The mechanical function of the patella has been well described. As an
integral part of the extensor mechanism, the patella acts as a spacer to
provide a mechanical advantage for the quadriceps by increasing the distance
between the muscle's line of action and the axis of rotation of the knee
joint1. Early
biomechanical representations of the patellofemoral joint were created with
the assumption that this articulation was a frictionless pulley system in
which the patella acted as a linkage to transmit the quadriceps force to the
patellar tendon without altering its
magnitude2-6.
However, more recent biomechanical studies of the patellofemoral joint have
demonstrated that the patella also acts as a complex lever system creating a
force differential between the quadriceps tendon and the patellar
tendon7-10—that
is, the patella has been thought to act as a balance beam, thereby altering
the lengths of the quadriceps and patellar tendon moment arms as a function of
the knee joint angle. The force differential between the quadriceps tendon and
the patellar tendon is thought to occur as a result of the varying geometry
and shape of the distal part of the femur and the patella as well as the
changing point of contact between the patella and femur as the knee flexes and
extends1,8,9.
Several in vitro studies have quantified the relationship between the force
in the quadriceps tendon (FQ) and the tension force in the patellar
tendon (FPT). The investigations have shown that the
FPT/FQ ratio ranges from 1.0 to 1.2 with the knee fully
extended to 0.6 to 0.8 with the knee flexed to
60°7-12.
Despite the variability in the reported FPT/FQ ratios
among studies, all authors have reported a similar trend of a decreasing ratio
(i.e., less patellar tendon tension relative to quadriceps force) with knee
flexion.
A limitation of previous studies is that it has been assumed that the force
transfer from the quadriceps tendon to the patellar tendon is not influenced
by the potential load-sharing behavior of soft tissues that cross the knee.
One structure that has the potential for load-sharing between the quadriceps
tendon and the patellar tendon is the peripatellar retinaculum. The normal
peripatellar retinaculum consists of layered, fibrous connective tissue that
traverses the medial and lateral margins of the patella with attachments to
the femur, tibia, patella, and patellar
tendon13.
Additionally, the superficial fibers of the retinaculum originate from the
vastus lateralis and vastus medialis fascia, linking the quadriceps to the
patella (Fig.
1)13,14.
This linkage is responsible for the dynamic influence of the quadriceps on the
patellofemoral joint during active knee motion.
The specific role of the peripatellar retinaculum as a frontal plane
stabilizer of the patellofemoral joint has been well
established15-18.
However, as a result of its unique orientation, the peripatellar retinaculum
also may play a complementary load-sharing role with respect to the patellar
tendon. Similar to the patellar tendon, the peripatellar retinaculum provides
distal inferior support for the patella through the medial and lateral
patellomeniscal ligament tendons that connect the patella to the
tibia14. However,
because of the transverse and oblique orientation of the retinaculum fibers,
it is conceivable that this structure plays a role in providing medial-lateral
stability for the patella and functionally unloading the patellar tendon by
resisting tensile forces created by the extensor mechanism.
The purpose of this study was to determine the extent to which the
peripatellar retinaculum affects the magnitude of forces experienced by the
patellar tendon. It was hypothesized that the peripatellar retinaculum acts as
a load-sharing structure to decrease the forces experienced by the patellar
tendon.
Experimental Setup
Ten fresh-frozen, unmatched human cadaver knees were used in this study.
Each was macroscopically intact and radiographically normal. The donors of the
specimens had ranged from sixty to eighty years of age at the time of death.
After thawing, the knees were dissected with care taken to keep the
retinaculum and the quadriceps tendon intact. Skin and subcutaneous fat were
removed, as were the muscles from the tibia and the posterior part of the
femur. The head of the fibula was secured to the tibia with a screw, while the
distal two-thirds of this bone was removed. Approximately 20 cm of tibial and
femoral length was left for mounting purposes.
The individual components of the extensor mechanism, vastus medialis,
vastus lateralis, vastus intermedius, and rectus femoris were separated from
each other, with the fascial planes between the muscles used as a guide.
Following dissection, the muscles were trimmed to accommodate the width of the
loading clamps. The muscle clamps measured 8.0 × 3.0 cm and were made of
stainless steel. Care was taken to select the portion of muscle that was
representative of the resultant force direction of all muscle fibers. The
muscles were clamped as close to their respective insertions as possible so
that tendinous fibers could be incorporated within the clamp. The vastus
intermedius and rectus femoris were clamped together since the direction of
the resultant force vector of these muscles with respect to the patella is
similar19,20.
Both the tibia and the femur were secured within 2-in (5-cm)-diameter
polyvinyl chloride tubing with use of diaphyseal bolts and locking pins. Each
bone was positioned within its tube such that the long axis of the cylinder
represented the long axis of the bone. After the tibia and femur had been
appropriately positioned, the plastic tubes were filled with dental plaster
and the diaphyseal bolts were removed.
