Characterization of Cruciate Ligament Impingement: The Influence

Characterization of Cruciate Ligament Impingement: The
Influence of Femoral or Tibial Tunnel Positioning at Different
Degrees of Knee Flexion
Diego Costa Astur, M.D., Ciro Veronese Santos, M.D., Vinicius Aleluia, M.D.,
Nelson Astur Neto, M.D., Gustavo Gonçalves Arliani, M.D., Camila Cohen Kaleka, M.D.,
Abdalla Skaf, M.D., and Moises Cohen, M.D., Ph.D
Purpose: We aimed to analyze how different positions of the tibial and femoral tunnels when used for anterior cruciate
ligament (ACL) reconstruction affect relations with the posterior cruciate ligament (PCL) at different degrees of knee
flexion. Information gained from this study may be helpful in determining optimal placement of the graft in ACL
reconstructive surgery. Methods: We divided 10 cadaveric knees into 2 groups of 5 and had either their femoral or tibial
ACL insertion detached. For each specimen, 16 different positions were reproduced during ACL reconstruction based on
a combination of 4 different tunnels in the tibia for group A (anterior-medial, anterior-lateral, posterior-medial, and
posterior-lateral) and 4 in the femur for group B (anterior-proximal, anterior-distal, posterior-proximal, and posteriordistal) with 4 of knee flexion for each (0 , 45 , 90 , and 135 ). We performed a magnetic resonance imaging (MRI) study
for each configuration and analyzed the cruciate ligament positioning. Results: We identified 3 different situations: no
contact between cruciate ligaments, contact without deformity, and contact with deformity. In group A, the degree of
flexion (P ¼ .003) and ligament insertion positioned in the posterior quadrants (P < .05) were statistically significant for
the presence of ACL impingement. Ligament contact with deformity was identified in 18 (22.5%) configurations, mostly
when the knee was flexed 45 and 90 and the ACL was in the posterior quadrants. For group B, “contact with deformity”
was identified in 23 MR images, mostly (12 cases) with the graft position being in the anterior-distal configuration, but it
was not significant for the occurrence of cruciate impingement. Conclusions: Impingement with ligament deformity is
greater when the graft is fixed at the posterior quadrants of the tibial footprint, regardless of the degree of knee flexion.
Although quite common, the ligament position in the femoral footprint was not a primary cause of ACL impingement
with deformity. Clinical Relevance: This study helps identify positions of the tibial or femoral tunnels during ACL
reconstruction to avoid impingement between cruciate ligaments.
T
he anterior cruciate ligament (ACL) is an important
structure in the knee because of both its anterior
and rotational stabilization properties. Proximally, it
From Departamento de Ortopedia e Traumatologia (D.C.A., G.G.A., M.C.)
da Escola Paulista de Medicina/UNIFESP, São Paulo; Department of
Orthopaedic Surgery (C.V.S., V.A.), Instituto Cohen, São Paulo; Department
of Orthopaedic Surgery (C.C.K.) Faculdade de Medicina da Santa Casa de
Misericórdia de São Paulo, São Paulo; Department of Radiology (A.S.), Alta
Institute, São Paulo, SP, Brazil; and Department of Orthopaedic Surgery
(N.A.N.), University of Tennessee, Memphis, TN, U.S.A.
The authors report that they have no conflicts of interest in the authorship
and publication of this article.
Received August 16, 2012; accepted January 3, 2013.
Address correspondence to Diego Costa Astur, M.D., Departamento de
Ortopedia e Traumatologia, Universidade Federal de São Paulo, Rua Borges
Lagoa 837, 5 andar, Vila Clementino, São Paulo, SP, Brazil. E-mail:
[email protected]
Ó 2013 by the Arthroscopy Association of North America
0749-8063/12541/$36.00
http://dx.doi.org/10.1016/j.arthro.2013.01.008
inserts on the medial aspect of the lateral femoral
condyle; distally, it inserts anteriorly and laterally to the
medial tibial spine.1-4
Anatomic reconstruction of the ACL is of paramount
importance to achieve optimal clinical outcomes after
surgery. The position of the tibial and femoral tunnels
for ACL reconstruction is well documented and is
one of the most important factors determining clinical
outcome.5 Anatomic positioning of the tunnel decreases
the incidence of complications, such as restriction of
knee motion, graft failure, or graft impingement against
the posterior cruciate ligament (PCL) or lateral femoral
condyle.
