Early generations of alumina ceramic bearings frequently were associated
with mechanical loosening of the socket and ceramic implant fracture, and
there were sporadic cases of excessive wear of the bearing surfaces. These
problems were attributed to suboptimal implant design, poor quality control,
and the intrinsic brittle nature of ceramic
materials1-3.
Alumina ceramic, now in its third generation, has been markedly improved in
terms of its mechanical properties, including purity, grain microstructure,
and burst strength, through the evolution of design features and manufacturing
processes over the past three decades and through the introduction of
proof-testing4-6.
Moreover, many different methods of socket fixation have been investigated in
order to obtain a secure link between the alumina insert and the
acetabulum2,7-9.
As a result, a press-fit metal-backed socket has been incorporated into the
majority of contemporary alumina-on-alumina hip prosthesis designs. Recent
clinical studies on improved alumina bearings in combination with modern
implant designs have shown very low wear rates and substantially reduced rates
of ceramic fracture and component
loosening10-14.
These favorable results have led to the increased use of an alumina-on-alumina
bearing as an alternative to a conventional metal-on-polyethylene
articulation, particularly for young, active patients. Nonetheless, because of
the lack of ductility of alumina ceramic, there is a concern regarding the
risk of fracture during insertion or in vivo
use15,16.
A ceramic bearing design incorporating a polyethylene-alumina composite liner
within a titanium-alloy shell was introduced in an attempt to address this
problem17.
We conducted a multicenter review of primary total hip arthroplasties that
had been performed with use of a polyethylene-ceramic composite liner combined
with a ceramic femoral head, with particular attention to failure of the
ceramic bearing.
From November 1998 to December 2001, 329 patients (367 hips) who had
undergone primary cementless total hip arthroplasty with an alumina-on-alumina
bearing with use of the SPH Contact acetabular component (Lima-Lto, Udine,
Italy) incorporating a polyethylene-ceramic composite liner within a
titanium-alloy shell were identified at four participating academic
institutions. Of these, eight patients (eight hips) had been lost to follow-up
and two patients (two hips) had died for reasons unrelated to the
arthroplasty, leaving 319 patients (357 hips) available for complete analysis.
All ten hips that were dropped from the study had been performing well and had
showed no evidence of prosthesis-related problems at the time of the last
follow-up. The design and protocol of this multicenter retrospective study
were approved by the institutional review board at each center, and all
patients provided informed consent.
The indications for the use of this composite alumina liner coupled with an
alumina femoral head varied among the centers. Two centers used this bearing
for the majority of patients who had a primary total hip arthroplasty, whereas
the other two centers reserved it for patients who were less than fifty years
old and who were engaged in a high level of activity. The average age of the
patients at the time of the index arthroplasty was 51.4 years (range, nineteen
to seventy-four years), and 195 (61.1%) of the patients were less than fifty
years old. The average duration of follow-up was 46.5 months (range,
thirty-six to seventy-two months). The demographic features of the patients,
including age, gender, weight, body-mass index, and primary diagnosis, are
presented in Table I.
One surgeon at each center (Y.-S.P., S.-K.H., W.-S.C., and Y.-S.K.)
performed all of the surgical procedures. The operative approach was
anterolateral in 230 hips (64.4%) and posterolateral in 127 hips (35.6%). All
patients received antibiotic prophylaxis perioperatively, and a standard
protocol of low-dose warfarin for prophylaxis against thromboembolism was used
selectively in high-risk patients. All patients were allowed partial
weight-bearing on the second or third postoperative day and full
weight-bearing after four to six weeks.
All patients who were enrolled in this series received the SPH Contact
acetabular component. This modular component consists of a preassembled,
polyethylene-alumina composite liner that is held in a titanium-alloy shell.
The three components are assembled into each other reciprocally by means of a
Morse taper system (Fig. 1).
