara-mummu, for her unfailing background
support.
I also want to thank my sisters, Sari and Niina, along with their families, parents-in-law, Kari
and Seija, and all my friends for their encouragement and interest in my work.
Finally, from all my heart, I thank my dear wife, Nina, for her endless and unselfish support
and love during the years of this study. During my work you have made a home where I have
experienced understanding, patience and love. She and our lovely children Emilia, Sofia,
Maria and Elias have given me a proper perspective on everyday life and also the strength to
complete the thesis. More importantly, you all are the greatest joy in my life.
This study was partly supported by the Medical Research Fund of Tampere University
Hospital, National Graduate School of Clinical Investigation and the Finnish Foundation for
Orthopaedical and Traumatological Research.
Jyväskylä, March 2004
Petteri Kousa
83
REFERENCES
Abate JA, Fadale PD, Hulstyn MJ and Walsh WR (1998): Initial fixation strength of
polylactic acid interference screws in anterior cruciate ligament reconstruction. Arthroscopy
14: 278-284.
Aglietti P, Buzzi R, Zaccherotti G and De Biase P (1994): Patellar tendon versus doubled
semitendinosus and gracilis tendons for anterior cruciate ligament reconstruction. Am J
Sports Med 22: 211-218.
Aglietti P, Buzzi R, Giron F, Simeone AJ, Zaccherotti G (1997): Arthroscopic-assisted
anterior cruciate ligament reconstruction with the central third patellar tendon. A 5-8-year
follow-up. Knee Surg Sports Traumatol Arthrosc 5: 138-144.
Allen CR, Giffin JR and Harner CD (2003): Revision anterior cruciate ligament
reconstruction. Orthop Clin N Am 34: 79-98.
Amis AA and Jacob RP (1998): Anterior cruciate ligament graft positioning, tensioning and
twisting. Knee Surg Sports Traumatol Arthrosc 6 (Suppl 1): 2-12.
Anderson MJ, Keyak JH and Skinner HB (1992): Compressive mechanical properties of
human cancellous bone after gamma irradiation. J Bone Joint Surg 74A: 747-752.
Arnoczky SP, Tarvin GB and Marshall JL (1982): Anterior cruciate ligament replacement
using patellar tendon. An evaluation of graft revascularization in the dog. J Bone Joint Surg
64A: 217-224.
84
Aune AK, Ekeland A and Cawley PW (1998): Interference screw fixation of hamstring vs
patellar tendon grafts for anterior cruciate ligament reconstruction. Knee Surg Sports
Traumatol Arthrosc 6: 99-102.
Aune AK, Holm I, Risberg MA, Jensen HK and Steen H (2001): Four-strand hamstring
tendon autograft compared with patellar tendon-bone autograft for anterior cruciate ligament
reconstruction. A randomized study with two-year follow-up. Am J Sports Med 29: 722-728.
Aydin H, Kerimoglu S, Çitlak A and Turhan AU (2004): Initial fixation strength of
interference nail fixation for anterior cruciate ligament reconstruction with patellar tendon
graft (experimental study). Knee Surg Sports Traumatol Arthrosc 12: 94-97.
Barber FA, Elrod BF, McGuire DA and Paulos L (1995): Preliminary results of an
absorbable interference screw. Arthroscopy 11: 537-548.
Bealle D and Johnson DL (1999): Technical pitfalls of anterior cruciate ligament surgery.
Clin Sports Med 18: 831-845.
Beard DJ, Anderson JL, Davies S, Price AJ and Dodd CA (2001): Hamstring vs. patella
tendon for anterior cruciate ligament reconstruction: A randomized controlled trial. Knee 8:
45-50.
Beevers DJ (2003): Metal vs bioabsorbable interference screws: initial fixation. J Eng Med
217: 59-75.
Beynnon BD and Amis AA (1998): In vitro testing protocols for the cruciate ligaments and
ligament reconstructions. Knee Surg Sports Traumatol Arthrosc 6 (Suppl 1): 70-76.
Beynnon BD, Fleming BC, Johnson RJ, Nichols CE, Renström PA and Pope MH (1995):
Anterior cruciate ligament strain behaviour during rehabilitation exercises in vivo. Am J
Sports Med 23: 24-34.
85
Beynnon BD, Johnson RJ, Fleming BC, Kannus P, Kaplan M, Samani J and Renström P
(2002): Anterior cruciate ligament replacement: comparison of bone patellar tendon-bone
grafts with two-strand hamstring grafts. J Bone Joint Surg 84A: 1503-1513.
Beynnon BD, Meriam CM, Ryder SH, Fleming BC and Johnson RJ (1998): The effect of
screw insertion torque on tendons fixed with spiked washers. Am J Sports Med 26: 536-539.
Black J: Editorial: Does corrosion matter? (1988): J Bone Joint Surg 70B: 517-520.
Black KP, Saunders MM, Stube KC, Moulton MJR and Jacobs CR (2000): Effects of
interference fit screw length on tibial tunnel fixation for anterior cruciate ligament
reconstruction. Am J Sports Med 28: 846-849.
Blevins FT, Hecker AT, Bigler GT, Boland AL and Hayes WC (1994): The effects of donor
age and strain rate on the biomechanical properties of bone-patellar tendon-bone allografts.
Am J Sports Med 22: 328-333.
Böstman OM (1991): Absorbable implants for the fixation of fractures: Current concepts
review. J Bone Joint Surg 73A: 148-153.
Boszotta H (1997): Arthroscopic anterior cruciate ligament reconstruction using a patellar
tendon graft in press-fit technique: Surgical technique and follow-up. Arthroscopy 13: 332339.
Brand J Jr, Hamilton D, Selby J, Pienkowski D, Caborn DNM and Johnson DL (2000a):
Biomechanical comparison of quadriceps tendon fixation with patellar tendon bone plug
interference fixation in cruciate ligament reconstruction. Arthroscopy 16: 805-812.
