The purpose of this scientific study is to clarify whether there is a relationship between the appearance of radiolucent lines in bone-cement interphase of the tibial component of the different models of the Oxford Uni Knee prosthesis and poor prognosis of these prostheses. This will be accomplished in a se-ries of three different Oxford Uni Knees.

The purpose of this scientific study is to clarify whether there is a relationship between the appearance of radiolucent lines in bone-cement interphase of the tibial component of the different models of the Oxford Uni Knee prosthesis and poor prognosis of these prostheses. This will be accomplished in a series of three different Oxford Uni Knees by studying:

1. Three way randomized comparison of migration (prosthetic fixation) by Model Based RSA with three different Oxford Uni Knee prosthesis (both tibial trays and femoral components will be assessed).
2. Fluoroscopy (screening) roentgenograms for determination of RLL around the tibial component and "overhang" of the tibial trays.
3. Changes in periprosthetic bone mineral density between the three randomized groups evaluated in the proximal tibia and distal femur on the surgical side and on the unaffected tibial condyle area (BMD).
4. Clinical function of the Uni Knee evaluated by a score system using a recent Danish adaptation of the Oxford Knee Score that consists of a questionnaire filled-out by the patient with 12 simple questions on pre-and postoperative assessment of the ADL.
5. Analysis of wear of the mobile polyethylene (meniscus) including a methodical study as to whether the technique requires weightbearing (standing) stereoroentgenograms or whether non-weightbearing stereoroentgenograms can be used (assessed with Model Based RSA).
6. If a fast bone-remodeling increases the risk of aseptic loosening (histomorphometric results combined with RSA results and possible revision)
7. Increased risk of aseptic loosening with cemented (two designs) compared to uncemented im-plants.
8. If bone-remodeling of the proximal tibia is dependent on gender and age.

The study would be considered a success if all designs of the prostheses are firmly fixed and remain so during the entire period of study, that is, no increasing migration is seen using RSA. It would be a success if the appearance of RLL beneath the implant in the proximal tibia is unrelated to bone loss and prosthesis migration.

The hypotheses are:

1. RLL in relation to the tibial component is unimportant for fixation of the Oxford Uni Knee as assessed with DEXA and RSA.
2. There is very little wear of the polyethylene in the course of a 5 year follow-up as assessed with RSA.
3. The femoral component is expected to remain fixed throughout the entire period of follow-up.
4. Function assessment as ROM (range of motion) after UKA is satisfactory and generally better than after TKA.
5. All three prosthetic designs are expected to migrate minimally and remain fixed throughout the follow-up period, but the uncemented design is expected to be superior to the cemented design.
6. No difference in clinical outcome between the three prosthetic types.
7. The bone-remodeling rate in the medial periarticular tibia bone increases the risk of aseptic loosening.

The primary parameters of effect are RLL, RSA and DEXA. The secondary parameters of effect are bone-remodeling, wear analysis, Oxford Knee Score (ADL) and ROM.

Inclusion Criteria:

1. Patients, both men and women, with anteromedial unicompartmental knee osteoarthritis
2. Patients found suitable for UKA with telos stress-radiographs
3. Patients with sufficient bone quality to allow insertion of a Uni Knee prosthesis
4. Patellofemoral osteoarthritis is not a contraindication if the patient does not have symptoms from the patellofemoral joint
5. Patients can only take part in one knee study
6. All age groups, in good condition and responsible adults
7. Informed, written compliance consent.

Exclusion Criteria:

1. Patients with neuromuscular or vascular diseases in the affected leg
2. Patients found to be unsuitable for UKA before or at surgery
3. Preoperative extensions defect greater than 10 degrees
4. Preoperative maximal flection of less than 100 degrees
5. Symptomatic patellofemoral osteoarthritis
6. Insufficiency with regard to ACL
7. Patients templated to a size XS or XL femoral component prior to surgery.
8. Patients with osteoporosis on the basis of preoperative x-rays or earlier diagnosis
9. Continuous medical treatment with vitamin K antagonist (Warfarin), known to reduce bone mass in general by a factor of 5
10. Patients with fracture sequelae (intraarticular fracture and all tibial condyle fractures)
11. Patients previous having PTO or other extensive knee surgery
12. Patients with metabolic bone disease
13. Patients with rheumatoid arthritis
14. Postmenopausal female patients being treated with systemic hormonal substitution (not taking into local vaginal treatment)
15. Patients who intermittently or continuously require treatment with systemic glucocorticoids
16. Non-Danish citizens
17. Insufficient command of the Danish language (read and speak)
18. Senile dementia
19. Misuse of drugs or Alcohol
20. Serious psychiatric illness
21. Disseminated malignant disease (cancer) and treatment with radiation or chemotherapy
22. Serious systemic disease (e.g.. hemiparesis and severe Parkinsonism)
23. Systemic hip or back condition
24. Employed by the department of orthopaedic surgery
25. On-going case involving work injury of the knee
26. Patients with poor dental status