Background: Osteoarthritis of the knee (KOA) is the main cause of disability in elderly people. Patients with KOA may often not achieve adequate pain control even after receiving all treatment modalities.

Objectives: The objective of this study was to examine the efficacy of ultrasound-guided radiofrequency ablation (RFA) as a treatment for moderate and severe KOA.

Study design: A prospective randomized controlled study.

Setting: The study was performed in the National Pain Management and Research Center of China-Japan Friendship Hospital.

Methods: Eligible participants were over 50 years old and had suffered from chronic knee joint pain for more than 6 months, scoring at least 4 on a numeric rating scale (NRS) and grade III-IV according to the Kellgren-Lawrence classification system. The target nerve selection principle was as follows: the superomedial genicular nerve (SMGN) branch and inferior medial genicular nerve (IMGN) branch of the saphenous nerve for medial knee pain, the superolateral genicular nerve (SLGN) branch of the femoral nerve for lateral pain, and the SMGN, IMGN, and SLGN branches for total knee pain. The main outcomes were the NRS pain score (including the most severe pain), the average pain, and the proportion of patients who had reached pain reduction of more than 2 points. The secondary outcome was the Western Ontario McMaster University Osteoarthritis Index (WOMAC) score. RFA at 70ºC was performed for 120 seconds per patient in the RFA group, and knee nerve blocks were performed in the control group.

Results: A total of 120 patients who met the inclusion criteria were selected in this study. The treatment groups showed significant differences in their mean NRS scores and worst pain during the first, third, and sixth months after treatment. There were significant differences in the mean WOMAC pain, physical function, and total scores between the treatment groups and over time. Between the treatment groups and over time, the mean WOMAC stiffness scores were not different. At each time point after treatment, the proportion of patients who needed analgesic drugs was significantly lower in the RFA group than in the control group. Univariate analysis showed that gender, age, pain course, and body mass index were not significantly correlated with the positive rate (NRS >= 2 score reduction). After we adjusted for multiple factors, the perceived beneficial effect of therapy was less when gonarthritis was more severe (P < 0.01).

Limitation: This study's limitation is that it was performed in only one unit of the National Pain Management and Research Center.

Conclusions: Ultrasound-guided RFA applied to knee nerves can significantly reduce KOA pain, improve knee joint function, improve patient satisfaction, and provide a feasible, safe, and effective minimally invasive procedure for moderate to severe KOA in elderly patients.