ISRCTN50216144ISRCTN09/07/2025University of OxfordGenicular embolisation for knee osteoarthritisGEKOGenicular artery embolisation for the symptomatic treatment of knee osteoarthritis refractory to conservative management (GEKO)31/07/2025Actual216Recruitinghttps://www.isrctn.com/ISRCTN50216144InterventionalRandomized; Interventional; Design type: Treatment, Imaging, Other (Treatment)Not ApplicableKnee osteoarthritisThe study will be a multi-centre two-arm randomised controlled trial with 1:1 allocation, comparing the clinical efficacy of genicular artery embolisation with a placebo procedure. Participants and follow-up assessors will be blinded to the randomisation allocation. Participants will be recruited from knee orthopaedic clinics. They will have the opportunity at their appointment to ask any questions they have about the study. Screening forms will be completed at each site. These will detail any reasons for exclusion and non-participation. On the day of the procedure, but prior to the procedure, possible eligibility will be re-confirmed and participants will complete a baseline questionnaire. Completing the baseline questionnaire will confirm whether the participant has moderate/severe knee pain measured by the pain Visual Analogue Scale (VAS) score (35-74 moderate, 75-100 severe) and therefore eligible for the study. Those who do not meet the eligibility criterion of moderate/severe pain would not undergo the procedure as pain is the main indication for treatment, and they would not proceed with the study. During the procedure, a medical image (angiogram) will be taken to check if the patient does have knee hypervascularity, and therefore, to confirm they are eligible for the study and can be randomised to the study. If the participant does not have knee hypervascularity, then they will not be able to have the intervention and they will not be randomised to the study. For these ineligible patients, a complications check will be made at 6 weeks after the procedure. Participants with knee hypervascularity will be randomised during the procedure to receive one of two possible interventions. Randomisation will function via the REDCap database online system. Group 1 will receive genicular artery embolisation and group 2 will have a placebo procedure with no embolisation. Participants will be blinded to their allocation. The participants will need a day at the hospital for the procedure and usual pre- and post-procedure care. Most participants will be able to go home on the same day if all goes well. Study follow-up consists of participant questionnaires at 6 weeks, 3 months, 6 months and 12 months post randomisation, and a contrast MRI scan at 3 months for a subgroup of 90 participants, and a standard MRI scan at 6 months for all participants. The questionnaires include a validated visual analogue scale (VAS), Knee Injury Osteoarthritis Outcome Score (KOOS), painDETECT, and EQ-5D-5L validated questionnaires, and bespoke complications and health research use questionnaires. There will be a check for complications for all patients randomised in the study on the procedure day and up to 12 months post-procedure. During the study we will consent participants for long-term follow-up and access to their routine NHS records (HES data linkage). This will allow for assessment of any long-term effects. This further follow-up would be subject to the receipt of additional funding. There is an embedded 8-month pilot phase within the study to assess: recruitment rate; ability to maintain assessor and patient blinding; adherence to the randomised procedure and early retention rates. In addition, we have embedded secondary mechanistic outcomes, taking advantage of the controlled placebo trial methodology to investigate the link between synovitis, hypervascularity and knee pain in osteoarthritis using standard and contrast-enhanced MRI and angiograms. Radiologists within the central team will review the first three patients who have a procedure with each site radiologist to assess the number of patients found to not have knee hypervascularity, and to assess whether the procedure went smoothly, as reported by the radiologists performing the procedure.031/12/2027NoData sharing statement to be made available at a later date.
Public
ScientificDebbieJewell
Surgical Intervention Trials Unit Botnar Research Centre Nuffield Department of Orthopaedics Rheumatology and Musculoskeletal Sciences (NDORMS) University of Oxford Windmill Road Headington
OxfordUnited KingdomOX3 7LD+44 (0)1865 613763debbie.jewell@ndorms.ox.ac.ukUnited KingdomEnglandScotlandWales1. Diagnosis of painful knee OA 2. Radiographic evidence of OA (KL 2-4) 3. Previously treated with NICE non-operative intervention, as determined by the treating clinician 4. Moderate/severe pain (to be confirmed when completing the baseline questionnaire ahead of the procedure measured by the pain Visual Analogue Scale [VAS] score) (35-74 moderate, 75-100 severe) 5. Not listed for or being considered as a candidate for joint replacement surgery 6. Aged 18 years or above 7. Knee hypervascularity suitable for embolisation (to be confirmed with an angiogram during the study procedure) 8. Patient willing and able to give informed consent18 Years110 YearsAll1. Patient previously participated in the GEKO trial (only one knee can be entered into the trial) 2. Received a steroid injection in the study knee in the past 6 weeks, or is scheduled to have a steroid injection prior to the study procedure 3. Infection or malignancy around the knee 4. Inflammatory arthropathy 5. History of acute injury to the knee (within 6 months) 6. Surgery to the involved knee in the past 6 months 7. Previous knee replacement (partial or total) in either limb 8. Severe allergic reaction to radiological contrast media, including iodine-based CT-contrast or Gadolinium-based MRI contrast 9. Objection (religious or personal) to the use of medical materials made from pigs 10. Allergy to gelatine from pigs (which is within the micro-beads used in the intervention) 11. Known significant renal impairment 12. Peripheral artery disease of the affected leg 13. Pregnant or lactating 14. Hepatic impairment 15. Clotting abnormality 16. OsteonecrosisMusculoskeletal DiseasesProcedure/SurgeryPain measured using the visual analogue scale (VAS) at 6 months follow-upMeasured at baseline, 6 weeks, 3, 6 and 12 months: 1. Pain measured using VAS 2. Knee function measured using the Knee Injury and Osteoarthritis Outcome Score (KOOS) 3. Pain measured using painDETECT 4. Health-related quality of life measured using EQ-5D-5L 5. Health resource use measured using a bespoke questionnaire324901Integrated Research Application System (IRAS)68970Central Portfolio Management System (CPMS)134096National Institute for Health and Care Research (NIHR)National Institute for Health and Care ResearchApproved 09/07/2025, Health and Social Care Research Ethics Committee A (HSC REC A; Office for Research Ethics Committees Northern Ireland (ORECNI), Business Services Organisation, Northern Ireland; +44 (0)28 95 361404; reca@hscni.net) ref: 25/NI/0081