Purpose: From 1985 to 2001, 31 total knee arthroplasties were performed for 17 adolescents or young adults with idiopathic juvenile osteoarthritis. The purpose of this work was to evaluate functional and radiological outcome.

Material and methods: Overall functional outcome was assessed with the Steinbrocker classification. Knee function was evaluated with the IKS score. Several types of prostheses were implanted: constrained GSB (n=14), cemented semi-constrained tri-CCC tri-compartment with a rotatory platform (n=10), non-cemented semi-constrained ROCC (n=1), LCS (n=2) including non-cemented, and FINN (n=2) (two custom-made rotation hinge prostheses implanted in the same patient). Fourteen prostheses involved bilateral implants, including three dual implantation procedures.

Results: Mean age at implantation was 20 years five months (14–29). There were fourteen girls and three boys. Eight had systemic idiopathic juvenile osteoarthritis and nine a polyarticular form. The Steinbrocker staging was: II (n=5,) III (n=6), IV or bedridden (n=4). Ten patients had two hip prostheses before bilateral knee arthroplasty. Mean follow-up was 4.5 years (1–12). Among the 31 operated knees, 16 were pain free, 14 minimally painful, and one painful due to loosening. The joint score was very good (n=18), good (n=4), and poor (n=5). Radiographically, normal alignment was found for 29 knees. Lucent lines were observed for 10 of the 14 GSB constrained prostheses. We did not observe any evidence of lucent lines for the non-cemented tri-compartment prostheses. Complications were: limited skin necrosis (n=1), bilateral supracondylar fracture one year after implantation (n=1).

Discussion: Outcome has been encouraging for total knee prostheses in patients with idiopathic juvenile osteoarthritis. These arthroplasties allow spectacular functional improvement. The few series reported have also reported very good results. Cemented tri-compartment semi-constrained implants appear to provide better stability at five years. Biologically sealed tri-compartment prostheses would be a very satisfactory solution due to the preservation of bone stock.

Purpose: We expose technical problems encountered for prosthesis replacement in osteoarthrosis juvenilis (OJ). The characteristic feature of this disease is early joint destruction during growth.

Material and methods: Total knee arthoplasty (TKA) was performed in 17 severely disabled patients (31 knees): Steinbrocker stage II=30%, stage III=30%, stage IV=40%. Mean age at operation was twenty years (14–29). Technical difficulties were related to the following combinations: 1) multidirectional malformations, generally in valgus (mean 16°, range 5–30°) in 30% of the knees associated with external rotation (mean 20°, range 5–50) and sagittal deformation with permanent flexion (mean 31°, range 5–60°) with external or posterior tibia dislocation; 2) limited joint motion: 71° (0–115°); 3) extraarticular deformations with permanent flexion or vicious hip rotation, tibial or femoral callus; 4) major condyle dysplasia due to growth deficiency (3/31 or necrosis (3/31); 5) low patella (100%) and subluxation; 6) weak bone and fragile skin related to corticosteroid therapy; 7) persistent growth cartilage in four patients. We tried to implant the most adapted prosthesis in each individual situation, favouring the least constrained implant possible.

Results: We used fifteen mobile plateau prostheses including five pure gliding TKA and ten posterostabilised TKA with a mobile plateau and 16 hinge prostheses with two rotators. Thirty of the 31 TKA were custom-made.

Discussion: Our different approaches enabled us to propose the following: correction of extra-articular deformations by TKA, after tenotomy and traction, or after concomitant osteotomy; primary approach after checking the vasculonervous bundle (popliteal sciatic); sub-periosteal dissection preserving the lateral ligaments searching to achieve ligament balance when possible in order to implant the least constrained implant possible; non-cemented implants, especially for “soft” or “fatty” bone; no patellar resurfacing when there is a risk of an overly thick low and subluxed patella.