The titanium-alloy shell contains screw-holes, slanted circumferential grooves
in the equatorial area, and a surface coating with plasma-sprayed pure
titanium. The socket has an outer diameter ranging from 46 to 62 mm, and in
each case it was press-fit into a 2-mm under-reamed acetabulum. If the socket
fixation was not rigid, one or more screws were inserted. All alumina inserts
had identical internal dimensions and were of the same thickness (4.8 mm),
whereas mating polyethylene shells varied in thickness to fit the
titanium-alloy shell.
On the femoral side, all patients who were enrolled in this series received
either a C2 stem (Lima-Lto) or an F2L stem (Lima-Lto); both were fixed without
cement after broaching with a rasp, and a 28-mm alumina head was always used.
The C2 stem is a straight, rectangular, tapered, titanium-alloy implant with a
rough, sandblasted surface. The neck-shaft angle is 131°. The neck has a
polished metal surface and a circular cross-sectional geometry. The diameter
of the taper is 12/14. The F2L stem is a modular, multi-necked, tapered,
titanium-alloy implant with a proximal hydroxyapatite coating. The neck-shaft
angle can be changed from 125° to 135° with use of modular neck
components, which are made in three types (straight, lateralized, and
anteverted) and two different lengths (short and long). The modular neck has a
polished metal surface and a circular cross-sectional geometry. The diameter
of the taper is 12/14. The C2 stem was implanted in 282 hips (79%), and the
F2L stem was implanted in 75 hips (21%). All ceramic implants were hot
isostatic pressed, laser-marked, and proof-tested third-generation BIOLOX
forte alumina (CeramTec, Plochingen, Germany).
Clinical and radiographic evaluations were performed at six weeks, three
and six months, and one year after surgery and then annually thereafter. Forms
with items related to the demographic characteristics of the patients, the
operative procedure, details regarding the implant, and intraoperative and
postoperative complications were completed by the contributing surgeons at
each center. Serial radiographs were examined by an independent observer who
was not involved in the clinical care of the patients at each center; only one
of these observers (S.-J.L.) was an author of the present study. Six-week
anteroposterior, frog-leg lateral, and cross-table lateral radiographs were
considered as the baseline for all comparisons. The radiographs were made from
a standard distance and with standard positioning of the patient. The position
of the acetabular component was determined according to the method of Woo and
Morrey18.
Anteversion was measured on the true lateral radiograph as the angle formed by
a line drawn tangential to the face of the acetabular component and a line
drawn perpendicular to the horizontal plane. Abduction was measured on the
anteroposterior pelvic radiograph as the angle formed by lines drawn
tangential to the acetabular component and tangential to the inferior margins
of the ischial tuberosities.
During the study period, ceramic failure occurred in six (1.7%) of the 357
hips that were evaluated. All six of the patients underwent an urgent revision
operation, and intraoperative findings were reported by each contributing
surgeon. All of the explanted alumina implants were examined by means of
visual inspection and with use of scanning electron microscopy equipped with
energy-dispersive x-ray spectrometry (XL30 ESEM-FEG; Philips, Eindhoven, The
Netherlands).
Statistical analysis was done with use of the SPSS statistical software
system (version 11.5; SPSS, Chicago, Illinois). The demographic data for the
overall group and the group with ceramic failure were compared with use of the
Mann-Whitney U test for continuous variables and the chi-square test (or, when
necessary, the Fisher exact test) for dichotomous values. The level of
significance was set at p < 0.05.
Two femoral heads fractured postoperatively at seven and eight months, and
four alumina liners fractured after an average of 36.8 months (range, sixteen
to fifty-eight months) in vivo. No evidence of fracture of the ceramic
components was seen at the time of the primary procedure as the alumina liner
and head were scrutinized after impaction. All of these failures occurred
during normal activities of daily living and were not related to unusual
traumatic events. All six of the ceramic failures occurred in men, and the
primary diagnosis leading to the hip arthroplasty was osteonecrosis in all six
hips. The average age and weight of the patients at the time of the index
arthroplasty was 51.4 years (range, nineteen to seventy-four years) and 61.4
kg (range, 41 to 86 kg), respectively, in the overall group and 50.0 years
(range, twenty-nine to sixty-four years) and 66.7 kg (range, 54 to 80 kg),
respectively, in the ceramic failure group. The average abduction angle and
anteversion of the acetabular component were 44.6° (range, 27° to
58°) and 14.1° (range, 0° to 23°), respectively, in the
overall group and 41.0° (range, 30° to 49°) and 12.7° (range,
8° to 16°), respectively, in the ceramic failure group. With the
numbers available, there was no significant difference between the overall
group and the ceramic failure group with regard to the gender (p = 0.091), age
(p = 0.433), or weight (p = 0.751) of the patients; the primary diagnosis (p =
0.141); or the abduction angle (p = 0.261) or anteversion (p = 0.270) of the
acetabular component. Data characterizing the six alumina failures are
summarized in Table II.