86
Brand JC Jr, Pienkowski D, Steenlage E, Hamilton D, Johnson DL and Caborn DNM
(2000b): Interference screw fixation strength of a quadrupled hamstring tendon graft is
directly related to bone mineral density and insertion torque. Am J Sports Med 28: 705-710.
Brand J Jr, Weiler A, Caborn DNM, Brown CH and Johnson DL (2000c): Graft fixation in
cruciate ligament reconstruction. Current concepts. Am J Sports Med 28: 761-774.
Brown CH and JH Sklar (1998): Endoscopic antrior cruciate ligament reconstruction using
quadrupled hamstring tendons and endobutton femoral fixation. Techniques in Orthopaedics
13: 281-298.
Brown CH Jr, Hecker AT, Hipp JA, Myers ER and Hayes WC (1993): The biomechanics of
interference screw fixation of patellar tendon anterior cruciate ligament grafts. Am J Sports
Med 21: 880-886.
Brown CH and Carson EW: Revision anterior cruciate ligament surgery (1999). Clin Sports
Med 18: 109-171.
Brown GA, Pena F, Grøntvedt T, Labadie D and Engebretsen L (1996): Fixation strength of
interference screw fixation in bovine, young human, and elderly human cadaver knees:
Influence of insertion torque, tunnel-bone block gap, and interference. Knee Surg Sports
Traumatol Arthrosc 3: 238-244.
Bryan JM, Bach BR, Bush-Joseph CA, Fisher IM and Hsu KY (1996): Comparison of
"inside-out" and "outside-in" interference screw fixation for anterior cruciate ligament
surgery in a bovine knee. Arthroscopy 12: 76-81.
Butler DL, Grood ES, Noyes FR, Olmstead ML, Hohn RB, Arnoczky SP and Siegel MG
(1989): Mechanical properties of primate vascularized vs. nonvascularized patellar tendon
grafts; changes over time. J Orthop Res 7: 68-79.
87
Butler JC, Branch TP and Hutton WC (1994): Optimal graft fixation -the effect of gap size
and screw size on bone plug fixation in ACL reconstruction. Arthroscopy 10: 524-529.
Caborn DNM, Urban WP Jr, Johnson DL, Nyland J and Pienkowksi D (1997):
Biomechanical comparison between Bioscrew and titanium alloy interference screws for
bone-patellar tendon-bone graft fixation in anterior cruciate ligament reconstruction.
Arthroscopy 13: 229-232.
Caborn DNM, Coen M, Neef R, Hamilton D, Nyland J and Johnson DL (1998): Quadrupled
semitendinosus-gracilis autograft fixation in the femoral tunnel: a comparison between a
metal and a bioabsorbable interference screw. Arthroscopy 14: 241-245.
Charlton WPH, Randolf DA Jr, Lemos S and Shields CL Jr (2003): Clinical outcome of
anterior cruciate ligament reconstruction with quadrupled hamstring tendon graft and
bioabsorbable interference screw fixation. Am J Sports Med 31: 518-521.
Clark R, Olsen RE, Larson BJ, Goble EM and Farrer R (1998): Cross-pin femoral fixation: A
new technique for hamstring anterior cruciate ligament reconstruction of the knee.
Arthroscopy 14: 258-267.
Corry IS, Webb JM, Clingeleffer AJ and Pinczewski LA (1999): Arthroscopic reconstruction
of the anterior cruciate ligament. A comparison of patellar tendon autograft and four-strand
hamstring tendon autograft. Am J Sports Med 27: 444-454.
Dalldorf PG, Alexander J and Lintner DM (1998): One- and two-incision anterior cruciate
ligament reconstruction: a biomechanical comparison including the effect of simulated
closed-chain exercise. Arthroscopy 14: 176-181.
Denti M, Bigoni M, Randelli P, Monteleone M, Cevenini A, Ghezzi A, Panni SA and
Trevisan C (1998): Graft-tunnel mismatch in endoscopic anterior cruciate ligament
88
reconstruction. Intraoperative and cadaver measurement of the intra-articular graft length and
the length of the patellar tendon. Knee Surg Sports Traumatol Arthrosc 6: 165-168.
Dienst M, Burks RT and Greis PE (2002): Anatomy and biomechanics of the anterior
cruciate ligament. Orthop Clin N Am 33: 605-620.
Ejerhed L, Kartus J, Sernert N, Köhler K and Karlsson J (2003): Patellar tendon or
semitendinosus tendon autografts for anterior cruciate ligament reconstruction? A
prospective randomized study with a two-year follow-up. Am J Sports Med 31: 19-25.
Eriksson K, Anderberg P, Hamberg P, Löfgren AC, Bredenberg M, Westman I and
Wredmark T (2001): A comparison of quadruple semitendinosus and patellar tendon grafts in
reconstruction of the anterior cruciate ligament. J Bone Joint Surg 83B: 348-354.
Feller JA and Webster KE (2003): A randomized comparison of patellar tendon and
hamstring tendon anterior cruciate ligament reconstruction. Am J Sports Med 31: 564573.
Ferretti A, Conteduca F, Morelli F, Monteleone L, Nanni F and Valente M (2003):
Biomechanics of anterior cruciate ligament reconstruction using twisted doubled hamstring
tendons. Int Orthop 27: 22-25.
Fowler BL and DiStefano VJ (1998): Tibial tunnel bone grafting: A new technique for
dealing with graft-tunnel mismatch in endoscopic anterior cruciate ligament reconstruction.
Atrhroscopy 14: 224-228.
Frank C, Amiel D and Akeson WH (1983): Healing of the medial collateral ligament of the
knee. A morphological and biomechanical assessment in rabbits. Acta Orthop Scand 54: 917923.