ACL impingement on the PCL has not been extensively studied. Previous studies have shown that contact
between the different ligamentous structures in the
knee might have a potential impact on joint pain,
decreased flexion, and increased instability.6,7 Thus, the
correct positioning of the tibial and femoral tunnels in
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol
-,
No
-
(Month), 2013: pp 1-7
1
2
D. C. ASTUR ET AL.
the coronal and sagittal planes in the ACL footprint is
necessary to achieve optimal results. Small variations in
these tunnels can affect the position of the graft and
result in impingement on the PCL.
The purpose of this study was to analyze how
different positions of both the tibial and femoral tunnels
in ACL reconstruction affect the relations with the PCL
at different degrees of knee flexion. Information gained
from this study may be helpful for defining optimal
placement of the graft during ACL reconstruction. We
hypothesized that small variations in these tunnels can
affect the position of the graft and result in impingement on the PCL.
Methods
After institutional review board approval, we dissected
10 human cadaveric knees in a cadaver laboratory. They
were between 27 and 43 years old without previous
ligament injuries or pre-existing arthritis.
We then took a medial parapatellar approach in each
specimen to access the ACL without affecting the
stability of the knee joint. We then used a surgical
marking pen to demarcate the femoral and tibial ACL
footprints before intentionally detaching the ACL at its
insertion on the femur or tibia, keeping its respective
counterpoint insertion intact.
We divided the specimens into 2 groups of 5 knees
each based on where the ACLs were detached from
their insertiondfrom the ACL tibial footprint (group
A) or from the ACL femoral footprint (group B). We
stitched the free end of the ACL with a No. 5 Ethibond
wire (Ethicon, Somerville, NJ) to make its manipulation and relocation in the desired position easier
(Fig 1).
We then divided the previously marked footprint of
the respective detached insertion of the ACL into 4
quadrants within the native ACL footprint for groups
A and Bdgroup A: anterior-medial, anterior-lateral,
posterior-medial, and posterior-lateral; group B: anteriorproximal, anterior-distal, posterior-proximal, and posteriordistal (Fig 2A).
For each quadrant in groups A and B, we drilled
a surgical tunnel in a standard fashion (ACL anatomic
reconstruction guides) with a 2-mm guidewire. We
placed an ACL tibial aimer set at 55 in the center of
each quadrant of the ACL tibial insertion site previously
marked. The position of the aimer on the tibial cortex
was between 3 and 6 cm medial to the tibial tubercle.
However, the tunnels were not drilled with the use of
a reamer. We just drilled the tunnels with the 2-mm
guidewire, avoiding bone fragility because of the fact
that all 4 tunnels were too close to each other. Next, we
turned attention to the femoral tunnels. We placed the
femoral tunnel in the center of each previously marked
insertion site and drilled each tunnel without an ACL
femoral aimer and with 90 of knee flexion.
Fig 1. Anterior cruciate ligament detached from the femoral
footprint stitched with an Ethibond suture wire to reposition
the ligament as desired.
Therefore, each knee in group A had 4 tibial tunnels
and each knee in group B had 4 femoral tunnels (Fig
2B). To place the free end of the detached ACL in the
desired position, we stitched the Ethibond wire to the
guidewire and pulled it through the respective tunnel.
We then fixed the Ethibond wire in the tibial shaft for
group A and in the femoral shaft for group B in a way
that kept the ACL tensioned (a bone tunnel perpendicular to the tibial and femoral shaft fixing the Ethibond wire to an EndoButton [Smith & Nephew
Endoscopy, Andover, MA] in the contralateral cortex).
After placing the free end of the ligament in the
desired quadrant in both groups, we created a polyvinyl
chloride pipe model to sustain the desired angle of knee
flexion (0 , 45 , 90 , or 135 ) (Fig 3). We performed
magnetic resonance imaging (MRI) for each specimen
and for every ligament positioning in the quadrants.
