# Starting up a cementless Oxford medial unicompartmental knee arthroplasty practice: a prospective cohort study of 200 knees

## TL;DR

This study examines the outcomes of a new medial unicompartmental knee arthroplasty practice, finding stable patient results and a learning curve for surgery duration.

## Contribution

The study provides empirical data on the learning curve and outcomes during the early implementation of cementless medial unicompartmental knee arthroplasty.

## Key findings

- Surgical duration reduced after 55 cases, indicating a learning curve.
- Patient-reported outcomes remained stable over 24 months.
- Implant survival at 5 years was 97.4% with good radiographic positioning.

## Abstract

Medial unicompartmental knee arthroplasty is widely used to treat anteromedial osteoarthritis, yet limited data exist on outcomes during its implementation phase. Therefore, this study aimed to evaluate the learning curve and the clinical and radiological outcomes during the early implementation of medial unicompartmental knee arthroplasty.

The first 200 medial unicompartmental knee arthroplasty procedures performed by two arthroplasty surgeons were analyzed to assess the relationship between outcomes and the cumulative number of cases. The primary outcome was the learning curve for the duration of surgery, while secondary outcomes included the Oxford Knee Score, the Forgotten Joint Score, and the Activity and Participation Questionnaire, which were assessed at 3, 12, and 24 months postoperatively. Implant survival and positioning were evaluated at the final follow-up.

Cumulative summation analysis showed a data-driven reduction in surgical duration after 55 cases. Median Oxford Knee Score was 41 (IQR 34–44) at 12 months and 42 (IQR 35–45) at 24 months. Implant survival at 5 years was 97.4% (95% CI: 95.1–99.7). Radiographically, 86.5% of patients had tibial implant valgus/varus within 5° of neutral, and no tibial implant overhang exceeded 2 mm.

Medial unicompartmental knee arthroplasty was associated with favorable clinical outcomes during early implementation. Surgical duration indicated a learning curve over 55 cases. Patient-reported outcome measures remained stable, showing reliable outcomes regardless of the learning phase. Surgical precision was maintained throughout, indicating proficient surgical outcomes even during the early phase.

The online version contains supplementary material available at 10.1007/s00402-026-06229-z.