The cylinders were then mounted on a custom knee test apparatus that
provided six degrees of freedom at the femur and five degrees of freedom at
the
tibia19,21.
This apparatus was fixed to a materials testing machine frame (model 1122;
Instron, Canton, Massachusetts), which was used to flex and extend the knee
(Fig. 2). Anatomically
positioned pulleys guided cables from the muscle clamps to the applied
load.
To quantify the force in the patellar tendon during testing, a buckle
transducer (NK Biotechnical, Minneapolis, Minnesota) was placed near the
tibial attachment (Fig. 2). The
buckle transducer was calibrated after data collection by applying known
collinear tensile forces through the patellar tendon. Recordings from the
transducer were linear (R2 = 0.99) throughout the range of
calibration loads used (10 to 600 N).
Muscle Loading
As described in a previous
publication19, the
extensor mechanism was loaded by applying forces through the individual heads
of the quadriceps along their principal muscle fiber orientation. Loading of
the individual muscles was important as the vastus medialis and the vastus
lateralis have the potential to impart additional forces to the patella
through their fascial expansion and interdigitation with the peripatellar
retinaculum14.
The distribution of extensor force across the various muscles was based on
cross-sectional area data reported by Wickiewicz et al. (vastus medialis = 67
N, vastus lateralis = 98 N, and rectus femoris/vastus intermedius = 111
N)22. Thus, the
total extensor muscle force used in this study was 276 N and represented a
submaximal load on the patellofemoral joint. Each pulley was adjusted so that
the force application of the respective muscles represented the primary fiber
direction and orientation (Fig.
2). Forces were applied to the respective muscle clamps through
the cable system and were controlled with use of LabVIEW pneumatic cylinders
(National Instruments, Austin, Texas).
Experimental Procedures
Each specimen was first positioned on the knee test apparatus in full
extension. Care was taken to ensure that the proximal cylinder (femur) and the
distal cylinder (tibia) were aligned so that the physiologic quadriceps angle
(Q angle) was maintained. The Q angle was determined through the use of a
standard goniometer prior to dissection. All degrees of freedom on the knee
test apparatus were locked and the patellar tendon buckle transducer was
zeroed prior to load application. Each pulley was then adjusted so that the
force application to the respective muscles represented the primary fiber
direction and orientation.
Following the initial setup, muscle loading commenced and the buckle
transducer force was recorded. After testing at 0° of knee flexion, all
locking bolts controlling the degrees of freedom were loosened, and the knee
was flexed to 20° by lowering the Instron crosshead. Following this
procedure, the locking bolts were tightened, and the pulley systems were
adjusted to accommodate the new knee flexion angle. The entire process, as
described above, was then repeated. For each knee flexion angle of 0°,
20°, 40°, and 60°, the measurements of the patellar tendon force
were repeated three times. Knee flexion angles were determined with use of a
digital inclinometer.
Once the tension of the patellar tendon was recorded with the retinaculum
intact, the lateral and medial retinacula were removed. The lateral
retinaculum was sectioned by performing a triangular incision beginning at the
lateral patellar margin to the level of the tibial tubercle and extending
proximally and laterally to the level of the lateral collateral ligament.
Similarly, the medial retinaculum was removed with a triangular incision
beginning at the medial patellar margin to the level of the tibial tubercle
and extending proximally and medially to the medial collateral ligament.
Following the removal of the retinaculum, the procedures, as described above,
were repeated. Throughout all testing procedures, the specimens were sprayed
with saline solution to keep the tissues moist.
Statistical Methods
A two-way repeated-measures analysis of variance was used to compare
patellar tendon tension between the retinaculum-intact and retinaculum-removed
conditions across the knee flexion angles. Statistical analysis was performed
with use of SPSS statistical software (SPSS, Chicago, Illinois) with a
significance level of p < 0.05.
For all knee flexion angles tested, the mean patellar tendon tension with
the retinaculum removed was greater than the mean patellar tendon tension with
the retinaculum intact (Fig.
3). Analysis of variance revealed a significant main effect for
the condition (p = 0.008) and a significant condition × knee flexion
angle interaction (p = 0.031). Post hoc analysis consisting of paired t tests
revealed that the mean patellar tendon tension with the retinaculum removed
was significantly greater than that with the retinaculum intact both at 0°
(206.0 ± 27.4 N compared with 176.6 ± 34.9 N; p = 0.006) and at
60° (124.3 ± 18.0 N compared with 113.4 ± 27.3 N; p = 0.018)
(Fig. 3). There was no
significant difference between the retinaculum-intact and retinaculum-removed
conditions at 20° or 40° of knee flexion.