We analyzed the distance between the ACL and PCL
and measured, in millimeters, the coronal, sagittal, and
oblique axial images after MRI was performed for every
quadrant situation and degree of knee flexion
described. We obtained T2-weighted and 1-mm-slice
MRI for each specimen and for each of their 16 possible
graft location/knee flexion combinations. We used the
major point of contact between the ACL and the PCL to
CHARACTERIZATION OF CRUCIATE LIGAMENT IMPINGEMENT
3
Fig 2. (A) Four different quadrants
inside the femoral and tibial footprints:
anterior-proximal tunnel (AP), anteriordistal tunnel (AD), posterior-proximal
tunnel (PP), posterior-distal tunnel (PD),
anterior-medial tunnel (AM), anteriorlateral tunnel (AL), posterior-medial
tunnel (PM), and posterior-lateral tunnel
(PL). (B) Different tunnel configurations
in the femur and tibia.
determine the distance between them. This is the
midpoint between proximal and distal insertions, which
corresponds to the MRI sagittal cut immediately below
the Blumensaat line. We used the oblique axial MRI
view to measure the distance between the ligaments or
the presence of deformity.
As a control group, we submitted 5 intact knees of
healthy individuals without ligament injuries or preexisting arthritis to MRI studies to determine the presence of contact between the cruciate ligaments.
established as 20% (power of 80%) to find at least
a 50% difference between different positions.
We used an exchangeable correlation matrix for all
analyses performed. To compare the presence of
deformity in different positions and degrees of flexion,
we used generalized estimating equations with logit
link for binary data. To determine the greater possibility
of impingement between the ligaments, we used the
Bonferroni multiple comparison method (P < .05).
Statistical Analyses
We performed sample size estimates before choosing
cadaveric knees. A type I error was pre-established as
5% (95% confidence interval), and a type II error was
Using the 10 knee specimens, we created 32 different
combinations considering 4 possible insertions in the
femoral ACL footprint, 4 possible insertions in the tibial
footprint, and 4 different grades of knee flexion. After
Results
Fig 3. Polyvinyl chloride pipe construct for fixed flexion positions of the knee. (A) Neutral; (B) 45 ; (C) 90 ; (D) 135 .
4
D. C. ASTUR ET AL.
performing a total of 160 MRI studies to examine every
specimen in every combination, we identified 3 relation
patterns between the ACL graft and the PCL: no contact
between the cruciate ligaments, contact without
deformity of the ligaments, and contact with deformity
(Fig 4).
The specimens that had their tibial ACL insertion
detached (group A) had 57 (71.25%) situations with
ligament contact but no deformity, most of them
(61.4%) when the graft was placed in the anterior
quadrants. Ligament contact with deformity was identified in 18 (22.5%) configurations, mostly when the
knee was flexed 45 and 90 and the ACL was in the
posterior quadrants (7 in the posteromedial and 11 in
the posterolateral quadrants). Finally, no contact was
seen in 5 (6.35%) MR images; 3 (60%) of these were in
45 of knee flexion. The distance between the ligaments
ranged from 1 to 2 mm.
In group B, when the femoral ACL insertion was
detached, “no contact” was identified in 32 (40%)
MRI studies, 19 (60%) of which were in the anteriorproximal quadrant in all different degrees of knee
flexion. The distance varied between 0.5 and 2.5 mm.
Contact without deformity was seen in 25 (31.25%) situations, most of them (56%) in 90 and 135 of flexion.
Contact with deformity was identified in 23 (28.75%) MR
images, mostly (53%) in the anterior-distal configuration. Although contact with deformity was more common at the femur than at the tibia, there were no
statistically significant results that would favor the
presence of impingement caused by a specific quadrant
of the femoral footprint.
In group A, the degree of flexion (P ¼ .003) was
statistically more significant for the presence of ACL
impingement than for the position of the ligament at
the tibial insertion (P ¼ .08). Knee positions of both
0 and 135 of flexion showed ligament deformity more
frequently than did the intermediate angles (P < .05),
and this deformity was more frequent when the graft
was placed at the posterior quadrants of the tibial
footprint (P < .05).
