DOI:10.2214/AJR.07.2921
AJR 2008; 191:W155-W159
© American Roentgen Ray Society
MRI Appearance of Posterior Cruciate Ligament Tears
William Rodriguez, Jr.1,
Emily N. Vinson1,
Clyde A. Helms1 and
Alison P. Toth2
1 Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC
27710.
2 Division of Orthopaedic Surgery, Duke University Medical Center, Durham,
NC.
Received July 23, 2007;
accepted after revision April 29, 2008.
Address correspondence to E. N. Vinson
(vinso003{at}mc.duke.edu).
Presented at the 2006 annual meeting of the American Roentgen Ray Society,
Vancouver, BC, Canada.
WEB
This is a Web exclusive article.
Abstract
OBJECTIVE. There is little in the radiology literature regarding the
MRI appearance of a torn posterior cruciate ligament (PCL). The purpose of
this study was to describe the MRI appearance of surgically proven PCL tears
and to emphasize previously unreported signs.
CONCLUSION. The PCL is usually injured as the result of stretching
deformation; on MRI, the ligament maintains continuity as a single structure
with apparent thickening. On sagittal T2-weighted images, an anteroposterior
diameter of 7 mm or more is highly suggestive of a torn PCL. Increased
intrasubstance signal intensity in the PCL on proton-density images with lower
signal intensity on T2-weighted images is another common feature.
Keywords: knee injury • MRI • posterior cruciate ligament
Introduction
As orthopedic surgeons gain more experience with posterior cruciate
ligament (PCL) reconstruction and acquire better understanding of the
long-term effects of PCL deficiency, it becomes increasingly important for
radiologists to identify PCL tears
[1,
2]. Knowledge of the diagnosis
and management of PCL tears lags behind that of anterior cruciate ligament
(ACL) tears. Inexperience with this injury is compounded by difficulty in
clinical identification of tears. In the acute phase of a tear, physical
examination of the PCL can be difficult because of the presence of a
hemarthrosis, pain, and concomitant ligamentous injuries
[3]. It is not unusual for the
findings at physical examination to be normal after acute tear of the PCL
[4], and patients rarely
describe hearing the pop common with ACL tears
[1]. As a result, many PCL
tears remain undiagnosed at the initial clinical evaluation
[5,
6]. In this setting,
radiologists may be in a position to make the index diagnosis. There is
relatively little in the radiology literature regarding the MRI appearance of
PCL tears
[7–9].
The purpose of this study was to describe the MRI appearance of surgically
proven PCL tears and to emphasize previously unreported signs.
Materials and Methods
Institutional review board approval was obtained for this study. The
records of 34 patients with a postoperative diagnosis of PCL tear and
available preoperative MR images of the knee were identified in a search of
surgical and imaging databases. The patient group consisted of 23 men (68%)
and 11 women (32%) with an average age of 35.2 years (range, 15–69
years). Sixteen of the injuries (47.1%) were the result of a motor vehicle
collision, seven (20.6%) were the result of a sports injury, and four (11.8%)
were the result of falls. Two patients (5.9%) were pedestrians injured by
moving vehicles. One patient (2.9%) fell from a moving vehicle, one twisted
his knee on a wet floor, one was injured walking on a beach, and one was
injured descending stairs. One patient had an unknown mechanism of injury. The
average time from date of injury to imaging was 54.8 days (range, 1–529
days) for 33 of the patients. The date of injury was unknown for one patient.
The average time from imaging to surgery was 60.1 days (range, 1–422
days).
All imaging studies were performed over a 4-year period on a 1.5-T unit
(Signa, GE Healthcare) according to our standard knee MRI protocol: sagittal
fat-suppressed spin-echo proton-density images (TR/TE, 1,500–2,000/20)
and sagittal, axial, and coronal fat-suppressed fast spin-echo T2-weighted
images (2,000–5,000/57–80). The number of signals acquired was two
for fast spin-echo sequences and one for proton-density sequences. The
echo-train length was 8–10 for the fast spin-echo sequences. A slice
thickness of 4 mm with a 0.4-mm interslice gap was used. The field of view was
16 cm2, and the matrix size was 256 x 192.
Images from the preoperative MRI examinations of the knee were
retrospectively evaluated in consensus by three musculoskeletal radiologists.
