Summary of Results in Original
Publication1
Two hundred consecutive, unselected patients (208 hips) with an average age
of sixty-one years (range, twenty-two to eighty-four years) underwent total
hip arthroplasty with a tapered, rectangular stem made of
titanium-aluminum-niobium alloy (Zweymüller Alloclassic; Zimmer,
Winterthur, Switzerland) and a conical screw cup (CSF; Zimmer [previously
Sulzer]) between October 1986 and November 1987. The arthroplasties were
performed by multiple surgeons in a single teaching hospital. One hundred and
thirty-three patients were still alive at a minimum of ten years
postoperatively1.
One hundred and twenty-three (92%) of those patients had not had a revision,
106 (86%) of the 123 had no hip-related pain, and 102 (83%) walked without a
limp. The average Harris hip score was 85.4 points (range, 46 to 100
points).
Radiographic findings included radiolucent lines and stress-shielding in
zones 1, 7, 8, and 14 as defined by Gruen et
al.2, but there was
no osteolysis around any cup or stem. No stem had a continuous radiolucent
line extending around its distal tip; therefore, no femoral component was
judged to be loose radiographically. Accordingly, all stems were classified as
stable-bone ingrown according to the criteria of Engh et
al.3.
At the time of the original
report1, nine cups
had been revised; five of those cups had been revised because of aseptic
loosening, at an average of eight years postoperatively. The probability of
survival of the stem, with any reoperation as the end point, was 0.99 (95%
confidence interval, 0.97 to 1.00) at a minimum of ten years.
In the present retrospective follow-up report, we extend the results of
that series to a mean of 15.5 years (range, fifteen to 17.25 years)
postoperatively.
We received institutional review board approval for this ongoing follow-up
evaluation of the results of Zweymüller Alloclassic total hip
arthroplasties. Of the 200 patients (208 hips) in the original study,
eighty-seven (eighty-nine hips) died less than fifteen years postoperatively.
Eleven patients (eleven hips) who had had the surgery at least fifteen years
prior to this review were lost to follow-up, leaving 102 living patients (108
hips). Information on fifteen patients (sixteen hips) who were unable to
attend our outpatient clinic was gathered with telephone interviews. The
remaining eighty-seven patients (ninety-two hips) were available for clinical
follow-up. Eighty-four patients (eighty-eight hips) had fifteen-year
radiographs available for evaluation.
Clinical Assessment
Patients were evaluated with use of the Harris hip
score4, with a
special emphasis on symptoms of thigh or groin pain.
Radiographic Assessment
Anteroposterior and frog-leg lateral radiographs of the hip were compared
with previous radiographs and were analyzed for radiolucent lines, osteolysis,
and polyethylene
wear5. The location
of radiolucent lines and osteolytic lesions around the stem were reported in
relation to the zones described by Gruen et
al.2 and those
adjacent to the cup, with the zones described by DeLee and
Charnley6. We
estimated the size of the osteolytic lesion with use of the method of Zicat et
al.7. Linear wear
data were recorded as the magnitude of femoral head penetration into the
acetabular component in millimeters as seen on an anteroposterior
radiograph8.
Radiographic criteria for cup loosening included radiolucent lines extending
around the entire cup as well as a progressive change in the cup position.
Statistical Analysis
Survivorship probabilities were estimated with the Kaplan-Meier
method9. The end
points were revision because of aseptic loosening and revision for any reason.
Any revision was also considered as censored at the date of the revision. When
the survival probabilities were estimated with failure defined as definite
radiographic signs of loosening of the implant, patients without radiographic
signs of loosening were censored at the date of the last radiographic
evaluation.
The average Harris hip score for the eighty-seven patients (ninety-two
hips) was 85.4 points (range, 46 to 100 points) at ten years and 88.6 points
(range, 48.7 to 100 points) at fifteen years. This difference was not
significant. Eighty-one patients were very satisfied or satisfied, two
patients were rather satisfied, two patients were dissatisfied with the
outcome of the operation, and two patients did not answer the question on the
questionnaire. One of the dissatisfied patients was confined to a wheelchair
with hemiplegia and was living in a nursing home. Two of four patients who had
reported isolated thigh pain at the ten-year follow-up evaluation continued to
have thigh pain at fifteen years.
A key finding of the ten-year report was the absence of osteolytic lesions.
At fifteen years, osteolysis was observed adjacent to the femoral component in
ten hips (11%), adjacent to the acetabular component in four (5%), and
adjacent to both the acetabular and the femoral component in two (2%).
Osteolytic lesions were noted only in the proximal part of the femur: in Gruen
zones 8 and 14 in seven hips and in zone 7 in five hips. In one hip, the
osteolysis extended distally to zones 3 and 5. The average size of the lesion
was 1.2 cm2 (range, 0.18 to 3.29 cm2).
Radiolucent lines of >1 mm in width were observed adjacent to the stem
on nineteen (22%) of the eighty-eight reviewed radiographs but were found in
Gruen zone 1 in fourteen hips, zone 7 in eight, zone 8 in twelve, and zone 14
in ten. In one hip, in a patient with primary hyperparathyroidism, there were
circumferential radiolucent lines (minimum width, 1 mm) around both the
acetabular and the femoral component. That patient underwent an exchange of
the cup and stem at 16.3 years postoperatively.
Radiolucent lines around the cup were observed on five (6%) of the
eighty-eight reviewed radiographs. Two cups were seen to be loose
radiographically, with circumferential radiolucent lines around the implant.
Polyethylene wear was evident on seventeen radiographs (19%). The average wear
was 0.2 mm/yr (range, 0 to 0.5 mm/yr).
Since the beginning of the original study, a total of twenty-two hips were
revised for any reason (Table
I). One stem (Case 1) was revised at five days postoperatively
because of varus malpositioning. Another stem (Case 3) was explanted with the
cup after multiple failed acetabular reconstructions, and a third stem (Case
9) required revision because of infection. None of the twenty-two revisions
were done because of aseptic loosening of the stem. Twenty cups were revised.
Nine polyethylene liners were exchanged because of progressive wear, and eight
cases of aseptic loosening of the titanium shell were recorded.
Since our previous
report1, there were
no stem revisions, but one stem was loose radiographically, with osteolytic
lesions in Gruen zones 1, 2, 3, 5, 6, and 7. In addition, ten acetabular
components were revised since our last report: three because of aseptic
loosening and seven because of polyethylene wear.
The probability of survival of the stem with revision for any reason as the
end point was 0.98 (95% confidence interval, 0.96 to 1.00) at fifteen years
(Fig. 1). The probability of
survival of the acetabular component with revision for any reason as the end
point was 0.85 (95% confidence interval, 0.79 to 0.91)
(Fig. 2). Kaplan-Meier curves
with aseptic loosening of the stem or cup as the end points are shown in
Figures 3 and
4.