Summary

Despite national registries indicating higher revision rates for UKAs and less satisfactory outcomes after conversion to TKA, our data demonstrates that TKAs performed after UKA show significantly improved outcomes and re-revision rates, surpassing previous reports and matching those of primary TKAs.

Abstract

Objectives: Joint replacement registries and previous studies have shown patients undergoing revision of UKA (Unicompartmental Knee Arthroplasty) to TKA (Total Knee Arthroplasty) to have Oxford scores, satisfaction scores and implant survivorship to be less satisfactory than primary TKA and similar to those of revision TKA patients. We hypothesised that the patient outcomes in our UKA revision to TKA would be similar to our primary TKA patients.

Method

Patients who presented to our clinic and underwent revision of a UKA to TKA were compared with primary TKA patients from the same period matched for age, Body Mass Index (BMI) and sex. We compared Oxford scores (OKS) and knee satisfaction scores, VR12 and revision rates for both groups with data collected preoperatively and at 1 year, and 5years or later after surgery.

Results

We analysed 90 patients with a UKA revision to TKA matched with 93 patients who underwent primary TKA at the same time. The average age (71.0yrs. vs 71.4yrs.), gender distribution (Females: 52% vs 57%) and mean BMI (29.7 kg/m2 vs 30.0 kg/m2) were similar in both groups. Re-revision rate in the revision UKA group was 4.4% with an 8-year survivorship of 95.6%. None of the primary TKA has had a revision.
Both groups had similar OKS (25.2 vs 26.0, p=0.52) prior to surgery. Lower preop OKS score in men was correlated with poor UKA implant survival time (r= -0.24, p=0.05). One year after surgery, there was improvement in all scores across both groups, with the primary TKA group showing a slightly better outcome in the OKS score (38.8 vs. 41.6, p=0.03). However, long term OKS for both groups showed no difference (40.0 vs 41.2, p=0.42). Over time, patients who had a UKA revision reported higher pain levels (27.3 vs. 9.14 on a 0-100 scale, p=0.01) and had lower physical health scores (40.2 vs. 43.6, p=0.04) compared to those who had a primary TKA. Patient satisfaction rates for both groups remained similar at all time points.

Conclusion

The past 15 years has seen a gradual reduction in the proportion of UKA to TKA. Whilst this is likely in part due to higher revision rates of UKAs as portrayed by national registries, prior studies suggest that the failure of a UKA will be followed by a less satisfied patient after revision to a TKA with a higher chance of re-revision. Our data runs contrary to this suggestion. We’ve shown that the outcomes and re-revision rates of TKAs performed following a prior UKA are better than previously reported and similar to those seen with primary TKA.