For group B, the graft position was not statistically
significant for the occurrence of ACL impingement on
the PCL. Contact between ligaments occurred more at
45 of flexion than at other grades of flexion (P < .05).
At 90 of flexion, deformity was more common than at
135 of flexion (P < .05). There was no difference
between 0 and 90 of flexion or between 0 and 135
of flexion.
In the same fashion, but keeping the ligaments intact,
we analyzed the 5 knees from the control group with
MRI. Each specimen underwent one MRI study for
every degree of flexion for a total of 20 studies. There
was no contact between the cruciate ligaments in 18
positions, and there was contact without deformity in 2
positions (both in flexion at 45 ) (Table 1).
Discussion
A potential complication of improper positioning of
the bone tunnels is contact between the new ACL and
the PCL at any point during full excursion of the knee
joint, which may result in instability and restriction in
knee flexion.5 This impingement may lead to knee pain
after ACL reconstruction.
Nishimori et al.8 reported that the majority of the
population with intact ACLs have no contact between
these 2 structures. However, the same authors claimed
that some healthy individuals do have some contact
between the native ligaments, but they do not report
the existence of deformity among ACL and PCL ligaments or pain symptoms. The results of the present
study suggest that some healthy individuals (10%)
have physiologic contact between the 2 ligaments
without clinical symptoms.
Cruciate impingement as a cause of painful symptoms
in patients with ACL reconstruction seems to be caused
not only by contact between the structures but also by
the impression of one ligament on the other and the
subsequent deformation of this neoligament.7 In our
study, several tunnel positions and certain degrees of
knee flexion were identified through MRI as potential
risks leading to cruciate contact and subsequent deformity. This seemed to occur significantly more
frequently in tunnels created through the posterior
quadrants of the ACL footprint on the tibia. When we
analyzed the position of the femoral tunnel as a risk for
ligamentous deformity, none of the quadrants were
significant risk factors.
Fig 4. Oblique axial magnetic resonance
image. (A) Contact with deformity of
both ligaments refers to the situation in
which there is loss of contour definition
of the cruciate ligaments, which causes
morphologic changes in both ligaments.
(B) Contact between the ligaments
without deformity of the ligaments. (C)
There is no contact between the ligaments. The arrows correspond to the site
of impingement.
Knee 1
Cadaveric Knees
Tibia
Anterior-medial
Anterior-lateral
Posterior-medial
Posterior-lateral
0
Knee 2
45
90
135
0
Knee 3
45
90
135
0
45
Knee 4
90
135
0
Knee 5
45
90
135
0
45
90
135
CWOD CWOD CWOD CWOD CWOD CWOD CWOD CWOD
1
CWOD CWOD CWOD CWOD CWOD CWOD CWOD
1.5
1.5 CWOD CWOD
CWOD CWOD CWOD CWOD CWOD CWOD CWOD CWOD CWOD
1
CWOD CWOD CWOD CWOD CWOD CWOD CWOD
2
CWOD CWOD
CWOD CWOD CWD CWOD CWOD CWD CWOD CWOD CWOD CWD CWD CWOD CWOD CWOD CWD CWOD CWOD CWD CWD CWOD
CWOD CWD CWD CWOD CWOD CWOD CWD CWOD CWOD CWD CWD CWOD CWOD CWD CWD CWD CWOD CWD CWD CWD
Knee 6
0
45
90
Knee 7
135
0
45
Knee 8
90
135
0
45
Knee 9
90
135
0
45
Knee 10
90
135
0
45
90
135
Femur
Anterior-proximal
1.5
2
1
2
1
1.5
CWOD
2
2
1.5
1.5
2
1.5
2.5
0.5
1.5
1.5
2.5
2
2.5
Anterior-distal
CWD CWD CWOD
2
CWD CWD CWOD CWOD CWD CWD CWD CWOD CWOD CWD CWD
1
CWD CWD CWD CWOD
Posterior-proximal CWOD
1
CWOD CWOD CWOD CWOD CWOD
1
1
CWOD
1
1
CWOD CWOD CWOD 1.5
1.5
CWOD CWOD
2
Posterior-distal
1.5
CWD CWD
2
CWOD CWD CWD CWOD
1
CWD CWOD CWD CWOD CWD CWD CWD CWOD CWD CWD CWOD
Knee 11
Control Group
Knee 12
Knee 13
Knee 14
Knee 15
0
45
90
135
0
45
90
135
0
45
90
135
0
45
90
135
0
45
90
135
1
CWOD
1.5
1.5
1
1
1
1
0.5
CWOD
1
1
0.5
0.5
1
1
1
1
1.5
1.5
The distance is measured in millimeters.