Frank CB and Jackson DW (1998): The science of reconstruction of the anterior cruciate
ligament. Current concepts review. J Bone Joint Surg 79A: 1556-1575.
89
Freedman KB, D’Amato MJ, Nedeff DD, Kaz A and Bach BR Jr (2003): Arthroscopic
anterior cruciate ligament reconstruction: A metaanalysis comparing patellar tendon and
hamstring tendon autografts. Am J Sports Med 31: 2-11.
Fu FH, Bennet CH, Lattermann C and Ma CB (1999): Current trends in anterior cruciate
ligament reconstruction. Part I: Biology and biomechanics of reconstruction. Current
concepts. Am J Sports Med 27: 821-830.
Fu FH, Bennet CH, Ma CB, Menetrey J and Lattermann C (2000): Current trends in anterior
cruciate ligament reconstruction. Part II: Operative procedures and clinical correlations.
Current concepts. Am J Sports Med 28: 124-130.
Gerich TG, Cassim A, Lattermann C and Lobenhoffer HP (1997): Pullout strength of tibial
graft fixation in anterior cruciate ligament replacement with a patellar tendon graft:
interference screw versus staple fixation in human knees. Knee Surg Sports Traumatol
Arthrosc 5: 84-88.
Girgis FG, Marshall JL and Al Monajem ARS (1975): The cruciate ligaments of the knee
joint. Anatomical, functional and experimental analysis. Clin Orthop 106: 216-231.
Giurea M, Zorilla P Amis AA and Aicroth P (1999): Comparative pull-out and cyclic-loading
strength tests of anchorage of hamstring tendon grafts in anterior cruciate ligament
reconstruction. Am J Sports Med 27: 621-625.
Gobbi A, Mahajan S, Zanazzo M and Tuy B (2003): Patellar tendon versus quadrupled bonesemitendinosus anterior cruciate ligament reconstruction: A prospective clinical investigation
in athletes. Arthroscopy 19: 592-601.
90
Goradia VK, Rochat MC, Kida M and Grana WA (2000): Natural history of a hamstring
tendon autograft used for anterior cruciate ligament reconstruction in a sheep model. Am J
Sports Med 28: 40-46.
Grana WA, Egle DM, Mahnken R and Goodhart CW (1994): An analysis of autograft
fixation after anterior cruciate ligament reconstruction in a rabbit model. Am J Sports Med
22: 344-351.
Hamner DL, Brown CH Jr, Steiner ME, Hecker AT and Hayes WC (1999): Hamstring
tendon grafts for reconstruction of the anterior cruciate ligament: Biomechanical evaluation
of the use of multiple strands and tensioning techniques. J Bone Joint Surg 81A: 549-557.
Harvey AR, Thomas NP and Amis AA (2003): The effect of screw length and position on
fixation of four-stranded hamstring grafts for anterior cruciate ligament reconstruction. The
Knee 10: 97-102.
Hoffmann RFG, Peine R, Bail HJ, Südkamp NP and Weiler A (1999): Initial fixation
strength of modified patellar tendon grafts for anatomic fixation in anterior cruciate ligament
reconstruction. Arthroscopy 15: 392-399.
Höher J, Livesay GA, Ma BC, Withrow JD, Fu FH and Woo S L-Y (1999): Hamstring graft
motion in the femoral bone tunnel when using titanium button/polyester tape fixation. Knee
Surg Sports Traumatol Arthrosc 7: 215-219.
Höher J, Scheffler SU, Withrow JD, Livesay GA, Debski RE, Fu FH and Woo SL-Y (2000):
Mechanical behavior of two hamstring graft constructs for reconstruction of the anterior
cruciate ligament. J Orthop Res 18: 456-461.
Höher J, Kanamori A, Zeminski J, Fu FH and Woo SL (2001): The position of the tibia
during graft fixation affects knee kinematics and graft forces for anterior cruciate ligament
reconstruction. Am J Sports Med 29: 771-776.
91
Honl M, Carrero V, Hille E, Schneider E and Morlock M (2002): Bone-patellar tendon-bone
grafts for anterior cruciate ligament reconstruction. An in vitro comparison of mechanical
behavior under failure tensile loading and cyclic submaximal tensile loading. Am J Sports
Med 30: 549-557.
Howell SM and Clark JA (1992): Tibial tunnel placement in anterior cruciate ligament
reconstructions and graft impingement. Clin Orthop 283: 187-195.
Hulstyn M, Fadale PD, Abate J and Walsh WR (1993): Biomechanical evaluation of
interference screw fixation in a bovine patellar bone-tendon-bone autograft complex for
anterior cruciate ligament reconstruction. Arthroscopy 9: 417-424.
Hutchinson MR and Bae TS (2001): Reproducibility of anatomic tibial landmarks for
anterior cruciate ligament reconstruction. Am J Sports Med 29: 777-780.
Imhoff AB, Martinek V, Schwamborn and Merl T (1998): Bioabsorbable interference screws
in ACL reconstruction: A prospective clinical and MRI study. Proc Europ Soc Sports
Traumatol, 8th Congress, Nice, France.
Ishibashi Y, Rudy TW, Livesay GA, Stone JD, Fu FH and Woo SL (1997): The effect of
anterior cruciate ligament graft fixation site at the tibia on knee stability: Evaluation using
robotic testing system. Arthroscopy 13: 177-182.
Jackson DW and Gasser SI (1994): Tibial tunnel placement in ACL reconstruction.
Arthroscopy 10: 124-131.
Jacobs JJ, Gilbert JL and Urban RM (1998): Corrosion of metal orthopaedic implants. J Bone
Joint Surg 80A: 268-282.
Jansson KA, Harilainen A, Sandelin J, Karjalainen PT, Aronen HJ and Tallroth K (1999):
Bone tunnel enlargement after anterior cruciate ligament reconstruction with the hamstring
92
autograft and endobutton fixation technique. A clinical, radiographic and magnetic resonance
imaging study with 2 years follow-up. Knee Surg Sports Traumatol Arthrosc 5: 234-238.
