Given the mixed outcome after revision total knee arthroplasty (TKA) for instability in the literature and the relative high recurrence of instability, we were interested in the outcome of a cohort of patients operated for various types of clinical instability and with different types of implants.
A total of 77 patients with unstable TKA were completely revised (19 hinged and 58 condylar implants). We classified the patients in 3 instability groups based on the literature: (1) anterior-posterior flexion instability (N = 29); (2) medial-lateral flexion instability (N = 16); and (3) multiplane instability (N = 32). Patients were evaluated up to 24 months postoperatively, concerning Knee Society clinical rating system, range of motion, visual analog scale (VAS) pain, and VAS satisfaction.
For the total group, all outcome scores improved, but substantial residual pain (VAS = 41) was reported. For type of instability, the clinical outcome was similar for all the groups. For type of implant, the hinged group had lower postoperative outcome scores but similar satisfaction scores compared with those in the condylar group. There was a considerable number of insert changes and secondary patellar resurfacing in the condylar group compared with no reoperations in the hinged group. Recurrent instability was not seen in the anterior-posterior flexion instability group and in patients who received a condylar constraint-type implant.
We recommend 3 options in revision TKA for instability: (1) hinged implants in cases with severe ligament instability in multiple planes or bone loss, (2) condylar implants with a posterior-stabilized insert in cases with isolated posterior cruciate ligament insufficiency, and (3) condylar implants with condylar constraints in all other cases.