260 / 460
Patellofemoral Arthroplasty
259
patellofemoral osteoarthritis can be reliably
assessed using conventional radiographs alone.
However, the exclusion of significant degene
rative changes in both femorotibial compart
ments is important, especially when patello
femoral arthroplasty is considered. For this
reason technetium-99m bone scans can be used.
A bone scan gives more accurate information
on femorotibial and PF involvement because of
its higher sensitivity for detecting OAcompared
to conventional radiographs. This is often very
useful in the decision making process.
Examining conventional radiographs of
patients with isolated patellofemoral osteo
arthritis, the majority (70-90%) showed lesions
on the lateral side of the trochlea [15, 28],
demonstrating degenerative changes due to
lateral malalignment with subluxation. A small
number of cases (<10%) showed medial facet
and medial trochlea disease. The precise
etiology of this condition is unknown but it
might suggest that there is some kind of medial
overload causing these changes, varus-valgus
alignment may influence which compartment
is affected [29]. The remaining cases
demonstrated symmetrical patterns of wear
affecting equally both the medial and lateral
facets of the patella and trochlea groove [28].
Implant designs for PF prosthesis to treat PFOA
already have seen a long history. The need for a
PF prosthesis was fed by unsuccessful non-
operative treatment modalities and insufficient
surgical procedures to address PFOA and PF
pain, such as patellectomy, lateral release and/
or realignment procedures. Joint-preserving
surgical treatment generally results in
insufficient, unpredictable, or only short-term
improvement [1]. The first attempt to replace
the PF joint surface was a Vitallium patella
resurfacing prosthesis introduced by McKeever
in 1948 [30]. Concerns about the trochlea in
patella resurfacing led to the development of
PFA as we know it today. In 1975, Lubinus
introduced the patella glide bearing total
prosthesis [31]. The early experiences showed
that the prosthesis did relieve retropatellar pain.
However, reports of medium-term results
described satisfactory, good or excellent results
in only 45% to 64% of cases, prompting some
authors to discontinue the use of this prosthesis
[28, 32]. The unconstrained anatomical implant,
which was narrow and short, was suspected of
making the patella susceptible to malalignment
and impingement [28]. Because of the relative
large radius of the curvature, placing the
trochlear component in flexion was often
necessary, leaving the proximal edge of the
prosthesis offset from the anterior femoral
shaft, resulting in snapping and catching [33].
Blazina
et al.
had started using the Richards
patellofemoral prosthesis in 1974 and reported
about the favorable short-term results of 57
replacements in 1979 [34]. The non-anatomic
trochlear component was highly constrained
with a deep central groove, and the polyethylene
patellar component had a longitudinal ridge. In
reviewing the failures, Blazina
et al.
noted
areas of concern, such as the tracking of the
patellar prosthesis when coming in and out of
the trochlear groove proximally and distally,
resulting in similar catching phenomena as
were seen with the Lubinus design. Suggestions
for a more shallow trochlear groove and adapt
the shape of the patellar component accordingly
were made [34]. Despite these concerns, the
Richards prosthesis has been widely used since
that time. The number of re-operations for
patellar maltracking is high in all reported
series [1, 35]. Recently, we published long-
termoutcomes of the Richards type II prosthesis
in two studies. With a median of 13.3 [1] and
9.2 [36] years of follow-up in 181 [1] and 33
[36] knees; the survival, with revision for any
reason as the endpoint, was determined at 84%
at 10 years and 69% at 20 years [1] and 88%
and 80% respectively [36]. In Table 1, we
summarized these results, which are comparable
to earlier published results on this PFA design
[1, 2, 3, 36].
The current concept of different designs of
PFA is to restore or recreate a PF joint with
preservation of as much bone stalk as possible
and not to make alterations to knee kinematics.
There is a wide diversity of PF implants, from
constrained types to less constraint designs.
All but one PFA prosthesis are cemented and
all consist of a polyethylene patellar component
and a metal (Cobalt-Chromium) trochlea
component. It is beyond the scope of this
overview to review all available PFA designs
in detail. It suffices to state that the problems