Jansson KA, Linko E, Sandelin J and Harilainen A (2003): A prospective randomized study
of patellar versus hamstring tendon autografts for anterior cruciate ligament reconstruction.
Am J Sports Med 31: 12-18.
Jaureguito and Paulos (1996): Why grafts fail. Clin Orthop 325: 25-41.
Johnson RJ, Beynnon BD, Nichols CE and Renström PAFH (1992): The treatment of injuries
of the anterior cruciate ligament. Current concepts review. J Bone Joint Surg 74A: 140-151.
Johnson LL and vanDyk GE (1996): Metal and biodegradable interference screws:
comparison of failure strength. Arthroscopy 12: 452-456.
Jomha NM, Raso VJ and Leung P (1993): Effect of varying angles on the pullout strength of
interference screw fixation. Arthroscopy 9: 580-583.
Kao TJ, Tibone JE and Shaffer B (1995): The pullout strength and use of tibial interference
screws during endoscopic ACL reconstruction surgery. Am J Knee Surg 8: 42-47.
Kartus J, Magnusson L, Stener S, Brandsson S, Eriksson BI and Karlsson J (1998):
Complications following arthroscopic anterior cruciate ligament reconstruction. A follow-up
of 604 patients. Proc Europ Soc Sports Traumatol, 8th Congress, Nice, France.
Kartus J, Movin T and Karlsson J (2001): Current concepts: Donor-site morbidity and
anterior knee problems after anterior cruciate ligament reconstruction using autografts.
Arthroscopy 17: 971-980.
Kocabey Y, Klein S, Nyland J and Caborn D (2004): Tibial fixation comparison of
semitendinosus-bone composite allografts fixed with bioabsorbable screws and bone-patella
93
tendon-bone grafts fixed with titanium screws. Knee Surg Sports Traumatol Arthrosc 12: 8893
Kohn D and Rose C (1994): Primary stability of interference screw fixation: Influence of
screw diameter and insertion torque. Am J Sports Med 22: 334-338.
Kurosaka M, Yoshiya S and Andrish JT (1987): A biomechanical comparison of different
surgical techniques of graft fixation in anterior cruciate ligament reconstruction. Am J Sports
Med 15: 225-229.
Kurzweil PR, Frogameni AD and Jackson DW (1995): Tibial interference screw remowal
following anterior cruciate ligament reconstruction. Arthroscopy 11: 289-291.
Lambert KL (1983): Vascularized patellar tendon graft with rigid internal fixation for
anterior cruciate ligament insufficiency. Clin Orthop 172: 85-89.
Lee MC, Jo H, Bae T-S, Jang JD and Seong SC (2003): Analysis of initial fixation strength
of press-fit fixation technique in anterior cruciate ligament reconstruction. A comparative
study with titanium and bioabsorbable interference screw using porcine lower limb. Knee
Surg Sports Traumatol Arthrosc 11: 91-98.
Lemos MJ, Jackson DW, Lee TQ and Simon TM (1995): Assessment of initial fixation of
endoscopic interference femoral screws with divergent and parallel placement. Arthroscopy
11: 37-41.
L'Insalata JC, Klatt B, Fu FH and Harner CD (1997): Tunnel expansion following anterior
cruciate ligament reconstruction: a comparison of hamstring and patellar tendon autografts.
Knee Surg Sports Traumatol Arthrosc 5: 234-238.
Liu SH, Kabo JM and Osti L (1995): Biomechanics of two types of bone-tendon-bone graft
for ACL reconstruction. J Bone Joint Surg 77B:232-235.
94
Macdonald P and Arneja S (2003): Biodegradable screw presents as a loose intra-articular
body after anterior cruciate ligament reconstruction. Arthroscopy 19: 22-24.
Magen HE, Howell SM and Hull ML (1999): Structural properties of six tibial fixation
methods for anterior cruciate ligament soft tissue grafts. Am J Sports Med 27: 35-43.
Marden RA, Raskind JR and Carroll M (1991): Prospective evaluation of arthroscopically
assisted anterior cruciate ligament reconstruction. Patellar tendon versus semitendinosus and
gracilis tendons. Am J Sports Med 19: 478-484.
Markolf
KL, Burchfield DM and Shapiro (1996): 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 78A: 17281734.
Markolf KL, Hame SL, Hunter DM, Oakes D and Gause P (2002): Biomechanical effects of
femoral notchplasty in anterior cruciate ligament reconstruction. Am J Sports Med 30: 83-89.
Marti C, Imhoff AB, Bahrs C and Romero (1997): Metallic versus bioabsorbable interference
screw for fixation of bone-patellar tendon-bone autograft in arthroscopic anterior cruciate
ligament reconstruction. A preliminary report. Knee Surg Sports Traumatol Arthrosc 5: 217221.
Martin SD, Martin TL and Brown CH (2002): Anterior cruciate ligament graft fixation.
Orthop Clin N Am 33: 685-696.
Martinek V and Friedrich N (1999): Tibial and pretibial cyst formation after anterior cruciate
ligament reconstruction with bioabsorbable interference screw fixation. Arthroscopy 15: 317320.
Matthews LS and Soffer SR (1989): Pitfalls in the use of interference screws for anterior
cruciate ligament reconstruction: Brief report. Arthroscopy 5: 225-226.
95
Matthews LS, Lawrence SJ, Yahiro MA and Siclair MR (1993): Fixation strengths of patellar
tendon-bone grafts. Arthroscopy 9: 76-81.
Matthews LS, Parks BG and Sabbagh RC (1998): Determination of fixation strength of largediameter interference screws. Arthroscopy 14: 70-74.