The results of this study support the hypothesis that the peripatellar
retinaculum plays a role in the transmission of forces within the extensor
mechanism. In particular, it was demonstrated that the patellar tendon
experienced less tension with the retinaculum intact than it did with the
retinaculum removed, under similar loading conditions
(Fig. 3). When averaged across
all knee flexion angles, the patellar tendon tension with the retinaculum
intact was 8.9% less than the tension with the retinaculum removed. However,
the decrease in patellar tendon tension varied across the tested knee flexion
angles, and significance was achieved at only the extremes of the range of
motion tested (0° and 60°).
Retinaculum removal had the largest influence on the transmission of forces
to the patellar tendon at 0°, where a 16.6% increase in patellar tendon
tension was measured. We postulated that the increase in the load-sharing
function of the peripatellar retinaculum with the knee extended could be the
result of the screw home mechanism of the knee complex. For example, it is
well-established that the tibia externally rotates relative to the femur
during knee
extension23-26.
The axial rotation of the tibia as the knee extends is thought to be the
result of the shape of the medial femoral condyle, the passive tension in the
anterior cruciate ligament, and the lateral pull of the quadriceps
muscle27. In their
study of the influence of tibial rotation on strain in the peripatellar
retinaculum, Lee et
al.21 demonstrated
that the largest increases in strain occurred with tibial external rotation at
0° of flexion. Thus, it is conceivable that increased passive tension of
the retinaculum as a result of tibial rotation could increase the load-sharing
behavior of this structure. Additional strain in the peripatellar retinaculum
with the knee extended also could result from the need for increased frontal
plane stability of the patella as the osseous support afforded by the
trochlear groove is minimal in this
position16,28.
Interestingly, there were no significant differences in patellar tendon
tension between the retinaculum-intact and retinaculum-removed conditions at
20° or 40° of flexion. Perhaps this was due to the positioning of the
patella within the deeper portion of the trochlear groove, thereby minimizing
the role of the peripatellar retinaculum as a frontal plane stabilizer of the
patella. In addition, as the knee flexes, the tibia internally rotates to a
more neutral
position27. Both of
these events could result in a decrease in the passive tension within the
retinaculum, thereby decreasing load-sharing.
At 60° of flexion, a 9.6% increase in patellar tendon tension was
observed in the retinaculum-removed condition. We hypothesized that, with
greater knee flexion angles, there is a gradual increase in the passive
tension within the peripatellar retinaculum, which may augment the
load-sharing function of this structure. This probably occurs as a result of
the retinaculum sharing some of the passive tension created by the lengthening
of the extensor
mechanism29-31.
Although patellar tendon tension was not quantified at knee flexion angles of
>60° in this investigation, we believe that this trend toward increased
load-sharing would continue with increasing knee flexion angles.
When considering the anatomy of the peripatellar retinaculum, it is easy to
visualize how quadriceps forces may be distributed to other structures rather
than just the patellar tendon (Fig.
4). The superficial fibers of the lateral retinaculum originate
from the iliotibial band and the vastus lateralis fascia and insert into the
lateral margin of the patella and the patellar tendon. Similarly, the
superficial fibers of the medial retinaculum originate from the vastus
medialis and the sartorius muscles and attach to the patella and the patellar
tendon medially14.
The deep layer of the lateral retinaculum consists of several structures,
including the transverse and lateral patellofemoral ligaments as well as the
patellotibial band, while the deep layer of the medial retinaculum is composed
of the medial patellofemoral, patellomeniscal, and medial patellotibial
ligaments13. These
structures subsequently connect the patella to the iliotibial tract, the
tibia, and the femur. Given this complex anatomic relationship, it is not
surprising that the peripatellar retinaculum would function as a
load-transmitting structure within the extensor mechanism.
A limitation of this study was that submaximal muscle forces were employed
during testing. This was done to maintain the integrity of the cadaver tissue.
Although 276 N may not represent the force generated by the quadriceps during
activities of daily living (e.g., walking), this load did allow adequate
comparison of the conditions under study. The fact that significant
differences were detected in patellar tendon tension under a submaximal
loading condition may imply that the trends observed in the current
investigation will be magnified when the retinaculum is under greater tension
with higher quadriceps forces. Future studies should address the effects of
increased loads on the force distribution within the extensor mechanism as a
function of knee position.
Our results have clinical implications in that any compromise of the
peripatellar retinaculum may impair its loadsharing capability and increase
the forces experienced by the patellar tendon. For example, in lateral
patellar dislocation, the medial retinaculum is frequently partially or
completely
torn32,33.
Similarly, the lateral peripatellar retinaculum is commonly surgically
disrupted in an effort to correct abnormal tracking of the
patella34. Given
that the patellofemoral joint reaction force is the result of the quadriceps
force vector and the patellar tendon vector, any increase in patellar tendon
tension would translate into greater joint compression. ?