Abstract
Background: Modular polyethylene inserts
have enabled surgeons to perform an isolated tibial insert exchange while
retaining well fixed components. The purpose of this study was to
review the results of insert revision and to clarify the role of
this option compared with that of revision total knee arthroplasty.
Methods: Fifty-six patients (sixty-three knees)
were managed with revision of a tibial polyethylene insert and retention
of well aligned and stable femoral and tibial components. The implants
had been in situ for an average of fifty-nine months
(range, two to 108 months) at the time of the insert exchange. The
inserts that were removed at the time of exchange were evaluated
with regard to wear of the articular surface according to the classification system
of Hood et al. and with regard to undersurface wear according to
the method described by Wasielewski et al. Forty-eight knees were
followed for an average of 7.4 years (range, 3.0 to 12.2 years)
after the insert exchange. Knees that did not require an additional
operation were considered to have had a successful exchange.
Results: Seven of the forty-eight exchanges
failed, at an average of fifty-four months, because of accelerated
wear of the new insert. All seven knees required complete revision
of all components. Of the twenty-two exchanges that were performed
because of severe wear of the primary insert, six (27 percent) failed
at an average of less than five years; thus, knees in which the
exchange was performed because of advanced wear were more likely
to fail again (p < 0.05). In addition, primary inserts that were
removed from knees in which the exchange procedure subsequently
failed had higher delamination scores than those that were removed
from knees in which the exchange was successful (p < 0.05).
Most of the primary inserts had substantial undersurface wear at
the time of the exchange procedure. Metallosis (thirty knees) and
osteolysis (nineteen knees) were unrelated to failure of the exchange.
Conclusions: An isolated revision of the tibial
polyethylene insert should not be performed when there is accelerated
wear of the insert with severe delamination and grade-3 or 4 undersurface
wear within ten years after the primary procedure. Because a variety
of patient-related, implant-related, and technical factors influence
polyethylene wear, the orthopaedist must consider multiple variables whenever
contemplating a limited revision.
The introduction of modular polyethylene inserts in the mid-1980s
gave surgeons the option of performing an isolated exchange of the
tibial insert while retaining well fixed components. Exchange of
the bearing surface offers several potential benefits compared with
revision total knee arthroplasty, including maintenance of bone
stock, diminished complexity, and lower cost. In addition, rehabilitation
is easier because an insert exchange involves less time and blood
loss than a complete revision. Moreover, because most patients with
polyethylene wear have minimal symptoms, they typically accept an
insert exchange but are often reluctant to undergo a total knee
revision. Lastly, whereas most failures of so-called first-generation,
condylar-type, nonmodular components occur because of aseptic loosening,
contemporary modular total knee implants rarely loosen1.
The most common indications for revision of modular total knee
implants are polyethylene wear and secondary osteolysis. In such
situations, the orthopaedist must decide whether to retain the stable
components - revising only the polyethylene insert - or to perform
a complete revision. Despite the widespread use of modular implants,
the outcome of isolated insert revision is largely unknown. We sought
to clarify the role of isolated insert exchange compared with that
of complete revision by reviewing the results of modular insert exchanges
that had been performed for a variety of indications.
The senior author (G. A. E.) performed an isolated exchange of
a tibial polyethylene insert with retention of the femoral and tibial
components in fifty-six patients (sixty-three knees) between February 1987
and September 1995. Seven patients had a bilateral insert exchange.
The primary implants had been in situ for an average
of fifty-nine months (range, two to 108 months) at the time of the
exchange. This option was chosen instead of complete revision if
the components were well aligned, showed no evidence of loosening,
and had no palpable roughening of the metallic bearing surfaces.
No other polyethylene insert exchanges were performed during this
interval.
Eight patients (eight knees) died of unrelated problems less
than twenty-four months after the revision (the minimum duration
of follow-up for the study. (However, two of these patients had
had a bilateral exchange and had at least twenty-four months of
follow-up on one side.) Seven patients (seven knees) responded to
a telephone survey only and were excluded from the analysis, although
all reported that they were doing well and had not had a revision.
