Some of the best evidence in sacroiliac joint pain sits behind procedures that are not performed in this clinic. Pretending otherwise would not serve you. This page sets out what those procedures are, what their trials show, and what has to be established before a referral is worth making — because a referral made without the groundwork usually comes back.

What is not done here

Referral, not in-clinic treatment

  • An injection whose target cannot be reached safely or reliably under ultrasound, or where placement inside the joint cavity has to be certain — CT-guided injection by a radiologist.
  • Controlled or comparative diagnostic blocks, and sacral lateral branch prognostic blocks — pain medicine specialist.
  • Sacral lateral branch radiofrequency ablation, cooled or conventional — pain medicine specialist.
  • Minimally invasive sacroiliac joint fusion — spinal surgeon.
  • Assessment and management of inflammatory sacroiliitis — rheumatologist.
  • Obstetric and pelvic floor management of persistent pregnancy-related pelvic girdle pain — women's health physiotherapy, and obstetric input where relevant.

Referrals are made to the pathway, not to a named provider chosen for us. Where you have a preference, or an existing specialist, that is accommodated.

Radiofrequency ablation — the strongest evidence in the field

The 2025 consensus guidelines describe strong evidence for sacral lateral branch radiofrequency ablation providing relief for at least six months in individuals with extra-articular pathology. That is the only 'strong evidence' statement anywhere in those guidelines for a treatment of this condition, and it belongs on a page like this whether or not the procedure is offered in-house.

The procedure interrupts the lateral branches of the sacral dorsal rami that supply the posterior sacroiliac complex. It treats the extra-articular structures — it does not denervate the joint cavity itself, which has a separate and less accessible supply.

Trial Design and n Result
Cohen 2008, AnesthesiologyRandomised placebo-controlled, n=28, injection-diagnosed SIJ pain, cooled probeAt 1, 3 and 6 months: 79, 64 and 57 per cent of treated patients had at least 50 per cent pain relief and significant functional improvement, versus 14 per cent of placebo patients at 1 month and none at 3 months. At 12 months only 2 of 14 (14 per cent) still had relief.
Cohen 2024, Regional Anesthesia & Pain MedicineRandomised multicentre comparative effectiveness, n=210, block-confirmed, cooled RFA versus standard medical managementAt 3 months, 52.3 per cent of the ablation group were responders (at least 2 points or 30 per cent relief) versus 4.3 per cent of the standard management group, p<0.0001. Disability and quality of life favoured ablation on the same scale.
Cohen 2026, Regional Anesthesia & Pain MedicineTwelve-month observational phase of the same study, including crossover patients57.4 per cent of the originally randomised ablation group and 55.6 per cent of the crossover group were still responders at 12 months. Mean disability improved by 12 to 14 points. No serious procedure-related adverse events.

Reading the durability honestly

  • The 2008 placebo-controlled trial and the 2024-26 comparative study disagree markedly about how long the effect lasts — 14 per cent still benefiting at a year in the first, 57 per cent in the second.
  • The later study is much larger and used a more aggressive lesioning strategy, which the guidelines note is associated with better results. It also had no placebo arm, comparing instead against standard medical management.
  • So the reasonable expectation is meaningful relief for at least six months in well-selected patients, with genuine uncertainty about the second half of the first year, and the understanding that the nerves regenerate and the procedure can be repeated.
  • Selection is what drives this. Every one of these trials required a positive block first.

Minimally invasive SIJ fusion

Fusion places titanium implants across the joint to eliminate movement. It is the most invasive option and the one where the gap between the trial results and the guideline verdict is widest.

The randomised trials report large effects. In the two-year trial of 148 patients, mean sacroiliac pain improved by 55.4 points on a 100-point scale in the surgical group, and 83 per cent achieved clinical improvement or substantial clinical benefit in pain at 24 months against under 10 per cent with non-surgical management. In the separate 103-patient trial, pain improved by 41.6 points against 14.0 with conservative management at 12 months.

The 2025 consensus guidelines nonetheless conclude that for carefully selected patients with intra-articular sacroiliac complex pain confirmed by controlled blocks who have failed conservative therapy, there is weak or very weak evidence that minimally invasive fusion provides benefit for at least one year.

Why the guideline verdict is so much more cautious than the trial results

  • Neither trial was blinded, and both used self-reported pain and disability as the primary outcomes. Unblinded surgical trials with subjective endpoints systematically overstate effect — the trial authors state this limitation themselves.
  • Employees of the implant manufacturer are listed as authors on both trials.
  • The comparator was non-surgical management in patients who had already failed non-surgical management, which is a low bar.
  • It is irreversible. Ablation wears off; a fusion does not.
  • Adverse events reported included nerve impingement, device loosening requiring revision, and postoperative haematoma.

None of that makes fusion wrong for the right person. It makes it a decision that should follow a controlled block, an adequately delivered rehabilitation programme, and usually a trial of ablation — not one taken early.

