High Prevalence of Radiographic Outliers and Revisions with Unicompartmental Knee Arthroplasty

Alignment outcomes and their impact on implant survival following unicompartmental knee arthroplasty (UKA) are unclear. The purpose of this study was to assess the implant survival and radiographic outcomes after UKA as well as the impact of component alignment and overhang on implant survival.

We performed a retrospective analysis of 253 primary fixed-bearing and mobile-bearing medial UKAs from a single academic center. All UKAs were performed by 2 high-volume fellowship-trained arthroplasty surgeons. UKAs comprised <10% of their knee arthroplasty practices, with an average of 14.2 medial UKAs per surgeon per year. Implant survival was assessed. Femoral coronal (FCA), femoral sagittal (FSA), tibial coronal (TCA), and tibial sagittal (TSA) angles as well as implant overhang were radiographically measured. Outliers were defined for FCA (>±10° deviation from neutral), FSA (>15° of flexion), TCA (>±5° deviation from neutral), and TSA (>±5° deviation from 7°). "Far outliers" were an additional >±2° of deviation. Outliers for overhang were identified as >3 mm for anterior overhang, >2 mm for posterior overhang, and >2 mm for medial overhang.

Among patients with a failed UKA, revision was performed at an average of 3.7 years (range, 0.03 to 8.7 years). The cumulative revision rate was 14.2%. Kaplan-Meier survival analysis demonstrated 5 and 10-year survival rates of 88.0% (95% confidence interval [CI] = 82.0% to 91.0%) and 70.0% (95% CI = 56.0% to 80.0%), respectively. Only 19.0% (48) of the UKAs met target alignment for all 4 alignment measures, and only 72.7% (184) met all 3 targets for overhang. Only 11.9% (30) fell within all alignment and overhang targets. The risk of implant failure was significantly impacted by outliers for FCA (failure rate = 15.4%, p = 0.036), FSA (16.2%, p = 0.028), TCA (17.9%, p = 0.020), and TSA (15.2%, p = 0.034) compared with implants with no alignment or overhang errors (0%); this was also true for far outliers (p < 0.05). Other risk factors for failure were posterior overhang (failure rate = 25.0%, p = 0.006) and medial overhang (38.2%, p < 0.001); anterior overhang was not a significant risk factor (10.0%, p = 0.090).

The proportions of UKA revisions and alignment outliers were greater than expected, even among high-volume arthroplasty surgeons performing an average of 14.2 UKAs per year (just below the high-volume UKA threshold of 15). Alignment and overhang outliers were significant risk factors for implant failure.

Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.