The Optetrak total knee arthroplasty system (Exactech, Gainesville,
Florida) is the fifth in a series of total knee arthroplasty designs developed
at the Hospital for Special
Surgery1-3.
The Optetrak implant retains the design concepts of the previous posterior
stabilized Insall-Burstein-I (IB-I PS; Johnson and Johnson, Braintree,
Massachusetts; Zimmer, Warsaw, Indiana) and Insall-Burstein-II (IB-II PS;
Zimmer) posterior stabilized total knee
implants2,4-7.
A constrained condylar knee version of the Optetrak design has an elevated
tibial post that, because of its higher and more squared design in comparison
with standard posterior stabilized designs, provides greater varus-valgus and
anteroposterior stability. The modular femoral and tibial components can
accommodate stems through a taper junction. Traditionally, constrained
condylar devices have been used with stem extensions to augment component
fixation. The stem transfers the bending and torsion loads to the diaphysis
and unloads the epiphyseal cancellous bone. From 1995 to 2006, 3886 of these
constrained condylar stems were implanted at our institution. We report the
cases of two patients who had failure in the male portion of the taper
junction of the femoral component and stem extension. This study was performed
under a protocol approved by the institutional review board of the Hospital
for Special Surgery. Both patients were informed that data concerning the
cases would be submitted for publication, and they consented.
Case 1. A man underwent a right total knee arthroplasty for
osteoarthritis in 1993 with use of the Insall-Burstein-II posterior stabilized
knee prosthesis. The patient was sixty-three-years old, weighed 91 kg, and was
1.6 m tall at the time of the operation. He did well for nine years, until
2002, at which time aseptic loosening necessitated revision total knee
arthroplasty with use of an Optetrak stemmed constrained condylar knee
prosthesis. A constrained insert and distal femoral (10-mm) augments were
used. The stems were uncemented. The patient did well for three years after
the revision arthroplasty, but then presented with a six-month history of
persistent pain in the right knee and an inability to walk more than one or
two blocks with the aid of a cane. He had a well-healed midline incision,
varus alignment of 15°, moderate varus-valgus and anterior-posterior
laxity, and a range of motion from 0° to 115°. Radiographs
demonstrated disruption of the stem and femoral component and radiolucencies
under the anterior flange of the femoral component and the tibial base-plate
(Figs. 1-A and 1-B). The
C-reactive protein level, erythrocyte sedimentation rate, and white blood-cell
count were all within normal limits. Aspiration of fluid from the knee joint
revealed clear fluid, a culture of which demonstrated no bacterial growth. The
patient underwent a second revision arthroplasty, during which gross loosening
of the femoral and tibial components was noted along with a fracture at the
male portion of the taper lock between the femoral component and the stem
extension (Fig. 1-C).
Microscopic evaluation of intraoperative specimens revealed evidence of
osteolysis as well as a chronic granulomatous reaction to polyethylene and
cement. Both components were revised to a hinged rotating-platform knee with
porouscoated press-fit stems and bone-defect-accommodating meta-physeal
sleeves (S-ROM; DePuy Orthopaedics, Warsaw, Indiana). The symptoms resolved
after surgery, and the patient could walk without the use of assistive devices
at the time of the six-week follow-up.
Case 2. A man underwent a left total knee arthroplasty for
osteoarthritis in 1991 with use of the Insall-Burstein-II posterior stabilized
knee prosthesis. The patient was sixty-six years old, weighed 100 kg, and was
1.8 m tall at the time of the operation. He did well for eight years, until
1999, at which time aseptic loosening necessitated revision total knee
arthroplasty with use of an Optetrak stemmed constrained condylar knee
prosthesis. A constrained insert and posterior (5-mm) and distal femoral
(10-mm and 5-mm) augments were used. The stems were uncemented. The patient
did well for five years after the revision arthroplasty, but then presented
with a four-month history of increasing pain in the left knee when walking and
an inability to walk more than one or two blocks with the aid of a cane. He
had a well-healed midline incision, normal alignment, moderate varus-valgus
and anterior-posterior laxity, and a range of motion from 0° to 120°.
Radiographs demonstrated a stress reaction at the femoral stem
(Fig. 2-A). A bone scan
demonstrated activity at the medial aspect of the femoral component. The
C-reactive protein level, erythrocyte sedimentation rate, and white blood-cell
count were all within normal limits. Aspiration of fluid from the knee joint
revealed clear fluid, a culture of which demonstrated no bacterial growth. The
patient underwent a second revision arthroplasty, during which gross loosening
of the femoral component was noted along with a fracture through the male
portion of the taper lock (Fig.
