Currently, nearly 1 million total knee arthroplasties (TKA) are performed yearly in the United States. Three million are projected to be performed in 2040. Between 15-30% of all patients who undergo TKA have continued pain, reduced quality of life and functional impairments that cannot be attributed to hardware failure/loosening or infection. Treatment options for persistent post TKA pain (failed TKA) are limited. There is a need for minimally invasive, and effective pain and disability modulating interventions for patients with failed TKA. Genicular radiofrequency ablation (GRFA) has been described, refined, and validated as an effective minimally invasive intervention to control refractory knee pain secondary to knee osteoarthritis (OA) as evidenced by three favorable meta-analyses published in 2021 alone.3-5 GRFA is a minimally invasive percutaneous procedure that utilizes thermal energy to coagulate nerves from the knee. Though sometimes used in practice, there is limited research describing and evaluating GRFA for patients with failed TKA. This will be the first trial to evaluate the safety and efficacy of GRFA in patients with failed TKA using a robust study design and up-to-date, evidence-based selection criteria and technique.

The general objective of this study is to determine the safety and efficacy of GRFA in patients with failed TKA. The central hypothesis is that GRFA is safe and more efficacious in improving pain and function than sham GRFA. A triple blinded randomized sham-controlled trial design will be used. Patients with failed TKA will be recruited and randomized into either a thermal or sham GRFA group. Patient selection criteria and GRFA procedural technique will incorporate refinements based on expanded understanding from recent research. Outcomes will be assessed prior to and at 3, 6 and 12 months post-GRFA. The primary outcome measures will be between group difference in the change in pain (NRS) between pre- and 6 months post. Secondary outcomes will include safety and changes in function (WOMAC; sit-stand test; strength; range of motion & balance), participants' impression of change, perceived need for revision arthroplasty, analgesic use, and quality-of-life. Crossover will be offered to participants in the sham group at 6 months post-treatment. Data will be analyzed using descriptive, linear mixed-effects model and Kaplan-Meier Survival Curve statistics. The proposed research is innovative and important as it uses a robust study design to evaluate the safety and efficacy of an emerging RFA intervention in a prevalent suffering patient population.

Inclusion Criteria:

• >1 year post TKA
• persistent knee pain > 4/10 in intensity on average over the prior week of one of the post-TKA knees
• refractory to conventional treatment (i.e., physiotherapy, medication, etc.)
• ≥80% pain relief with a single fluoroscopically guided local anesthetic block using 0.5 mL of 2% lidocaine for the medial and lateral branches of the nerve to the vastus intermedialis, nerve to the vastus lateralis*, inferior medial genicular nerve, infrapatellar branch of the saphenous nerve, and the recurrent fibular nerve. 0.5 mL will be injected over 2 spots (1.5 cm apart) for each of the superior medial and lateral genicular nerves corresponding to the sites of the RF cannulae for the bipolar strip lesions. *Blocks will be combined fluoroscopy/ultrasound guided.

Exclusion Criteria:

• Infection (peri-prosthetic, joint, systemic, skin/soft tissue overlying the knee to be treated)
• Prosthetic loosening or failure, periprosthetic fracture
• Severe psychiatric disorder
• Non-genicular lower extremity pain source from the spine (i.e., radiculopathy or neurogenic claudication) or peripheral joints (i.e., hip osteoarthritis)
• Peripheral vascular disease causing vascular claudication
• Connective tissue/inflammatory joint disease/widespread soft tissue pain disorder
• Poor tolerance of injection procedures (as observed from the prognostic local anesthetic blocks)
• Allergy to local anesthetic, synovial expansion into the anticipated path(s) of the local anesthetic needle/RF cannula as determined by ultrasound scanning
• Uncontrolled bleeding diathesis
• Pregnancy
• Pacemaker or neurostimulator
• Inability to complete Patient-Reported Outcome Measures due to cognitive or language limitations.

Triple blinding will be accomplished as follows: operator (RF generator operated by assistant out of sight of operator); participant (local anesthetic administration and cannula placement identical; RF generator makes the same beeping sound for the same duration); assessor (only the study coordinator is aware of group assignment).