Forty-three patients (forty-eight knees) had a complete follow-up
after a minimum of thirty-six months (average, 7.4 years; range,
3.0 to 12.2 years). We evaluated the clinical records, the fourteen
by seventeen-inch (thirty-six by forty-three-centimeter) anteroposterior
and lateral radiographs, and the failed components of these patients.
The average age of the patients was sixty-three years (range,
thirty-seven to eighty-four years) at the time of the primary total
knee arthroplasty and sixty-eight years (range, forty-three to ninety years)
at the time of the insert exchange. Twenty-three patients (twenty-eight
knees) were male and twenty patients (twenty knees) were female.
The underlying diagnosis was osteoarthritis in forty-one knees,
rheumatoid arthritis in six, and septic arthritis in one. The average
body-mass index of all patients was twenty-nine kilograms per square meter
(range, twenty to forty-three kilograms per square meter).
The senior author performed forty-five of the forty-eight primary
total knee arthroplasties that subsequently were revised with an
isolated insert exchange. During the same eight-year period, this surgeon
performed a total of 1298 primary total knee arthroplasties. Therefore,
3.5 percent of the primary total knee arthroplasties that were performed
by this surgeon subsequently were revised with an insert exchange.
The implants that were used for the primary arthroplasties included
118 Porous-Coated Anatomic implants (PCA; Howmedica, Rutherford,
New Jersey), 204 Synatomic implants (DePuy, Warsaw, Indiana), and
976 Anatomic Modular Knee implants (AMK; DePuy). The senior author
performed 275 knee revisions during this time. Consequently, the
forty-eight insert exchanges represented 17 percent of the revision
procedures that were performed by this surgeon.
The main indications for revision of the tibial insert were severe
polyethylene wear seen on radiographs (twenty patients, twenty-two
knees) (Fig. 1) and wear and damage of the insert noted during revision
of a failed metal-backed patellar component (nineteen patients,
twenty-one knees) (Fig. 2). (One patient with bilateral knee replacement underwent
insert exchange because of severe polyethylene wear in one knee
and because of a failed metal-backed patellar component in the other knee.)
The remaining five exchanges were performed in conjunction with
d衲idement for acute infection (two patients, two knees), treatment
of arthrofibrosis (one knee), repair of a ruptured quadriceps tendon
(one knee), and treatment of patellar dislocation (one knee).
Polyethylene wear was identified on standing anteroposterior
radiographs without magnification markers. In sixteen knees, extensive
polyethylene wear was indicated by a reduction in the space between
the femoral component and the tibial base-plate to less than two
millimeters. Despite the polyethylene wear seen on radiographs,
all patients were asymptomatic. The indication for revision when
wear-through had not occurred was the anticipation that wear-through
was pending and would damage the femoral component, necessitating
total knee revision. Five other patients (five knees) had full-thickness
wear of the insert and reported stiffness, low-grade pain, and catching
or giving-way of the knee.
Prior to the revision of the insert, osteolysis had been detected
on the anteroposterior and lateral radiographs of ten knees that
were revised because of severe polyethylene wear and on the radiographs
of nine knees that were revised because of failure of a metal-backed
patellar component. The five knees that were revised for other reasons
had not exhibited osteolysis.
The forty-eight knees in the present study had been treated with
a variety of modular implants at the time of the primary procedure;
these implants included twenty-three Synatomic implants, fifteen AMK
implants, eight PCA implants, one Miller-Galante I implant (Zimmer,
Warsaw, Indiana), and one Coordinate implant (DePuy). In all knees
but one, which had an AMK implant, both the tibial and the femoral
component were fixed without cement. The Synatomic and Coordinate
tibial inserts were curved in both the anteroposterior and the mediolateral
plane. The AMK inserts were curved in the anteroposterior plane.