2-B). The tibial component was well fixed. Microscopic evaluation
of intraoperative specimens revealed evidence of chronic inflammation as well
as a mononuclear and giant-cell reaction to polyethylene and cement. The
femoral component was revised to an Optetrak cemented stemmed component with
distal femoral (15-mm) and posterior femoral (5-mm and 10-mm) augments. The
symptoms resolved after surgery, and the patient could walk comfortably
without the use of assistive devices at the time of the six-week
follow-up.
Light and scanning electron microscopic examination of both stems revealed
that the fractures initiated medially, near the junction at which the male
taper portion of the stem overlaps with the female part. In the prosthesis in
each of our patients, the fracture initiated within 0.5 mm of the rim of the
female portion and then propagated laterally and distally (Figs.
1-C and
2-B). The tibial posts from
each prosthesis were grossly deformed medially and had loss of polyethylene
posteriorly, which was consistent with lateral condylar lift-off and allowed
the internal corner of the femoral condyle to impinge on the post. The
anterior sides of the posts from each patient exhibited a
"bow-tie" pattern of wear, much more severe on the medial edge,
suggestive of impact with the anterior edge of the intercondylar box of the
femoral component in hyperextension and internal rotation. This damage was
consistent with the application of large torques at the interface between the
tibial post and the femoral box, which may have contributed to loosening of
the femoral component. After the femoral component became loose, the torsion
loads were likely transferred to the junction between the stem and the taper,
causing damage that would be consistent with the fracture pattern seen in both
prostheses.
The literature contains examples of dissociation of modular taper junctions
in shoulder, knee, and hip implants secondary to fluid contamination, galvanic
corrosion, fretting, or
trauma8-16;
however, we are aware of only a few reports that have described a fatigue
fracture at such a
joint9. Botti et al.
reported a fracture of the femoral neck trunnion in a modular cementless
anatomic medullary locking hip prosthesis (AML; DePuy
Orthopaedics)9,
which occurred without a history of trauma seven years after revision for
polyethylene wear. The taper surfaces had evidence of crevice corrosion,
suggesting that the fracture was caused both by damage to the trunnion during
surgery and by an increased susceptibility to crevice corrosion associated
with a modular femoral component with a long skirted
neck9.
The fractures in our patients also occurred without apparent trauma, which
suggests fatigue failure associated with loosening. Total knee arthroplasty
designs with increased constraint can serve as an adjunct in stabilizing the
joint17-20,
but the centralized tibial post and femoral box constraint mechanism is at a
mechanical disadvantage to resisting the bending and torsion loads applied
across the joint19.
Stem extensions can aid in resisting the applied loads and can protect the
cancellous bone21,
but they are unable to resist the entire load should no osseous support
exist17,20.
In our two patients, there was gross and microscopic evidence of osteolysis
and bone loss at the distal portion of the femur, which may have contributed
to aseptic loosening of the femoral component. Furthermore, stress-shielding
by larger-diameter stemmed femoral components may result in distal bone loss
secondary to more proximal load
transfer7,22-25.
Aseptic loosening of the femoral components would likely increase stresses
throughout the stem and the taper connection. In this particular stem design,
the male portion of the taper joint is smaller in diameter relative to the
diameter of the stem. In addition, a substantial length of the male portion of
the taper junction remains exposed after engagement with the sleeve
(Fig. 1-C). The taper lock may
therefore be a weak area in the implant and may be unable to withstand the
concentrated stresses that occur with osteolysis and loosening of the femoral
component.
A finite-element analysis of a taper lock in a modular implant demonstrated
that the joint is subjected to high stresses under physiologic cyclical
loading. These loads across the taper connection can induce a gap to open
between the cone and the sleeve, allowing ingress of joint fluid and
subsequent corrosion and fracture of the
implant26. These
loads can also cause fretting, leading to scratching, increased local stresses
at the base of the scratches, and
pitting27-29.
Potential solutions to the problem of stem fracture associated with
component loosening are to increase the diameter of the male portion of the
taper joint and/or minimize the length of the male portion that is exposed
after sleeve engagement. Thickening the male portion of the taper junction
and/or increasing the overlap with the female segment so that the length of
exposed male taper is minimized will likely increase the strength of the taper
joint in the Optetrak implant. ?