At the time of revision surgery, composite acetabular liners and modular
femoral heads were removed and component fixation was assessed. All acetabular
and femoral components were found to be well-fixed and therefore were left in
place after an attempt at complete débridement and synovectomy to
remove as much of the alumina debris as possible. In two hips, a
cobalt-chromium head was inserted on the scratched Morse taper and a
polyethylene liner was inserted to replace the damaged polyethylene-alumina
composite liner. A new alumina-on-alumina bearing was implanted in the
remaining four hips, which had no macroscopically apparent damage to the Morse
taper. Of the six hips that were treated with modular exchange of the
composite acetabular liner and femoral head, one underwent repeat revision of
an unrevised acetabular cup because of recurrent dislocation after revision
surgery with a new cobalt-chromium head and a polyethylene liner for the
treatment of alumina head fracture. The remaining five hips did not require an
additional reoperation, and the four revision alumina heads did not fracture
during the remainder of the follow-up period (range, six to thirty-six
months).
Intraoperative examination of the two hips with fracture of the ceramic
femoral head (Cases 1 and 2) showed multiple alumina head fragments of varying
sizes as well as a black stain and curvilinear crack on the articulating
surface of the alumina insert (Fig.
2). The top surface of the Morse taper was scratched in both hips.
These findings indicated that the alumina insert had been contacted by the
metal taper following the fracture of the alumina head.
The four hips with fracture of the ceramic liner (Cases 3 through 6) had
multiple small alumina liner fragments along with slight black staining of the
surrounding tissues. The alumina inserts had dissociated from the polyethylene
shells (Fig. 3-A), which
remained well-seated within the metal cup. The superior portions of the
polyethylene shells were grossly deformed. The retrieved alumina inserts
showed extensive rim fractures in an arc of varying degrees and a conspicuous
black stain on the surface of the unbroken rims
(Fig. 3-B). Scanning electron
microscopy revealed multiple pits on the inner edge of the unbroken rims
(Fig. 3-C), and
energy-dispersive x-ray spectrometry demonstrated that the surface-deposited
materials on the black-stained areas were titanium particles. The surfaces of
the retrieved alumina heads showed some scratches and cracks on visual
inspection and on scanning electron microscopy (see Appendix). A deep notch
corresponding to impingement was identified on the neck of the femoral
component in one case (see Appendix), but the top of the Morse taper was not
damaged macroscopically in any case.
Recent clinical studies on the current generation of alumina-on-alumina hip
prostheses have shown mostly excellent results at intermediate-term
follow-up10-14,
and some manufacturers have emphasized that the risk of alumina chipping or
fracture seems to be a negligible problem. Our results did not, however,
support these observations because a relatively high rate of alumina bearing
surface failure (1.7%, six of 357) was identified at short-term follow-up. Our
findings were similar to those in sporadic cases in which alumina liner
fracture has been reported without trauma after total hip arthroplasty with
use of a polyethylene-alumina composite
liner19,20.
All four of the alumina liners that were retrieved in our series showed
evidence of rim contact with the metal neck of the femoral component, as
demonstrated by extensive rim fractures as well as multiple pits and the
conspicuous black stain on the surface of the unbroken rim. Composition
analysis of the black-stained areas with use of scanning electron microscopy
with energy-dispersive spectroscopy confirmed that the surface-deposited
materials were titanium particles that had been transferred from the femoral
component.