The PCA and Miller-Galante inserts were flat in both planes. The
Synatomic, AMK, and Coordinate tibial inserts were secured with
a tongue-in-groove locking mechanism and were further stabilized
with a central locking pin. The PCA inserts were captured and secured
with a locking tab on the tibial tray. The Miller-Galante I insert
was secured with a snap-fit with a full peripheral capture on the
tibial tray. All of the primary and revision inserts were sterilized with
low-dose gamma radiation in air. The primary and revision Synatomic,
AMK, and Coordinate inserts were machined from ram-extruded bar stock,
the PCA inserts were heat-pressed, and the Miller-Galante I inserts
were compression molded. Seven inserts were the largest size available from
the manufacturer, and none were the smallest size.
At the time of the index procedure (the insert exchange), twenty-five
metal-backed patellar components were revised, eight patellar components were
left in situ, three patellar components were removed
without resurfacing the patella, and two patellae were resurfaced
for the first time.
All knees had an extensive synovectomy at the time of the insert
exchange, and thirty (63 percent) had clearly visible metallosis.
The twenty-one knees with a failed metal-backed patellar component
had discoloration of the synovial tissue secondary to metallic debris.
Another five knees had metallosis because severe polyethylene wear
had resulted in metal-on-metal contact between the femoral component
and the tibial base-plate. This contact had caused faint surface
scratching and slight burnishing of the femoral component; however,
none of these five components had palpable surface roughness. Four
other knees had metallosis because the femoral intercondylar notch
impinged on the locking pin that was used to stabilize the Synatomic
tibial component. These four knees were revised with use of an insert
that had a lower central eminence and a recessed pin that did not impinge
and wear against the intercondylar notch of the femoral component.
The average thickness of the tibial insert increased from ten
millimeters (range, seven to eighteen millimeters) at the time of
the primary procedure to twelve millimeters (range, nine to twenty-five
millimeters) at the time of the exchange procedure. In each knee,
the thickness of the new insert was determined intraoperatively
to provide balanced flexion and extension gaps and varus-valgus
stability following appropriate ligament releases. The thickness
of the polyethylene insert remained the same in seven knees. It
increased by two millimeters in thirty knees and by four millimeters
in eleven knees. A Hylamer-M insert was selected for the exchange
procedure in four of the forty-eight knees; all four of these knees
were in the group in which the revision was performed because of
failure of the original tibial insert.
Forty-two (88 percent) of the forty-eight revised tibial inserts
were evaluated with regard to region and mode of wear. (The other
six inserts had undergone destructive testing in an unrelated study without
having been graded for articular surface wear.) The inserts were
divided into anterior and posterior regions, which were subdivided
into medial, central, and lateral zones. The six zones were assessed
with regard to the seven modes of wear defined by Hood et al.6: surface deformation, pitting, embedded
cement debris, scratching, burnishing, abrasion, and delamination.
Each mode of wear in each zone was assigned a severity score of
0 points (no wear), 1 point (less than 10 percent wear), 2 points (10
to 50 percent wear), or 3 points (more than 50 percent wear). The
overall wear score was the sum of the scores for the individual
zones; thus, the maximum wear score was 126 points.
The same forty-two inserts were evaluated with regard to wear
of the undersurface (so-called backside wear) and evidence of plastic
deformation. The inserts were graded according to the method described
by Wasielewski et al.16,17, which
assesses polyethylene cold flow into the screw-holes on the tibial
component (Fig. 3). Five grades of wear were possible: grade 0 (no
screw-hole impression visible), grade 1 (partial screw-hole impression
visible), grade 2 (entire perimeter of screw-hole impression visible),
grade 3 (palpable screw-hole impression, 0.10 to 0.49 millimeter deep,
with abrasive wear completely eliminating the surrounding machine
lines), and grade 4 (screw-hole impression at least 0.50 millimeter deep
or severe linear abrasive wear due to gross relative motion between
the insert and the metal backing). The screw-hole impressions were
measured with a digital caliper (Mitutoyo Digimatic Caliper, Kanagawa,
Japan) to the nearest 0.01 millimeter. The overall grade was the
highest grade of all screw-hole impressions. PCA implants, which have
only one central screw-hole for tibial tray stabilization, were
subjectively evaluated with regard to undersurface wear by examination
of the extent of polishing of machine marks and polyethylene cold
flow around the outer edges. Lastly, wear patterns of the inserts
that initially failed were compared with those of their replacements
that later failed.