The PCL was assessed on the sagittal T2-weighted and proton-density images for
anteroposterior diameter, intrasubstance signal intensity, and the presence or
absence of complete disruption of the ligament, defined on the basis of fluid
signal intensity traversing the full thickness of the PCL on sagittal
T2-weighted images. In cases of disruption or avulsion of the PCL, the site
was noted. The anteroposterior diameter of the vertical segment of the PCL
between the tibial attachment and genu of the PCL was measured on the sagittal
T2-weighted images. The measurement perpendicular to the fibers of the PCL was
obtained.
The images of a group of control patients (108 men, 92 women; average age,
42.5 years; range, 17–81 years) without a history of trauma who
underwent 200 consecutive MRI examinations of the knee over a 2-month period
were retrospectively evaluated for signal intensity of the PCL on proton
density– and T2-weighted images and for anteroposterior diameter of the
vertical segment of the PCL measured on sagittal T2-weighted images. All
observations and measurements of the study and control patients were made in
consensus by unblinded observers. Although axial and coronal T2-weighted
images were available for review, the assessments of PCL thickness,
continuity, and signal characteristics were made on the sagittal proton
density– and T2-weighted images.
Results
In the 200 consecutive control MRI examinations of the knee, 184 PCLs (92%)
measured 6 mm or less in anteroposterior diameter (Fig.
1A,
1B). The range was 4–8
mm. The signal intensity of the PCL in these patients was low on the
T2-weighted and proton-density images.
View larger version (185K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —54-year-old man with normal posterior cruciate ligament
(arrows). Sagittal fat-suppressed proton density–weighted MR
image (TR/TE, 1,800/17) shows smooth, thin, continuous band of low signal
intensity typical of normal posterior cruciate ligament.
|
|
View larger version (162K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —54-year-old man with normal posterior cruciate ligament
(arrows). Sagittal fat-suppressed T2-weighted MR image (3,500/52) in
same plane as A shows smooth, thin band of low signal intensity.
Anteroposterior measurement of vertical segment is 4 mm, which is within range
of normal.
|
|
In the 34 patients with surgically documented PCL tears, 32 PCLs (94%)
measured 7 mm or more in anteroposterior diameter on sagittal T2-weighted
images. The average thickness was 9.6 mm, with a range of 4–15 mm (Fig.
2A,
2B,
2C). Also, in 32 of the 34
cases, the anteroposterior diameter of the distal half of the ligament
increased with no relation to the location of the tear. In 30 of the 34 PCL
tears (88%), the PCL intrasubstance signal intensity on the proton-density
images was higher than that on T2-weighted images. All 34 patients had
abnormal intermediate intrasubstance or fluid signal intensity on
proton-density images (Fig. 3A,
3B,
3C,
3D). The differences in
distribution of PCL thicknesses between the control group and the group of
patients with PCL tears was statistically significant (p < 0.001,
Wilcoxon's test). The estimated sensitivity and specificity of several PCL
thicknesses (7 mm and greater, 8 mm and greater, and 9 mm and greater) in the
diagnosis of PCL tear are summarized in
Table 1. The 95% CIs were based
on the Wilson score test.
View larger version (173K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —43-year-old woman with knee pain after injury sustained while
playing softball. Sagittal fat-suppressed proton-density MR image (TR/TE,
2,000/20) shows abnormally thickened posterior cruciate ligament (PCL)
(arrows) with increased signal intensity within ligament.
|
|
View larger version (166K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —43-year-old woman with knee pain after injury sustained while
playing softball. Sagittal fat-suppressed T2-weighted MR image (3,500/69.1)
shows abnormally thickened PCL (arrows) with anteroposterior
measurement of 10 mm, which appears to maintain continuity as one structure.
Signal intensity of PCL is relatively low in contrast to appearance in
A.
|
|
View larger version (169K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2C —43-year-old woman with knee pain after injury sustained while
playing softball. Coronal fat-suppressed T2-weighted MR image (3,000/69.2)
shows markedly thickened PCL (arrows) in region of genu. At surgery,
PCL was present and intact to probing though lax on posterior drawer test
performed during arthroscopy. One operative diagnosis was PCL tear.
|
|
View larger version (183K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —17-year-old girl with knee pain after motor vehicle crash.
Sagittal fat-suppressed proton density–weighted MR image (TR/TE,
2,000/20) of posterior cruciate ligament (PCL) near femoral attachment shows
abnormal thickening (arrow) with intrasubstance fluid signal
intensity.
|
|
View larger version (186K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —17-year-old girl with knee pain after motor vehicle crash.