We believe that impingement occurs in some hips, especially those in which
the components are suboptimally positioned and those with a high range of
motion. Furthermore, repetitive impingement can occur during normal activities
of daily living, particularly in Asian patients, who frequently squat or sit
in the cross-legged position. Evidence of femoral neck-to-acetabular rim
contact has been recognized as a common occurrence following total hip
replacement, with impingement being seen in 39% of 111 retrieved polyethylene
acetabular liners in one
report21. We
believe that interposing a polyethylene layer between the alumina liner and
the metal shell provides little benefit, except for a reduced risk of liner
chipping during implantation. Unfortunately, we believe that this design
modification results in a thinner alumina insert, which increases the
likelihood of a peripheral chip fracture and subsequent crack propagation
through the brittle alumina material under impingement conditions. In
addition, the second modular interface may unnecessarily introduce the
potential for dissociation between the alumina and the polyethylene
(Fig. 4).
Recently, some manufacturers of ceramic-on-ceramic hip systems have changed
the shape of the neck of the stem into a trapezoidal cross-sectional geometry
in an effort to minimize femoral neck impingement against either the ceramic
insert or the outer cup itself. All of the patients in the present study had
received a stem with a circular neck cross-section combined with a
28-mm-diameter femoral head, resulting in a relatively low head-to-neck ratio,
which may have contributed to impingement of the stem neck with the alumina
insert, as described by Barrack et
al.22.
It is also notable that two ceramic femoral heads fractured during the
first postoperative year in the absence of any specific traumatic event, even
though a proof-tested, third-generation alumina had been used. Recently
reported data have indicated that the fracture rate for currently available
alumina ball heads has been reduced to a level of one fracture per 10,000
implants4,5,23,24.
However, we believe that, despite this extremely low rate, it may not be
possible to eliminate the actual risk of alumina head fracture, and therefore
patients should be informed about the potential for this calamitous
complication before receiving an alumina-on-alumina bearing.
At the time of revision surgery following a ceramic implant fracture, the
use of a new ceramic head on a damaged Morse taper is not recommended because
it creates an area of high point pressure and acts as a stress-riser that
ultimately may lead to a fracture of the newly implanted ceramic
head4,5,25.
Although complete exchange of a femoral component avoids this problem,
controversy exists over whether to revise a stable femoral component at the
time of revision surgery following the failure of a ceramic
implant15,26.
We believe that removal of a well-fixed stem can be a complicated surgical
procedure associated with increased blood loss, operative time, and damage to
the bone stock. Therefore, in the two hips with an alumina head fracture, we
implanted a new cobalt-chromium head on the scratched Morse taper, leaving the
well-fixed acetabular and femoral components in situ, after complete
débridement and synovectomy in order to remove as many of the alumina
particles as possible. In the four hips with fracture of the alumina liner and
no macroscopic damage to the Morse taper, we performed a modular exchange with
a new alumina-on-alumina bearing because we believed that the new ceramic head
would be harder than a new cobalt-chromium head and therefore would be better
able to resist the abrasive wear that can be caused by undetectable
microscopic ceramic debris. There was no fracture or abnormal wear of the
reimplanted femoral heads during our relatively short follow-up period.
The number of failed alumina implants was small, which precluded any
statistical analysis of our results to determine the prognostic factors
associated with the time to alumina implant failure. Another limitation of the
present study was a lack of uniform indications for the use of this alumina
ceramic bearing at the four participating centers.
This relatively high rate of catastrophic alumina bearing surface failure
at short-term follow-up prompted us to discontinue the use of this type of
alumina bearing design. If it is used, we recommend that every effort should
be made to prevent impingement by meticulous component positioning and,
whenever possible, the use of a larger femoral head combined with the optimal
design of acetabular liner and femoral neck to realize the potential long-term
durability of a total hip replacement with an alumina-on-alumina bearing.
Images of additional hips are available with the electronic versions of
this article, on our web site at
(go to
the article citation and click on "Supplementary Material") and on
our quarterly CD-ROM (call our subscription department, at 781-449-9780, to
order the CD-ROM).