The aim of the present study was to assess the outcome of revision surgery, using semiconstrained implants, in the management of tibiofemoral instability complicating primary total knee arthroplasty.

Between Feb 1987 and Oct 2000, 177 primary, unconstrained, surface replacement total knee arthroplasties were revised at our institution. Instability was the commonest reason for revision surgery and accounted for 22.6 % of overall revisions. Excluding tertiary referrals, instability necessitated revision surgery in 0.31% of 1918 primary total knee arthroplasties performed ‘in-house’ during the same period.

The results of 17 revision total knee arthroplasties using semiconstrained prostheses are presented. Six operations were performed for sagittal plane instability, 5 for coronal and 6 for multiplanar instability. 10 revisions were performed using the PFC ‘stabilised plus’ prostheses, and a further 7 with TC3 prostheses. 17 patients (13 F: 4 M), aged 48–83 years (average 67.8 years) underwent revisions, between 9–132 m from the date of the index arthroplasty. At an average follow-up of 36m, the Knee Society score had risen from 31.2 points preoperatively to 60.9 at last follow-up (LFU) [Joint score from 47.5 preop to 81.5 at LFU/Function score: from 14.4 to 39.7 at LFU]. Radiolucency rates were insignificant and at LFU no joints showed evidence of osteolysis, implant subsidence or polyethylene wear. One unresurfaced patella spontaneously fractured 10m postoperatively and one patella showed persistent subluxation. There were no other significant complications.

The present short term study attests to the efficacy of semi-constrained implants in the revision of unstable primary arthroplasties. However, in the longer term, implant longevity remains undetermined.