McCalden RW, McGeough JA, Barker MB and Court-Brown CM (1993): Age-related
changes in the tensile properties of cortical bone. The relative importance of changes in
porosity, mineralization, and microstructure. J Bone Joint Surg 75A: 1193-1205.
McGuire DA, Barber FA, Elrod BF and Paulos LE (1999): Bioabsorbable interference
screws for graft fixation in anterior cruciate ligament reconstruction. Arthroscopy 15: 463473.
Miller CM, Tibone JE, Hewitt M, Kharrazi FD and Elattrache NS (2002): Interference screw
divergence in femoral tunnel fixation during endoscopic anterior cruciate ligament
reconstruction using hamstring grafts. Arthroscopy 18: 510-514.
Miller LS and Gladstone JN (2002): Graft selection in anterior cruciate ligament
reconstruction. Orthop Clin N Am 33: 675-683.
Miyasaka KC, Daniel DM, Stone ML and Hirshman P (1991): The incidence of knee
ligament injuries in the general population. Am J Knee Surg 4: 3-8.
Miyata K, Yasuda K, Kondo E, Nakano H, Kimura S and Hara N (2000): Biomechanical
comparison of anterior cruciate ligament: reconstruction procedures with flexor tendon graft.
J Orthop Sci 5: 585-592.
Moholkar K, Taylor D, O'Reagan M and Fenelon G (2002): A biomechanical analysis of four
different methods of harvesting bone-patellar tendon-bone graft in porcine knees. J Bone
96
Joint Surg 84A: 1782-1787.
Morgan CD, Kalman VR and Grawl DM (1995): Isometry testing for anterior cruciate
ligament reconstruction revisited. Arthroscopy 11: 647-659.
Musahl V, Abramowitch SD, Gabriel MT, Debski RE, Hertel P, Fu FH and Woo SL-Y
(2003): Tensile properties of an anterior cruciate ligament graft after bone-patellar tendonbone press-fit fixation. Knee Surg Sports Traumatol Arthrosc 11: 68-74.
Nagarkatti DG, McKeon BP, Donahue BS and Fulkerson JP (2001): Mechanical evaluation
of a soft tissue interference screw in free tendon anterior cruciate ligament graft fixation. Am
J Sports Med 29: 67-71.
Nakano H, Yasuda K, Tohyama H, Yamanaka M, Wada T and Kaneda K (2000):
Interference screw fixation of doubled flexor tendon graft in anterior cruciate ligament
reconstruction - biomechanical evaluation with cyclic elongation. Clin Biomech 15: 188-195.
Nebelung W, Becker R, Merkel M and Ropke M (1998): Bone tunnel enlargement after
anterior cruciate ligament reconstruction with semitendinosus tendon using endobutton
fixation on the femoral side: Arthroscopy 14: 9-14.
Nebelung W, Becker R, Urbach D, Ropke M and Roessner A (2003) Histological findings of
tendon-bone healing following anterior cruciate ligament reconstruction with hamstring
grafts. Arch Orthop Trauma 123: 158-163.
Noyes FR, Butler DL, Grood ES, Zernicke RF and Hefzy MS (1984): Biomechanical
analysis of human ligament grafts used in knee-ligament repairs and reconstructions. J Bone
Joint Surg 66A: 344-352.
97
Novak PJ, Wexler GM, Williams JS Jr, Bach BR Jr and Bush-Joseph CA (1996):
Comparison of screw post fixation and free bone block interference fixation for anterior
cruciate ligament soft tissue grafts: Biomechanical considerations. Arthroscopy 12: 470-473.
Numazaki H, Tohyama H, Nakano H, Kikuchi S and Yasuda K (2002): The effect of initial
graft tension in anterior cruciate ligament reconstruction on the mechanical behaviors of the
femur-graft-tibia complex during cyclic loading. Am J Sports Med 30: 800-805.
Nurmi JT, Järvinen TLN, Kannus P, Sievänen H, Toukosalo J and Järvinen M (2002):
Compaction versus extraction drilling for fixation of the hamstring graft in ACL
reconstruction. Am J Sports Med 30: 167-173.
Nurmi JT, Kannus P, Sievänen H, Järvinen M and Järvinen TLN (2003): Compaction drilling
does not increase the initial fixation strength of the hamstring tendon graft in anterior cruciate
ligament reconstruction in a cadaveric model. Am J Sports Med 31: 353-358.
Nurmi JT, Kannus P, Sievänen H, Järvelä T, Järvinen M and Järvinen TLN (2004a):
Interference screw fixation of soft tissue grafts in anterior cruciate ligament reconstruction.
Part I: Effect of tunnel compaction by serial dilators versus extraction drilling on the initial
fixation strength. Am J Sports Med 32: 411-417.
Nurmi JT, Kannus P, Sievänen H, Järvelä T, Järvinen M and Järvinen TLN (2004b):
Interference screw fixation of soft tissue grafts in anterior cruciate ligament reconstruction.
Part II: Effect of preconditioning on graft tension during and after screw insertion. Am J
Sports Med 32: 418-424.
Nurmi JT, Sievänen H, Kannus P, Järvinen M and Järvinen TLN (2004c): Porcine tibia is a
poor substitute for human cadaver tibia for evaluating interference screw fixation. Am J
Sports Med 32: In Press.
98
Olszewski AD, Miller MD and Ritchie JR (1998): Ideal tibial tunnel length for endoscopic
anterior cruciate ligament reconstruction. Arthroscopy 15: 115-117.
O’Neil DB (1996): Arthroscopically assisted reconstruction of the anterior cruciate ligament.
A prospective randomized analysis of three techniques. J Bone Joint Surg 78A: 803-813.
Otero AL and Hutcheson L (1993): A comparison of the doubled semitendinosus/gracilis and
central third of the patellar tendon in arthroscopic anterior cruciate ligament reconstruction.
Arthroscopy 9: 143-148.