Sagittal fat-suppressed proton density–weighted MR image (2,000/20) of
PCL one image plane lateral to A reveals abnormal thickening
(arrows) with longitudinally oriented striations of fluid signal
intensity.
|
|
View larger version (186K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —17-year-old girl with knee pain after motor vehicle crash.
Sagittal fat-suppressed proton density–weighted MR image (2,000/20) of
PCL tibial attachment immediately lateral to B shows abnormal
thickening with intrasubstance striations (arrows) of fluid signal
intensity. Lack of complete ligamentous disruption of this torn PCL is
evident.
|
|
View larger version (184K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3D —17-year-old girl with knee pain after motor vehicle crash.
Sagittal fat-suppressed T2-weighted MR image (3,500/69) of PCL in same imaging
plane as B reveals abnormal thickening with striations
(arrows) of intermediate signal intensity. Intrasubstance signal
intensity appears relatively lower than in B. Anteroposterior
measurement of this vertical segment is 10 mm, which is abnormal. At surgery,
this PCL was described as torn. Patient underwent anterior cruciate ligament,
PCL, and posterolateral corner reconstructions.
|
|
In 21 of 34 cases (62%), the PCL was not completely disrupted or avulsed.
Five of the 13 ligaments (15% of the total) that appeared completely disrupted
or avulsed were avulsed from the femoral attachment, five (15%) tears were
midsubstance, and three ligaments (9%) were avulsed from the tibial
attachment. Two of the three avulsions at the tibial attachment were tibial
avulsion fractures in which a small fragment of cortical bone was distracted
proximally from the tibia (Fig.
4). Of the 21 cases without complete disruption, intrasubstance
fluid signal intensity within the PCL on the T2-weighted images was present in
only four cases (19% of the 21). Another relatively common feature of the torn
PCL was a striated appearance, somewhat similar to the normal ACL appearance,
with longitudinally oriented lines of increased intrasubstance signal
intensity. Twenty-four of the 34 tears (71%) had this striated appearance on
proton density– or T2-weighted images. The characteristics of this
striation were variable with regard to location, signal intensity, thickness
of lines, and length of area of signal intensity within the ligament. Laxity
or a wavy appearance of the PCL was apparent in eight of the 34 cases (24%).
Two PCLs that appeared as one continuous structure at MRI were described as
completely disrupted at surgery.
View larger version (183K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4 —59-year-old woman with knee pain after motor vehicle crash.
Sagittal fat-suppressed T2-weighted MR image (TR/TE, 4,000/70) of posterior
cruciate ligament (PCL) shows tibial avulsion fracture. Thin piece of cortical
bone (arrow) is avulsed from tibia at insertion of PCL. PCL is
redundant and retracted proximally.
|
|
Only two of the 34 PCL tears were described as isolated injuries in the
operative notes; in most of the cases, multiple ligament injuries were
reported. The most common associated injuries were ACL tears in 20 of the
patients (59%), medial collateral ligament tears in 12 patients (35%), and
injuries to one or more structures in the posterolateral corner in 17 patients
(50%).
Discussion
The incidence of PCL injuries is difficult to assess because of the large
percentage that remain undiagnosed in the acute setting, a reported
1–44% [10]. Schultz et
al. [5] retrospectively studied
the epidemiologic characteristics of 494 cases of PCL insufficiency. Those
investigators found that most of the injuries were related to traffic
accidents (45%) and sports injuries (40%). The mechanism of injury was
dashboard type in 35%, fall on a flexed knee (blow to the anterior proximal
tibia) with the foot plantar-flexed in 24%, and hyperextension in 12% of
cases. Hyperflexion injury has been described as a less-common mechanism of
PCL injury [3,
11]. Sports-related PCL
injuries are significantly more likely to be isolated and more likely to be
diagnosed in the acute phase
[5,
11]. Overall, more than 50% of
patients presented more than 1 year after injury
[5].
Because the PCL is the primary restraint to posterior tibial translation
relative to the femur [12],
PCL tears are diagnosed and graded clinically with the posterior drawer test,
which is used to evaluate the degree of posterior translation of the tibia
relative to the femur with the knee flexed 90°. Tears therefore are
defined on the basis of functional competency of the ligament rather than the
presence or absence of complete disruption. Laxity to probing at arthroscopy
is diagnostic of a tear.
