Abstract
Introduction
Unicompartmental knee arthroplasty (UKA) is a successful procedure for medial compartment osteoarthritis (OA). Recent studies using the same implant report a revision rate of 2.9%. Other centers have reported revision rates as high as 10.3%. The purpose of this study was to retrospectively review the clinical results of Oxford Phase 3 UKA's performed in the setting of isolated medial compartment OA and to compare our results to the previous mid-term studies. Our secondary goal was to determine reasons for revision and evaluate selected independent predictors of failure.
Methods
A retrospective review of 465 Oxford Phase 3 medial UKA's performed on 386 patients (222 female; 164 male) with isolated medial compartment OA. The average age at surgery was 69.5 years (40–88). Outcome measures included: Knee Society Scores(KSS), Oxford Knee Scores(OKS), SF-12, WOMAC, revision rates, and patient satisfaction. We evaluated independently predictors of failure including: gender, body mass index(BMI), number of previous surgeries, implant sizes, cement technique (simultaneous vs staged), cement type. Revision rates based upon the polyethylene thickness (defined as thin 3–4 mm; medium 5–6 mm; thick 7–9 mm). The need for stems and augments and the degree of constraint required at revision to a total knee arthroplasty (TKA) were evaluated.
Results
At a mean follow-up of 60.7 months (11–114) OKS improved from 21 to 37 points (p<.05). Latest SF-12 score was 43.8 points (16.8–64.7 points; SD, 10.5) and WOMAC was 80 points (23–100 points; SD, 18). The overall revision rate was 6.9% (32/465 knees). Mean time to revision in 25 knees was 34.5 months (7–96), and revision was most commonly performed for lateral compartment OA (10). Eight knees were revised for tibial loosening, femoral loosening (6), and PCL failure (1). Revision implants included posterior stabilized in 13 knees (52%), cruciate retaining in 9 knees (36%), and cruciate substituting/dished in 3 knees (12%). Five revisions (20%) required tibial augments and 2(8%) had cemented tibial stems. The mean revision polyethylene thickness was 12 mm (range, 9–19 mm) and one knee required a constrained polyethylene. Three knees are pending revision to TKA. Four knees underwent poly exchange for bearing dislocation and 3 knees had further arthroscopic procedures. Eighty-four percent of the patella were resurfaced at revision. Three quarters (76%) of the patients were extremely or very satisfied with their surgery. Over 90% would have had their surgeries again. Gender, BMI, number of previous surgeries, femoral or tibial sizing, poly thickness, cementing technique or type did not predict revision, the need for constraint, or the need for stems or augments.
Conclusion
Our revision rate of 6.9% was comparable to other midterm studies from independent centers but not as low as recently reported results from Oxford. Progression to lateral compartment OA was the most common reason for revision. We could not find any independent predictors of failures in this group of 465 knees.