CWD, ligamental contact with deformity; CWOD, ligamental contact without deformity.
*The measurements correspond to the value of each knee examined.
CHARACTERIZATION OF CRUCIATE LIGAMENT IMPINGEMENT
Table 1. Results Obtained for the Different Positions of the Tibial and Femoral Tunnels with Different Degrees of Knee Flexion*
5
6
D. C. ASTUR ET AL.
Some studies report that impingement between the
neo-ACL and the PCL is more significant when the knee
is in full extension.6,9,10 Simmons et al.11 considers that
impingement is greater with the knee in flexion. This
study demonstrates that the degree of flexion of the
knee is a predictor of impingement.
With the tibial tunnels, when the knee was in extension or flexed 135 , contact without deformity of the
ligaments was frequently observed, regardless of the
quadrant insertion position. However, when the tibial
tunnels were drilled at the posterior quadrants, contact
with deformity was very common, particularly when the
knee was flexed at 45 and 90 . Previous studies reported that the clinical diagnosis of contact, with or
without deformity, during arthroscopic procedures is
difficult.6,10 This is one of the reasons for the misdiagnosis of a possible impingement during ACL reconstruction screw fixation, which usually occurs at 90 of
knee flexion, thereby requiring MRI for the diagnosis.
When we analyzed the femoral tunnels, ligamentous
contact was seldom seen in extension or at 135 of
flexion. In this same group, we observed impingement
most often at 45 of flexion.
This study focused on the influence of the exact
location of the entry point of the femoral or tibial
tunnel inside its respective ACL footprint as a cause of
postoperative ligament impingement. Although other
studies analyzed different ACL reconstruction techniques and the incidence of impingement, our results
may guide a more precise positioning of the tunnel
during surgery. Furthermore, graft choice during ACL
reconstruction can also increase ligament impingement
and deformity. Studies concluded that the average area
and diameter in the midportion of the semitendinosus
and gracilis tendon graft and the bone-patella tendonbone graft are greater than those of the native ACL.12,13
The analysis of multiple knee flexion grades helped
define the relationship between the cruciate ligaments
along the full excursion of the knee range of motion.
Furthermore, this study identified not only the concept
of contact but also how deformity of the ligaments after
contact may play a role in postoperative ligament
impingement. This would be a potential cause for the
appearance of symptoms after ACL reconstruction.
Because this was a cadaveric study, clinical outcome
and type of ligament relationship could not be
analyzed. Avoiding impingement with ligament deformity during ACL reconstruction seems to be necessary
to avoid adverse patient symptoms such as instability
and knee flexion deficits.5,14,15 Knowledge of safe zones
within the footprint to avoid PCL impingement can
minimize the chance of these symptoms occurring.
Limitations
There are several limitations to this study. Cadaveric
studies do not always reproduce the physiology of a live
joint. The main limitation was the lack of a third group
that would assess ligament detachment on the tibia and
femur and favor the crossing of 4 different positions of
the femur and the tibia tunnels. In this case, 16
different situations for each knee could possibly indicate
tunnel positions more prone to ligament contact as well
as positions that are safer to use to avoid ligament
contact. Moreover, we did not assess the relation
between the ligament impingement and the roof
impingement. Positioning the ACL on the tibia may be
one of the reasons for these types of impingements,
leading to limitation of range of motion and pain.16 In
addition, the flexion angle for the fixation might have
an impact on graft rotation and therefore an influence
on graft impingement.
Conclusions
Impingement with ligament deformity is more likely
when the graft is fixed in the posterior quadrants of the
tibial footprint, regardless of the degree of knee flexion.
