A quarter of a million NHS patients each year will be offered a chance to take part in a trial that tests whether burning a tiny nerve in the lower back actually relieves chronic pain. Chronic low back pain affects millions of people in the UK, and many do not improve with standard treatments like exercise or painkillers. One common next step is radiofrequency denervation (RFD), where a heated needle is used to disable the nerve branches thought to be causing the pain. But the evidence for RFD is weak, and the procedure is costly and carries risks. The RADICAL trial will randomly assign 250 patients to receive either real RFD or a sham version where the needle is inserted but not heated. Neither the patient nor the doctor will know which they received. If the real procedure does not produce a meaningful reduction in pain after three months, patients can opt for a second, blinded procedure with the opposite treatment. If RFD proves no better than placebo, the NHS could stop funding a procedure that currently costs millions each year and exposes patients to unnecessary risk. If it works, the trial will provide the first robust evidence to support its use, potentially changing clinical guidelines and saving the health service money by reducing reliance on more invasive surgeries or long-term pain management.
View original technical description
Design: Double-blind, parallel group, superiority RCT with internal pilot, qualitative research and health economic analysis. Setting: 20 pain and spinal clinics providing RFD to NHS patients. Inclusion criteria: Adults with chronic moderate to severe LBP referred to secondary care after insufficient improvement with NICE recommended conservative management; clinical features suggestive of facet joint involvement; positive response to single diagnostic medial branch block. Exclusion criteria: Pregnancy; infection; unwilling or unable to tolerate procedure; psychological barriers to recovery (severe depression or widespread pain); previous RFD; pacemaker or implantable cardioverter defibrillator; malignancy as the reason for LBP Health technologies being assessed: RFD of the lumbar medial branches of the dorsal rami. Using the C-arm of a fluoroscope, a needle will be inserted adjacent to the nerve and an electrode heated to make a RF lesion. Placebo RFD will follow the same protocol, but the electrode tip temperature will not be raised. Prior to commencing the trial, a survey and PI meeting will be undertaken to develop a protocol that specifies the variations in the RFD procedure that will be permitted and prohibited within the trial. Blinded re-intervention with alternative treatment: Participants who do not experience a clinically meaningful improvement in pain 3 months after randomization will be offered “repeat RFD” but with the alternative intervention to the one provided at the outset without disclosing the original allocation. Measurement of outcomes: The primary clinical outcome will be pain severity, measured using a 0-10 pain NRS, at 3 months after randomization. Secondary outcomes will be functional disability, health-related quality of life, psychological distress, time to pain recovery, satisfaction, frequency of uptake of offer of repeat RFD, adverse events, work outcomes and healthcare utilisation. Pain severity will be assessed at 2, 4, 6, 8 and 10 weeks after randomization, with the EQ-5D-5L at 6 weeks. Questionnaires, assessing all outcomes, will be administered at baseline and at 3, 6, 12, 18 and 24 months after randomization. Cost-effectiveness analysis: The primary economic evaluation will take an NHS and social services perspective and estimate the discounted cost per quality adjusted life year and incremental net benefit of RFD over the 2 year follow up period. Cost-effectiveness acceptability curves will estimate the probability of cost-effectiveness at different values of willingness-to-pay. In secondary analyses we will estimate cost-effectiveness at 3 month follow up and include patient costs and productivity gains/losses in a societal perspective 2 year analysis. Sample size: Assuming that scores at 3 months will be adjusted for baseline scores, with a correlation of 0.3 between scores, a sample size of 226 participants is required to detect the target difference of 0.84 in the NRS with 90% power and 5% 2-tailed significance (standard deviation 2; standardised difference=0.42). Allowing for up to 10% attrition at 3 months gives a total of 250 participants (125 per group). Project timetable: 60 months comprising 9 months set up, 18 months recruitment (including 12 month pilot), 25 months follow-up, and 8 months data analyses and reporting.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know