Of the 118 patients (150 knees), twenty patients (thirty knees) died and
fourteen patients (fifteen knees) were lost to follow-up. Clinical and
radiographic follow-up was thus available for eighty-four patients (105 knees)
with a mean duration of follow-up of twelve years (range, ten to thirteen
years). The preoperative diagnosis in this cohort of eighty-four patients was
osteoarthritis in seventy-seven patients (ninety-four knees), rheumatoid
arthritis in four patients (eight knees), and osteonecrosis in three patients
(three knees).
Knee Society Score
The mean Knee Society functional score at the time of the latest follow-up
for the eighty-four patients (105 knees) was 73 points (range, 50 to 100
points), and the mean clinical score was 88 points (range, 50 to 100 points).
The overall clinical result was excellent for seventy-nine knees (75%) in
sixty-three patients (75%), good for fifteen knees (14%) in thirteen patients
(15%), fair for six knees (6%) in four patients (5%), and poor for five knees
(5%) in four patients (5%). Hence, a good or excellent result was attained in
ninety-four knees (90%) in seventy-six patients (90%).
Patient-Assessment Questionnaire
Among the 105 knees, ninety (86%) had no pain, eleven (10%) had minor
anterior pain, and four (4%) had moderate pain. Sixty-eight (81%) of the
eighty-four patients were able to ascend or descend stairs in a reciprocal
fashion without support.
Overall, eighty patients (95%) were fully satisfied and four (5%) were only
partially satisfied because of pain during walking. Fifty-one (61%) of the
eighty-four patients were participating in recreational sports such as walking
1 to 3 mi (1.6 to 4.8 km) per day, golf, doubles tennis, swimming (thirty
minutes per day), or stationary bicycling (thirty minutes per day) at the time
of the latest follow-up.
Range of Motion
The mean range of motion at the time of the latest follow-up was 108°
(range, 60° to 135°). Eleven knees (10%) had <90° of motion,
twenty-two knees (21%) had 91° to 110°, and seventy-two knees (69%)
had >110°. Five knees had a flexion contracture of 5°.
Patellofemoral Articulation
There were no instances of patellar fracture, patellar dislocation,
patellar component loosening, or patellar clunk syndrome. Mild anterior knee
pain was present in eight patients (eleven knees; 10%). No patient had
persisting delayed-onset disabling anterior knee pain as described in the
previous
report8.
Radiographic Results
The mean duration of radiographic follow-up of the overall cohort of
survivors was twelve years (range, ten to thirteen years). Fifty-two (50%) of
the 105 knees had an incomplete radiolucent line around the tibial component
(twenty-six knees; 25%), the femoral component (fifteen knees; 14%), or both
components (eleven knees; 10%). None of the knees had radiolucent lines in
more than two contiguous zones. Two knees had progressive radiolucent lines
around the tibial component in zones 1 and 2. Both of these knees also had
expansive medial femoral condylar osteolysis (Figs.
1 and
2) and subsequently underwent
revision. No other knee had radiographic evidence of loosening of any
component.
Revision Operations
Five knees were revised in five patients. Two knees were revised because of
infection, and one was revised because of dislocation, as reported
earlier8. Two
additional knees were revised at six and eight years because of femoral
osteolysis. Both of the knees had had failure of the locking mechanism of the
polyethylene insert due to fracture of the locking tab. The top side of the
polyethylene liners had delamination and yellow discoloration due to
oxidation. Burnished tibial baseplates and the absence of manufacturer
markings on the insert suggested so-called backside polyethylene wear. Both
knees had well-fixed femoral and tibial components at the time of revision.
The femoral components were removed because of extensive bone loss under the
components. After removal of the components, the metaphyseal bone was
reconstructed with bone graft and wedges. Press Fit Condylar TC-3 implants
(Johnson and Johnson) were then inserted to achieve knee stability in
flexion.
Survivorship Analysis
With revision or recommended revision for any reason as the end point, the
twelve-year survival rate for the entire cohort was 94.6% ± 4.0% in the
best-case scenario and 84.9% ± 5.9% in the worst-case scenario
(Fig. 3). With mechanical
failure as the end point, the twelve-year survival rate was 98.3% ±
2.4% in the best-case scenario and 88.3% ± 5.3% in the worst-case
scenario (Fig. 4). Forty-five
knees in thirty-five patients were at risk after twelve years of follow-up. As
none of the knees had radiographic evidence of loosening at the time of the
last review, the radiographic survival rate was the same as the clinical
survival rate.
The patient-assessment questionnaire is 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). ?