Panni AS, Milano G, Lucania L and Fabbriciani C (1997): Graft healing after anterior
cruciate ligament reconstruction in rabbits. Clin Orthop 343: 203-212.
Papageorgiou CD, Ma CB, Abramowitch SD, Clineff TD and Woo SL-Y (2001): A
multidisciplinary study of the healing of an intraarticular anterior cruciate ligament graft in a
goat model. Am J Sports Med 29: 620-626.
Paschal SO, Seeman MD, Ashman RB, Allard RN and Mongomery JB (1994): Interference
fixation versus postfixation of bone-patellar tendon-bone grafts for anterior cruciate ligament
reconstruction: A biomechanical comparative study in porcine knees. Clin Orthop 300: 281287.
Pelker RR, Friedlaender G, Markham TC, Panjabi MM and Moen CJ (1984): Effects of
freezing and freeze-drying on the biomechanical properties of rat bone: J Orthop Res 1: 405411.
Pena F, Grøntvedt T, Brown GA, Aune AK and Engebretsen L (1996): Comparison of failure
strength between metallic and absorbable interference screws: Influence of insertion torque,
tunnel-bone block gap, bone mineral density, and interference. Am J Sports Med 24: 329334.
99
Pierz K, Baltz M and Fulkerson J (1995): The effect of Kurosaka screw divergence on the
holding strength of bone-tendon-bone grafts. Am J Sports Med 23: 332-335.
Pinczewski LA, Deehan DJ, Salmon LJ, Russell VJ and Clingeleffer A (2002): A five-year
comparison of patellar tendon versus four-strand hamstring tendon autograft for arthroscopic
reconstruction of the anterior cruciate ligament. Am J Sports Med 30: 523-536.
Pohjonen T and Törmälä P (1996): In vitro hydrolysis of self-reinforced polylactide
composites. Med Biol Eng Compu 34 (Suppl 1, Part 1): 127-128.
Pomeroy G, Baltz M, Pierz K, Nowak M, Post W and Fulkerson JP (1998): The effects of
bone plug length and screw diameter on the holding strength of bone-tendon-bone grafts.
Arthroscopy 14: 148-152.
Reznik AM, Davis JL and Daniel DM (1990): Optimizing interference fixation for cruciate
ligament reconstruction. Trans Orthop Res Soc 15: 519.
Rittmeister ME, Noble PC, Bocell JR Jr, Alexander JW, Conditt MA and Kohl HW III
(2001): Interactive effects of tunnel dilation on the mechanical properties of hamstring grafts
fixed in the tibia with interference screws. Knee Surg Sports Traumatol Arthrosc 9: 267-271.
Rittmeister ME, Noble PC, Bocell JR Jr, Alexander JW, Conditt MA and Kohl HW III
(2002): Components of laxity in interference fit fixation of quadrupled hamstring grafts. Acta
Orthop Scand 73: 65-71.
Robertson DB, Daniel DM and Biden E (1986): Soft tissue fixation to bone. Am J Sports
Med 14: 398-403.
Rodeo SA, Arnoczky SP, Torzilli PA, Hidaka C and Warren RF (1993): Tendon-healing in a
bone tunnel. A biomechanical and histological study in the dog. J Bone Joint Surg 75A:
1795-1803.
100
Rokkanen P, Böstman O, Vainionpää K, Vihtonen K, Törmälä P, Laiho J, Kilpikari J and
Tamminmäki M (1985): Biodegradable implants in fracture fixation: Early results of
treatment of fractures of the ankle. Lancet: 1422-1424.
Roos H and Karlsson J (1998): Anterior cruciate ligament instability and reconstruction.
Review of current trends in treatment. Scand J Med Sci Sports 8: 426-431.
Rowden NJ, Sher D, Rogers GJ and Schindhelm K (1997): Anterior cruciate ligament graft
fixation. Initial comparison of patellar tendon and semitendinosus autografts in young fresh
cadavers. Am J Sports Med 25: 472-478.
Ruluff D and McIntyre PE (1982): Some current uses for metals in, on and around your
body. Mech Eng 4: 40-47.
Rupp S, Krauss PW and Fritsch EW (1997): Fixation strength of a biodegradable interference
screw and a press-fit technique in anterior cruciate ligament reconstruction with a BPTB
graft. Arthroscopy 13: 61-65.
Rupp S, Seil R, Krauß PW and Kohn DM (1998): Cortical versus cancellous interference
fixation for bone-patellar tendon-bone grafts. Arthroscopy 14: 484-488.
Rupp S, Hopf T, Hess T, Seil R and Kohn DM (1999a): Resulting tensile forces in the human
bone-patellar tendon-bone graft: Direct force measurements in vitro. Arthroscopy 15: 179184.
Rupp S, Seil R, Schneider A and Kohn DM (1999b): Ligament graft initial fixation strength
using biodegradable interference screws. J Biomed Mater Res (Appl Biomater) 48: 70-74.
Sachs RA, Daniel DM, Stone ML and Garfein RF (1989): Patellofemoral problems after
anterior cruciate ligament reconstruction. Am J Sports Med 17: 760-765.
101
Safran and Harner (1996): Technical considerations of revision anterior cruciate ligament
surgery. Clin Orthop 323: 50-64.
Sakane M, Fox RJ, Woo SL-Y, Livesay GA, Li G and Fu FH (1997): In situ forces in the
anterior cruciate ligament and its bundles in response to anterior tibial loads. J Orthop Res
15: 285-293.
Scranton PE, Lanzer WL, Ferguson MS, Kirkman TR and Pflaster DS (1998): Mechanisms
of anterior cruciate ligament neovascularization and ligamentization. Arthroscopy 14: 702716.
Schatzmann L, Brunner P and Stäubli HU (1998): Effect of cyclic preconditioning on the
tensile properties of human quadriceps tendons and patellar ligaments. Knee Surg Sports
Traumatol Arthrosc 6 (Suppl 1): 56-61.
