# Facet joint pain: the threshold that defines it was set in a workshop

slug: facet-joint-syndrome · https://miscsubjects.com/a/facet-joint-syndrome · updated 2026-08-05T09:19:48.793Z

At the back of every level of your spine sit two small joints, one on each side, linking each vertebra to the one below. They are ordinary synovial joints — cartilage, capsule, lubricating fluid — and like every other joint of that kind they wear, they get arthritic, and they can hurt.

The pain they produce has a shape. It is worse leaning backwards and worse twisting, because both movements load and compress them. It eases leaning forward, which opens them. It sits low in the back, often to one side, and can refer into the buttock and the back of the thigh — but it stops above the knee, which is the line that separates it from a nerve root problem.

That is the clinical picture. The problem with this condition is not describing it. The problem is proving it.

## Why nobody can diagnose this from a scan

Facet joints look arthritic on imaging in enormous numbers of people, including people with no back pain at all — the same trap that runs through spinal stenosis and rotator cuff findings. Arthritic facets on a report tell you the person has a spine of a certain age.

Because the picture cannot settle it, the field settled on a functional test instead: anaesthetise the small nerve branches that supply a specific joint and see whether the pain goes. If it does, that joint was the source. If it does not, it was not.

This is where the honest difficulty of the field lives, and a 2021 guideline development workshop shows it. Of the criteria the group considered mandatory for selecting patients, one needed defining outright: what counts as a positive response to a diagnostic block.

[[embed:source:s1]]

They settled on at least 60% pain relief.

Sit with that. The threshold defining whether a person has this condition was, as recently as 2021, a thing that had to be agreed in a workshop rather than something the evidence supplied. That is not a scandal — it is a fair description of how diagnosis works when the picture and the symptoms do not line up — but it should shape how much confidence anyone puts in the label.

## What predicts whether the block will work

If the diagnostic block is the gateway, it is worth knowing who responds to it. A 2026 study compared responders with non-responders and found the difference was not primarily about the joints themselves.

[[embed:source:s2]]

Responders were younger and had lower body mass index, and specific measures of pelvic and spinal alignment were associated with a negative response. The authors suggest adding biomechanical assessment to patient selection.

The useful reading: if your spine and pelvis sit in a particular alignment, your back pain is less likely to be coming purely from a facet joint, and a block is less likely to give you the answer you were hoping for. It is a reason to have the alignment looked at before agreeing to a diagnostic procedure.

## Burning the nerve: what it does, and how confident anyone should be

If blocks confirm the joint is the source, the usual next step is radiofrequency denervation — heating the small nerve branches so they stop carrying pain signals from that joint. The nerves regrow, so the effect is measured in months to a couple of years.

A 2026 systematic review of the whole field gives the fairest summary available.

[[embed:source:s3]]

Small improvements in pain and function against sham, on low to moderate certainty evidence. Not nothing, and not a solution.

That word "sham" matters here more than in most procedures. A sham-controlled comparison is the only way to separate the effect of the treatment from the effect of having a procedure, and procedures produce large expectation effects. Small improvements over sham means small improvements over the considerable benefit of being treated at all.

Practice reports look better than that, and it is worth understanding why. A prospective series found significant reductions in back pain, radiating pain and disability against the patients' own baseline, plus reduced painkiller use.

[[embed:source:s4]]

Against baseline is not against sham. Both findings are true. The one that tells you what the heating itself contributed is the sham-controlled one.

## The alternative to heat, and where it currently stands

Cryoneurolysis — freezing the nerve rather than heating it — has been compared head to head with radiofrequency for chronic back pain, using a global impression of change as the primary measure four weeks after the procedure.

[[embed:source:s5]]

Note the design detail sitting in the middle of that trial: physical therapy was added for all groups after four weeks. That is good clinical practice and it means anything measured later than four weeks includes the therapy. It is a reminder that the procedure is a window, not a treatment on its own.

## Where steroid injections into the joint sit

Injecting steroid directly into a facet joint is offered widely. As of 2024 a Cochrane review was underway specifically to establish whether it helps and harms, and specifically to test whether the answer depends on how the diagnosis was made.

[[embed:source:s6]]

That second aim is the interesting one, and it is the theme of this whole page. If a treatment is given to people diagnosed by a scan and people diagnosed by a block, and the two groups respond differently, then the treatment's reputation depends on who is being selected for it — not only on what it does.

## Ten people with facet joint pain

Ten people have back pain that worsens leaning back and eases leaning forward. Most have arthritic-looking facet joints on imaging, and so do large numbers of people with no pain at all. To find out which ten actually have facet-driven pain, each needs a diagnostic block, with 60% relief as the agreed threshold — a threshold set by consensus, not derived from evidence. The younger and lighter of them, with certain spinal alignments, are the likelier responders. Those who go on to have the nerves burned get small improvements over a sham procedure on low-to-moderate certainty evidence, and considerably larger improvements over where they started, and both statements are true at once.

## An arthritic joint is not a tendon

No repair peptide has randomised human trial evidence in facet joint pain. The mechanical objection is worth stating: the pain is generated by an arthritic synovial joint under load, so a compound would have to act on joint cartilage and capsule, not on tendon or nerve. Where such a case exists, it belongs on a page naming both the compound and this condition, with its evidence tier stated. The operator of this site has a commercial interest in compounds of that kind, which is why this page states the absence and the objection rather than talking around them.

## Questions worth asking before the needle

1. **Check the movement pattern.** Worse leaning back and twisting, better leaning forward, pain stopping above the knee. Pain travelling below the knee points at a nerve root instead.
2. **Do not accept a scan as the diagnosis.** Arthritic facets are close to universal with age and appear in people with no pain.
3. **If a diagnostic block is proposed, ask what threshold counts as positive.** The agreed answer is at least 60% relief, and it is a consensus figure rather than an evidence-derived one.
4. **Ask about your spinal and pelvic alignment first.** It was associated with whether the block works, and it is measurable before anyone puts a needle in.
5. **Ask for the sham-controlled numbers, not the before-and-after ones.** Against baseline the results look strong; against a sham procedure they are small.
6. **Treat the procedure as a window.** In the trials, physical therapy was added afterwards for everyone. The procedure creates the conditions for the work; it is not the work.


## Sources

1. Radiofrequency denervation of the lumbar facet joints: guidelines for the RADICAL randomised controlled trial — https://pubmed.ncbi.nlm.nih.gov/34377456/
2. Spinopelvic Alignment as an Associated Factor of Short-Term Diagnostic Response to Lumbar Medial Branch Block: A Prospective Study — https://pubmed.ncbi.nlm.nih.gov/42279214/
3. Current evidence on radiofrequency denervation for chronic low back pain: a systematic review and meta-analysis — https://pubmed.ncbi.nlm.nih.gov/42463214/
4. Prospective analysis of radiofrequency denervation in patients with chronic low back pain — https://pubmed.ncbi.nlm.nih.gov/33447672/
5. Cryoneurolysis versus radiofrequency ablation outcome on pain experience in chronic low back pain (COPE): a single-blinded randomised controlled trial — https://pubmed.ncbi.nlm.nih.gov/38724261/
6. Glucocorticoid facet joint injection for chronic back or neck pain — https://pubmed.ncbi.nlm.nih.gov/39569679/