The normal appearance of the PCL on MR images is a well-defined continuous
band of low signal intensity in all pulse sequences. As found in our
evaluation of 200 consecutive MR examinations of the knee without a known
diagnosis of a PCL tear, a normal PCL usually measures no more than 6 mm in
anteroposterior diameter on sagittal T2-weighted images, with the measurement
perpendicular to the fibers of the vertical segment. A torn PCL typically
becomes abnormally enlarged. To our knowledge, abnormal thickening of a torn
PCL on MR images has not been previously described. In 94% of our 34 patients
with PCL tear, the PCL measured 7 mm or greater. On the basis of a sensitivity
of 94% and specificity of 92%, we advocate using 6 mm as the upper limit of
normal PCL thickness, thicknesses of 7 mm and greater being suggestive of a
tear. This finding corroborates those of Cho et al.
[13] in a sonographic
investigation. Those investigators measured the distal half of the PCL
(vertical segment) and evaluated the ligament in the sagittal plane. Using the
other knee as a control, they found that the normal PCL measured less than 6
mm (range, 4–6 mm). The mean measurement of torn PCLs in that study was
15.6 mm (range, 12–22 mm). The thickness of the distal half of the PCL
increased whether the tear occurred proximally or distally. Cho et al.
concluded that the increase in thickness of the ligament is the most important
criterion among the sonographic findings of torn PCL.
As we found in this investigation, an acutely torn PCL usually maintains
continuity as a single structure and is typically injured as a result of
stretching deformity. In 62% of the 34 cases of PCL tear in this study, the
PCL maintained continuity as a single structure on images. Similar findings of
ligament continuity were discussed by Akisue et al.
[14], who evaluated acutely
torn PCLs clinically and with MRI. In 75% (36 of 48) of their cases of torn
PCL, the PCL maintained continuity as one structure on MRI. Those
investigators concluded that a high percentage of acutely injured PCLs are
likely to develop somewhat slack but continuous ligamentlike tissue. It has
been suggested that continuity of the torn PCL on MR images has a favorable
prognosis for nonoperative treatment
[1]. Surgeons, however,
consider many variables when evaluating PCL injuries for possible
reconstruction, such as associated ligament, chondral, and meniscal injuries
and subjective instability
[11].
We found the proton-density (short TE) sequence extremely important in the
diagnosis of PCL injuries. With reliance on T2-weighted images alone, one may
not recognize subtle pathologic changes in the PCL, resulting in lower
sensitivity. In all of our cases of torn PCL, the ligament had abnormally
increased intrasubstance signal intensity on proton-density images. In 88% of
our cases, the signal intensity on proton-density images was increased
relative to the signal intensity on T2-weighted images. If the ligament has
homogeneously low signal intensity on proton-density images, acute PCL injury
is highly unlikely. In evaluation of the PCL, the negative predictive value of
MRI performed with a short-TE sequence is reported to approach 100%
[9]. The proton-density
sequence, however, should not be used for obtaining the anteroposterior
measurement of the ligament because the intrasubstance signal intensity of the
torn PCL on the proton-density images often approximates that of the adjacent
edematous soft tissues, making the margins of the ligament difficult to
discern. For this reason, sagittal T2-weighted images should be used for
measuring the PCL.
Unlike other ligament tears, PCL tears rarely have intrasubstance fluid
signal intensity on T2-weighted images. In the 13 cases in this study labeled
as discontinuous at imaging, increased T2 signal intensity was present in the
expected course of the torn ligament. Eight of these patients had avulsion of
the ligament from either the femur or the tibia. However, of the 21 cases in
this study in which the PCL appeared as one continuous structure, only four
exhibited intrasubstance fluid signal intensity within the PCL on T2-weighted
images. The cause of this observation is not certain. Perhaps proteinaceous
fluid within the torn PCL has increased signal intensity on proton-density
images, whereas the T2 relaxation time is not particularly lengthened owing to
the relatively restricted environment of the protons within the contiguous,
though functionally torn, fibers of the PCL.
Although we did not assess bone marrow contusions in this study, other
authors have described contusion patterns that can lead radiologists to
suspect PCL injury. Mair et al.
[15] evaluated the bone
contusion pattern related to acute PCL tears and found that 83% of their 35
patients had a contusion in at least one location. A contusion involving the
anterior proximal tibia is a known common finding related to a direct blow
[16,
17]. Hyperextension injury
resulting in kissing contusions with focal bone marrow edema on the
antero-inferior femoral condyle and proximal anterior tibia also may be seen
on MR images [16].