Scheffler SU, Südkamp NP, Göckenjan A, Hoffmann RFG and Weiler A (2002):
Biomechanical comparison of hamstring and patellar tendon graft anterior cruciate ligament
reconstruction techniques: The impact of fixation level and fixation method under cyclic
loading. Arthroscopy 18: 304-315.
Seil R, Rupp S, Krauss PW, Benz A and Kohn DM (1998): Comparison of initial fixation
strength between biodegradable and metallic interference screws and a press-fit fixation
technique in a porcine model. Am J Sports Med 26: 815-819.
Seitz H, Vécsei V, Menth-Chiari WA, Pichl W, Wielke B and Marlovits S (1999):
Comparison of femoral and tibial pullout forces in bone-patellar tendon-bone anterior
cruciate ligament reconstructions with a new interference fixation device. Arthroscopy 15:
173-178.
102
Selby JB, Johnson DL, Hester P and Caborn DNM (2001): Effect of screw length on
bioabsorbable interference screw fixation in a tibial bone tunnel. Am J Sports Med 29: 614619
Shaffer B, Gow W and Tibone JE (1993): Graft-tunnel mismatch in endoscopic anterior
cruciate ligament reconstruction: A new technique of intraarticular measurement and
modified graft harvesting. Arthroscopy 9: 633-646.
Shaieb MD, Kan DM, Chang SK, Marumoto JM and Richardson AB (2002): A prospective
randomized comparison of patellar tendon versus semitendinosus and gracilis tendon
autografts for anterior cruciate ligament reconstruction. Am J Sports Med 30: 214-220.
Shapiro JD, Cohn BT, Jackson DW, Postak PD, Parker RD and Greenwald AS (1992): The
biomechanical effects of geometric configuration of bone-tendon-bone autografts in anterior
cruciate ligament reconstruction. Arthroscopy 8: 453-458.
Shapiro JD, Jackson DW, Aberman HM, Lee TQ and Simon TM (1995): Comparison of
pullout strength for seven- and nine-millimeter diameter interference screw size as used in
anterior cruciate ligament reconstruction. Arthroscopy 11: 596-599.
Shelbourne KD and Trumper RV (1997): Preventing anterior knee pain after anterior cruciate
ligament reconstruction. Am J Sports Med 25: 41-47.
Shelbourne KD and Nitz (1990): Accelerated rehabilitation after anterior cruciate ligament
reconstruction. Am J Sports Med 18: 292-299.
Shellock FG, Mink JH, Curtin S and Friedman MJ (1992): MR imaging and metallic
implants for anterior cruciate ligament reconstruction: assessment of ferromagnetism and
artifact. J Magn Reson Imaging 2: 225-228.
103
Shino K and Pflaster DS (2000): Comparison of eccentric and concentric screw placement
for hamstring graft fixation in the tibial tunnel. Knee Surg Sports Traumatol Arthrosc 8: 7375.
Simonian PT, Sussmann PS, Baldini TH, Crockett HC and Wickiewicz TL (1998):
Interference screw position and hamstring graft location for anterior cruciate ligament
reconstruction. Arthroscopy 14: 459-464.
Singhatat W, Lawhorn KW, Howell SM and Hull ML (2002): How four weeks of
implantation affect the strength and stiffness of a tendon graft in a bone tunnel: a study of
two fixation devices in an extraarticular model in ovine. Am J Sports Med 30: 506-513.
Stadelmaier DM, Lowe WR, Ilahi OA, Noble PC and Kohl HW III (1999): Cyclic pull-out
strength of hamstring tendon graft fixation with soft tissue interference screws. Influence of
screw length. Am J Sports Med 27: 778-783.
Stähelin A, Weiler A, Rüftnacht H, Hoffmann R, Geissmann A and Feinstein R (1997):
Clinical degradation and biocompatibility of different bioabsorbable interference screws: A
report of six cases. Arthroscopy 13: 238-244.
Stapelton TR, Waldrop JI, Ruder CR, Parrish TA and Kuivila TE (1998): Graft fixation
strength with arthroscopic anterior cruciate ligament reconstruction. Two-incision rear entry
technique compared with one-incision technique. Am J Sports Med 26: 442-445.
Starch DW, Alexander JW, Noble PC, Reddy S and Litner DM (2003): Multistranded
hamstring tendon graft fixation with a central four-quadrant or a standard tibial interference
screw for anterior cruciate ligament reconstruction. Am J Sports Med 31: 338-344.
Steenlage E, Brand JC, Johnson DL and Caborn DNM (2002): Correlation of bone tunnel
diameter with quadrupled hamstring graft fixation strength using a biodegradable
interference screw. Arthroscopy 18: 901-907.
104
Stein DA, Hunt SA, Rosen JE and Sherman OH (2002): The incidence and outcome of
patella fractures after anterior cruciate ligament reconstruction. Arthroscopy 18: 578-583.
Steiner ME, Hecker AT, Brown CH Jr and Hayes WC (1994): Anterior cruciate ligament
graft fixation: comparison of hamstring and patellar tendon grafts. Am J Sports Med 22: 240247.
Suh JS, Jeong EK, Shin KH, Cho JH, Na J-B, Kim D-H and Han C-D (1998): Minimizing
artifacts caused by metallic implants at MR imaging: Experimental and clinical studies. Am J
Roentgenol 171: 1207-1213.
Swenson TM and Harner CD (1995): Knee ligament and meniscal injuries: Current concepts.
Orthop Clin North Am 26: 529-546.
Taylor DE, Dervin GF, Keene GCR (1996): Femoral bone block recession in endoscopic
anterior cruciate ligament reconstruction. Technical note. Arthroscopy 12: 513-515.