Data were analyzed with use of SPSS for Windows statistical software
(version 8.0; SPSS, Chicago, Illinois). Fisher's exact test was
used to detect significant differences with regard to osteolysis,
gender, and implant fixation between the group in which the exchange
failed and the group in which the exchange was successful. The nonparametric
Mann-Whitney U test was used to compare the two groups with regard
to the average patient height, the overall wear score, the delamination
score, and the tibiofemoral angle. Kaplan-Meier analysis with log-rank
testing was used to examine the survivorship of the replacement
inserts in knees that had had the index procedure because of polyethylene
wear and that of the inserts in knees that had had the procedure
because of failure of a metal-backed patellar component.
Isolated exchange of the tibial polyethylene insert failed in
eight (17 percent) of the forty-eight knees at an average of fifty-four
months (range, fifteen to ninety-six months). All eight knees subsequently were
treated with total knee revision. The exchange failed in six (27
percent) of the twenty-two knees in which the index procedure had
been performed because of accelerated polyethylene wear; all six
failures were due to advanced wear of the revision insert after
an average of fifty-four months (range, thirty-seven to ninety-six
months) in situ. The exchange failed in one (5
percent) of the twenty-one knees in which the index procedure had
been performed because of the failure of a metal-backed patellar
component; this knee had a repeat revision ninety-two months after
the index procedure because of wear of the revision insert and osteolysis.
Therefore, seven of the eight failures were due to rapid wear of
the new insert. The eighth failure occurred, fifteen months after the
insert exchange, because of recurrent rotational instability and
secondary patellofemoral subluxation. This patient was excluded
from further analyses because the failure was unrelated to polyethylene
wear.
Kaplan-Meier survivorship analysis revealed that at six years
the probability of survival of the insert was 100 percent after
insert exchanges that had been performed because of the failure
of a metal-backed patellar component. After procedures that had
been performed because of polyethylene wear, survival was only 64
percent. This difference was significant (p < 0.05, log-rank
test) (Fig. 4).
Forty knees (83 percent) did not require an additional operation
after the insert exchange. At the time of the most recent follow-up
examination, performed at an average of eighty-six months (range,
thirty-five to 146 months), these forty knees had an average flexion
contracture of 1.6 degrees (range, 0 to 20 degrees) and an average
flexion arc of 110 degrees (range, 60 to 135 degrees). Thirty-six
knees had a good or excellent clinical score according to the classification
system of the Knee Society8 or that of The
Hospital for Special Surgery7.
Three of the remaining four knees had a fair score because of limited
function with advanced age and generalized arthritis. The fourth
knee continued to have a poor knee score and a very limited range
of motion (from a 20-degree flexion contracture to 60 degrees of
flexion) after insert exchange and excision of scar tissue for severe
fibrous ankylosis.
Factors Associated with
Failure of the Exchange
Six of the seven patients who subsequently required a total knee
revision because of polyethylene wear were male. The patients who
had a failed exchange were taller than those who had a successful
exchange (average height, 178 compared with 160 centimeters; p =
0.057), and they were marginally younger at the time of the index
procedure (average age, sixty-four compared with sixty-eight years;
p > 0.05). Insert failure was not related to implant design.