Corticosteroid injection — where it still fits

The 2025 guidelines found that both intra-articular and extra-articular steroid injections provide at least four weeks of relief in well-selected patients, with the evidence slightly stronger for the extra-articular route.

Four weeks is a real but modest claim, and it is the honest one. Corticosteroid has a place in settling pain enough to make rehabilitation possible, and in the Kim trial it performed identically to prolotherapy at two weeks before falling away by 15 months. It is not a long-term strategy, and repeated corticosteroid injection carries its own tissue concerns. See the corticosteroid injection page.

Rheumatology, and the pattern that should trigger it

Axial spondyloarthritis presents at the sacroiliac joints and is routinely mistaken for mechanical back pain for years. Referral should be considered where the history shows onset before age 45, morning stiffness lasting more than 30 minutes, pain improving with activity and worsening with rest, night pain in the second half of the night, a good response to anti-inflammatories, or associated features such as psoriasis, inflammatory bowel disease, uveitis or enthesitis elsewhere.

This pathway is separate from everything else on this page. Injections and shockwave are not the treatment for inflammatory sacroiliitis, and delay in starting appropriate disease-modifying treatment has consequences. See ankylosing spondylitis.

What a good referral contains

  1. 1 A stated working diagnosis with its uncertainty Including whether intra-articular or extra-articular pain is suspected, because that determines whether ablation or fusion is even the right conversation.
  2. 2 The provocation cluster result, with the number of positive tests And the result of long dorsal sacroiliac ligament palpation, which the cluster does not cover.
  3. 3 Documented rehabilitation, with duration and supervision stated Three months of supervised progressive loading is the reference point. 'Tried physio' is not enough information for a specialist to act on.
  4. 4 Baseline and current outcome measures Oswestry Disability Index and a numeric pain rating, recorded separately for morning, evening and provoked pain, plus the active straight leg raise.
  5. 5 What has already been injected, when, and what happened Including the duration and magnitude of any response, since prior injection response is part of the selection criteria for what comes next.
  6. 6 The alternatives considered and excluded Hip, lumbar, gluteal tendon, deep gluteal and inflammatory causes, with the reasoning.

Related on this site

Common questions

Does radiofrequency ablation work for sacroiliac joint pain?
It has the strongest evidence of any treatment for this condition. The 2025 multispecialty consensus guidelines describe strong evidence for sacral lateral branch ablation providing relief for at least six months in people with extra-articular pain. A randomised comparative study of 210 patients found 52 per cent were responders at three months against 4 per cent with standard medical management, and 57 per cent were still responders at twelve months. An earlier placebo-controlled trial found relief in 57 per cent at six months but only 14 per cent at one year, so durability beyond six months is genuinely uncertain. It is not performed in this clinic and requires referral to a pain medicine specialist.
Should I have my sacroiliac joint fused?
That is a decision for a spinal surgeon, and it should come late rather than early. The randomised trials report large improvements, but neither was blinded, both used self-reported outcomes, and employees of the implant manufacturer are authors on both. The 2025 consensus guidelines conclude there is weak or very weak evidence that fusion provides benefit for at least a year, in carefully selected patients whose pain has been confirmed by controlled blocks and who have failed conservative treatment. It is also irreversible, which ablation is not.
Do I need a diagnostic block before ablation or surgery?
Yes. Every trial of ablation and fusion required block confirmation before enrolment, and the guidelines make selection the central issue. The most commonly used threshold is 50 per cent relief; three of the participating societies dissented and would require 75 per cent relief from two separate blocks before ablation. Blocks are not performed in this clinic and are arranged by referral.
Can you refer me for these procedures?
Yes. What makes a referral useful is the groundwork — a characterised diagnosis, documented supervised rehabilitation with its duration stated, baseline and current outcome measures, and a record of what has already been injected and what happened. That is what is prepared before the referral goes out.
How long does a cortisone injection last in the sacroiliac joint?
The 2025 consensus guidelines conclude that both intra-articular and extra-articular steroid injections provide at least four weeks of relief in well-selected patients, with the evidence slightly stronger for the extra-articular route. In the one randomised comparison against prolotherapy, corticosteroid performed identically at two weeks but only 10 per cent of patients still had at least half their pain relief at 15 months, against 59 per cent with prolotherapy.

Evidence reviewed

Every clinical statement on this page traces to one of the sources below, including the ones that point the other way. Where a study is small, unblinded or authored by people with a commercial interest in the result, that is noted alongside it rather than left out.

Last reviewed 2026-09-04. This is general information about a condition and its treatments, not personal medical advice, and it is no substitute for assessment by a clinician who has examined you.

Ready for a clearer plan for your back or musculoskeletal pain?

Book an assessment with Dr Joshua Hatch.

Your assessment focuses on understanding the likely source of your pain and the most appropriate non-surgical options for your diagnosis, with the aim of reducing pain and improving function.

Book an appointment
Book an appointment with Back Pain Doctor
Book now Call