To JT, Howell SM and Hull ML (1999): Contributions of femoral fixation methods to the
stiffness of anterior cruciate ligament replacements at implantation. Arthroscopy 15: 379387.
Toutoungi DE, Lu TW, Leardini A, Ctani F and O'Connor JJ (2000): Cruciate ligament
forces in the human knee during rehabilitation exercises. Clin Biomech 15: 176-187.
Tsuda E, Fukuda Y, Loh JC, Debski RE, Fu FH and Woo SL (2002): The effect of soft-tissue
graft fixation in anterior cruciate ligament reconstruction on graft-tunnel motion under
anterior tibial loading. Arthroscopy 18: 960-967.
105
Weiler A, Hoffman RFG, Stähelin AC, Bail HJ, Siepe CJ and Südkamp NP (1998a):
Hamstring tendon fixation using interference screws: a biomechanical study in calf tibial
bone. Arthroscopy 14: 29-37.
Weiler A, Windhagen HJ, Raschke MJ, Laumeyer A and Hoffmann RFG (1998b):
Biodegradable interference screw fixation exhibits pull-out force and stiffness similar to
titanium screws. Am J Sports Med 26: 119-128.
Weiler A, Hoffman RFG, Siepe CJ, Kolbeck SF and Südkamp NP (2000): The influence of
screw geometry on hamstring tendon interference fit fixation. Am J Sports Med 28: 356-359.
Weiler A, Richter M, Schmidmaier G, Kandziora F and Südkamp NP (2001): The EndoPearl
device increases fixation strength and eliminates construct slippage of hamstring tendon
grafts with interference screw fixation. Arthroscopy 17: 353-359.
Weiler A, Hoffmann RFG, Bail HJ, Rehm O and Südkamp NP (2002a): Tendon healing in
bone tunnel. Part II: Histologic analysis after biodegradable interference fit fixation in a
model of anterior cruciate ligament reconstruction in sheep. Arthroscopy 18: 124-135.
Weiler A, Peine R, Pashmineh-Azar A, Abel C, Südkamp NP and Hoffmann RFG (2002b):
Tendon healing in a bone tunnel. Part I: Biomechanical results sfter biodegradable
interference fit fixation in a sheep model of anterior cruciate ligament reconstruction in
sheep. Arthroscopy 18:113-123.
Wilson TW, Zafuta MP and Zobitz (1999): A biomechanical analysis of matched bonepatellar tendon-bone and double-looped semitendinosus and gracilis tendon grafts. Am J
Sports Med 27: 202-207.
Woo SL-Y, Orlando CA, Camp JF and Akeson WH (1986): Effects of post-mortem storage
by freezing on ligaments: A comparative study with two existing methods. J Biomech 19:
399- 404.
106
Woo SL, Hollis JM, Adams DJ; Lyon RM and Takai S (1991): Tensile properties of the
human femur-anterior cruciate ligament-tibia complex. The effects of specimen age and
orientation. Am J Sports Med 19: 217-225.
Woo SL-Y, Debski RE, Withrow JD and Janaushek MA (1999): Biomechanics of knee
ligaments. Current concepts. Am J Sports Med 27: 533-543.
Woo SL, Kanamori A, Zeminski J, Yagi M, Papageorgiou C and Fu FH (2002): the
effectiveness of reconstruction of the anterior cruciate ligament with hamstrings and patellar
tendon. A cadaveric study comparing anterior tibial and rotational loads. J Bone Joint Surg
84A: 907-914.
Vuori I, Heinonen A, Sievänen H, Kannus P, Pasanen M and Oja P (1994): Effects of
unilateral strength training and detraining on bone mineral density and content in young
women. A study of mechanical loading and deloading on human bones. Calcif Tissue Int 55:
59-67.
Yagi M, Wong EK, Kanamori A, Debski RE, Fu FH and Woo SL-Y (2002): Biomechanical
analysis of an anatomic anterior cruciate ligament reconstruction. Am J Sports Med 30: 660666.
Yamakado K, Kitaoka K, Yamada H, Hashiba K, Nakamura R and Tomita K (2002): The
influence of mechanical stress on graft healing in a bone tunnel. Arthroscopy 18: 82-90.
Yamanaka M, Yasuda K, Tohyama H, Nakano H and Wada T (1999): The effect of cyclic
displacement on the biomechanical characteristics of anterior cruciate ligament
reconstructions. Am J Sports Med 27: 772-777.
107
Yasuda K, Tsujino J, Tanabe Y and Kaneda K (1997): Effects of initial graft tension on
clinical outcome after anterior cruciate ligament reconstruction: Autogenous doubled
hamstring tendons connected in series with polyester tapes. Am J Sports Med 25: 99-106.
Yoshiya S, Andrish JT, Manley MT and Bauer TW (1987): Graft tension in anterior cruciate
ligament reconstruction. An in vivo study in dogs. Am J Sports Med 15: 464-470.
Yoshiya S, Nagano M, Kurosaka M, Muratsu H and Mizuno K (2000): Graft heling in the
bone tunnel in anterior cruciate ligament reconstruction. Clin Orthop 376: 278-286.
Yoshiya S, Kurosaka M, Ouchi K, Kuroda R and Mizuno K (2002): Graft tension and knee
stability after anterior cruciate ligament recontruction. Clin Orthop 394: 154-160.
Yunes M, Richmond JC, Engels EA and Pinczewski LA (2001): Patellar tendon versus
hamstring tendons in anterior cruciate ligament reconstruction: A meta-analysis. Arthroscopy
17: 248-257.
Zeminski J, Yagi M, Kanamori A, Fu FH and Woo SL-Y (2000): High initial graft tension
has no apparent benefit on the biomechanical outcome of ACL reconstruction. Presented at
the 46th annual meeting of the Orthopaedic Research Society, Orlando, FL: 484.
108
ORIGINAL PUBLICATIONS
109