We also examined the relationship between evidence of metallosis
at the time of the index procedure and subsequent failure of the
exchange. Metallosis was observed in five knees in which a PCA insert
was exchanged because of full-thickness wear-through to the tibial
base-plate. The retained cobalt-chromium femoral components had slight
visible scratching and no visible burnishing. In three of the five
knees, the revision insert had accelerated wear and a wear pattern
that was similar to that of the original insert. The twenty-one knees
that had gross metallosis due to a failed metal-backed patellar
component had mild scratching of the femoral component in the region
of the patellar groove. However, in this group of twenty-one knees,
no association was found between gross metallosis and failure. Similarly,
no such association was found in the group of four knees that had
metallosis due to impingement of the intercondylar eminence. Overall,
only one knee that had metallosis for a reason other than full-thickness
wear of the insert required an additional operation after the insert
exchange.
Articular Surface and Undersurface Wear
The forty-two primary inserts that were examined with regard
to region and mode of articular surface wear had an average wear
score of 21 points (range, 2 to 43 points) of a possible 126 points. The
average wear score for the primary inserts in knees in which the
exchange failed was 25 points (range, 8 to 38 points), which was
slightly higher than the wear score for the primary inserts in knees in
which the exchange was successful (21 points; range, 10 to 43 points)
(p = 0.22).
The average delamination score for all forty-two primary inserts
was 7 points (range, 0 to 17 points) of a possible 18 points. The
average delamination score for the primary inserts in knees in which
the exchange failed (12 points) was significantly higher than that
for the primary inserts in knees in which the exchange was successful
(7 points) (p < 0.05). The average delamination score for the
five inserts in knees that were revised for a reason other than
polyethylene wear or failure of a metal-backed patellar component
was only 1 point (0, 1, or 2 points).
Most of the primary inserts had undersurface wear at the time
of the exchange procedure. Specifically, thirty-five (83 percent)
of the forty-two primary inserts had grade-3 or 4 wear and seven
had grade-0, 1, or 2 wear. No significant difference in the undersurface
wear of the primary insert could be detected between the knees in
which the exchange failed and those in which it was successful;
however, none of the seven knees in which the primary insert had
grade-0, 1, or 2 wear went on to have failure of the revision insert.
When the primary inserts that failed were compared in a visual
manner with their replacements that subsequently failed, the two
inserts in each pair usually revealed similar wear patterns. Six matched
pairs of inserts were available for comparison. Five pairs were
from knees in which both the primary and the revision insert failed
because of accelerated wear. This group included two pairs in which
the two inserts had symmetrical wear patterns, two pairs in which
the two inserts had identical medial wear patterns, and one pair
in which the two inserts had matching lateral wear patterns. The
sixth matched pair was from a knee that initially was revised because
of the failure of a metal-backed patellar component and subsequently
was revised because of severe wear and osteolysis. In this pair,
the initial insert had symmetrical wear but the revised insert had
mostly medial wear.
Radiographic Findings
The average tibiofemoral angle was 4 degrees of valgus (range,
1 to 6 degrees of valgus) in the seven knees in which the exchange
failed and 6 degrees of valgus (range, 1 to 13 degrees of valgus) in
the forty knees in which the exchange was successful. The tibiofemoral
angle was between 1 and 10 degrees of valgus for all but two of
the forty-seven knees. Both of the malaligned knees, which had tibiofemoral
angles of 12 and 13 degrees of valgus, had a successful insert exchange.
In the seven knees in which the exchange failed, the average
angle between the tibial component and the tibial axis was 2 degrees
of varus (1, 2, or 3 degrees of varus) and the average angle between the
femoral component and the femoral axis was 6 degrees of valgus (5,
6, or 7 degrees of valgus). In the forty knees in which the exchange
was successful, the average angle between the tibial component and
the tibial axis was 1 degree of varus (range, 4 degrees of varus
to 1 degree of valgus) and the average angle between the femoral
component and the femoral axis was 8 degrees of valgus (range, 3
to 13 degrees of valgus).
Nineteen knees had osteolytic lesions at the time of the insert
exchange. Only one of these knees had a complication secondary to
the osteolysis. Fourteen of the nineteen knees had a successful
insert exchange, with no evidence of osteolytic progression. The
other five knees had failure of the exchange. In two of these five
knees, the osteolytic defects healed after being treated with morseled allograft
bone at the time of total knee revision. In another two knees, the
osteolytic lesions showed no evidence of progression after treatment
with curettage. The fifth knee, which received one of the four Hylamer-M
inserts at the time of the exchange, had progressive femoral osteolysis.