The PCL usually is injured as a result of a stretching deformation, whereby
the ligament maintains continuity as a thickened structure, rather than frank
rupture. Apparent ligamentous continuity was seen in 62% of our cases of
surgically proven tears. Abnormally increased signal intensity within the PCL
on proton-density images is highly suggestive of PCL tear and was seen in all
of the patients in this study who had a surgical diagnosis of PCL tear. With
use of sagittal T2-weighted images, a 7-mm or greater anteroposterior
measurement of the vertical portion of the PCL can be considered evidence of a
torn PCL with a high degree of certainty. Recognition of these findings should
improve radiologists' ability to diagnose PCL tears preoperatively.
Acknowledgments
We thank David M. DeLong for assistance with statistical analysis.
References
- Wind WM Jr, Bergfeld JA, Parker RD. Evaluation and treatment of
posterior cruciate ligament injuries: revisited. Am J Sports
Med 2004; 32:1765
–1775[Abstract/Free Full Text]
- Strobel MJ, Weiler A, Schulz MS, Russe K, Eichhorn HJ. Arthroscopic
evaluation of articular cartilage lesions in
posterior-cruciate-ligament-deficient knees.
Arthroscopy 2003;19
: 262–268[Medline]
- Margheritini F, Mariani PP. Diagnostic evaluation of posterior
cruciate ligament injuries. Knee Surg Sports Traumatol
Arthrosc 2003; 11:282
–288[CrossRef][Medline]
- Burger RS, Larson RL. Acute ligamentous injury. In: Larson RL,
Grana WA, eds. The knee: form, function, pathology, and
treatment. Philadelphia, PA: Saunders, 1993:565
–576
- Schulz MS, Russe K, Weiler A, Eichhorn HJ, Strobel MJ. Epidemiology
of posterior cruciate ligament injuries. Arch Orthop Trauma
Surg 2003; 123:186
–191[Medline]
- Dandy DJ, Pusey RJ. The long-term results of unrepaired tears of
the posterior cruciate ligament. J Bone Joint Surg Br1982; 64:92
–94[Medline]
- Sonin AH, Fitzgerald SW, Friedman H, Hoff FL, Hendrix RW, Rogers
LF. Posterior cruciate ligament injury: MR imaging diagnosis and patterns of
injury. Radiology 1994;190
: 455–458[Abstract/Free Full Text]
- Niitsu M, Anno I, Fukubayashi T, Shimojo H, Kuno S, Akisada M.
Tears of cruciate ligaments and menisci: evaluation with cine MR imaging.
Radiology 1991;178
: 859–864[Abstract/Free Full Text]
- Grover JS, Bassett LW, Gross ML, Seeger LL, Finerman GA. Posterior
cruciate ligament: MR imaging. Radiology1990; 174:527
–530[Abstract/Free Full Text]
- Shelbourne KD, Davis TJ, Patel DV. The natural history of acute,
isolated, nonoperatively treated posterior cruciate ligament injuries: a
prospective study. Am J Sports Med 1999;27
: 276–283[Abstract/Free Full Text]
- Shino K, Horibe S, Nakata K, Maeda A, Hamada M, Nakamura N.
Conservative treatment of isolated injuries to the posterior cruciate ligament
in athletes. J Bone Joint Surg Br 1995;77
: 895–900[Medline]
- Amis AA, Gupte CM, Bull AM, Edwards A. Anatomy of the posterior
cruciate ligament and the meniscofemoral ligaments. Knee Surg
Sports Traumatol Arthrosc 2006;14
: 257–263[CrossRef][Medline]
- Cho KH, Lee DC, Chhem RK, et al. Normal and acutely torn posterior
cruciate ligament of the knee at US evaluation: preliminary experience.
Radiology 2001;219
: 375–380[Abstract/Free Full Text]
- Akisue T, Kurosaka M, Yoshiya S, Kuroda R, Mizuno K. Evaluation of
healing of the injured posterior cruciate ligament: analysis of instability
and magnetic resonance imaging. Arthroscopy2001; 17:264
–269[Medline]
- Mair SD, Schlegel TF, Gill TJ, Hawkins RJ, Steadman JR. Incidence
and location of bone bruises after acute posterior cruciate ligament injury.
Am J Sports Med 2004;32
:1681
–1687[Abstract/Free Full Text]
- Sanders TG, Miller MD. A systematic approach to magnetic resonance
imaging interpretation of sports medicine injuries of the knee. Am
J Sports Med 2005; 33:131
–148[Abstract/Free Full Text]
- Ross G, Chapman AW, Newberg AR, Scheller AD Jr. Magnetic resonance
imaging for the evaluation of acute posterolateral complex injuries of the
knee. Am J Sports Med 1997;25
: 444–448[Abstract/Free Full Text]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
