Question: In patients undergoing primary total knee arthroplasty
(TKA), are better clinical, functional, and satisfaction results achieved with
resurfacing or nonresurfacing of the patella?
Design: Randomized (allocation
concealed)*, blinded
(patients, data collectors, outcome assessors, data analysts, and data safety
and monitoring
committee)*,
controlled trial with 10-year follow-up.
Information provided by author.
Setting: A tertiary-care university center in London, Ontario,
Canada.
Patients: 90 patients (mean age, 70 y; 57% women) (100 knees) having
TKA. Exclusion criteria included previous patellectomy, inflammatory
arthritis, patellar fracture or instability, previous extensor mechanism
procedures, high tibial osteotomy, severe valgus or varus deformity, previous
unicondylar knee replacement, and history of septic arthritis or
osteomyelitis. 83 patients (92%) (90 knees) were available for follow-up.
Intervention: TKA featured an anatomically designed cruciate
retaining femoral trochlear groove and intercondylar notch acceptable for both
resurfaced and nonresurfaced patellofemoral articulation. After the femoral
and tibial cuts were done, patients were allocated to patellar resurfacing, in
which an attempt was made to restore the baseline composite height of the
patella (n = 42), or for the patella to remain unresurfaced, in which case a
patelloplasty was done (n = 48).
Main outcome measures: Reoperation rate, Knee Society clinical
rating score, functional outcomes, patient satisfaction, anterior knee pain,
and radiographic assessment.
Main results: 45 patients (50 knees) were alive at 10-year
follow-up. The resurfaced and nonresurfaced groups did not differ for
reoperation rate (5% vs 15%; p = 0.166). Both the resurfaced and nonresurfaced
groups improved in Knee Society clinical rating scores
(Table). Total function scores
decreased in both groups compared with assessment at 2 years (p < 0.045).
Pain and total knee scores did not decrease from 2 years. The mean range of
motion at the most recent follow-up was 109° in both groups. 85% of the
resurfaced group and 93% of the nonresurfaced group responded as
"extremely satisfied" or "very satisfied" on the
patient questionnaire. The resurfaced and nonresurfaced groups did not differ
for the number of stairs climbed in 30 seconds (mean, 31 vs 21; p = 0.072).
The groups did not differ for flexion (29.79 vs 36.92 nm; p = 0.62) or
extension (55.67 vs 63.68 nm; p = 0.36) torque. Patient-reported functional
outcomes such as getting in and out of a car or chair or going up and down
stairs did not differ between groups. Radiographic assessment for all patients
who did not have knee revision showed similar results between resurfaced and
nonresurfaced groups. There was no relation between grade of patellar
cartilage loss and anterior knee pain in the nonresurfaced group at 10
years.
Conclusion: In patients undergoing primary total knee arthroplasty,
long-term clinical, functional, and satisfaction outcomes were not different
if the patella was resurfaced or not resurfaced.
The study by Burnett and colleagues attempts to address a controversial
aspect of TKA. The implant chosen for this study was the Anatomical Medullary
Knee (DePuy, Warsaw, Indiana) which features an anatomically designed femoral
groove and intercondylar notch designed for both a resurfaced and
nonresurfaced patellofemoral articulation. In the unresurfaced knees,
osteophytes and fibrillated or torn cartilage were removed with no drilling of
eburnated bone. The authors specifically address the issue of symptoms
relative to the patellofemoral joint by analyzing anterior knee pain by
history, physical examination, and having patients respond to a patellofemoral
questionnaire.
One may certainly ask why a 10-year minimum follow-up is necessary.
Interestingly enough, the authors note that in their 2-year
report1, knees with
unresurfaced patellas had better flexion torques than knees with resurfaced
patellas, and both scores improved from preoperative measures. When they
repeated these measures at a minimum of 10 years of clinical follow-up, they
found that the improvement in both flexion and extension torque strength had
been lost in both groups, such that a difference no longer remained from the
original preoperative scores in both groups. An important limitation of the
study is the fact that no power analysis was performed; however, the authors
clearly recognize this limitation and discuss it appropriately.
Two additional comments should be considered. First, one may ask whether
there are situations in which the patella should never be resurfaced. The
authors believe that those with inadequate or poor quality bone stock should
not be resurfaced, and indeed this is a reasonable recommendation. Second is
the senior author's comment that they frequently resurface the patella despite
the findings of their study. They often will consider not resurfacing the
patella in patients who do not have anterior knee pain, have well-preserved
articular cartilage, and have normal patellar tracking and anatomy. Once
again, I believe many surgeons would agree that a patient meeting these
criteria is a potential candidate for not resurfacing the patella.
Bourne RB, Rorabeck CH, Vaz M, Kramer J,
Hardie R, Robertson D. Resurfacing versus not resurfacing the patella during
total knee replacement. Clin Orthop Relat Res.1995;321:
156-61.321156
1995
[PubMed]