This patient sustained a supracondylar fracture that necessitated
two femoral head allografts with total knee revision.
The rates of polyethylene wear and osteolysis - increasingly
common mechanisms of failure in patients who have had a total knee
arthroplasty2-5,9,11,14,15 - have
been on the rise since the advent of modular knee designs, as evidenced
by the paucity of clinical case reports published before 199213. It is clear that revision arthroplasty
is necessary when the tibial and femoral components are loose, but
the solution is not as obvious when the tibial and femoral components
are well fixed and the polyethylene fails. The surgeon must decide
when it is appropriate to revise stable tibial or femoral components
and when it is appropriate to simply exchange the polyethylene insert.
The findings of this report begin to clarify the indications
for polyethylene insert exchange with retention of the femoral and
tibial components. Our results suggest that polyethylene insert
exchange should not be performed when the primary insert has (1)
severe wear within ten years after insertion, (2) severe delamination,
(3) full-thickness wear-through, or (4) extensive backside wear.
However, because this study included a limited number of knees (forty-eight)
and a small number of failed exchanges (eight), additional studies
are necessary to confirm these impressions.
Wear of a modular polyethylene tibial insert is influenced by
a combination of implant characteristics (polyethylene thickness,
type of locking mechanism, and design and finish of the bearing surfaces)
and patient-related factors (age, weight, height, activity level,
length of time with the implant in situ, and component
alignment). Insert exchange can address only some of the implant
features that contribute to accelerated polyethylene wear. New inserts
could be thicker5, manufactured
from machined or molded polyethylene instead of heat-pressed material,
or sterilized without use of gamma radiation in air; however, surgeons
must determine whether changing such variables would reduce wear
or if other variables (such as micromotion of the tibial tray, ligamentous
balance, third-body wear, and damaged or poorly functioning locking
mechanisms) still would make the results of insert exchange unpredictable.
The assessment of the failed primary inserts provided insight
into the polyethylene wear characteristics that predispose an insert
exchange to failure. The inserts that were removed from knees in which
the exchange subsequently failed had notable backside wear and higher
delamination scores than those that were removed from knees in which the
exchange was successful, suggesting that surgeons should assess
delamination of the articulating surface and backside wear of the
polyethylene insert when considering an isolated insert exchange.
Wasielewski et al.17 identified
backside wear as a major source of polyethylene debris contributing
to tibial osteolysis. Just as a damaged or flawed locking mechanism
permits micromotion and subsequent wear, micromotion of an insert
against an unpolished tibial tray (as occurred in the report by
Wasielewski et al. and the present series) accelerates wear. The
potential for substantial micromotion of the tibial insert has been
reported in association with a variety of snap-fit and tongue-in-groove
tibial tray capture mechanisms12.
So-called counterface roughness of the femoral component also
greatly influences polyethylene wear2.
Scratches on the femoral component increase surface roughness and
accelerate the wear of the polyethylene insert. Dwyer et al.2 found that retrieved femoral components
that had been revised in association with severe polyethylene wear
had significantly greater surface roughness than control components
that had not been implanted (p < 0.005). With use of scanning
electron microscopy, Dwyer et al. detected scratches, grooves, and
particulate matter within the surface defects of the retrieved components.
However, it remains unclear how much damage necessitates a complete
revision of a well fixed femoral component to avoid the accelerated
wear that is associated with the retention of a damaged component.
Femoral components that have metal-on-metal contact because of
full-thickness wear-through of the insert are of special concern.