Although quite common, the ligament position in the
femoral footprint was not found to be a primary cause
of ACL impingement with deformity.
References
1. Cohen M, Astur Dda C, Kaleka CC, et al. Introducing
3-dimensional stereoscopic imaging to the study of
musculoskeletal anatomy. Arthroscopy 2011;27:593-596.
2. Duthon VB, Barea C, Abrassart S, Fasel JH, Fritschy D,
Menetrey J. Anatomy of the anterior cruciate ligament.
Knee Surg Sports Traumatol Arthrosc 2006;14:204-213.
3. Matsumoto H, Suda Y, Otani T, Niki Y, Seedhom BB,
Fujikawa K. Roles of the anterior cruciate ligament and
the medial collateral ligament in preventing valgus
instability. J Orthop Sci 2001;6:28-32.
4. Sakane M, Fox RJ, Woo SL, Livesay GA, Li G, Fu FH. In
situ forces in the anterior cruciate ligament and its
bundles in response to anterior tibial loads. J Orthop Res
1997;15:285-293.
5. Howell SM, Hull ML. Checkpoints for judging tunnel and
anterior cruciate ligament graft placement. J Knee Surg
2009;22:161-170.
6. Fujimoto E, Sumen Y, Deie M, Yasumoto M, Kobayashi K,
Ochi M. Anterior cruciate ligament graft impingement
against the posterior cruciate ligament: diagnosis using
MRI plus three-dimensional reconstruction software.
Magn Reson Imaging 2004;22:1125-1129.
7. Maak TG, Bedi A, Raphael BS, et al. Effect of femoral
socket position on graft impingement after anterior
cruciate ligament reconstruction. Am J Sports Med 2011;39:
1018-1023.
8. Nishimori M, Sumen Y, Sakaridani K, Nakamura M. An
evaluation of reconstructed ACL impingement on PCL
using MRI. Magn Reson Imaging 2007;25:722-726.
9. Iriuchishima T, Tajima G, Ingham SJ, Shen W, Smolinski P,
Fu FH. Impingement pressure in the anatomical and
nonanatomical anterior cruciate ligament reconstruction:
a cadaver study. Am J Sports Med 2010;38:1611-1617.
CHARACTERIZATION OF CRUCIATE LIGAMENT IMPINGEMENT
10. Strobel MJ, Castillo RJ, Weiler A. Reflex extension loss after
anterior cruciate ligament reconstruction due to femoral
“high noon” graft placement. Arthroscopy 2001;17:408-411.
11. Simmons R, Howell SM, Hull ML. Effect of the angle of
the femoral and tibial tunnels in the coronal plane and
incremental excision of the posterior cruciate ligament on
tension of an anterior cruciate ligament graft: an in vitro
study. J Bone Joint Surg Am 2003;85-A:1018-1029.
12. Iriuchima T, Shirakura K, Yorifuji H, Aizawa S, Fu FH.
Size comparison of ACL footprint and reconstructed
autograft. Knee Surg Sports Traumatol Arthrosc. 2012 Mar 10
[Epub ahead of print].
13. Pujol N, Queinnec S, Boisrenoult P, Magdes A, Beaufils P.
Anatomy of the anterior cruciate ligament related to
7
hamstring tendon grafts. A cadaveric study. Knee. 2012
Nov 14 [Epub ahead of print].
14. Iriushima T, Tajima G, Ingham SJ, Shirakura K, Fu FH.
PCL to graft impingement pressure after anatomical or
non-anatomical single-bundle ACL reconstruction. Knee
Surg Sports Traumatol Arthrosc 2012;20:964-969.
15. Liu SH, Osti L, Dorey F, Yao L. Anterior cruciate ligament
tear. A new diagnostic index on magnetic resonance
imaging. Clin Orthop Relat Res 1994;302:147-150.
16. Kropf EJ, Shen W, van Eck CF, Musahl V, Irrgang JJ,
Fu FH. ACL-PCL and intercondylar notch impingement:
magnetic resonance imaging of native and double-bundle
ACL reconstructed knees. Knee Surg Sports Traumatol
Arthrosc. 2012 May 24 [Epub ahead of print].