The goal of this pragmatic, open, (1:1) randomized, multi-centre, non-inferiority trial is to to determine whether knee-joint distraction (KD) is non-inferior on patient reported effectiveness as compared to a knee-prosthesis (KP; i.e. usual care) for relatively young patients with end-stage knee Osteoarthritis (OA).

The main question[s] it aims to answer are:

• Is KD non-inferior to KP regarding pain, function and stiffness (as indicated by the total score on the total Western Ontario and McMaster Universities Osteoarthritis Index' (WOMAC), with a non-inferiority limit of 15 points) at 2 years?
• Is KD non-inferior to KP regarding Quality of Life (36-Item Short Form Health Survey, (SF36), Physical- and Mental- Component Summary (PCS/MCS), with a non-inferiority limit of 10 points) at 2 years?
• Does KD lead to regeneration of tissue (increase in minimum joint-space-width > 0.05mm on x-ray) over 2 years?

Participants will be allocated to undergo either a knee joint distraction or prosthesis (total- or unicompartmental KP according to orthopedic surgeon/patients discretion), and groups will be compared using will be compared between groups using (multivariable) random effects (mixed) modelling to account for the nested and longitudinal structure of the data over the 24 months follow-up. The stratification factors for randomization, center (using a random intercept) and gender, as well as a limited number of a priori defined prognostic factors (i.e. baseline WOMAC total score, age, BMI, severity of cartilage damage) will be accounted for in this analysis. The difference in mean total WOMAC score at 24 months between treatment groups will be estimated from this model with a 95% confidence interval (CI) and non-inferiority will be determined using the lower limit of this confidence interval.

Inclusion Criteria:

• Clinical diagnosis of knee OA
• Age ≤ 65 years and ≥ 18 years
• Persistent, refractory pain, insufficiently responding to conservative or previous surgical therapy
• Structural OA joint damage, indicated by a K&L grade of at least 2 as determined by the orthopedic surgeon in line with NOV recommendations.
• Able to wear an external fixator and care for it for 6 weeks
• Accepting that the maximal effect of KJD is not present directly after removal of the frame but may take months after frame removal
• Sufficient joint stability (according to the orthopedic surgeon's judgement)
• Flexion (>100 degrees) and extension range (<10 degrees)
• Weight and BMI <120 kg and <35 kg/m2, respectively
• Sufficient understanding of the Dutch language
• Signed informed consent

Exclusion criteria:

• Surgical intervention in last 6 months
• Leg-axis deviation > 10 degrees (as determined by the orthopedic surgeon)
• Serious osteopenia making placing bone-pins and wearing a frame into a risk (according to the orthopedic surgeon's judgement)
• Coagulation problems making occurrence of thrombosis or embolies into a risk (according to the orthopedic surgeon's judgement)
• Existing endoprosthesis at any other joint (e.g. hip or contralateral knee) to prevent infection of existing prosthesis
• History or presence of joint infection/inflammation
• Hypersensitivity to antibiotics
• Presence of systemic inflammatory disease, like rheumatoid arthritis