In the present study, three of five knees that were revised in the presence
of full-thickness wear-through of the primary insert had subsequent
failure of the revision insert. Our experience has indicated that
when a femoral component has articulated with a metal tibial base-plate
it is likely that the component is damaged and that the new insert
will have accelerated wear. In contrast, Knight et al.10 reported no adverse effects after
retention of eleven femoral components that were associated with full-thickness
wear-through and seven femoral components that had burnishing because
of metal-on-metal contact. However, the short duration of follow-up
in that study (average, twenty-four months; range, five to thirty-six
months) prevents definitive conclusions from being drawn.
The presence of metallosis in association with a failed metal-backed
patellar component would seem to necessitate total knee revision.
Prior to insert revision, more than 50 percent (twenty-six) of the
forty-eight knees in the present study had extensive metallosis
with the inherent potential for abrasive third-body wear. The most
frequent sources of metal debris in these knees were a failed metal-backed
patellar component (twenty-one knees) and metal-on-metal contact
of the femoral and tibial components (five knees). Most of the knees
were treated with insert exchange combined with revision of the
metal-backed patellar component, and only one knee subsequently
had failure of the revision insert. Additional studies are needed
to refine the indications and contraindications for retaining components
that have been exposed to third-body debris.
The most noteworthy patient-related factors associated with accelerated
implant wear in this study were gender and height. Six of the seven
insert exchanges that failed because of insert wear were in male
patients. Heck et al.5 reported
a preponderance of male patients in their study of twelve knees
with gross polyethylene failure. Those authors also found that patients
who had polyethylene failure were significantly taller than those
who did not (average height, 176 compared with 161 centimeters;
p = 0.003); this finding was duplicated in our study (p = 0.057).
These findings may be evidence of the association between the variables
of gender and height.
Abnormal loading patterns also can contribute to increased rates
of polyethylene wear. For instance, if the knee remains malaligned
and imbalanced, the revision insert will be subjected to the same forces
that contributed to the accelerated wear of the primary insert.
In our study, five of the six matched pairs displayed similar wear
patterns on both the primary and the revision insert.
As an alternative to complete revision in knees that have reasonable
but suboptimal alignment in the presence of well fixed components,
Shaw14 suggested revising the
polyethylene insert with one that is thicker on the undercorrected
side of the deformity. It is possible that this option could reduce
wear by correcting alignment and thereby changing the kinematics
and stresses on the polyethylene insert. Larger sample sizes and
longer-term follow-up are needed to validate the efficacy of such
custom bearings.
An important factor in predicting the success of a revision insert
seems to be the length of time that the primary insert performed
acceptably before failing. A replacement insert likely will perform well
for a reasonable period if the initial insert did the same. We concluded
that insert exchange with retention of the tibial and femoral components may
be appropriate when the primary insert has had acceptable wear performance.
We therefore believe simple insert exchange is indicated for patients
who have late failure of the primary tibial insert. We also revise
inserts that have mild-to-moderate wear when knee surgery is being
performed for other reasons.
Conversely, a replacement insert is likely to show wear within
a relatively short time if the initial insert failed rapidly. We
are reluctant to perform an isolated insert exchange for patients
who have polyethylene wear (particularly delamination) within a
relatively short time, have full-thickness wear of the insert, or
have advanced undersurface wear of the insert or damage to the tibial
base-plate (both of which are suggestive of design-specific problems
with the locking mechanism of the tibial tray). On occasion, we
perform an isolated insert exchange for an elderly, low-demand patient
in order to avoid a more complex procedure involving revision of
well fixed tibial and femoral components. However, whenever a damaged
femoral component is retained or a potentially poor modular tibial
locking mechanism is present with resultant backside wear, increased
clinical and radiographic surveillance is necessary in order to identify
extensive polyethylene wear and osteolysis.
Bugbee, W. D.; Ammeen, D. S.; Parks, N. L.; and Engh, G. A.: 4- to 10-year results with the anatomic modular total
knee. Clin. Orthop.,348: 158-165, 1998.348158
1998
[PubMed]
Dwyer, K. A.; Topoleski, L. D. T.; Bauk, D. J.; Nakielny, R.; and Engh, G. A.: The neglected side of the wear couple: analysis of surface
morphology of retrieved femoral components. Trans. Orthop. Res. Soc.,18: 82, 1993.1882
1993
Engh, G. A.; Dwyer, K. A.; and Hanes, C. K.: Polyethylene wear of metal-backed tibial components in
total and unicompartmental knee prostheses. J Bone Joint Surg,74-B(1): 9-17, 1992.74-B(1)9
1992
Feng, E. L.; Stulberg, S. D.; and Wixson, R. L.: Progressive subluxation and polyethylene wear in total
knee replacements with flat articular surfaces. Clin. Orthop.,229: 60-71, 1994.22960
1994
Heck, D. A.; Clingman, J. K.; and Kettelkamp, D. G.: Gross polyethylene failure in total knee arthroplasty. Orthopedics,15: 23-28, 1992.1523
1992
[PubMed]
Hood, R. W.; Wright, T. M.; and Burstein, A. H.: Retrieval analysis of total knee prostheses: a method
and its application to 48 total condylar prostheses. J. Biomed. Mater. Res.,17: 829-842, 1983.17829
1983
[PubMed]
Insall, J. N.; Ranawat, C. S.; Aglietti, P.; and Shine, J.: A comparison of four models of total knee replacement
prostheses. J Bone Joint Surg,58-A: 754-765, Sept. 1976.58-A754
1976
Insall, J. N.; Dorr, L. D.; Scott, R. D.; and Scott, W. N.: Rationale of the Knee Society clinical rating system. Clin. Orthop.,248: 13-14, 1989.24813
1989
[PubMed]
Jones, S. M. G.; Pinder, I. M.; Moran, C. G.; and Malcolm, A. J.: Polyethylene wear in uncemented knee replacements. J Bone Joint Surg,74-B(1): 18-22, 1992.74-B(1)18
1992
Knight, J. L.; Gorai, P. A.; Atwater, R. D.; and Grothaus, L.: Tibial polyethylene failure after primary porous-coated
anatomic total knee arthroplasty. Aids to diagnosis and revision. J. Arthroplasty,10: 748-757, 1995.10748
1995
[PubMed]
Lewis, P.; Rorabeck, C. H.; Bourne, R. B.; and Devane, P.: Posteromedial tibial polyethylene failure in total knee
replacements. Clin. Orthop.,299: 11-17, 1994.29911
1994
[PubMed]
Parks, N. L.; Engh, G. A.; Topoleski, L. D. T.; and Emperado, J.: Modular tibial insert micromotion. A concern with contemporary
knee implants. Clin. Orthop.,356: 10-15, 1998.35610
1998
[PubMed]
Peters, P. C., Jr.; Engh, G. A.; Dwyer, K. A.; and Vinh, T. N.: Osteolysis after total knee arthroplasty without cement. J Bone Joint Surg,74-A: 864-876, July 1992.74-A864
1992
Shaw, J. A.: Angled bearing inserts in total knee arthroplasty. A brief
technical note. J. Arthroplasty,7: 211-216, 1992.7211
1992
[PubMed]
Tsao, A.; Mintz, L.; McRae, C. R.; Stulberg, S. D.; and Wright, T.: Failure of the porous-coated anatomic prostheses in total
knee arthroplasty due to severe polyethylene wear. J Bone Joint Surg,75-A: 19-26, Jan. 1993.75-A19
1993
Wasielewski, R. C.; Galante, J. O.; Leighty, R. M.; Natarajan, R. N.; and Rosenberg, A. G.: Wear patterns on retrieved polyethylene tibial inserts
and their relationship to technical considerations during total
knee arthroplasty. Clin. Orthop.,229: 31-43, 1994.22931
1994
Wasielewski, R. C.; Parks, N.; Williams, I.; Surprenant, H.; Collier, J. P.; and Engh, G.: Tibial insert undersurface as a contributing source of
polyethylene wear debris. Clin. Orthop.,345: 53-59, 1997.34553
1997
[PubMed]